Staging: virtual block device driver (ramzswap)

Creates RAM based block devices (/dev/ramzswapX) which can be
used (only) as swap disks. Pages swapped to these are compressed
and stored in memory itself.

The module is called ramzswap.ko. It depends on:
 - xvmalloc memory allocator (compiled with this driver)
 - lzo_compress.ko
 - lzo_decompress.ko

See ramzswap.txt for usage details.

Signed-off-by: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

drivers/staging/Kconfig

@@ -123,6 +123,8 @@ source "drivers/staging/sep/Kconfig"
 
 source "drivers/staging/iio/Kconfig"
 
+source "drivers/staging/ramzswap/Kconfig"
+
 source "drivers/staging/strip/Kconfig"
 
 source "drivers/staging/arlan/Kconfig"

drivers/staging/Makefile

@@ -43,6 +43,7 @@ obj-$(CONFIG_VME_BUS)		+= vme/
 obj-$(CONFIG_RAR_REGISTER)	+= rar/
 obj-$(CONFIG_DX_SEP)		+= sep/
 obj-$(CONFIG_IIO)		+= iio/
+obj-$(CONFIG_RAMZSWAP)		+= ramzswap/
 obj-$(CONFIG_STRIP)		+= strip/
 obj-$(CONFIG_ARLAN)		+= arlan/
 obj-$(CONFIG_WAVELAN)		+= wavelan/

drivers/staging/ramzswap/Kconfig

@@ -0,0 +1,21 @@
+config RAMZSWAP
+	tristate "Compressed in-memory swap device (ramzswap)"
+	depends on SWAP
+	select LZO_COMPRESS
+	select LZO_DECOMPRESS
+	default n
+	help
+	  Creates virtual block devices which can be used (only) as a swap
+	  disks. Pages swapped to these disks are compressed and stored in
+	  memory itself.
+
+	  See ramzswap.txt for more information.
+	  Project home: http://compcache.googlecode.com/
+
+config RAMZSWAP_STATS
+	bool "Enable ramzswap stats"
+	depends on RAMZSWAP
+	default y
+	help
+	  Enable statistics collection for ramzswap. This adds only a minimal
+	  overhead. In unsure, say Y.

drivers/staging/ramzswap/Makefile

@@ -0,0 +1,3 @@
+ramzswap-objs	:=	ramzswap_drv.o xvmalloc.o
+
+obj-$(CONFIG_RAMZSWAP)	+=	ramzswap.o

drivers/staging/ramzswap/ramzswap_drv.h

@@ -0,0 +1,171 @@
+/*
+ * Compressed RAM based swap device
+ *
+ * Copyright (C) 2008, 2009  Nitin Gupta
+ *
+ * This code is released using a dual license strategy: BSD/GPL
+ * You can choose the licence that better fits your requirements.
+ *
+ * Released under the terms of 3-clause BSD License
+ * Released under the terms of GNU General Public License Version 2.0
+ *
+ * Project home: http://compcache.googlecode.com
+ */
+
+#ifndef _RAMZSWAP_DRV_H_
+#define _RAMZSWAP_DRV_H_
+
+#include "ramzswap_ioctl.h"
+#include "xvmalloc.h"
+
+/*
+ * Some arbitrary value. This is just to catch
+ * invalid value for num_devices module parameter.
+ */
+static const unsigned max_num_devices = 32;
+
+/*
+ * Stored at beginning of each compressed object.
+ *
+ * It stores back-reference to table entry which points to this
+ * object. This is required to support memory defragmentation or
+ * migrating compressed pages to backing swap disk.
+ */
+struct zobj_header {
+#if 0
+	u32 table_idx;
+#endif
+};
+
+/*-- Configurable parameters */
+
+/* Default ramzswap disk size: 25% of total RAM */
+static const unsigned default_disksize_perc_ram = 25;
+static const unsigned default_memlimit_perc_ram = 15;
+
+/*
+ * Max compressed page size when backing device is provided.
+ * Pages that compress to size greater than this are sent to
+ * physical swap disk.
+ */
+static const unsigned max_zpage_size_bdev = PAGE_SIZE / 2;
+
+/*
+ * Max compressed page size when there is no backing dev.
+ * Pages that compress to size greater than this are stored
+ * uncompressed in memory.
+ */
+static const unsigned max_zpage_size_nobdev = PAGE_SIZE / 4 * 3;
+
+/*
+ * NOTE: max_zpage_size_{bdev,nobdev} sizes must be
+ * less than or equal to:
+ *   XV_MAX_ALLOC_SIZE - sizeof(struct zobj_header)
+ * since otherwise xv_malloc would always return failure.
+ */
+
+/*-- End of configurable params */
+
+#define SECTOR_SHIFT		9
+#define SECTOR_SIZE		(1 << SECTOR_SHIFT)
+#define SECTORS_PER_PAGE_SHIFT	(PAGE_SHIFT - SECTOR_SHIFT)
+#define SECTORS_PER_PAGE	(1 << SECTORS_PER_PAGE_SHIFT)
+
+/* Debugging and Stats */
+#if defined(CONFIG_RAMZSWAP_STATS)
+#define stat_inc(stat)	((stat)++)
+#define stat_dec(stat)	((stat)--)
+#else
+#define stat_inc(x)
+#define stat_dec(x)
+#endif
+
+/* Flags for ramzswap pages (table[page_no].flags) */
+enum rzs_pageflags {
+	/* Page is stored uncompressed */
+	RZS_UNCOMPRESSED,
+
+	/* Page consists entirely of zeros */
+	RZS_ZERO,
+
+	__NR_RZS_PAGEFLAGS,
+};
+
+/*-- Data structures */
+
+/*
+ * Allocated for each swap slot, indexed by page no.
+ * These table entries must fit exactly in a page.
+ */
+struct table {
+	struct page *page;
+	u16 offset;
+	u8 count;	/* object ref count (not yet used) */
+	u8 flags;
+} __attribute__((aligned(4)));;
+
+/*
+ * Swap extent information in case backing swap is a regular
+ * file. These extent entries must fit exactly in a page.
+ */
+struct ramzswap_backing_extent {
+	pgoff_t phy_pagenum;
+	pgoff_t num_pages;
+} __attribute__((aligned(4)));
+
+struct ramzswap_stats {
+	/* basic stats */
+	size_t compr_size;	/* compressed size of pages stored -
+				 * needed to enforce memlimit */
+	/* more stats */
+#if defined(CONFIG_RAMZSWAP_STATS)
+	u64 num_reads;		/* failed + successful */
+	u64 num_writes;		/* --do-- */
+	u64 failed_reads;	/* can happen when memory is too low */
+	u64 failed_writes;	/* should NEVER! happen */
+	u64 invalid_io;		/* non-swap I/O requests */
+	u32 pages_zero;		/* no. of zero filled pages */
+	u32 pages_stored;	/* no. of pages currently stored */
+	u32 good_compress;	/* % of pages with compression ratio<=50% */
+	u32 pages_expand;	/* % of incompressible pages */
+	u64 bdev_num_reads;	/* no. of reads on backing dev */
+	u64 bdev_num_writes;	/* no. of writes on backing dev */
+#endif
+};
+
+struct ramzswap {
+	struct xv_pool *mem_pool;
+	void *compress_workmem;
+	void *compress_buffer;
+	struct table *table;
+	struct mutex lock;
+	struct request_queue *queue;
+	struct gendisk *disk;
+	int init_done;
+	/*
+	 * This is limit on compressed data size (stats.compr_size)
+	 * Its applicable only when backing swap device is present.
+	 */
+	size_t memlimit;	/* bytes */
+	/*
+	 * This is limit on amount of *uncompressed* worth of data
+	 * we can hold. When backing swap device is provided, it is
+	 * set equal to device size.
+	 */
+	size_t disksize;	/* bytes */
+
+	struct ramzswap_stats stats;
+
+	/* backing swap device info */
+	struct ramzswap_backing_extent *curr_extent;
+	struct list_head backing_swap_extent_list;
+	unsigned long num_extents;
+	char backing_swap_name[MAX_SWAP_NAME_LEN];
+	struct block_device *backing_swap;
+	struct file *swap_file;
+};
+
+/*-- */
+
+#endif
+

drivers/staging/ramzswap/ramzswap_ioctl.h

@@ -0,0 +1,49 @@
+/*
+ * Compressed RAM based swap device
+ *
+ * Copyright (C) 2008, 2009  Nitin Gupta
+ *
+ * This code is released using a dual license strategy: BSD/GPL
+ * You can choose the licence that better fits your requirements.
+ *
+ * Released under the terms of 3-clause BSD License
+ * Released under the terms of GNU General Public License Version 2.0
+ *
+ * Project home: http://compcache.googlecode.com
+ */
+
+#ifndef _RAMZSWAP_IOCTL_H_
+#define _RAMZSWAP_IOCTL_H_
+
+#define MAX_SWAP_NAME_LEN 128
+
+struct ramzswap_ioctl_stats {
+	char backing_swap_name[MAX_SWAP_NAME_LEN];
+	u64 memlimit;		/* only applicable if backing swap present */
+	u64 disksize;		/* user specified or equal to backing swap
+				 * size (if present) */
+	u64 num_reads;		/* failed + successful */
+	u64 num_writes;		/* --do-- */
+	u64 failed_reads;	/* can happen when memory is too low */
+	u64 failed_writes;	/* should NEVER! happen */
+	u64 invalid_io;		/* non-swap I/O requests */
+	u32 pages_zero;		/* no. of zero filled pages */
+	u32 good_compress_pct;	/* no. of pages with compression ratio<=50% */
+	u32 pages_expand_pct;	/* no. of incompressible pages */
+	u32 pages_stored;
+	u32 pages_used;
+	u64 orig_data_size;
+	u64 compr_data_size;
+	u64 mem_used_total;
+	u64 bdev_num_reads;	/* no. of reads on backing dev */
+	u64 bdev_num_writes;	/* no. of writes on backing dev */
+} __attribute__ ((packed, aligned(4)));
+
+#define RZSIO_SET_DISKSIZE_KB	_IOW('z', 0, size_t)
+#define RZSIO_SET_MEMLIMIT_KB	_IOW('z', 1, size_t)
+#define RZSIO_SET_BACKING_SWAP	_IOW('z', 2, unsigned char[MAX_SWAP_NAME_LEN])
+#define RZSIO_GET_STATS		_IOR('z', 3, struct ramzswap_ioctl_stats)
+#define RZSIO_INIT		_IO('z', 4)
+#define RZSIO_RESET		_IO('z', 5)
+
+#endif