These patches optionally improve the multi-threading peformance

of Squashfs by adding parallel decompression, and direct
 decompression into the page cache, eliminating an intermediate
 buffer (removing memcpy overhead and lock contention).
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.4.12 (GNU/Linux)
 
 iQIcBAABAgAGBQJSjP25AAoJEJAch/D1fbHUsFAQAJjpCfWyv7JfNtJSUk20UgbC
 kQvpMUbISwrLnszW6ooWBJxQ4OCyQ3AWN5yrAk6hb86oR0SN33WAJjqW5hR5htOy
 5ZMs3OnzE3haUej+Xxw/VTK61FoOq/PjK8UZ6NBfhdnihfE/fQykrgDhHznJ68iq
 0wWeqTY68sF5ZBQ2kKhBSfF+lGlWeLqhiFGOq68MAv4Rd8dGZsiLIFG7JQsrwmAn
 cswmiCTQppGGz/+6FBWvaEEpD+nUCX/h/1XKwMhuzWwZZTFWPM+BkEfMOKv78txW
 GWn/o1/kWA/u1f5V+nZlUhNtj+KCU11YZfTAJ30Ie1erzKCh8DGcLhCyfC0N+Hw/
 Na5vxyjEnTdJoBnRbcPpHcGwPB0J5Q2nCzu1b/3blUGdpXQrNp/zZ4hg53fYEKHy
 2KAf9j5rqs85IqoKwrzeod/V1WakjMQJPntoJ2r7ILP4lKfvOHt6m1D5/7tVodxZ
 mGa8eaQtH5SrtnLldKo4vGgh65/ViQ2cVlAbGC7I9rfXJ0fITYO8PvKBTcXvtOHc
 +rjCnoOHtSv8FvFf1G9qfbBMwaC+3n95rYSac0Ibl6O7x2pdQusUmiuyUf1NXsDg
 V4ENspn/DTrltZbTbBTgI3LizxvJOMtf72zo+Bhghitp09yeIFfieVqM/kuR74Ym
 O4EaVGcFdXJJc3UmK/69
 =lV/M
 -----END PGP SIGNATURE-----

Merge tag 'squashfs-updates' of git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-next

Pull squashfs updates from Phillip Lougher:
 "These patches optionally improve the multi-threading peformance of
  Squashfs by adding parallel decompression, and direct decompression
  into the page cache, eliminating an intermediate buffer (removing
  memcpy overhead and lock contention)"

* tag 'squashfs-updates' of git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-next:
  Squashfs: Check stream is not NULL in decompressor_multi.c
  Squashfs: Directly decompress into the page cache for file data
  Squashfs: Restructure squashfs_readpage()
  Squashfs: Generalise paging handling in the decompressors
  Squashfs: add multi-threaded decompression using percpu variable
  squashfs: Enhance parallel I/O
  Squashfs: Refactor decompressor interface and code

fs/squashfs/Kconfig

@@ -25,6 +25,78 @@ config SQUASHFS
 
 	  If unsure, say N.
 
+choice
+	prompt "File decompression options"
+	depends on SQUASHFS
+	help
+	  Squashfs now supports two options for decompressing file
+	  data.  Traditionally Squashfs has decompressed into an
+	  intermediate buffer and then memcopied it into the page cache.
+	  Squashfs now supports the ability to decompress directly into
+	  the page cache.
+
+	  If unsure, select "Decompress file data into an intermediate buffer"
+
+config SQUASHFS_FILE_CACHE
+	bool "Decompress file data into an intermediate buffer"
+	help
+	  Decompress file data into an intermediate buffer and then
+	  memcopy it into the page cache.
+
+config SQUASHFS_FILE_DIRECT
+	bool "Decompress files directly into the page cache"
+	help
+	  Directly decompress file data into the page cache.
+	  Doing so can significantly improve performance because
+	  it eliminates a memcpy and it also removes the lock contention
+	  on the single buffer.
+
+endchoice
+
+choice
+	prompt "Decompressor parallelisation options"
+	depends on SQUASHFS
+	help
+	  Squashfs now supports three parallelisation options for
+	  decompression.  Each one exhibits various trade-offs between
+	  decompression performance and CPU and memory usage.
+
+	  If in doubt, select "Single threaded compression"
+
+config SQUASHFS_DECOMP_SINGLE
+	bool "Single threaded compression"
+	help
+	  Traditionally Squashfs has used single-threaded decompression.
+	  Only one block (data or metadata) can be decompressed at any
+	  one time.  This limits CPU and memory usage to a minimum.
+
+config SQUASHFS_DECOMP_MULTI
+	bool "Use multiple decompressors for parallel I/O"
+	help
+	  By default Squashfs uses a single decompressor but it gives
+	  poor performance on parallel I/O workloads when using multiple CPU
+	  machines due to waiting on decompressor availability.
+
+	  If you have a parallel I/O workload and your system has enough memory,
+	  using this option may improve overall I/O performance.
+
+	  This decompressor implementation uses up to two parallel
+	  decompressors per core.  It dynamically allocates decompressors
+	  on a demand basis.
+
+config SQUASHFS_DECOMP_MULTI_PERCPU
+	bool "Use percpu multiple decompressors for parallel I/O"
+	help
+	  By default Squashfs uses a single decompressor but it gives
+	  poor performance on parallel I/O workloads when using multiple CPU
+	  machines due to waiting on decompressor availability.
+
+	  This decompressor implementation uses a maximum of one
+	  decompressor per core.  It uses percpu variables to ensure
+	  decompression is load-balanced across the cores.
+
+endchoice
+
 config SQUASHFS_XATTR
 	bool "Squashfs XATTR support"
 	depends on SQUASHFS

fs/squashfs/Makefile

@@ -5,6 +5,11 @@
 obj-$(CONFIG_SQUASHFS) += squashfs.o
 squashfs-y += block.o cache.o dir.o export.o file.o fragment.o id.o inode.o
 squashfs-y += namei.o super.o symlink.o decompressor.o
+squashfs-$(CONFIG_SQUASHFS_FILE_CACHE) += file_cache.o
+squashfs-$(CONFIG_SQUASHFS_FILE_DIRECT) += file_direct.o page_actor.o
+squashfs-$(CONFIG_SQUASHFS_DECOMP_SINGLE) += decompressor_single.o
+squashfs-$(CONFIG_SQUASHFS_DECOMP_MULTI) += decompressor_multi.o
+squashfs-$(CONFIG_SQUASHFS_DECOMP_MULTI_PERCPU) += decompressor_multi_percpu.o
 squashfs-$(CONFIG_SQUASHFS_XATTR) += xattr.o xattr_id.o
 squashfs-$(CONFIG_SQUASHFS_LZO) += lzo_wrapper.o
 squashfs-$(CONFIG_SQUASHFS_XZ) += xz_wrapper.o

fs/squashfs/block.c

@@ -36,6 +36,7 @@
 #include "squashfs_fs_sb.h"
 #include "squashfs.h"
 #include "decompressor.h"
+#include "page_actor.h"
 
 /*
  * Read the metadata block length, this is stored in the first two
@@ -86,16 +87,16 @@ static struct buffer_head *get_block_length(struct super_block *sb,
  * generated a larger block - this does occasionally happen with compression
  * algorithms).
  */
-int squashfs_read_data(struct super_block *sb, void **buffer, u64 index,
+int squashfs_read_data(struct super_block *sb, u64 index, int length,
-			int length, u64 *next_index, int srclength, int pages)
+		u64 *next_index, struct squashfs_page_actor *output)
 {
 	struct squashfs_sb_info *msblk = sb->s_fs_info;
 	struct buffer_head **bh;
 	int offset = index & ((1 << msblk->devblksize_log2) - 1);
 	u64 cur_index = index >> msblk->devblksize_log2;
-	int bytes, compressed, b = 0, k = 0, page = 0, avail;
+	int bytes, compressed, b = 0, k = 0, avail, i;
 
-	bh = kcalloc(((srclength + msblk->devblksize - 1)
+	bh = kcalloc(((output->length + msblk->devblksize - 1)
 		>> msblk->devblksize_log2) + 1, sizeof(*bh), GFP_KERNEL);
 	if (bh == NULL)
 		return -ENOMEM;
@@ -111,9 +112,9 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index,
 			*next_index = index + length;
 
 		TRACE("Block @ 0x%llx, %scompressed size %d, src size %d\n",
-			index, compressed ? "" : "un", length, srclength);
+			index, compressed ? "" : "un", length, output->length);
 
-		if (length < 0 || length > srclength ||
+		if (length < 0 || length > output->length ||
 				(index + length) > msblk->bytes_used)
 			goto read_failure;
 
@@ -145,7 +146,7 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index,
 		TRACE("Block @ 0x%llx, %scompressed size %d\n", index,
 				compressed ? "" : "un", length);
 
-		if (length < 0 || length > srclength ||
+		if (length < 0 || length > output->length ||
 					(index + length) > msblk->bytes_used)
 			goto block_release;
 
@@ -158,9 +159,15 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index,
 		ll_rw_block(READ, b - 1, bh + 1);
 	}
 
+	for (i = 0; i < b; i++) {
+		wait_on_buffer(bh[i]);
+		if (!buffer_uptodate(bh[i]))
+			goto block_release;
+	}
+
 	if (compressed) {
-		length = squashfs_decompress(msblk, buffer, bh, b, offset,
+		length = squashfs_decompress(msblk, bh, b, offset, length,
-			 length, srclength, pages);
+			output);
 		if (length < 0)
 			goto read_failure;
 	} else {
@@ -168,22 +175,20 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index,
 		 * Block is uncompressed.
 		 */
 		int in, pg_offset = 0;
+		void *data = squashfs_first_page(output);
 
 		for (bytes = length; k < b; k++) {
 			in = min(bytes, msblk->devblksize - offset);
 			bytes -= in;
-			wait_on_buffer(bh[k]);
-			if (!buffer_uptodate(bh[k]))
-				goto block_release;
 			while (in) {
 				if (pg_offset == PAGE_CACHE_SIZE) {
-					page++;
+					data = squashfs_next_page(output);
 					pg_offset = 0;
 				}
 				avail = min_t(int, in, PAGE_CACHE_SIZE -
 						pg_offset);
-				memcpy(buffer[page] + pg_offset,
+				memcpy(data + pg_offset, bh[k]->b_data + offset,
-						bh[k]->b_data + offset, avail);
+						avail);
 				in -= avail;
 				pg_offset += avail;
 				offset += avail;
@@ -191,6 +196,7 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index,
 			offset = 0;
 			put_bh(bh[k]);
 		}
+		squashfs_finish_page(output);
 	}
 
 	kfree(bh);

fs/squashfs/cache.c

@@ -56,6 +56,7 @@
 #include "squashfs_fs.h"
 #include "squashfs_fs_sb.h"
 #include "squashfs.h"
+#include "page_actor.h"
 
 /*
  * Look-up block in cache, and increment usage count.  If not in cache, read
@@ -119,9 +120,8 @@ struct squashfs_cache_entry *squashfs_cache_get(struct super_block *sb,
 			entry->error = 0;
 			spin_unlock(&cache->lock);
 
-			entry->length = squashfs_read_data(sb, entry->data,
+			entry->length = squashfs_read_data(sb, block, length,
-				block, length, &entry->next_index,
+				&entry->next_index, entry->actor);
-				cache->block_size, cache->pages);
 
 			spin_lock(&cache->lock);
 
@@ -220,6 +220,7 @@ void squashfs_cache_delete(struct squashfs_cache *cache)
 				kfree(cache->entry[i].data[j]);
 			kfree(cache->entry[i].data);
 		}
+		kfree(cache->entry[i].actor);
 	}
 
 	kfree(cache->entry);
@@ -280,6 +281,13 @@ struct squashfs_cache *squashfs_cache_init(char *name, int entries,
 				goto cleanup;
 			}
 		}
+
+		entry->actor = squashfs_page_actor_init(entry->data,
+						cache->pages, 0);
+		if (entry->actor == NULL) {
+			ERROR("Failed to allocate %s cache entry\n", name);
+			goto cleanup;
+		}
 	}
 
 	return cache;
@@ -410,6 +418,7 @@ void *squashfs_read_table(struct super_block *sb, u64 block, int length)
 	int pages = (length + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
 	int i, res;
 	void *table, *buffer, **data;
+	struct squashfs_page_actor *actor;
 
 	table = buffer = kmalloc(length, GFP_KERNEL);
 	if (table == NULL)
@@ -421,19 +430,28 @@ void *squashfs_read_table(struct super_block *sb, u64 block, int length)
 		goto failed;
 	}
 
+	actor = squashfs_page_actor_init(data, pages, length);
+	if (actor == NULL) {
+		res = -ENOMEM;
+		goto failed2;
+	}
+
 	for (i = 0; i < pages; i++, buffer += PAGE_CACHE_SIZE)
 		data[i] = buffer;
 
-	res = squashfs_read_data(sb, data, block, length |
+	res = squashfs_read_data(sb, block, length |
-		SQUASHFS_COMPRESSED_BIT_BLOCK, NULL, length, pages);
+		SQUASHFS_COMPRESSED_BIT_BLOCK, NULL, actor);
 
 	kfree(data);
+	kfree(actor);
 
 	if (res < 0)
 		goto failed;
 
 	return table;
 
+failed2:
+	kfree(data);
 failed:
 	kfree(table);
 	return ERR_PTR(res);

fs/squashfs/decompressor.c

@@ -30,6 +30,7 @@
 #include "squashfs_fs_sb.h"
 #include "decompressor.h"
 #include "squashfs.h"
+#include "page_actor.h"
 
 /*
  * This file (and decompressor.h) implements a decompressor framework for
@@ -37,29 +38,29 @@
  */
 
 static const struct squashfs_decompressor squashfs_lzma_unsupported_comp_ops = {
-	NULL, NULL, NULL, LZMA_COMPRESSION, "lzma", 0
+	NULL, NULL, NULL, NULL, LZMA_COMPRESSION, "lzma", 0
 };
 
 #ifndef CONFIG_SQUASHFS_LZO
 static const struct squashfs_decompressor squashfs_lzo_comp_ops = {
-	NULL, NULL, NULL, LZO_COMPRESSION, "lzo", 0
+	NULL, NULL, NULL, NULL, LZO_COMPRESSION, "lzo", 0
 };
 #endif
 
 #ifndef CONFIG_SQUASHFS_XZ
 static const struct squashfs_decompressor squashfs_xz_comp_ops = {
-	NULL, NULL, NULL, XZ_COMPRESSION, "xz", 0
+	NULL, NULL, NULL, NULL, XZ_COMPRESSION, "xz", 0
 };
 #endif
 
 #ifndef CONFIG_SQUASHFS_ZLIB
 static const struct squashfs_decompressor squashfs_zlib_comp_ops = {
-	NULL, NULL, NULL, ZLIB_COMPRESSION, "zlib", 0
+	NULL, NULL, NULL, NULL, ZLIB_COMPRESSION, "zlib", 0
 };
 #endif
 
 static const struct squashfs_decompressor squashfs_unknown_comp_ops = {
-	NULL, NULL, NULL, 0, "unknown", 0
+	NULL, NULL, NULL, NULL, 0, "unknown", 0
 };
 
 static const struct squashfs_decompressor *decompressor[] = {
@@ -83,10 +84,11 @@ const struct squashfs_decompressor *squashfs_lookup_decompressor(int id)
 }
 
 
-void *squashfs_decompressor_init(struct super_block *sb, unsigned short flags)
+static void *get_comp_opts(struct super_block *sb, unsigned short flags)
 {
 	struct squashfs_sb_info *msblk = sb->s_fs_info;
-	void *strm, *buffer = NULL;
+	void *buffer = NULL, *comp_opts;
+	struct squashfs_page_actor *actor = NULL;
 	int length = 0;
 
 	/*
@@ -94,23 +96,46 @@ void *squashfs_decompressor_init(struct super_block *sb, unsigned short flags)
 	 */
 	if (SQUASHFS_COMP_OPTS(flags)) {
 		buffer = kmalloc(PAGE_CACHE_SIZE, GFP_KERNEL);
-		if (buffer == NULL)
+		if (buffer == NULL) {
-			return ERR_PTR(-ENOMEM);
+			comp_opts = ERR_PTR(-ENOMEM);
+			goto out;
+		}
 
-		length = squashfs_read_data(sb, &buffer,
+		actor = squashfs_page_actor_init(&buffer, 1, 0);
-			sizeof(struct squashfs_super_block), 0, NULL,
+		if (actor == NULL) {
-			PAGE_CACHE_SIZE, 1);
+			comp_opts = ERR_PTR(-ENOMEM);
+			goto out;
+		}
+
+		length = squashfs_read_data(sb,
+			sizeof(struct squashfs_super_block), 0, NULL, actor);
 
 		if (length < 0) {
-			strm = ERR_PTR(length);
+			comp_opts = ERR_PTR(length);
-			goto finished;
+			goto out;
 		}
 	}
 
-	strm = msblk->decompressor->init(msblk, buffer, length);
+	comp_opts = squashfs_comp_opts(msblk, buffer, length);
 
-finished:
+out:
+	kfree(actor);
 	kfree(buffer);
-
+	return comp_opts;
-	return strm;
+}
+
+
+void *squashfs_decompressor_setup(struct super_block *sb, unsigned short flags)
+{
+	struct squashfs_sb_info *msblk = sb->s_fs_info;
+	void *stream, *comp_opts = get_comp_opts(sb, flags);
+
+	if (IS_ERR(comp_opts))
+		return comp_opts;
+
+	stream = squashfs_decompressor_create(msblk, comp_opts);
+	if (IS_ERR(stream))
+		kfree(comp_opts);
+
+	return stream;
 }

fs/squashfs/decompressor.h

@@ -24,28 +24,22 @@
  */
 
 struct squashfs_decompressor {
-	void	*(*init)(struct squashfs_sb_info *, void *, int);
+	void	*(*init)(struct squashfs_sb_info *, void *);
+	void	*(*comp_opts)(struct squashfs_sb_info *, void *, int);
 	void	(*free)(void *);
-	int	(*decompress)(struct squashfs_sb_info *, void **,
+	int	(*decompress)(struct squashfs_sb_info *, void *,
-		struct buffer_head **, int, int, int, int, int);
+		struct buffer_head **, int, int, int,
+		struct squashfs_page_actor *);
 	int	id;
 	char	*name;
 	int	supported;
 };
 
-static inline void squashfs_decompressor_free(struct squashfs_sb_info *msblk,
+static inline void *squashfs_comp_opts(struct squashfs_sb_info *msblk,
-	void *s)
+							void *buff, int length)
 {
-	if (msblk->decompressor)
+	return msblk->decompressor->comp_opts ?
-		msblk->decompressor->free(s);
+		msblk->decompressor->comp_opts(msblk, buff, length) : NULL;
-}
-
-static inline int squashfs_decompress(struct squashfs_sb_info *msblk,
-	void **buffer, struct buffer_head **bh, int b, int offset, int length,
-	int srclength, int pages)
-{
-	return msblk->decompressor->decompress(msblk, buffer, bh, b, offset,
-		length, srclength, pages);
 }
 
 #ifdef CONFIG_SQUASHFS_XZ

fs/squashfs/decompressor_multi.c

@@ -0,0 +1,198 @@
+/*
+ *  Copyright (c) 2013
+ *  Minchan Kim <minchan@kernel.org>
+ *
+ *  This work is licensed under the terms of the GNU GPL, version 2. See
+ *  the COPYING file in the top-level directory.
+ */
+#include <linux/types.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/buffer_head.h>
+#include <linux/sched.h>
+#include <linux/wait.h>
+#include <linux/cpumask.h>
+
+#include "squashfs_fs.h"
+#include "squashfs_fs_sb.h"
+#include "decompressor.h"
+#include "squashfs.h"
+
+/*
+ * This file implements multi-threaded decompression in the
+ * decompressor framework
+ */
+
+
+/*
+ * The reason that multiply two is that a CPU can request new I/O
+ * while it is waiting previous request.
+ */
+#define MAX_DECOMPRESSOR	(num_online_cpus() * 2)
+
+
+int squashfs_max_decompressors(void)
+{
+	return MAX_DECOMPRESSOR;
+}
+
+
+struct squashfs_stream {
+	void			*comp_opts;
+	struct list_head	strm_list;
+	struct mutex		mutex;
+	int			avail_decomp;
+	wait_queue_head_t	wait;
+};
+
+
+struct decomp_stream {
+	void *stream;
+	struct list_head list;
+};
+
+
+static void put_decomp_stream(struct decomp_stream *decomp_strm,
+				struct squashfs_stream *stream)
+{
+	mutex_lock(&stream->mutex);
+	list_add(&decomp_strm->list, &stream->strm_list);
+	mutex_unlock(&stream->mutex);
+	wake_up(&stream->wait);
+}
+
+void *squashfs_decompressor_create(struct squashfs_sb_info *msblk,
+				void *comp_opts)
+{
+	struct squashfs_stream *stream;
+	struct decomp_stream *decomp_strm = NULL;
+	int err = -ENOMEM;
+
+	stream = kzalloc(sizeof(*stream), GFP_KERNEL);
+	if (!stream)
+		goto out;
+
+	stream->comp_opts = comp_opts;
+	mutex_init(&stream->mutex);
+	INIT_LIST_HEAD(&stream->strm_list);
+	init_waitqueue_head(&stream->wait);
+
+	/*
+	 * We should have a decompressor at least as default
+	 * so if we fail to allocate new decompressor dynamically,
+	 * we could always fall back to default decompressor and
+	 * file system works.
+	 */
+	decomp_strm = kmalloc(sizeof(*decomp_strm), GFP_KERNEL);
+	if (!decomp_strm)
+		goto out;
+
+	decomp_strm->stream = msblk->decompressor->init(msblk,
+						stream->comp_opts);
+	if (IS_ERR(decomp_strm->stream)) {
+		err = PTR_ERR(decomp_strm->stream);
+		goto out;
+	}
+
+	list_add(&decomp_strm->list, &stream->strm_list);
+	stream->avail_decomp = 1;
+	return stream;
+
+out:
+	kfree(decomp_strm);
+	kfree(stream);
+	return ERR_PTR(err);
+}
+
+
+void squashfs_decompressor_destroy(struct squashfs_sb_info *msblk)
+{
+	struct squashfs_stream *stream = msblk->stream;
+	if (stream) {
+		struct decomp_stream *decomp_strm;
+
+		while (!list_empty(&stream->strm_list)) {
+			decomp_strm = list_entry(stream->strm_list.prev,
+						struct decomp_stream, list);
+			list_del(&decomp_strm->list);
+			msblk->decompressor->free(decomp_strm->stream);
+			kfree(decomp_strm);
+			stream->avail_decomp--;
+		}
+		WARN_ON(stream->avail_decomp);
+		kfree(stream->comp_opts);
+		kfree(stream);
+	}
+}
+
+
+static struct decomp_stream *get_decomp_stream(struct squashfs_sb_info *msblk,
+					struct squashfs_stream *stream)
+{
+	struct decomp_stream *decomp_strm;
+
+	while (1) {
+		mutex_lock(&stream->mutex);
+
+		/* There is available decomp_stream */
+		if (!list_empty(&stream->strm_list)) {
+			decomp_strm = list_entry(stream->strm_list.prev,
+				struct decomp_stream, list);
+			list_del(&decomp_strm->list);
+			mutex_unlock(&stream->mutex);
+			break;
+		}
+
+		/*
+		 * If there is no available decomp and already full,
+		 * let's wait for releasing decomp from other users.
+		 */
+		if (stream->avail_decomp >= MAX_DECOMPRESSOR)
+			goto wait;
+
+		/* Let's allocate new decomp */
+		decomp_strm = kmalloc(sizeof(*decomp_strm), GFP_KERNEL);
+		if (!decomp_strm)
+			goto wait;
+
+		decomp_strm->stream = msblk->decompressor->init(msblk,
+						stream->comp_opts);
+		if (IS_ERR(decomp_strm->stream)) {
+			kfree(decomp_strm);
+			goto wait;
+		}
+
+		stream->avail_decomp++;
+		WARN_ON(stream->avail_decomp > MAX_DECOMPRESSOR);
+
+		mutex_unlock(&stream->mutex);
+		break;
+wait:
+		/*
+		 * If system memory is tough, let's for other's
+		 * releasing instead of hurting VM because it could
+		 * make page cache thrashing.
+		 */
+		mutex_unlock(&stream->mutex);
+		wait_event(stream->wait,
+			!list_empty(&stream->strm_list));
+	}
+
+	return decomp_strm;
+}
+
+
+int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh,
+	int b, int offset, int length, struct squashfs_page_actor *output)
+{
+	int res;
+	struct squashfs_stream *stream = msblk->stream;
+	struct decomp_stream *decomp_stream = get_decomp_stream(msblk, stream);
+	res = msblk->decompressor->decompress(msblk, decomp_stream->stream,
+		bh, b, offset, length, output);
+	put_decomp_stream(decomp_stream, stream);
+	if (res < 0)
+		ERROR("%s decompression failed, data probably corrupt\n",
+			msblk->decompressor->name);
+	return res;
+}

fs/squashfs/decompressor_multi_percpu.c

@@ -0,0 +1,97 @@
+/*
+ * Copyright (c) 2013
+ * Phillip Lougher <phillip@squashfs.org.uk>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/percpu.h>
+#include <linux/buffer_head.h>
+
+#include "squashfs_fs.h"
+#include "squashfs_fs_sb.h"
+#include "decompressor.h"
+#include "squashfs.h"
+
+/*
+ * This file implements multi-threaded decompression using percpu
+ * variables, one thread per cpu core.
+ */
+
+struct squashfs_stream {
+	void		*stream;
+};
+
+void *squashfs_decompressor_create(struct squashfs_sb_info *msblk,
+						void *comp_opts)
+{
+	struct squashfs_stream *stream;
+	struct squashfs_stream __percpu *percpu;
+	int err, cpu;
+
+	percpu = alloc_percpu(struct squashfs_stream);
+	if (percpu == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	for_each_possible_cpu(cpu) {
+		stream = per_cpu_ptr(percpu, cpu);
+		stream->stream = msblk->decompressor->init(msblk, comp_opts);
+		if (IS_ERR(stream->stream)) {
+			err = PTR_ERR(stream->stream);
+			goto out;
+		}
+	}
+
+	kfree(comp_opts);
+	return (__force void *) percpu;
+
+out:
+	for_each_possible_cpu(cpu) {
+		stream = per_cpu_ptr(percpu, cpu);
+		if (!IS_ERR_OR_NULL(stream->stream))
+			msblk->decompressor->free(stream->stream);
+	}
+	free_percpu(percpu);
+	return ERR_PTR(err);
+}
+
+void squashfs_decompressor_destroy(struct squashfs_sb_info *msblk)
+{
+	struct squashfs_stream __percpu *percpu =
+			(struct squashfs_stream __percpu *) msblk->stream;
+	struct squashfs_stream *stream;
+	int cpu;
+
+	if (msblk->stream) {
+		for_each_possible_cpu(cpu) {
+			stream = per_cpu_ptr(percpu, cpu);
+			msblk->decompressor->free(stream->stream);
+		}
+		free_percpu(percpu);
+	}
+}
+
+int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh,
+	int b, int offset, int length, struct squashfs_page_actor *output)
+{
+	struct squashfs_stream __percpu *percpu =
+			(struct squashfs_stream __percpu *) msblk->stream;
+	struct squashfs_stream *stream = get_cpu_ptr(percpu);
+	int res = msblk->decompressor->decompress(msblk, stream->stream, bh, b,
+		offset, length, output);
+	put_cpu_ptr(stream);
+
+	if (res < 0)
+		ERROR("%s decompression failed, data probably corrupt\n",
+			msblk->decompressor->name);
+
+	return res;
+}
+
+int squashfs_max_decompressors(void)
+{
+	return num_possible_cpus();
+}

fs/squashfs/decompressor_single.c

@@ -0,0 +1,85 @@
+/*
+ * Copyright (c) 2013
+ * Phillip Lougher <phillip@squashfs.org.uk>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/types.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/buffer_head.h>
+
+#include "squashfs_fs.h"
+#include "squashfs_fs_sb.h"
+#include "decompressor.h"
+#include "squashfs.h"
+
+/*
+ * This file implements single-threaded decompression in the
+ * decompressor framework
+ */
+
+struct squashfs_stream {
+	void		*stream;
+	struct mutex	mutex;
+};
+
+void *squashfs_decompressor_create(struct squashfs_sb_info *msblk,
+						void *comp_opts)
+{
+	struct squashfs_stream *stream;
+	int err = -ENOMEM;
+
+	stream = kmalloc(sizeof(*stream), GFP_KERNEL);
+	if (stream == NULL)
+		goto out;
+
+	stream->stream = msblk->decompressor->init(msblk, comp_opts);
+	if (IS_ERR(stream->stream)) {
+		err = PTR_ERR(stream->stream);
+		goto out;
+	}
+
+	kfree(comp_opts);
+	mutex_init(&stream->mutex);
+	return stream;
+
+out:
+	kfree(stream);
+	return ERR_PTR(err);
+}
+
+void squashfs_decompressor_destroy(struct squashfs_sb_info *msblk)
+{
+	struct squashfs_stream *stream = msblk->stream;
+
+	if (stream) {
+		msblk->decompressor->free(stream->stream);
+		kfree(stream);
+	}
+}
+
+int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh,
+	int b, int offset, int length, struct squashfs_page_actor *output)
+{
+	int res;
+	struct squashfs_stream *stream = msblk->stream;
+
+	mutex_lock(&stream->mutex);
+	res = msblk->decompressor->decompress(msblk, stream->stream, bh, b,
+		offset, length, output);
+	mutex_unlock(&stream->mutex);
+
+	if (res < 0)
+		ERROR("%s decompression failed, data probably corrupt\n",
+			msblk->decompressor->name);
+
+	return res;
+}
+
+int squashfs_max_decompressors(void)
+{
+	return 1;
+}

fs/squashfs/file.c

@@ -370,77 +370,15 @@ static int read_blocklist(struct inode *inode, int index, u64 *block)
 	return le32_to_cpu(size);
 }
 
-
+/* Copy data into page cache  */
-static int squashfs_readpage(struct file *file, struct page *page)
+void squashfs_copy_cache(struct page *page, struct squashfs_cache_entry *buffer,
+	int bytes, int offset)
 {
 	struct inode *inode = page->mapping->host;
 	struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info;
-	int bytes, i, offset = 0, sparse = 0;
-	struct squashfs_cache_entry *buffer = NULL;
 	void *pageaddr;
-
+	int i, mask = (1 << (msblk->block_log - PAGE_CACHE_SHIFT)) - 1;
-	int mask = (1 << (msblk->block_log - PAGE_CACHE_SHIFT)) - 1;
+	int start_index = page->index & ~mask, end_index = start_index | mask;
-	int index = page->index >> (msblk->block_log - PAGE_CACHE_SHIFT);
-	int start_index = page->index & ~mask;
-	int end_index = start_index | mask;
-	int file_end = i_size_read(inode) >> msblk->block_log;
-
-	TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
-				page->index, squashfs_i(inode)->start);
-
-	if (page->index >= ((i_size_read(inode) + PAGE_CACHE_SIZE - 1) >>
-					PAGE_CACHE_SHIFT))
-		goto out;
-
-	if (index < file_end || squashfs_i(inode)->fragment_block ==
-					SQUASHFS_INVALID_BLK) {
-		/*
-		 * Reading a datablock from disk.  Need to read block list
-		 * to get location and block size.
-		 */
-		u64 block = 0;
-		int bsize = read_blocklist(inode, index, &block);
-		if (bsize < 0)
-			goto error_out;
-
-		if (bsize == 0) { /* hole */
-			bytes = index == file_end ?
-				(i_size_read(inode) & (msblk->block_size - 1)) :
-				 msblk->block_size;
-			sparse = 1;
-		} else {
-			/*
-			 * Read and decompress datablock.
-			 */
-			buffer = squashfs_get_datablock(inode->i_sb,
-								block, bsize);
-			if (buffer->error) {
-				ERROR("Unable to read page, block %llx, size %x"
-					"\n", block, bsize);
-				squashfs_cache_put(buffer);
-				goto error_out;
-			}
-			bytes = buffer->length;
-		}
-	} else {
-		/*
-		 * Datablock is stored inside a fragment (tail-end packed
-		 * block).
-		 */
-		buffer = squashfs_get_fragment(inode->i_sb,
-				squashfs_i(inode)->fragment_block,
-				squashfs_i(inode)->fragment_size);
-
-		if (buffer->error) {
-			ERROR("Unable to read page, block %llx, size %x\n",
-				squashfs_i(inode)->fragment_block,
-				squashfs_i(inode)->fragment_size);
-			squashfs_cache_put(buffer);
-			goto error_out;
-		}
-		bytes = i_size_read(inode) & (msblk->block_size - 1);
-		offset = squashfs_i(inode)->fragment_offset;
-	}
 
 	/*
 	 * Loop copying datablock into pages.  As the datablock likely covers
@@ -451,7 +389,7 @@ static int squashfs_readpage(struct file *file, struct page *page)
 	for (i = start_index; i <= end_index && bytes > 0; i++,
 			bytes -= PAGE_CACHE_SIZE, offset += PAGE_CACHE_SIZE) {
 		struct page *push_page;
-		int avail = sparse ? 0 : min_t(int, bytes, PAGE_CACHE_SIZE);
+		int avail = buffer ? min_t(int, bytes, PAGE_CACHE_SIZE) : 0;
 
 		TRACE("bytes %d, i %d, available_bytes %d\n", bytes, i, avail);
 
@@ -475,11 +413,75 @@ skip_page:
 		if (i != page->index)
 			page_cache_release(push_page);
 	}
+}
 
-	if (!sparse)
+/* Read datablock stored packed inside a fragment (tail-end packed block) */
-		squashfs_cache_put(buffer);
+static int squashfs_readpage_fragment(struct page *page)
+{
+	struct inode *inode = page->mapping->host;
+	struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info;
+	struct squashfs_cache_entry *buffer = squashfs_get_fragment(inode->i_sb,
+		squashfs_i(inode)->fragment_block,
+		squashfs_i(inode)->fragment_size);
+	int res = buffer->error;
 
+	if (res)
+		ERROR("Unable to read page, block %llx, size %x\n",
+			squashfs_i(inode)->fragment_block,
+			squashfs_i(inode)->fragment_size);
+	else
+		squashfs_copy_cache(page, buffer, i_size_read(inode) &
+			(msblk->block_size - 1),
+			squashfs_i(inode)->fragment_offset);
+
+	squashfs_cache_put(buffer);
+	return res;
+}
+
+static int squashfs_readpage_sparse(struct page *page, int index, int file_end)
+{
+	struct inode *inode = page->mapping->host;
+	struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info;
+	int bytes = index == file_end ?
+			(i_size_read(inode) & (msblk->block_size - 1)) :
+			 msblk->block_size;
+
+	squashfs_copy_cache(page, NULL, bytes, 0);
 	return 0;
+}
+
+static int squashfs_readpage(struct file *file, struct page *page)
+{
+	struct inode *inode = page->mapping->host;
+	struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info;
+	int index = page->index >> (msblk->block_log - PAGE_CACHE_SHIFT);
+	int file_end = i_size_read(inode) >> msblk->block_log;
+	int res;
+	void *pageaddr;
+
+	TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
+				page->index, squashfs_i(inode)->start);
+
+	if (page->index >= ((i_size_read(inode) + PAGE_CACHE_SIZE - 1) >>
+					PAGE_CACHE_SHIFT))
+		goto out;
+
+	if (index < file_end || squashfs_i(inode)->fragment_block ==
+					SQUASHFS_INVALID_BLK) {
+		u64 block = 0;
+		int bsize = read_blocklist(inode, index, &block);
+		if (bsize < 0)
+			goto error_out;
+
+		if (bsize == 0)
+			res = squashfs_readpage_sparse(page, index, file_end);
+		else
+			res = squashfs_readpage_block(page, block, bsize);
+	} else
+		res = squashfs_readpage_fragment(page);
+
+	if (!res)
+		return 0;
 
 error_out:
 	SetPageError(page);

fs/squashfs/file_cache.c

@@ -0,0 +1,38 @@
+/*
+ * Copyright (c) 2013
+ * Phillip Lougher <phillip@squashfs.org.uk>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/fs.h>
+#include <linux/vfs.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/pagemap.h>
+#include <linux/mutex.h>
+
+#include "squashfs_fs.h"
+#include "squashfs_fs_sb.h"
+#include "squashfs_fs_i.h"
+#include "squashfs.h"
+
+/* Read separately compressed datablock and memcopy into page cache */
+int squashfs_readpage_block(struct page *page, u64 block, int bsize)
+{
+	struct inode *i = page->mapping->host;
+	struct squashfs_cache_entry *buffer = squashfs_get_datablock(i->i_sb,
+		block, bsize);
+	int res = buffer->error;
+
+	if (res)
+		ERROR("Unable to read page, block %llx, size %x\n", block,
+			bsize);
+	else
+		squashfs_copy_cache(page, buffer, buffer->length, 0);
+
+	squashfs_cache_put(buffer);
+	return res;
+}

fs/squashfs/file_direct.c

@@ -0,0 +1,173 @@
+/*
+ * Copyright (c) 2013
+ * Phillip Lougher <phillip@squashfs.org.uk>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/fs.h>
+#include <linux/vfs.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/pagemap.h>
+#include <linux/mutex.h>
+
+#include "squashfs_fs.h"
+#include "squashfs_fs_sb.h"
+#include "squashfs_fs_i.h"
+#include "squashfs.h"
+#include "page_actor.h"
+
+static int squashfs_read_cache(struct page *target_page, u64 block, int bsize,
+	int pages, struct page **page);
+
+/* Read separately compressed datablock directly into page cache */
+int squashfs_readpage_block(struct page *target_page, u64 block, int bsize)
+
+{
+	struct inode *inode = target_page->mapping->host;
+	struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info;
+
+	int file_end = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT;
+	int mask = (1 << (msblk->block_log - PAGE_CACHE_SHIFT)) - 1;
+	int start_index = target_page->index & ~mask;
+	int end_index = start_index | mask;
+	int i, n, pages, missing_pages, bytes, res = -ENOMEM;
+	struct page **page;
+	struct squashfs_page_actor *actor;
+	void *pageaddr;
+
+	if (end_index > file_end)
+		end_index = file_end;
+
+	pages = end_index - start_index + 1;
+
+	page = kmalloc(sizeof(void *) * pages, GFP_KERNEL);
+	if (page == NULL)
+		return res;
+
+	/*
+	 * Create a "page actor" which will kmap and kunmap the
+	 * page cache pages appropriately within the decompressor
+	 */
+	actor = squashfs_page_actor_init_special(page, pages, 0);
+	if (actor == NULL)
+		goto out;
+
+	/* Try to grab all the pages covered by the Squashfs block */
+	for (missing_pages = 0, i = 0, n = start_index; i < pages; i++, n++) {
+		page[i] = (n == target_page->index) ? target_page :
+			grab_cache_page_nowait(target_page->mapping, n);
+
+		if (page[i] == NULL) {
+			missing_pages++;
+			continue;
+		}
+
+		if (PageUptodate(page[i])) {
+			unlock_page(page[i]);
+			page_cache_release(page[i]);
+			page[i] = NULL;
+			missing_pages++;
+		}
+	}
+
+	if (missing_pages) {
+		/*
+		 * Couldn't get one or more pages, this page has either
+		 * been VM reclaimed, but others are still in the page cache
+		 * and uptodate, or we're racing with another thread in
+		 * squashfs_readpage also trying to grab them.  Fall back to
+		 * using an intermediate buffer.
+		 */
+		res = squashfs_read_cache(target_page, block, bsize, pages,
+								page);
+		goto out;
+	}
+
+	/* Decompress directly into the page cache buffers */
+	res = squashfs_read_data(inode->i_sb, block, bsize, NULL, actor);
+	if (res < 0)
+		goto mark_errored;
+
+	/* Last page may have trailing bytes not filled */
+	bytes = res % PAGE_CACHE_SIZE;
+	if (bytes) {
+		pageaddr = kmap_atomic(page[pages - 1]);
+		memset(pageaddr + bytes, 0, PAGE_CACHE_SIZE - bytes);
+		kunmap_atomic(pageaddr);
+	}
+
+	/* Mark pages as uptodate, unlock and release */
+	for (i = 0; i < pages; i++) {
+		flush_dcache_page(page[i]);
+		SetPageUptodate(page[i]);
+		unlock_page(page[i]);
+		if (page[i] != target_page)
+			page_cache_release(page[i]);
+	}
+
+	kfree(actor);
+	kfree(page);
+
+	return 0;
+
+mark_errored:
+	/* Decompression failed, mark pages as errored.  Target_page is
+	 * dealt with by the caller
+	 */
+	for (i = 0; i < pages; i++) {
+		if (page[i] == target_page)
+			continue;
+		flush_dcache_page(page[i]);
+		SetPageError(page[i]);
+		unlock_page(page[i]);
+		page_cache_release(page[i]);
+	}
+
+out:
+	kfree(actor);
+	kfree(page);
+	return res;
+}
+
+
+static int squashfs_read_cache(struct page *target_page, u64 block, int bsize,
+	int pages, struct page **page)
+{
+	struct inode *i = target_page->mapping->host;
+	struct squashfs_cache_entry *buffer = squashfs_get_datablock(i->i_sb,
+						 block, bsize);
+	int bytes = buffer->length, res = buffer->error, n, offset = 0;
+	void *pageaddr;
+
+	if (res) {
+		ERROR("Unable to read page, block %llx, size %x\n", block,
+			bsize);
+		goto out;
+	}
+
+	for (n = 0; n < pages && bytes > 0; n++,
+			bytes -= PAGE_CACHE_SIZE, offset += PAGE_CACHE_SIZE) {
+		int avail = min_t(int, bytes, PAGE_CACHE_SIZE);
+
+		if (page[n] == NULL)
+			continue;
+
+		pageaddr = kmap_atomic(page[n]);
+		squashfs_copy_data(pageaddr, buffer, offset, avail);
+		memset(pageaddr + avail, 0, PAGE_CACHE_SIZE - avail);
+		kunmap_atomic(pageaddr);
+		flush_dcache_page(page[n]);
+		SetPageUptodate(page[n]);
+		unlock_page(page[n]);
+		if (page[n] != target_page)
+			page_cache_release(page[n]);
+	}
+
+out:
+	squashfs_cache_put(buffer);
+	return res;
+}

fs/squashfs/lzo_wrapper.c

@@ -31,13 +31,14 @@
 #include "squashfs_fs_sb.h"
 #include "squashfs.h"
 #include "decompressor.h"
+#include "page_actor.h"
 
 struct squashfs_lzo {
 	void	*input;
 	void	*output;
 };
 
-static void *lzo_init(struct squashfs_sb_info *msblk, void *buff, int len)
+static void *lzo_init(struct squashfs_sb_info *msblk, void *buff)
 {
 	int block_size = max_t(int, msblk->block_size, SQUASHFS_METADATA_SIZE);
 
@@ -74,22 +75,16 @@ static void lzo_free(void *strm)
 }
 
 
-static int lzo_uncompress(struct squashfs_sb_info *msblk, void **buffer,
+static int lzo_uncompress(struct squashfs_sb_info *msblk, void *strm,
-	struct buffer_head **bh, int b, int offset, int length, int srclength,
+	struct buffer_head **bh, int b, int offset, int length,
-	int pages)
+	struct squashfs_page_actor *output)
 {
-	struct squashfs_lzo *stream = msblk->stream;
+	struct squashfs_lzo *stream = strm;
-	void *buff = stream->input;
+	void *buff = stream->input, *data;
 	int avail, i, bytes = length, res;
-	size_t out_len = srclength;
+	size_t out_len = output->length;
-
-	mutex_lock(&msblk->read_data_mutex);
 
 	for (i = 0; i < b; i++) {
-		wait_on_buffer(bh[i]);
-		if (!buffer_uptodate(bh[i]))
-			goto block_release;
-
 		avail = min(bytes, msblk->devblksize - offset);
 		memcpy(buff, bh[i]->b_data + offset, avail);
 		buff += avail;
@@ -104,24 +99,24 @@ static int lzo_uncompress(struct squashfs_sb_info *msblk, void **buffer,
 		goto failed;
 
 	res = bytes = (int)out_len;
-	for (i = 0, buff = stream->output; bytes && i < pages; i++) {
+	data = squashfs_first_page(output);
-		avail = min_t(int, bytes, PAGE_CACHE_SIZE);
+	buff = stream->output;
-		memcpy(buffer[i], buff, avail);
+	while (data) {
-		buff += avail;
+		if (bytes <= PAGE_CACHE_SIZE) {
-		bytes -= avail;
+			memcpy(data, buff, bytes);
+			break;
+		} else {
+			memcpy(data, buff, PAGE_CACHE_SIZE);
+			buff += PAGE_CACHE_SIZE;
+			bytes -= PAGE_CACHE_SIZE;
+			data = squashfs_next_page(output);
+		}
 	}
+	squashfs_finish_page(output);
 
-	mutex_unlock(&msblk->read_data_mutex);
 	return res;
 
-block_release:
-	for (; i < b; i++)
-		put_bh(bh[i]);
-
 failed:
-	mutex_unlock(&msblk->read_data_mutex);
-
-	ERROR("lzo decompression failed, data probably corrupt\n");
 	return -EIO;
 }
 

fs/squashfs/page_actor.c

@@ -0,0 +1,100 @@
+/*
+ * Copyright (c) 2013
+ * Phillip Lougher <phillip@squashfs.org.uk>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/pagemap.h>
+#include "page_actor.h"
+
+/*
+ * This file contains implementations of page_actor for decompressing into
+ * an intermediate buffer, and for decompressing directly into the
+ * page cache.
+ *
+ * Calling code should avoid sleeping between calls to squashfs_first_page()
+ * and squashfs_finish_page().
+ */
+
+/* Implementation of page_actor for decompressing into intermediate buffer */
+static void *cache_first_page(struct squashfs_page_actor *actor)
+{
+	actor->next_page = 1;
+	return actor->buffer[0];
+}
+
+static void *cache_next_page(struct squashfs_page_actor *actor)
+{
+	if (actor->next_page == actor->pages)
+		return NULL;
+
+	return actor->buffer[actor->next_page++];
+}
+
+static void cache_finish_page(struct squashfs_page_actor *actor)
+{
+	/* empty */
+}
+
+struct squashfs_page_actor *squashfs_page_actor_init(void **buffer,
+	int pages, int length)
+{
+	struct squashfs_page_actor *actor = kmalloc(sizeof(*actor), GFP_KERNEL);
+
+	if (actor == NULL)
+		return NULL;
+
+	actor->length = length ? : pages * PAGE_CACHE_SIZE;
+	actor->buffer = buffer;
+	actor->pages = pages;
+	actor->next_page = 0;
+	actor->squashfs_first_page = cache_first_page;
+	actor->squashfs_next_page = cache_next_page;
+	actor->squashfs_finish_page = cache_finish_page;
+	return actor;
+}
+
+/* Implementation of page_actor for decompressing directly into page cache. */
+static void *direct_first_page(struct squashfs_page_actor *actor)
+{
+	actor->next_page = 1;
+	return actor->pageaddr = kmap_atomic(actor->page[0]);
+}
+
+static void *direct_next_page(struct squashfs_page_actor *actor)
+{
+	if (actor->pageaddr)
+		kunmap_atomic(actor->pageaddr);
+
+	return actor->pageaddr = actor->next_page == actor->pages ? NULL :
+		kmap_atomic(actor->page[actor->next_page++]);
+}
+
+static void direct_finish_page(struct squashfs_page_actor *actor)
+{
+	if (actor->pageaddr)
+		kunmap_atomic(actor->pageaddr);
+}
+
+struct squashfs_page_actor *squashfs_page_actor_init_special(struct page **page,
+	int pages, int length)
+{
+	struct squashfs_page_actor *actor = kmalloc(sizeof(*actor), GFP_KERNEL);
+
+	if (actor == NULL)
+		return NULL;
+
+	actor->length = length ? : pages * PAGE_CACHE_SIZE;
+	actor->page = page;
+	actor->pages = pages;
+	actor->next_page = 0;
+	actor->pageaddr = NULL;
+	actor->squashfs_first_page = direct_first_page;
+	actor->squashfs_next_page = direct_next_page;
+	actor->squashfs_finish_page = direct_finish_page;
+	return actor;
+}

fs/squashfs/page_actor.h

@@ -0,0 +1,81 @@
+#ifndef PAGE_ACTOR_H
+#define PAGE_ACTOR_H
+/*
+ * Copyright (c) 2013
+ * Phillip Lougher <phillip@squashfs.org.uk>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ */
+
+#ifndef CONFIG_SQUASHFS_FILE_DIRECT
+struct squashfs_page_actor {
+	void	**page;
+	int	pages;
+	int	length;
+	int	next_page;
+};
+
+static inline struct squashfs_page_actor *squashfs_page_actor_init(void **page,
+	int pages, int length)
+{
+	struct squashfs_page_actor *actor = kmalloc(sizeof(*actor), GFP_KERNEL);
+
+	if (actor == NULL)
+		return NULL;
+
+	actor->length = length ? : pages * PAGE_CACHE_SIZE;
+	actor->page = page;
+	actor->pages = pages;
+	actor->next_page = 0;
+	return actor;
+}
+
+static inline void *squashfs_first_page(struct squashfs_page_actor *actor)
+{
+	actor->next_page = 1;
+	return actor->page[0];
+}
+
+static inline void *squashfs_next_page(struct squashfs_page_actor *actor)
+{
+	return actor->next_page == actor->pages ? NULL :
+		actor->page[actor->next_page++];
+}
+
+static inline void squashfs_finish_page(struct squashfs_page_actor *actor)
+{
+	/* empty */
+}
+#else
+struct squashfs_page_actor {
+	union {
+		void		**buffer;
+		struct page	**page;
+	};
+	void	*pageaddr;
+	void    *(*squashfs_first_page)(struct squashfs_page_actor *);
+	void    *(*squashfs_next_page)(struct squashfs_page_actor *);
+	void    (*squashfs_finish_page)(struct squashfs_page_actor *);
+	int	pages;
+	int	length;
+	int	next_page;
+};
+
+extern struct squashfs_page_actor *squashfs_page_actor_init(void **, int, int);
+extern struct squashfs_page_actor *squashfs_page_actor_init_special(struct page
+							 **, int, int);
+static inline void *squashfs_first_page(struct squashfs_page_actor *actor)
+{
+	return actor->squashfs_first_page(actor);
+}
+static inline void *squashfs_next_page(struct squashfs_page_actor *actor)
+{
+	return actor->squashfs_next_page(actor);
+}
+static inline void squashfs_finish_page(struct squashfs_page_actor *actor)
+{
+	actor->squashfs_finish_page(actor);
+}
+#endif
+#endif

fs/squashfs/squashfs.h

@@ -28,8 +28,8 @@
 #define WARNING(s, args...)	pr_warning("SQUASHFS: "s, ## args)
 
 /* block.c */
-extern int squashfs_read_data(struct super_block *, void **, u64, int, u64 *,
+extern int squashfs_read_data(struct super_block *, u64, int, u64 *,
-				int, int);
+				struct squashfs_page_actor *);
 
 /* cache.c */
 extern struct squashfs_cache *squashfs_cache_init(char *, int, int);
@@ -48,7 +48,14 @@ extern void *squashfs_read_table(struct super_block *, u64, int);
 
 /* decompressor.c */
 extern const struct squashfs_decompressor *squashfs_lookup_decompressor(int);
-extern void *squashfs_decompressor_init(struct super_block *, unsigned short);
+extern void *squashfs_decompressor_setup(struct super_block *, unsigned short);
+
+/* decompressor_xxx.c */
+extern void *squashfs_decompressor_create(struct squashfs_sb_info *, void *);
+extern void squashfs_decompressor_destroy(struct squashfs_sb_info *);
+extern int squashfs_decompress(struct squashfs_sb_info *, struct buffer_head **,
+	int, int, int, struct squashfs_page_actor *);
+extern int squashfs_max_decompressors(void);
 
 /* export.c */
 extern __le64 *squashfs_read_inode_lookup_table(struct super_block *, u64, u64,
@@ -59,6 +66,13 @@ extern int squashfs_frag_lookup(struct super_block *, unsigned int, u64 *);
 extern __le64 *squashfs_read_fragment_index_table(struct super_block *,
 				u64, u64, unsigned int);
 
+/* file.c */
+void squashfs_copy_cache(struct page *, struct squashfs_cache_entry *, int,
+				int);
+
+/* file_xxx.c */
+extern int squashfs_readpage_block(struct page *, u64, int);
+
 /* id.c */
 extern int squashfs_get_id(struct super_block *, unsigned int, unsigned int *);
 extern __le64 *squashfs_read_id_index_table(struct super_block *, u64, u64,

fs/squashfs/squashfs_fs_sb.h

@@ -50,6 +50,7 @@ struct squashfs_cache_entry {
 	wait_queue_head_t	wait_queue;
 	struct squashfs_cache	*cache;
 	void			**data;
+	struct squashfs_page_actor	*actor;
 };
 
 struct squashfs_sb_info {
@@ -63,10 +64,9 @@ struct squashfs_sb_info {
 	__le64					*id_table;
 	__le64					*fragment_index;
 	__le64					*xattr_id_table;
-	struct mutex				read_data_mutex;
 	struct mutex				meta_index_mutex;
 	struct meta_index			*meta_index;
-	void					*stream;
+	struct squashfs_stream			*stream;
 	__le64					*inode_lookup_table;
 	u64					inode_table;
 	u64					directory_table;

fs/squashfs/super.c

@@ -98,7 +98,6 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent)
 	msblk->devblksize = sb_min_blocksize(sb, SQUASHFS_DEVBLK_SIZE);
 	msblk->devblksize_log2 = ffz(~msblk->devblksize);
 
-	mutex_init(&msblk->read_data_mutex);
 	mutex_init(&msblk->meta_index_mutex);
 
 	/*
@@ -206,13 +205,14 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent)
 		goto failed_mount;
 
 	/* Allocate read_page block */
-	msblk->read_page = squashfs_cache_init("data", 1, msblk->block_size);
+	msblk->read_page = squashfs_cache_init("data",
+		squashfs_max_decompressors(), msblk->block_size);
 	if (msblk->read_page == NULL) {
 		ERROR("Failed to allocate read_page block\n");
 		goto failed_mount;
 	}
 
-	msblk->stream = squashfs_decompressor_init(sb, flags);
+	msblk->stream = squashfs_decompressor_setup(sb, flags);
 	if (IS_ERR(msblk->stream)) {
 		err = PTR_ERR(msblk->stream);
 		msblk->stream = NULL;
@@ -336,7 +336,7 @@ failed_mount:
 	squashfs_cache_delete(msblk->block_cache);
 	squashfs_cache_delete(msblk->fragment_cache);
 	squashfs_cache_delete(msblk->read_page);
-	squashfs_decompressor_free(msblk, msblk->stream);
+	squashfs_decompressor_destroy(msblk);
 	kfree(msblk->inode_lookup_table);
 	kfree(msblk->fragment_index);
 	kfree(msblk->id_table);
@@ -383,7 +383,7 @@ static void squashfs_put_super(struct super_block *sb)
 		squashfs_cache_delete(sbi->block_cache);
 		squashfs_cache_delete(sbi->fragment_cache);
 		squashfs_cache_delete(sbi->read_page);
-		squashfs_decompressor_free(sbi, sbi->stream);
+		squashfs_decompressor_destroy(sbi);
 		kfree(sbi->id_table);
 		kfree(sbi->fragment_index);
 		kfree(sbi->meta_index);

fs/squashfs/xz_wrapper.c

@@ -32,44 +32,70 @@
 #include "squashfs_fs_sb.h"
 #include "squashfs.h"
 #include "decompressor.h"
+#include "page_actor.h"
 
 struct squashfs_xz {
 	struct xz_dec *state;
 	struct xz_buf buf;
 };
 
-struct comp_opts {
+struct disk_comp_opts {
 	__le32 dictionary_size;
 	__le32 flags;
 };
 
-static void *squashfs_xz_init(struct squashfs_sb_info *msblk, void *buff,
+struct comp_opts {
-	int len)
+	int dict_size;
+};
+
+static void *squashfs_xz_comp_opts(struct squashfs_sb_info *msblk,
+	void *buff, int len)
 {
-	struct comp_opts *comp_opts = buff;
+	struct disk_comp_opts *comp_opts = buff;
-	struct squashfs_xz *stream;
+	struct comp_opts *opts;
-	int dict_size = msblk->block_size;
+	int err = 0, n;
-	int err, n;
+
+	opts = kmalloc(sizeof(*opts), GFP_KERNEL);
+	if (opts == NULL) {
+		err = -ENOMEM;
+		goto out2;
+	}
 
 	if (comp_opts) {
 		/* check compressor options are the expected length */
 		if (len < sizeof(*comp_opts)) {
 			err = -EIO;
-			goto failed;
+			goto out;
 		}
 
-		dict_size = le32_to_cpu(comp_opts->dictionary_size);
+		opts->dict_size = le32_to_cpu(comp_opts->dictionary_size);
 
 		/* the dictionary size should be 2^n or 2^n+2^(n+1) */
-		n = ffs(dict_size) - 1;
+		n = ffs(opts->dict_size) - 1;
-		if (dict_size != (1 << n) && dict_size != (1 << n) +
+		if (opts->dict_size != (1 << n) && opts->dict_size != (1 << n) +
 						(1 << (n + 1))) {
 			err = -EIO;
-			goto failed;
+			goto out;
 		}
-	}
+	} else
+		/* use defaults */
+		opts->dict_size = max_t(int, msblk->block_size,
+							SQUASHFS_METADATA_SIZE);
 
-	dict_size = max_t(int, dict_size, SQUASHFS_METADATA_SIZE);
+	return opts;
+
+out:
+	kfree(opts);
+out2:
+	return ERR_PTR(err);
+}
+
+
+static void *squashfs_xz_init(struct squashfs_sb_info *msblk, void *buff)
+{
+	struct comp_opts *comp_opts = buff;
+	struct squashfs_xz *stream;
+	int err;
 
 	stream = kmalloc(sizeof(*stream), GFP_KERNEL);
 	if (stream == NULL) {
@@ -77,7 +103,7 @@ static void *squashfs_xz_init(struct squashfs_sb_info *msblk, void *buff,
 		goto failed;
 	}
 
-	stream->state = xz_dec_init(XZ_PREALLOC, dict_size);
+	stream->state = xz_dec_init(XZ_PREALLOC, comp_opts->dict_size);
 	if (stream->state == NULL) {
 		kfree(stream);
 		err = -ENOMEM;
@@ -103,42 +129,37 @@ static void squashfs_xz_free(void *strm)
 }
 
 
-static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void **buffer,
+static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void *strm,
-	struct buffer_head **bh, int b, int offset, int length, int srclength,
+	struct buffer_head **bh, int b, int offset, int length,
-	int pages)
+	struct squashfs_page_actor *output)
 {
 	enum xz_ret xz_err;
-	int avail, total = 0, k = 0, page = 0;
+	int avail, total = 0, k = 0;
-	struct squashfs_xz *stream = msblk->stream;
+	struct squashfs_xz *stream = strm;
-
-	mutex_lock(&msblk->read_data_mutex);
 
 	xz_dec_reset(stream->state);
 	stream->buf.in_pos = 0;
 	stream->buf.in_size = 0;
 	stream->buf.out_pos = 0;
 	stream->buf.out_size = PAGE_CACHE_SIZE;
-	stream->buf.out = buffer[page++];
+	stream->buf.out = squashfs_first_page(output);
 
 	do {
 		if (stream->buf.in_pos == stream->buf.in_size && k < b) {
 			avail = min(length, msblk->devblksize - offset);
 			length -= avail;
-			wait_on_buffer(bh[k]);
-			if (!buffer_uptodate(bh[k]))
-				goto release_mutex;
-
 			stream->buf.in = bh[k]->b_data + offset;
 			stream->buf.in_size = avail;
 			stream->buf.in_pos = 0;
 			offset = 0;
 		}
 
-		if (stream->buf.out_pos == stream->buf.out_size
+		if (stream->buf.out_pos == stream->buf.out_size) {
-							&& page < pages) {
+			stream->buf.out = squashfs_next_page(output);
-			stream->buf.out = buffer[page++];
+			if (stream->buf.out != NULL) {
-			stream->buf.out_pos = 0;
+				stream->buf.out_pos = 0;
-			total += PAGE_CACHE_SIZE;
+				total += PAGE_CACHE_SIZE;
+			}
 		}
 
 		xz_err = xz_dec_run(stream->state, &stream->buf);
@@ -147,23 +168,14 @@ static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void **buffer,
 			put_bh(bh[k++]);
 	} while (xz_err == XZ_OK);
 
-	if (xz_err != XZ_STREAM_END) {
+	squashfs_finish_page(output);
-		ERROR("xz_dec_run error, data probably corrupt\n");
-		goto release_mutex;
-	}
 
-	if (k < b) {
+	if (xz_err != XZ_STREAM_END || k < b)
-		ERROR("xz_uncompress error, input remaining\n");
+		goto out;
-		goto release_mutex;
-	}
 
-	total += stream->buf.out_pos;
+	return total + stream->buf.out_pos;
-	mutex_unlock(&msblk->read_data_mutex);
-	return total;
-
-release_mutex:
-	mutex_unlock(&msblk->read_data_mutex);
 
+out:
 	for (; k < b; k++)
 		put_bh(bh[k]);
 
@@ -172,6 +184,7 @@ release_mutex:
 
 const struct squashfs_decompressor squashfs_xz_comp_ops = {
 	.init = squashfs_xz_init,
+	.comp_opts = squashfs_xz_comp_opts,
 	.free = squashfs_xz_free,
 	.decompress = squashfs_xz_uncompress,
 	.id = XZ_COMPRESSION,

fs/squashfs/zlib_wrapper.c

@@ -32,8 +32,9 @@
 #include "squashfs_fs_sb.h"
 #include "squashfs.h"
 #include "decompressor.h"
+#include "page_actor.h"
 
-static void *zlib_init(struct squashfs_sb_info *dummy, void *buff, int len)
+static void *zlib_init(struct squashfs_sb_info *dummy, void *buff)
 {
 	z_stream *stream = kmalloc(sizeof(z_stream), GFP_KERNEL);
 	if (stream == NULL)
@@ -61,44 +62,37 @@ static void zlib_free(void *strm)
 }
 
 
-static int zlib_uncompress(struct squashfs_sb_info *msblk, void **buffer,
+static int zlib_uncompress(struct squashfs_sb_info *msblk, void *strm,
-	struct buffer_head **bh, int b, int offset, int length, int srclength,
+	struct buffer_head **bh, int b, int offset, int length,
-	int pages)
+	struct squashfs_page_actor *output)
 {
-	int zlib_err, zlib_init = 0;
+	int zlib_err, zlib_init = 0, k = 0;
-	int k = 0, page = 0;
+	z_stream *stream = strm;
-	z_stream *stream = msblk->stream;
 
-	mutex_lock(&msblk->read_data_mutex);
+	stream->avail_out = PAGE_CACHE_SIZE;
-
+	stream->next_out = squashfs_first_page(output);
-	stream->avail_out = 0;
 	stream->avail_in = 0;
 
 	do {
 		if (stream->avail_in == 0 && k < b) {
 			int avail = min(length, msblk->devblksize - offset);
 			length -= avail;
-			wait_on_buffer(bh[k]);
-			if (!buffer_uptodate(bh[k]))
-				goto release_mutex;
-
 			stream->next_in = bh[k]->b_data + offset;
 			stream->avail_in = avail;
 			offset = 0;
 		}
 
-		if (stream->avail_out == 0 && page < pages) {
+		if (stream->avail_out == 0) {
-			stream->next_out = buffer[page++];
+			stream->next_out = squashfs_next_page(output);
-			stream->avail_out = PAGE_CACHE_SIZE;
+			if (stream->next_out != NULL)
+				stream->avail_out = PAGE_CACHE_SIZE;
 		}
 
 		if (!zlib_init) {
 			zlib_err = zlib_inflateInit(stream);
 			if (zlib_err != Z_OK) {
-				ERROR("zlib_inflateInit returned unexpected "
+				squashfs_finish_page(output);
-					"result 0x%x, srclength %d\n",
+				goto out;
-					zlib_err, srclength);
-				goto release_mutex;
 			}
 			zlib_init = 1;
 		}
@@ -109,29 +103,21 @@ static int zlib_uncompress(struct squashfs_sb_info *msblk, void **buffer,
 			put_bh(bh[k++]);
 	} while (zlib_err == Z_OK);
 
-	if (zlib_err != Z_STREAM_END) {
+	squashfs_finish_page(output);
-		ERROR("zlib_inflate error, data probably corrupt\n");
+
-		goto release_mutex;
+	if (zlib_err != Z_STREAM_END)
-	}
+		goto out;
 
 	zlib_err = zlib_inflateEnd(stream);
-	if (zlib_err != Z_OK) {
+	if (zlib_err != Z_OK)
-		ERROR("zlib_inflate error, data probably corrupt\n");
+		goto out;
-		goto release_mutex;
-	}
 
-	if (k < b) {
+	if (k < b)
-		ERROR("zlib_uncompress error, data remaining\n");
+		goto out;
-		goto release_mutex;
-	}
 
-	length = stream->total_out;
+	return stream->total_out;
-	mutex_unlock(&msblk->read_data_mutex);
-	return length;
-
-release_mutex:
-	mutex_unlock(&msblk->read_data_mutex);
 
+out:
 	for (; k < b; k++)
 		put_bh(bh[k]);