Merge branch 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus

Pull MIPS updates from Ralf Baechle:
 "This is the MIPS pull request for the next kernel:

   - Zubair's patch series adds CMA support for MIPS.  Doing so it also
     touches ARM64 and x86.
   - remove the last instance of IRQF_DISABLED from arch/mips
   - updates to two of the MIPS defconfig files.
   - cleanup of how cache coherency bits are handled on MIPS and
     implement support for write-combining.
   - platform upgrades for Alchemy
   - move MIPS DTS files to arch/mips/boot/dts/"

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus: (24 commits)
  MIPS: ralink: remove deprecated IRQF_DISABLED
  MIPS: pgtable.h: Implement the pgprot_writecombine function for MIPS
  MIPS: cpu-probe: Set the write-combine CCA value on per core basis
  MIPS: pgtable-bits: Define the CCA bit for WC writes on Ingenic cores
  MIPS: pgtable-bits: Move the CCA bits out of the core's ifdef blocks
  MIPS: DMA: Add cma support
  x86: use generic dma-contiguous.h
  arm64: use generic dma-contiguous.h
  asm-generic: Add dma-contiguous.h
  MIPS: BPF: Add new emit_long_instr macro
  MIPS: ralink: Move device-trees to arch/mips/boot/dts/
  MIPS: Netlogic: Move device-trees to arch/mips/boot/dts/
  MIPS: sead3: Move device-trees to arch/mips/boot/dts/
  MIPS: Lantiq: Move device-trees to arch/mips/boot/dts/
  MIPS: Octeon: Move device-trees to arch/mips/boot/dts/
  MIPS: Add support for building device-tree binaries
  MIPS: Create common infrastructure for building built-in device-trees
  MIPS: SEAD3: Enable DEVTMPFS
  MIPS: SEAD3: Regenerate defconfigs
  MIPS: Alchemy: DB1300: Add touch penirq support
  ...

arch/arm64/include/asm/Kbuild

@@ -9,6 +9,7 @@ generic-y += current.h
 generic-y += delay.h
 generic-y += div64.h
 generic-y += dma.h
+generic-y += dma-contiguous.h
 generic-y += early_ioremap.h
 generic-y += emergency-restart.h
 generic-y += errno.h

arch/arm64/include/asm/dma-contiguous.h

@@ -1,28 +0,0 @@
-/*
- * Copyright (c) 2013, The Linux Foundation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 and
- * only version 2 as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- */
-
-#ifndef _ASM_DMA_CONTIGUOUS_H
-#define _ASM_DMA_CONTIGUOUS_H
-
-#ifdef __KERNEL__
-#ifdef CONFIG_DMA_CMA
-
-#include <linux/types.h>
-
-static inline void
-dma_contiguous_early_fixup(phys_addr_t base, unsigned long size) { }
-
-#endif
-#endif
-
-#endif

arch/mips/Kconfig

@@ -29,6 +29,7 @@ config MIPS
 	select GENERIC_ATOMIC64 if !64BIT
 	select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
 	select HAVE_DMA_ATTRS
+	select HAVE_DMA_CONTIGUOUS
 	select HAVE_DMA_API_DEBUG
 	select GENERIC_IRQ_PROBE
 	select GENERIC_IRQ_SHOW
@@ -353,6 +354,7 @@ config MIPS_SEAD3
 	bool "MIPS SEAD3 board"
 	select BOOT_ELF32
 	select BOOT_RAW
+	select BUILTIN_DTB
 	select CEVT_R4K
 	select CSRC_R4K
 	select CSRC_GIC
@@ -742,6 +744,7 @@ config CAVIUM_OCTEON_SOC
 	select ARCH_SPARSEMEM_ENABLE
 	select SYS_SUPPORTS_SMP
 	select NR_CPUS_DEFAULT_16
+	select BUILTIN_DTB
 	help
 	  This option supports all of the Octeon reference boards from Cavium
 	  Networks. It builds a kernel that dynamically determines the Octeon
@@ -2482,6 +2485,9 @@ config USE_OF
 	select OF_EARLY_FLATTREE
 	select IRQ_DOMAIN
 
+config BUILTIN_DTB
+	bool
+
 endmenu
 
 config LOCKDEP_SUPPORT

arch/mips/Makefile

@@ -333,6 +333,16 @@ endif
 
 CLEAN_FILES += vmlinux.32 vmlinux.64
 
+# device-trees
+core-$(CONFIG_BUILTIN_DTB) += arch/mips/boot/dts/
+
+%.dtb %.dtb.S %.dtb.o: | scripts
+	$(Q)$(MAKE) $(build)=arch/mips/boot/dts arch/mips/boot/dts/$@
+
+PHONY += dtbs
+dtbs: scripts
+	$(Q)$(MAKE) $(build)=arch/mips/boot/dts dtbs
+
 archprepare:
 ifdef CONFIG_MIPS32_N32
 	@echo '  Checking missing-syscalls for N32'
@@ -367,6 +377,7 @@ define archhelp
 	echo '  vmlinuz.srec         - SREC zboot image'
 	echo '  uImage               - U-Boot image'
 	echo '  uImage.gz            - U-Boot image (gzip)'
+	echo '  dtbs                 - Device-tree blobs for enabled boards'
 	echo
 	echo '  These will be default as appropriate for a configured platform.'
 endef

arch/mips/alchemy/devboards/db1300.c

@@ -21,6 +21,7 @@
 #include <linux/mtd/partitions.h>
 #include <linux/platform_device.h>
 #include <linux/smsc911x.h>
+#include <linux/wm97xx.h>
 
 #include <asm/mach-au1x00/au1000.h>
 #include <asm/mach-au1x00/au1100_mmc.h>
@@ -711,6 +712,46 @@ static struct platform_device db1300_lcd_dev = {
 
 /**********************************************************************/
 
+static void db1300_wm97xx_irqen(struct wm97xx *wm, int enable)
+{
+	if (enable)
+		enable_irq(DB1300_AC97_PEN_INT);
+	else
+		disable_irq_nosync(DB1300_AC97_PEN_INT);
+}
+
+static struct wm97xx_mach_ops db1300_wm97xx_ops = {
+	.irq_enable	= db1300_wm97xx_irqen,
+	.irq_gpio	= WM97XX_GPIO_3,
+};
+
+static int db1300_wm97xx_probe(struct platform_device *pdev)
+{
+	struct wm97xx *wm = platform_get_drvdata(pdev);
+
+	/* external pendown indicator */
+	wm97xx_config_gpio(wm, WM97XX_GPIO_13, WM97XX_GPIO_IN,
+			   WM97XX_GPIO_POL_LOW, WM97XX_GPIO_STICKY,
+			   WM97XX_GPIO_WAKE);
+
+	/* internal "virtual" pendown gpio */
+	wm97xx_config_gpio(wm, WM97XX_GPIO_3, WM97XX_GPIO_OUT,
+			   WM97XX_GPIO_POL_LOW, WM97XX_GPIO_NOTSTICKY,
+			   WM97XX_GPIO_NOWAKE);
+
+	wm->pen_irq = DB1300_AC97_PEN_INT;
+
+	return wm97xx_register_mach_ops(wm, &db1300_wm97xx_ops);
+}
+
+static struct platform_driver db1300_wm97xx_driver = {
+	.driver.name	= "wm97xx-touch",
+	.driver.owner	= THIS_MODULE,
+	.probe		= db1300_wm97xx_probe,
+};
+
+/**********************************************************************/
+
 static struct platform_device *db1300_dev[] __initdata = {
 	&db1300_eth_dev,
 	&db1300_i2c_dev,
@@ -755,6 +796,9 @@ int __init db1300_dev_setup(void)
 	i2c_register_board_info(0, db1300_i2c_devs,
 				ARRAY_SIZE(db1300_i2c_devs));
 
+	if (platform_driver_register(&db1300_wm97xx_driver))
+		pr_warn("DB1300: failed to init touch pen irq support!\n");
+
 	/* Audio PSC clock is supplied by codecs (PSC1, 2) */
 	__raw_writel(PSC_SEL_CLK_SERCLK,
 	    (void __iomem *)KSEG1ADDR(AU1300_PSC1_PHYS_ADDR) + PSC_SEL_OFFSET);
@@ -762,9 +806,10 @@ int __init db1300_dev_setup(void)
 	__raw_writel(PSC_SEL_CLK_SERCLK,
 	    (void __iomem *)KSEG1ADDR(AU1300_PSC2_PHYS_ADDR) + PSC_SEL_OFFSET);
 	wmb();
-	/* I2C uses internal 48MHz EXTCLK1 */
+	/* I2C driver wants 50MHz, get as close as possible */
 	c = clk_get(NULL, "psc3_intclk");
 	if (!IS_ERR(c)) {
+		clk_set_rate(c, 50000000);
 		clk_prepare_enable(c);
 		clk_put(c);
 	}

arch/mips/alchemy/devboards/db1550.c

@@ -34,12 +34,9 @@ static void __init db1550_hw_setup(void)
 	void __iomem *base;
 	unsigned long v;
 
-	/* complete SPI setup: link psc0_intclk to a 48MHz source,
-	 * and assign GPIO16 to PSC0_SYNC1 (SPI cs# line) as well as PSC1_SYNC
-	 * for AC97 on PB1550.
+	/* complete pin setup: assign GPIO16 to PSC0_SYNC1 (SPI cs# line)
+	 * as well as PSC1_SYNC for AC97 on PB1550.
 	 */
-	v = alchemy_rdsys(AU1000_SYS_CLKSRC);
-	alchemy_wrsys(v | 0x000001e0, AU1000_SYS_CLKSRC);
 	v = alchemy_rdsys(AU1000_SYS_PINFUNC);
 	alchemy_wrsys(v | 1 | SYS_PF_PSC1_S1, AU1000_SYS_PINFUNC);
 
@@ -586,11 +583,13 @@ int __init db1550_dev_setup(void)
 
 	c = clk_get(NULL, "psc0_intclk");
 	if (!IS_ERR(c)) {
+		clk_set_rate(c, 50000000);
 		clk_prepare_enable(c);
 		clk_put(c);
 	}
 	c = clk_get(NULL, "psc2_intclk");
 	if (!IS_ERR(c)) {
+		clk_set_rate(c, db1550_spi_platdata.mainclk_hz);
 		clk_prepare_enable(c);
 		clk_put(c);
 	}

arch/mips/alchemy/devboards/platform.c

@@ -11,6 +11,7 @@
 #include <linux/pm.h>
 
 #include <asm/bootinfo.h>
+#include <asm/idle.h>
 #include <asm/reboot.h>
 #include <asm/mach-au1x00/au1000.h>
 #include <asm/mach-db1x00/bcsr.h>
@@ -53,6 +54,8 @@ static void db1x_power_off(void)
 {
 	bcsr_write(BCSR_RESETS, 0);
 	bcsr_write(BCSR_SYSTEM, BCSR_SYSTEM_PWROFF | BCSR_SYSTEM_RESET);
+	while (1)		/* sit and spin */
+		cpu_wait();
 }
 
 static void db1x_reset(char *c)

arch/mips/boot/.gitignore

@@ -5,3 +5,4 @@ zImage
 zImage.tmp
 calc_vmlinuz_load_addr
 uImage
+*.dtb

arch/mips/boot/dts/Makefile

@@ -0,0 +1,20 @@
+dtb-$(CONFIG_CAVIUM_OCTEON_SOC)		+= octeon_3xxx.dtb octeon_68xx.dtb
+dtb-$(CONFIG_DT_EASY50712)		+= easy50712.dtb
+dtb-$(CONFIG_DT_XLP_EVP)		+= xlp_evp.dtb
+dtb-$(CONFIG_DT_XLP_SVP)		+= xlp_svp.dtb
+dtb-$(CONFIG_DT_XLP_FVP)		+= xlp_fvp.dtb
+dtb-$(CONFIG_DT_XLP_GVP)		+= xlp_gvp.dtb
+dtb-$(CONFIG_DTB_RT2880_EVAL)		+= rt2880_eval.dtb
+dtb-$(CONFIG_DTB_RT305X_EVAL)		+= rt3052_eval.dtb
+dtb-$(CONFIG_DTB_RT3883_EVAL)		+= rt3883_eval.dtb
+dtb-$(CONFIG_DTB_MT7620A_EVAL)		+= mt7620a_eval.dtb
+dtb-$(CONFIG_MIPS_SEAD3)		+= sead3.dtb
+
+obj-y		+= $(patsubst %.dtb, %.dtb.o, $(dtb-y))
+
+targets		+= dtbs
+targets		+= $(dtb-y)
+
+dtbs: $(addprefix $(obj)/, $(dtb-y))
+
+clean-files	+= *.dtb *.dtb.S

arch/mips/cavium-octeon/.gitignore

@@ -1,2 +0,0 @@
-*.dtb.S
-*.dtb

arch/mips/cavium-octeon/Makefile

@@ -20,13 +20,3 @@ obj-y += executive/
 obj-$(CONFIG_MTD)		      += flash_setup.o
 obj-$(CONFIG_SMP)		      += smp.o
 obj-$(CONFIG_OCTEON_ILM)	      += oct_ilm.o
-
-DTS_FILES = octeon_3xxx.dts octeon_68xx.dts
-DTB_FILES = $(patsubst %.dts, %.dtb, $(DTS_FILES))
-
-obj-y += $(patsubst %.dts, %.dtb.o, $(DTS_FILES))
-
-# Let's keep the .dtb files around in case we want to look at them.
-.SECONDARY:  $(addprefix $(obj)/, $(DTB_FILES))
-
-clean-files += $(DTB_FILES) $(patsubst %.dtb, %.dtb.S, $(DTB_FILES))

arch/mips/configs/sead3_defconfig

@@ -31,8 +31,8 @@ CONFIG_IP_PNP_BOOTP=y
 # CONFIG_IPV6 is not set
 # CONFIG_WIRELESS is not set
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_DEVTMPFS=y
 CONFIG_MTD=y
-CONFIG_MTD_CHAR=y
 CONFIG_MTD_BLOCK=y
 CONFIG_MTD_CFI=y
 CONFIG_MTD_CFI_INTELEXT=y

arch/mips/configs/sead3micro_defconfig

@@ -32,8 +32,8 @@ CONFIG_IP_PNP_BOOTP=y
 # CONFIG_IPV6 is not set
 # CONFIG_WIRELESS is not set
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_DEVTMPFS=y
 CONFIG_MTD=y
-CONFIG_MTD_CHAR=y
 CONFIG_MTD_BLOCK=y
 CONFIG_MTD_CFI=y
 CONFIG_MTD_CFI_INTELEXT=y

arch/mips/include/asm/Kbuild

@@ -1,6 +1,7 @@
 # MIPS headers
 generic-y += cputime.h
 generic-y += current.h
+generic-y += dma-contiguous.h
 generic-y += emergency-restart.h
 generic-y += hash.h
 generic-y += irq_work.h

arch/mips/include/asm/cpu-features.h

@@ -231,6 +231,16 @@
 #define cpu_has_clo_clz	cpu_has_mips_r
 #endif
 
+/*
+ * MIPS32 R2, MIPS64 R2, Loongson 3A and Octeon have WSBH.
+ * MIPS64 R2, Loongson 3A and Octeon have WSBH, DSBH and DSHD.
+ * This indicates the availability of WSBH and in case of 64 bit CPUs also
+ * DSBH and DSHD.
+ */
+#ifndef cpu_has_wsbh
+#define cpu_has_wsbh		cpu_has_mips_r2
+#endif
+
 #ifndef cpu_has_dsp
 #define cpu_has_dsp		(cpu_data[0].ases & MIPS_ASE_DSP)
 #endif

arch/mips/include/asm/cpu-info.h

@@ -79,6 +79,11 @@ struct cpuinfo_mips {
 #define NUM_WATCH_REGS 4
 	u16			watch_reg_masks[NUM_WATCH_REGS];
 	unsigned int		kscratch_mask; /* Usable KScratch mask. */
+	/*
+	 * Cache Coherency attribute for write-combine memory writes.
+	 * (shifted by _CACHE_SHIFT)
+	 */
+	unsigned int		writecombine;
 } __attribute__((aligned(SMP_CACHE_BYTES)));
 
 extern struct cpuinfo_mips cpu_data[];

arch/mips/include/asm/mach-au1x00/cpu-feature-overrides.h

@@ -8,6 +8,12 @@
 #define __ASM_MACH_AU1X00_CPU_FEATURE_OVERRIDES_H
 
 #define cpu_has_tlb			1
+#define cpu_has_tlbinv			0
+#define cpu_has_segments		0
+#define cpu_has_eva			0
+#define cpu_has_htw			0
+#define cpu_has_rixiex			0
+#define cpu_has_maar			0
 #define cpu_has_4kex			1
 #define cpu_has_3k_cache		0
 #define cpu_has_4k_cache		1
@@ -28,6 +34,8 @@
 #define cpu_has_mdmx			0
 #define cpu_has_mips3d			0
 #define cpu_has_smartmips		0
+#define cpu_has_rixi			0
+#define cpu_has_mmips			0
 #define cpu_has_vtag_icache		0
 #define cpu_has_dc_aliases		0
 #define cpu_has_ic_fills_f_dc		1
@@ -50,4 +58,8 @@
 #define cpu_dcache_line_size()		32
 #define cpu_icache_line_size()		32
 
+#define cpu_has_perf_cntr_intr_bit	0
+#define cpu_has_vz			0
+#define cpu_has_msa			0
+
 #endif /* __ASM_MACH_AU1X00_CPU_FEATURE_OVERRIDES_H */

arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h

@@ -57,6 +57,7 @@
 #define cpu_has_vint		0
 #define cpu_has_veic		0
 #define cpu_hwrena_impl_bits	0xc0000000
+#define cpu_has_wsbh            1
 
 #define cpu_has_rixi		(cpu_data[0].cputype != CPU_CAVIUM_OCTEON)
 

arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h

@@ -59,4 +59,6 @@
 #define cpu_has_watch		1
 #define cpu_has_local_ebase	0
 
+#define cpu_has_wsbh		IS_ENABLED(CONFIG_CPU_LOONGSON3)
+
 #endif /* __ASM_MACH_LOONGSON_CPU_FEATURE_OVERRIDES_H */

arch/mips/include/asm/pgtable-bits.h

@@ -224,40 +224,54 @@ static inline uint64_t pte_to_entrylo(unsigned long pte_val)
 #if defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)
 
 #define _CACHE_CACHABLE_NONCOHERENT 0
+#define _CACHE_UNCACHED_ACCELERATED _CACHE_UNCACHED
 
 #elif defined(CONFIG_CPU_SB1)
 
 /* No penalty for being coherent on the SB1, so just
    use it for "noncoherent" spaces, too.  Shouldn't hurt. */
 
-#define _CACHE_UNCACHED		    (2<<_CACHE_SHIFT)
-#define _CACHE_CACHABLE_COW	    (5<<_CACHE_SHIFT)
 #define _CACHE_CACHABLE_NONCOHERENT (5<<_CACHE_SHIFT)
-#define _CACHE_UNCACHED_ACCELERATED (7<<_CACHE_SHIFT)
 
 #elif defined(CONFIG_CPU_LOONGSON3)
 
 /* Using COHERENT flag for NONCOHERENT doesn't hurt. */
 
-#define _CACHE_UNCACHED             (2<<_CACHE_SHIFT)  /* LOONGSON       */
 #define _CACHE_CACHABLE_NONCOHERENT (3<<_CACHE_SHIFT)  /* LOONGSON       */
 #define _CACHE_CACHABLE_COHERENT    (3<<_CACHE_SHIFT)  /* LOONGSON-3     */
-#define _CACHE_UNCACHED_ACCELERATED (7<<_CACHE_SHIFT)  /* LOONGSON       */
 
-#else
+#elif defined(CONFIG_MACH_JZ4740)
 
-#define _CACHE_CACHABLE_NO_WA	    (0<<_CACHE_SHIFT)  /* R4600 only	  */
-#define _CACHE_CACHABLE_WA	    (1<<_CACHE_SHIFT)  /* R4600 only	  */
-#define _CACHE_UNCACHED		    (2<<_CACHE_SHIFT)  /* R4[0246]00	  */
-#define _CACHE_CACHABLE_NONCOHERENT (3<<_CACHE_SHIFT)  /* R4[0246]00	  */
-#define _CACHE_CACHABLE_CE	    (4<<_CACHE_SHIFT)  /* R4[04]00MC only */
-#define _CACHE_CACHABLE_COW	    (5<<_CACHE_SHIFT)  /* R4[04]00MC only */
-#define _CACHE_CACHABLE_COHERENT    (5<<_CACHE_SHIFT)  /* MIPS32R2 CMP	  */
-#define _CACHE_CACHABLE_CUW	    (6<<_CACHE_SHIFT)  /* R4[04]00MC only */
-#define _CACHE_UNCACHED_ACCELERATED (7<<_CACHE_SHIFT)  /* R10000 only	  */
+/* Ingenic uses the WA bit to achieve write-combine memory writes */
+#define _CACHE_UNCACHED_ACCELERATED (1<<_CACHE_SHIFT)
 
 #endif
 
+#ifndef _CACHE_CACHABLE_NO_WA
+#define _CACHE_CACHABLE_NO_WA		(0<<_CACHE_SHIFT)
+#endif
+#ifndef _CACHE_CACHABLE_WA
+#define _CACHE_CACHABLE_WA		(1<<_CACHE_SHIFT)
+#endif
+#ifndef _CACHE_UNCACHED
+#define _CACHE_UNCACHED			(2<<_CACHE_SHIFT)
+#endif
+#ifndef _CACHE_CACHABLE_NONCOHERENT
+#define _CACHE_CACHABLE_NONCOHERENT	(3<<_CACHE_SHIFT)
+#endif
+#ifndef _CACHE_CACHABLE_CE
+#define _CACHE_CACHABLE_CE		(4<<_CACHE_SHIFT)
+#endif
+#ifndef _CACHE_CACHABLE_COW
+#define _CACHE_CACHABLE_COW		(5<<_CACHE_SHIFT)
+#endif
+#ifndef _CACHE_CACHABLE_CUW
+#define _CACHE_CACHABLE_CUW		(6<<_CACHE_SHIFT)
+#endif
+#ifndef _CACHE_UNCACHED_ACCELERATED
+#define _CACHE_UNCACHED_ACCELERATED	(7<<_CACHE_SHIFT)
+#endif
+
 #define __READABLE	(_PAGE_SILENT_READ | _PAGE_ACCESSED | (cpu_has_rixi ? 0 : _PAGE_READ))
 #define __WRITEABLE	(_PAGE_WRITE | _PAGE_SILENT_WRITE | _PAGE_MODIFIED)
 

arch/mips/include/asm/pgtable.h

@@ -366,6 +366,16 @@ static inline pgprot_t pgprot_noncached(pgprot_t _prot)
 	return __pgprot(prot);
 }
 
+static inline pgprot_t pgprot_writecombine(pgprot_t _prot)
+{
+	unsigned long prot = pgprot_val(_prot);
+
+	/* cpu_data[0].writecombine is already shifted by _CACHE_SHIFT */
+	prot = (prot & ~_CACHE_MASK) | cpu_data[0].writecombine;
+
+	return __pgprot(prot);
+}
+
 /*
  * Conversion functions: convert a page and protection to a page entry,
  * and a page entry and page directory to the page they refer to.

arch/mips/include/uapi/asm/swab.h

@@ -13,12 +13,16 @@
 
 #define __SWAB_64_THRU_32__
 
-#if defined(__mips_isa_rev) && (__mips_isa_rev >= 2)
+#if (defined(__mips_isa_rev) && (__mips_isa_rev >= 2)) ||		\
+    defined(_MIPS_ARCH_LOONGSON3A)
 
 static inline __attribute_const__ __u16 __arch_swab16(__u16 x)
 {
 	__asm__(
+	"	.set	push			\n"
+	"	.set	arch=mips32r2		\n"
 	"	wsbh	%0, %1			\n"
+	"	.set	pop			\n"
 	: "=r" (x)
 	: "r" (x));
 
@@ -29,8 +33,11 @@ static inline __attribute_const__ __u16 __arch_swab16(__u16 x)
 static inline __attribute_const__ __u32 __arch_swab32(__u32 x)
 {
 	__asm__(
+	"	.set	push			\n"
+	"	.set	arch=mips32r2		\n"
 	"	wsbh	%0, %1			\n"
 	"	rotr	%0, %0, 16		\n"
+	"	.set	pop			\n"
 	: "=r" (x)
 	: "r" (x));
 
@@ -46,8 +53,11 @@ static inline __attribute_const__ __u32 __arch_swab32(__u32 x)
 static inline __attribute_const__ __u64 __arch_swab64(__u64 x)
 {
 	__asm__(
-	"	dsbh	%0, %1\n"
-	"	dshd	%0, %0"
+	"	.set	push			\n"
+	"	.set	arch=mips64r2		\n"
+	"	dsbh	%0, %1			\n"
+	"	dshd	%0, %0			\n"
+	"	.set	pop			\n"
 	: "=r" (x)
 	: "r" (x));
 
@@ -55,5 +65,5 @@ static inline __attribute_const__ __u64 __arch_swab64(__u64 x)
 }
 #define __arch_swab64 __arch_swab64
 #endif /* __mips64 */
-#endif /* MIPS R2 or newer  */
+#endif /* MIPS R2 or newer or Loongson 3A */
 #endif /* _ASM_SWAB_H */

arch/mips/kernel/cpu-probe.c

@@ -27,6 +27,7 @@
 #include <asm/msa.h>
 #include <asm/watch.h>
 #include <asm/elf.h>
+#include <asm/pgtable-bits.h>
 #include <asm/spram.h>
 #include <asm/uaccess.h>
 
@@ -764,6 +765,7 @@ static inline void cpu_probe_legacy(struct cpuinfo_mips *c, unsigned int cpu)
 			break;
 		case PRID_REV_LOONGSON3A:
 			c->cputype = CPU_LOONGSON3;
+			c->writecombine = _CACHE_UNCACHED_ACCELERATED;
 			__cpu_name[cpu] = "ICT Loongson-3";
 			set_elf_platform(cpu, "loongson3a");
 			break;
@@ -798,67 +800,83 @@ static inline void cpu_probe_legacy(struct cpuinfo_mips *c, unsigned int cpu)
 
 static inline void cpu_probe_mips(struct cpuinfo_mips *c, unsigned int cpu)
 {
+	c->writecombine = _CACHE_UNCACHED_ACCELERATED;
 	switch (c->processor_id & PRID_IMP_MASK) {
 	case PRID_IMP_4KC:
 		c->cputype = CPU_4KC;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 4Kc";
 		break;
 	case PRID_IMP_4KEC:
 	case PRID_IMP_4KECR2:
 		c->cputype = CPU_4KEC;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 4KEc";
 		break;
 	case PRID_IMP_4KSC:
 	case PRID_IMP_4KSD:
 		c->cputype = CPU_4KSC;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 4KSc";
 		break;
 	case PRID_IMP_5KC:
 		c->cputype = CPU_5KC;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 5Kc";
 		break;
 	case PRID_IMP_5KE:
 		c->cputype = CPU_5KE;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 5KE";
 		break;
 	case PRID_IMP_20KC:
 		c->cputype = CPU_20KC;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 20Kc";
 		break;
 	case PRID_IMP_24K:
 		c->cputype = CPU_24K;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 24Kc";
 		break;
 	case PRID_IMP_24KE:
 		c->cputype = CPU_24K;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 24KEc";
 		break;
 	case PRID_IMP_25KF:
 		c->cputype = CPU_25KF;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 25Kc";
 		break;
 	case PRID_IMP_34K:
 		c->cputype = CPU_34K;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 34Kc";
 		break;
 	case PRID_IMP_74K:
 		c->cputype = CPU_74K;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 74Kc";
 		break;
 	case PRID_IMP_M14KC:
 		c->cputype = CPU_M14KC;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS M14Kc";
 		break;
 	case PRID_IMP_M14KEC:
 		c->cputype = CPU_M14KEC;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS M14KEc";
 		break;
 	case PRID_IMP_1004K:
 		c->cputype = CPU_1004K;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 1004Kc";
 		break;
 	case PRID_IMP_1074K:
 		c->cputype = CPU_1074K;
+		c->writecombine = _CACHE_UNCACHED;
 		__cpu_name[cpu] = "MIPS 1074Kc";
 		break;
 	case PRID_IMP_INTERAPTIV_UP:
@@ -932,6 +950,7 @@ static inline void cpu_probe_sibyte(struct cpuinfo_mips *c, unsigned int cpu)
 {
 	decode_configs(c);
 
+	c->writecombine = _CACHE_UNCACHED_ACCELERATED;
 	switch (c->processor_id & PRID_IMP_MASK) {
 	case PRID_IMP_SB1:
 		c->cputype = CPU_SB1;
@@ -1063,6 +1082,7 @@ static inline void cpu_probe_ingenic(struct cpuinfo_mips *c, unsigned int cpu)
 	switch (c->processor_id & PRID_IMP_MASK) {
 	case PRID_IMP_JZRISC:
 		c->cputype = CPU_JZRISC;
+		c->writecombine = _CACHE_UNCACHED_ACCELERATED;
 		__cpu_name[cpu] = "Ingenic JZRISC";
 		break;
 	default:
@@ -1169,6 +1189,7 @@ void cpu_probe(void)
 	c->processor_id = PRID_IMP_UNKNOWN;
 	c->fpu_id	= FPIR_IMP_NONE;
 	c->cputype	= CPU_UNKNOWN;
+	c->writecombine = _CACHE_UNCACHED;
 
 	c->processor_id = read_c0_prid();
 	switch (c->processor_id & PRID_COMP_MASK) {

arch/mips/kernel/setup.c

@@ -24,6 +24,8 @@
 #include <linux/debugfs.h>
 #include <linux/kexec.h>
 #include <linux/sizes.h>
+#include <linux/device.h>
+#include <linux/dma-contiguous.h>
 
 #include <asm/addrspace.h>
 #include <asm/bootinfo.h>
@@ -476,6 +478,7 @@ static void __init bootmem_init(void)
  *  o bootmem_init()
  *  o sparse_init()
  *  o paging_init()
+ *  o dma_continguous_reserve()
  *
  * At this stage the bootmem allocator is ready to use.
  *
@@ -609,6 +612,7 @@ static void __init request_crashkernel(struct resource *res)
 
 static void __init arch_mem_init(char **cmdline_p)
 {
+	struct memblock_region *reg;
 	extern void plat_mem_setup(void);
 
 	/* call board setup routine */
@@ -675,6 +679,11 @@ static void __init arch_mem_init(char **cmdline_p)
 	sparse_init();
 	plat_swiotlb_setup();
 	paging_init();
+
+	dma_contiguous_reserve(PFN_PHYS(max_low_pfn));
+	/* Tell bootmem about cma reserved memblock section */
+	for_each_memblock(reserved, reg)
+		reserve_bootmem(reg->base, reg->size, BOOTMEM_DEFAULT);
 }
 
 static void __init resource_init(void)

arch/mips/lantiq/Kconfig

@@ -30,6 +30,7 @@ choice
 config DT_EASY50712
 	bool "Easy50712"
 	depends on SOC_XWAY
+	select BUILTIN_DTB
 endchoice
 
 config PCI_LANTIQ

arch/mips/lantiq/Makefile

@@ -6,8 +6,6 @@
 
 obj-y := irq.o clk.o prom.o
 
-obj-y += dts/
-
 obj-$(CONFIG_EARLY_PRINTK) += early_printk.o
 
 obj-$(CONFIG_SOC_TYPE_XWAY) += xway/

arch/mips/lantiq/dts/Makefile

@@ -1 +0,0 @@
-obj-$(CONFIG_DT_EASY50712) := easy50712.dtb.o

arch/mips/lib/csum_partial.S

@@ -277,9 +277,12 @@ LEAF(csum_partial)
 #endif
 
 	/* odd buffer alignment? */
-#ifdef CONFIG_CPU_MIPSR2
+#if defined(CONFIG_CPU_MIPSR2) || defined(CONFIG_CPU_LOONGSON3)
+	.set	push
+	.set	arch=mips32r2
 	wsbh	v1, sum
 	movn	sum, v1, t7
+	.set	pop
 #else
 	beqz	t7, 1f			/* odd buffer alignment? */
 	 lui	v1, 0x00ff
@@ -726,9 +729,12 @@ LEAF(csum_partial)
 	addu	sum, v1
 #endif
 
-#ifdef CONFIG_CPU_MIPSR2
+#if defined(CONFIG_CPU_MIPSR2) || defined(CONFIG_CPU_LOONGSON3)
+	.set	push
+	.set	arch=mips32r2
 	wsbh	v1, sum
 	movn	sum, v1, odd
+	.set	pop
 #else
 	beqz	odd, 1f			/* odd buffer alignment? */
 	 lui	v1, 0x00ff

arch/mips/mm/dma-default.c

@@ -16,6 +16,7 @@
 #include <linux/string.h>
 #include <linux/gfp.h>
 #include <linux/highmem.h>
+#include <linux/dma-contiguous.h>
 
 #include <asm/cache.h>
 #include <asm/cpu-type.h>
@@ -128,23 +129,30 @@ static void *mips_dma_alloc_coherent(struct device *dev, size_t size,
 	dma_addr_t * dma_handle, gfp_t gfp, struct dma_attrs *attrs)
 {
 	void *ret;
+	struct page *page = NULL;
+	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
 
 	if (dma_alloc_from_coherent(dev, size, dma_handle, &ret))
 		return ret;
 
 	gfp = massage_gfp_flags(dev, gfp);
 
-	ret = (void *) __get_free_pages(gfp, get_order(size));
+	if (IS_ENABLED(CONFIG_DMA_CMA) && !(gfp & GFP_ATOMIC))
+		page = dma_alloc_from_contiguous(dev,
+					count, get_order(size));
+	if (!page)
+		page = alloc_pages(gfp, get_order(size));
 
-	if (ret) {
-		memset(ret, 0, size);
-		*dma_handle = plat_map_dma_mem(dev, ret, size);
+	if (!page)
+		return NULL;
 
-		if (!plat_device_is_coherent(dev)) {
-			dma_cache_wback_inv((unsigned long) ret, size);
-			if (!hw_coherentio)
-				ret = UNCAC_ADDR(ret);
-		}
+	ret = page_address(page);
+	memset(ret, 0, size);
+	*dma_handle = plat_map_dma_mem(dev, ret, size);
+	if (!plat_device_is_coherent(dev)) {
+		dma_cache_wback_inv((unsigned long) ret, size);
+		if (!hw_coherentio)
+			ret = UNCAC_ADDR(ret);
 	}
 
 	return ret;
@@ -164,6 +172,8 @@ static void mips_dma_free_coherent(struct device *dev, size_t size, void *vaddr,
 {
 	unsigned long addr = (unsigned long) vaddr;
 	int order = get_order(size);
+	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	struct page *page = NULL;
 
 	if (dma_release_from_coherent(dev, order, vaddr))
 		return;
@@ -173,7 +183,10 @@ static void mips_dma_free_coherent(struct device *dev, size_t size, void *vaddr,
 	if (!plat_device_is_coherent(dev) && !hw_coherentio)
 		addr = CAC_ADDR(addr);
 
-	free_pages(addr, get_order(size));
+	page = virt_to_page((void *) addr);
+
+	if (!dma_release_from_contiguous(dev, page, count))
+		__free_pages(page, get_order(size));
 }
 
 static inline void __dma_sync_virtual(void *addr, size_t size,

arch/mips/mti-sead3/Makefile

@@ -19,9 +19,5 @@ obj-y				+= sead3-i2c-dev.o sead3-i2c.o \
 
 obj-$(CONFIG_EARLY_PRINTK)	+= sead3-console.o
 obj-$(CONFIG_USB_EHCI_HCD)	+= sead3-ehci.o
-obj-$(CONFIG_OF)		+= sead3.dtb.o
 
 CFLAGS_sead3-setup.o = -I$(src)/../../../scripts/dtc/libfdt
-
-$(obj)/%.dtb: $(obj)/%.dts
-	$(call if_changed,dtc)

arch/mips/net/bpf_jit.c

@@ -163,6 +163,19 @@ do {							\
 	(ctx)->idx++;					\
 } while (0)
 
+/*
+ * Similar to emit_instr but it must be used when we need to emit
+ * 32-bit or 64-bit instructions
+ */
+#define emit_long_instr(ctx, func, ...)			\
+do {							\
+	if ((ctx)->target != NULL) {			\
+		u32 *p = &(ctx)->target[ctx->idx];	\
+		UASM_i_##func(&p, ##__VA_ARGS__);	\
+	}						\
+	(ctx)->idx++;					\
+} while (0)
+
 /* Determine if immediate is within the 16-bit signed range */
 static inline bool is_range16(s32 imm)
 {
@@ -218,13 +231,6 @@ static inline void emit_ori(unsigned int dst, unsigned src, u32 imm,
 	}
 }
 
-
-static inline void emit_daddu(unsigned int dst, unsigned int src1,
-			      unsigned int src2, struct jit_ctx *ctx)
-{
-	emit_instr(ctx, daddu, dst, src1, src2);
-}
-
 static inline void emit_daddiu(unsigned int dst, unsigned int src,
 			       int imm, struct jit_ctx *ctx)
 {
@@ -283,11 +289,7 @@ static inline void emit_xori(ptr dst, ptr src, u32 imm, struct jit_ctx *ctx)
 
 static inline void emit_stack_offset(int offset, struct jit_ctx *ctx)
 {
-	if (config_enabled(CONFIG_64BIT))
-		emit_instr(ctx, daddiu, r_sp, r_sp, offset);
-	else
-		emit_instr(ctx, addiu, r_sp, r_sp, offset);
-
+	emit_long_instr(ctx, ADDIU, r_sp, r_sp, offset);
 }
 
 static inline void emit_subu(unsigned int dst, unsigned int src1,
@@ -365,10 +367,7 @@ static inline void emit_store_stack_reg(ptr reg, ptr base,
 					unsigned int offset,
 					struct jit_ctx *ctx)
 {
-	if (config_enabled(CONFIG_64BIT))
-		emit_instr(ctx, sd, reg, offset, base);
-	else
-		emit_instr(ctx, sw, reg, offset, base);
+	emit_long_instr(ctx, SW, reg, offset, base);
 }
 
 static inline void emit_store(ptr reg, ptr base, unsigned int offset,
@@ -381,10 +380,7 @@ static inline void emit_load_stack_reg(ptr reg, ptr base,
 				       unsigned int offset,
 				       struct jit_ctx *ctx)
 {
-	if (config_enabled(CONFIG_64BIT))
-		emit_instr(ctx, ld, reg, offset, base);
-	else
-		emit_instr(ctx, lw, reg, offset, base);
+	emit_long_instr(ctx, LW, reg, offset, base);
 }
 
 static inline void emit_load(unsigned int reg, unsigned int base,
@@ -458,10 +454,7 @@ static inline void emit_load_ptr(unsigned int dst, unsigned int src,
 				     int imm, struct jit_ctx *ctx)
 {
 	/* src contains the base addr of the 32/64-pointer */
-	if (config_enabled(CONFIG_64BIT))
-		emit_instr(ctx, ld, dst, imm, src);
-	else
-		emit_instr(ctx, lw, dst, imm, src);
+	emit_long_instr(ctx, LW, dst, imm, src);
 }
 
 /* load a function pointer to register */
@@ -483,10 +476,7 @@ static inline void emit_load_func(unsigned int reg, ptr imm,
 /* Move to real MIPS register */
 static inline void emit_reg_move(ptr dst, ptr src, struct jit_ctx *ctx)
 {
-	if (config_enabled(CONFIG_64BIT))
-		emit_daddu(dst, src, r_zero, ctx);
-	else
-		emit_addu(dst, src, r_zero, ctx);
+	emit_long_instr(ctx, ADDU, dst, src, r_zero);
 }
 
 /* Move to JIT (32-bit) register */
@@ -623,10 +613,7 @@ static void save_bpf_jit_regs(struct jit_ctx *ctx, unsigned offset)
 	if (ctx->flags & SEEN_MEM) {
 		if (real_off % (RSIZE * 2))
 			real_off += RSIZE;
-		if (config_enabled(CONFIG_64BIT))
-			emit_daddiu(r_M, r_sp, real_off, ctx);
-		else
-			emit_addiu(r_M, r_sp, real_off, ctx);
+		emit_long_instr(ctx, ADDIU, r_M, r_sp, real_off);
 	}
 }
 
@@ -1241,7 +1228,7 @@ jmp_cmp:
 			emit_half_load(r_A, r_skb, off, ctx);
 #ifdef CONFIG_CPU_LITTLE_ENDIAN
 			/* This needs little endian fixup */
-			if (cpu_has_mips_r2) {
+			if (cpu_has_wsbh) {
 				/* R2 and later have the wsbh instruction */
 				emit_wsbh(r_A, r_A, ctx);
 			} else {

arch/mips/netlogic/Kconfig

@@ -4,6 +4,7 @@ if NLM_XLP_BOARD
 config DT_XLP_EVP
 	bool "Built-in device tree for XLP EVP boards"
 	default y
+	select BUILTIN_DTB
 	help
 	  Add an FDT blob for XLP EVP boards into the kernel.
 	  This DTB will be used if the firmware does not pass in a DTB
@@ -13,6 +14,7 @@ config DT_XLP_EVP
 config DT_XLP_SVP
 	bool "Built-in device tree for XLP SVP boards"
 	default y
+	select BUILTIN_DTB
 	help
 	  Add an FDT blob for XLP VP boards into the kernel.
 	  This DTB will be used if the firmware does not pass in a DTB
@@ -22,6 +24,7 @@ config DT_XLP_SVP
 config DT_XLP_FVP
 	bool "Built-in device tree for XLP FVP boards"
 	default y
+	select BUILTIN_DTB
 	help
 	  Add an FDT blob for XLP FVP board into the kernel.
 	  This DTB will be used if the firmware does not pass in a DTB
@@ -31,6 +34,7 @@ config DT_XLP_FVP
 config DT_XLP_GVP
 	bool "Built-in device tree for XLP GVP boards"
 	default y
+	select BUILTIN_DTB
 	help
 	  Add an FDT blob for XLP GVP board into the kernel.
 	  This DTB will be used if the firmware does not pass in a DTB

arch/mips/netlogic/Makefile

@@ -1,4 +1,3 @@
 obj-$(CONFIG_NLM_COMMON)	+=	common/
 obj-$(CONFIG_CPU_XLR)		+=	xlr/
 obj-$(CONFIG_CPU_XLP)		+=	xlp/
-obj-$(CONFIG_CPU_XLP)		+=	dts/

arch/mips/netlogic/dts/Makefile

@@ -1,4 +0,0 @@
-obj-$(CONFIG_DT_XLP_EVP) := xlp_evp.dtb.o
-obj-$(CONFIG_DT_XLP_SVP) += xlp_svp.dtb.o
-obj-$(CONFIG_DT_XLP_FVP) += xlp_fvp.dtb.o
-obj-$(CONFIG_DT_XLP_GVP) += xlp_gvp.dtb.o

arch/mips/ralink/Kconfig

@@ -42,18 +42,22 @@ choice
 	config DTB_RT2880_EVAL
 		bool "RT2880 eval kit"
 		depends on SOC_RT288X
+		select BUILTIN_DTB
 
 	config DTB_RT305X_EVAL
 		bool "RT305x eval kit"
 		depends on SOC_RT305X
+		select BUILTIN_DTB
 
 	config DTB_RT3883_EVAL
 		bool "RT3883 eval kit"
 		depends on SOC_RT3883
+		select BUILTIN_DTB
 
 	config DTB_MT7620A_EVAL
 		bool "MT7620A eval kit"
 		depends on SOC_MT7620
+		select BUILTIN_DTB
 
 endchoice
 

arch/mips/ralink/Makefile

@@ -16,5 +16,3 @@ obj-$(CONFIG_SOC_RT3883) += rt3883.o
 obj-$(CONFIG_SOC_MT7620) += mt7620.o
 
 obj-$(CONFIG_EARLY_PRINTK) += early_printk.o
-
-obj-y += dts/

arch/mips/ralink/dts/Makefile

@@ -1,4 +0,0 @@
-obj-$(CONFIG_DTB_RT2880_EVAL) := rt2880_eval.dtb.o
-obj-$(CONFIG_DTB_RT305X_EVAL) := rt3052_eval.dtb.o
-obj-$(CONFIG_DTB_RT3883_EVAL) := rt3883_eval.dtb.o
-obj-$(CONFIG_DTB_MT7620A_EVAL) := mt7620a_eval.dtb.o

arch/mips/ralink/timer.c

@@ -58,7 +58,7 @@ static irqreturn_t rt_timer_irq(int irq, void *_rt)
 
 static int rt_timer_request(struct rt_timer *rt)
 {
-	int err = request_irq(rt->irq, rt_timer_irq, IRQF_DISABLED,
+	int err = request_irq(rt->irq, rt_timer_irq, 0,
 						dev_name(rt->dev), rt);
 	if (err) {
 		dev_err(rt->dev, "failed to request irq\n");

arch/x86/include/asm/Kbuild

@@ -6,6 +6,7 @@ genhdr-y += unistd_x32.h
 
 generic-y += clkdev.h
 generic-y += cputime.h
+generic-y += dma-contiguous.h
 generic-y += early_ioremap.h
 generic-y += mcs_spinlock.h
 generic-y += scatterlist.h

include/asm-generic/dma-contiguous.h

@@ -1,7 +1,5 @@
-#ifndef ASMX86_DMA_CONTIGUOUS_H
-#define ASMX86_DMA_CONTIGUOUS_H
-
-#ifdef __KERNEL__
+#ifndef _ASM_GENERIC_DMA_CONTIGUOUS_H
+#define _ASM_GENERIC_DMA_CONTIGUOUS_H
 
 #include <linux/types.h>
 
@@ -9,4 +7,3 @@ static inline void
 dma_contiguous_early_fixup(phys_addr_t base, unsigned long size) { }
 
 #endif
-#endif