Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable: (25 commits)
  Btrfs: forced readonly mounts on errors
  btrfs: Require CAP_SYS_ADMIN for filesystem rebalance
  Btrfs: don't warn if we get ENOSPC in btrfs_block_rsv_check
  btrfs: Fix memory leak in btrfs_read_fs_root_no_radix()
  btrfs: check NULL or not
  btrfs: Don't pass NULL ptr to func that may deref it.
  btrfs: mount failure return value fix
  btrfs: Mem leak in btrfs_get_acl()
  btrfs: fix wrong free space information of btrfs
  btrfs: make the chunk allocator utilize the devices better
  btrfs: restructure find_free_dev_extent()
  btrfs: fix wrong calculation of stripe size
  btrfs: try to reclaim some space when chunk allocation fails
  btrfs: fix wrong data space statistics
  fs/btrfs: Fix build of ctree
  Btrfs: fix off by one while setting block groups readonly
  Btrfs: Add BTRFS_IOC_SUBVOL_GETFLAGS/SETFLAGS ioctls
  Btrfs: Add readonly snapshots support
  Btrfs: Refactor btrfs_ioctl_snap_create()
  btrfs: Extract duplicate decompress code
  ...

fs/btrfs/Kconfig

@@ -4,6 +4,8 @@ config BTRFS_FS
 	select LIBCRC32C
 	select ZLIB_INFLATE
 	select ZLIB_DEFLATE
+	select LZO_COMPRESS
+	select LZO_DECOMPRESS
 	help
 	  Btrfs is a new filesystem with extents, writable snapshotting,
 	  support for multiple devices and many more features.

fs/btrfs/Makefile

@@ -6,5 +6,5 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
 	   transaction.o inode.o file.o tree-defrag.o \
 	   extent_map.o sysfs.o struct-funcs.o xattr.o ordered-data.o \
 	   extent_io.o volumes.o async-thread.o ioctl.o locking.o orphan.o \
-	   export.o tree-log.o acl.o free-space-cache.o zlib.o \
+	   export.o tree-log.o acl.o free-space-cache.o zlib.o lzo.o \
 	   compression.o delayed-ref.o relocation.o

fs/btrfs/acl.c

@@ -60,8 +60,10 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 		size = __btrfs_getxattr(inode, name, value, size);
 		if (size > 0) {
 			acl = posix_acl_from_xattr(value, size);
-			if (IS_ERR(acl))
+			if (IS_ERR(acl)) {
+				kfree(value);
 				return acl;
+			}
 			set_cached_acl(inode, type, acl);
 		}
 		kfree(value);

fs/btrfs/btrfs_inode.h

@@ -157,7 +157,7 @@ struct btrfs_inode {
 	/*
 	 * always compress this one file
 	 */
-	unsigned force_compress:1;
+	unsigned force_compress:4;
 
 	struct inode vfs_inode;
 };

fs/btrfs/compression.c

@@ -62,6 +62,9 @@ struct compressed_bio {
 	/* number of bytes on disk */
 	unsigned long compressed_len;
 
+	/* the compression algorithm for this bio */
+	int compress_type;
+
 	/* number of compressed pages in the array */
 	unsigned long nr_pages;
 
@@ -173,11 +176,12 @@ static void end_compressed_bio_read(struct bio *bio, int err)
 	/* ok, we're the last bio for this extent, lets start
 	 * the decompression.
 	 */
-	ret = btrfs_zlib_decompress_biovec(cb->compressed_pages,
-					cb->start,
-					cb->orig_bio->bi_io_vec,
-					cb->orig_bio->bi_vcnt,
-					cb->compressed_len);
+	ret = btrfs_decompress_biovec(cb->compress_type,
+				      cb->compressed_pages,
+				      cb->start,
+				      cb->orig_bio->bi_io_vec,
+				      cb->orig_bio->bi_vcnt,
+				      cb->compressed_len);
 csum_failed:
 	if (ret)
 		cb->errors = 1;
@@ -588,6 +592,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 
 	cb->len = uncompressed_len;
 	cb->compressed_len = compressed_len;
+	cb->compress_type = extent_compress_type(bio_flags);
 	cb->orig_bio = bio;
 
 	nr_pages = (compressed_len + PAGE_CACHE_SIZE - 1) /
@@ -677,3 +682,317 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	bio_put(comp_bio);
 	return 0;
 }
+
+static struct list_head comp_idle_workspace[BTRFS_COMPRESS_TYPES];
+static spinlock_t comp_workspace_lock[BTRFS_COMPRESS_TYPES];
+static int comp_num_workspace[BTRFS_COMPRESS_TYPES];
+static atomic_t comp_alloc_workspace[BTRFS_COMPRESS_TYPES];
+static wait_queue_head_t comp_workspace_wait[BTRFS_COMPRESS_TYPES];
+
+struct btrfs_compress_op *btrfs_compress_op[] = {
+	&btrfs_zlib_compress,
+	&btrfs_lzo_compress,
+};
+
+int __init btrfs_init_compress(void)
+{
+	int i;
+
+	for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
+		INIT_LIST_HEAD(&comp_idle_workspace[i]);
+		spin_lock_init(&comp_workspace_lock[i]);
+		atomic_set(&comp_alloc_workspace[i], 0);
+		init_waitqueue_head(&comp_workspace_wait[i]);
+	}
+	return 0;
+}
+
+/*
+ * this finds an available workspace or allocates a new one
+ * ERR_PTR is returned if things go bad.
+ */
+static struct list_head *find_workspace(int type)
+{
+	struct list_head *workspace;
+	int cpus = num_online_cpus();
+	int idx = type - 1;
+
+	struct list_head *idle_workspace	= &comp_idle_workspace[idx];
+	spinlock_t *workspace_lock		= &comp_workspace_lock[idx];
+	atomic_t *alloc_workspace		= &comp_alloc_workspace[idx];
+	wait_queue_head_t *workspace_wait	= &comp_workspace_wait[idx];
+	int *num_workspace			= &comp_num_workspace[idx];
+again:
+	spin_lock(workspace_lock);
+	if (!list_empty(idle_workspace)) {
+		workspace = idle_workspace->next;
+		list_del(workspace);
+		(*num_workspace)--;
+		spin_unlock(workspace_lock);
+		return workspace;
+
+	}
+	if (atomic_read(alloc_workspace) > cpus) {
+		DEFINE_WAIT(wait);
+
+		spin_unlock(workspace_lock);
+		prepare_to_wait(workspace_wait, &wait, TASK_UNINTERRUPTIBLE);
+		if (atomic_read(alloc_workspace) > cpus && !*num_workspace)
+			schedule();
+		finish_wait(workspace_wait, &wait);
+		goto again;
+	}
+	atomic_inc(alloc_workspace);
+	spin_unlock(workspace_lock);
+
+	workspace = btrfs_compress_op[idx]->alloc_workspace();
+	if (IS_ERR(workspace)) {
+		atomic_dec(alloc_workspace);
+		wake_up(workspace_wait);
+	}
+	return workspace;
+}
+
+/*
+ * put a workspace struct back on the list or free it if we have enough
+ * idle ones sitting around
+ */
+static void free_workspace(int type, struct list_head *workspace)
+{
+	int idx = type - 1;
+	struct list_head *idle_workspace	= &comp_idle_workspace[idx];
+	spinlock_t *workspace_lock		= &comp_workspace_lock[idx];
+	atomic_t *alloc_workspace		= &comp_alloc_workspace[idx];
+	wait_queue_head_t *workspace_wait	= &comp_workspace_wait[idx];
+	int *num_workspace			= &comp_num_workspace[idx];
+
+	spin_lock(workspace_lock);
+	if (*num_workspace < num_online_cpus()) {
+		list_add_tail(workspace, idle_workspace);
+		(*num_workspace)++;
+		spin_unlock(workspace_lock);
+		goto wake;
+	}
+	spin_unlock(workspace_lock);
+
+	btrfs_compress_op[idx]->free_workspace(workspace);
+	atomic_dec(alloc_workspace);
+wake:
+	if (waitqueue_active(workspace_wait))
+		wake_up(workspace_wait);
+}
+
+/*
+ * cleanup function for module exit
+ */
+static void free_workspaces(void)
+{
+	struct list_head *workspace;
+	int i;
+
+	for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
+		while (!list_empty(&comp_idle_workspace[i])) {
+			workspace = comp_idle_workspace[i].next;
+			list_del(workspace);
+			btrfs_compress_op[i]->free_workspace(workspace);
+			atomic_dec(&comp_alloc_workspace[i]);
+		}
+	}
+}
+
+/*
+ * given an address space and start/len, compress the bytes.
+ *
+ * pages are allocated to hold the compressed result and stored
+ * in 'pages'
+ *
+ * out_pages is used to return the number of pages allocated.  There
+ * may be pages allocated even if we return an error
+ *
+ * total_in is used to return the number of bytes actually read.  It
+ * may be smaller then len if we had to exit early because we
+ * ran out of room in the pages array or because we cross the
+ * max_out threshold.
+ *
+ * total_out is used to return the total number of compressed bytes
+ *
+ * max_out tells us the max number of bytes that we're allowed to
+ * stuff into pages
+ */
+int btrfs_compress_pages(int type, struct address_space *mapping,
+			 u64 start, unsigned long len,
+			 struct page **pages,
+			 unsigned long nr_dest_pages,
+			 unsigned long *out_pages,
+			 unsigned long *total_in,
+			 unsigned long *total_out,
+			 unsigned long max_out)
+{
+	struct list_head *workspace;
+	int ret;
+
+	workspace = find_workspace(type);
+	if (IS_ERR(workspace))
+		return -1;
+
+	ret = btrfs_compress_op[type-1]->compress_pages(workspace, mapping,
+						      start, len, pages,
+						      nr_dest_pages, out_pages,
+						      total_in, total_out,
+						      max_out);
+	free_workspace(type, workspace);
+	return ret;
+}
+
+/*
+ * pages_in is an array of pages with compressed data.
+ *
+ * disk_start is the starting logical offset of this array in the file
+ *
+ * bvec is a bio_vec of pages from the file that we want to decompress into
+ *
+ * vcnt is the count of pages in the biovec
+ *
+ * srclen is the number of bytes in pages_in
+ *
+ * The basic idea is that we have a bio that was created by readpages.
+ * The pages in the bio are for the uncompressed data, and they may not
+ * be contiguous.  They all correspond to the range of bytes covered by
+ * the compressed extent.
+ */
+int btrfs_decompress_biovec(int type, struct page **pages_in, u64 disk_start,
+			    struct bio_vec *bvec, int vcnt, size_t srclen)
+{
+	struct list_head *workspace;
+	int ret;
+
+	workspace = find_workspace(type);
+	if (IS_ERR(workspace))
+		return -ENOMEM;
+
+	ret = btrfs_compress_op[type-1]->decompress_biovec(workspace, pages_in,
+							 disk_start,
+							 bvec, vcnt, srclen);
+	free_workspace(type, workspace);
+	return ret;
+}
+
+/*
+ * a less complex decompression routine.  Our compressed data fits in a
+ * single page, and we want to read a single page out of it.
+ * start_byte tells us the offset into the compressed data we're interested in
+ */
+int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
+		     unsigned long start_byte, size_t srclen, size_t destlen)
+{
+	struct list_head *workspace;
+	int ret;
+
+	workspace = find_workspace(type);
+	if (IS_ERR(workspace))
+		return -ENOMEM;
+
+	ret = btrfs_compress_op[type-1]->decompress(workspace, data_in,
+						  dest_page, start_byte,
+						  srclen, destlen);
+
+	free_workspace(type, workspace);
+	return ret;
+}
+
+void __exit btrfs_exit_compress(void)
+{
+	free_workspaces();
+}
+
+/*
+ * Copy uncompressed data from working buffer to pages.
+ *
+ * buf_start is the byte offset we're of the start of our workspace buffer.
+ *
+ * total_out is the last byte of the buffer
+ */
+int btrfs_decompress_buf2page(char *buf, unsigned long buf_start,
+			      unsigned long total_out, u64 disk_start,
+			      struct bio_vec *bvec, int vcnt,
+			      unsigned long *page_index,
+			      unsigned long *pg_offset)
+{
+	unsigned long buf_offset;
+	unsigned long current_buf_start;
+	unsigned long start_byte;
+	unsigned long working_bytes = total_out - buf_start;
+	unsigned long bytes;
+	char *kaddr;
+	struct page *page_out = bvec[*page_index].bv_page;
+
+	/*
+	 * start byte is the first byte of the page we're currently
+	 * copying into relative to the start of the compressed data.
+	 */
+	start_byte = page_offset(page_out) - disk_start;
+
+	/* we haven't yet hit data corresponding to this page */
+	if (total_out <= start_byte)
+		return 1;
+
+	/*
+	 * the start of the data we care about is offset into
+	 * the middle of our working buffer
+	 */
+	if (total_out > start_byte && buf_start < start_byte) {
+		buf_offset = start_byte - buf_start;
+		working_bytes -= buf_offset;
+	} else {
+		buf_offset = 0;
+	}
+	current_buf_start = buf_start;
+
+	/* copy bytes from the working buffer into the pages */
+	while (working_bytes > 0) {
+		bytes = min(PAGE_CACHE_SIZE - *pg_offset,
+			    PAGE_CACHE_SIZE - buf_offset);
+		bytes = min(bytes, working_bytes);
+		kaddr = kmap_atomic(page_out, KM_USER0);
+		memcpy(kaddr + *pg_offset, buf + buf_offset, bytes);
+		kunmap_atomic(kaddr, KM_USER0);
+		flush_dcache_page(page_out);
+
+		*pg_offset += bytes;
+		buf_offset += bytes;
+		working_bytes -= bytes;
+		current_buf_start += bytes;
+
+		/* check if we need to pick another page */
+		if (*pg_offset == PAGE_CACHE_SIZE) {
+			(*page_index)++;
+			if (*page_index >= vcnt)
+				return 0;
+
+			page_out = bvec[*page_index].bv_page;
+			*pg_offset = 0;
+			start_byte = page_offset(page_out) - disk_start;
+
+			/*
+			 * make sure our new page is covered by this
+			 * working buffer
+			 */
+			if (total_out <= start_byte)
+				return 1;
+
+			/*
+			 * the next page in the biovec might not be adjacent
+			 * to the last page, but it might still be found
+			 * inside this working buffer. bump our offset pointer
+			 */
+			if (total_out > start_byte &&
+			    current_buf_start < start_byte) {
+				buf_offset = start_byte - buf_start;
+				working_bytes = total_out - start_byte;
+				current_buf_start = buf_start + buf_offset;
+			}
+		}
+	}
+
+	return 1;
+}

fs/btrfs/compression.h

@@ -19,24 +19,27 @@
 #ifndef __BTRFS_COMPRESSION_
 #define __BTRFS_COMPRESSION_
 
-int btrfs_zlib_decompress(unsigned char *data_in,
-			  struct page *dest_page,
-			  unsigned long start_byte,
-			  size_t srclen, size_t destlen);
-int btrfs_zlib_compress_pages(struct address_space *mapping,
-			      u64 start, unsigned long len,
-			      struct page **pages,
-			      unsigned long nr_dest_pages,
-			      unsigned long *out_pages,
-			      unsigned long *total_in,
-			      unsigned long *total_out,
-			      unsigned long max_out);
-int btrfs_zlib_decompress_biovec(struct page **pages_in,
-			      u64 disk_start,
-			      struct bio_vec *bvec,
-			      int vcnt,
-			      size_t srclen);
-void btrfs_zlib_exit(void);
+int btrfs_init_compress(void);
+void btrfs_exit_compress(void);
+
+int btrfs_compress_pages(int type, struct address_space *mapping,
+			 u64 start, unsigned long len,
+			 struct page **pages,
+			 unsigned long nr_dest_pages,
+			 unsigned long *out_pages,
+			 unsigned long *total_in,
+			 unsigned long *total_out,
+			 unsigned long max_out);
+int btrfs_decompress_biovec(int type, struct page **pages_in, u64 disk_start,
+			    struct bio_vec *bvec, int vcnt, size_t srclen);
+int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
+		     unsigned long start_byte, size_t srclen, size_t destlen);
+int btrfs_decompress_buf2page(char *buf, unsigned long buf_start,
+			      unsigned long total_out, u64 disk_start,
+			      struct bio_vec *bvec, int vcnt,
+			      unsigned long *page_index,
+			      unsigned long *pg_offset);
+
 int btrfs_submit_compressed_write(struct inode *inode, u64 start,
 				  unsigned long len, u64 disk_start,
 				  unsigned long compressed_len,
@@ -44,4 +47,37 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
 				  unsigned long nr_pages);
 int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 				 int mirror_num, unsigned long bio_flags);
+
+struct btrfs_compress_op {
+	struct list_head *(*alloc_workspace)(void);
+
+	void (*free_workspace)(struct list_head *workspace);
+
+	int (*compress_pages)(struct list_head *workspace,
+			      struct address_space *mapping,
+			      u64 start, unsigned long len,
+			      struct page **pages,
+			      unsigned long nr_dest_pages,
+			      unsigned long *out_pages,
+			      unsigned long *total_in,
+			      unsigned long *total_out,
+			      unsigned long max_out);
+
+	int (*decompress_biovec)(struct list_head *workspace,
+				 struct page **pages_in,
+				 u64 disk_start,
+				 struct bio_vec *bvec,
+				 int vcnt,
+				 size_t srclen);
+
+	int (*decompress)(struct list_head *workspace,
+			  unsigned char *data_in,
+			  struct page *dest_page,
+			  unsigned long start_byte,
+			  size_t srclen, size_t destlen);
+};
+
+extern struct btrfs_compress_op btrfs_zlib_compress;
+extern struct btrfs_compress_op btrfs_lzo_compress;
+
 #endif

fs/btrfs/ctree.c

@@ -105,6 +105,8 @@ noinline void btrfs_clear_path_blocking(struct btrfs_path *p,
 /* this also releases the path */
 void btrfs_free_path(struct btrfs_path *p)
 {
+	if (!p)
+		return;
 	btrfs_release_path(NULL, p);
 	kmem_cache_free(btrfs_path_cachep, p);
 }
@@ -2514,6 +2516,9 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
 	btrfs_assert_tree_locked(path->nodes[1]);
 
 	right = read_node_slot(root, upper, slot + 1);
+	if (right == NULL)
+		return 1;
+
 	btrfs_tree_lock(right);
 	btrfs_set_lock_blocking(right);
 
@@ -2764,6 +2769,9 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 	btrfs_assert_tree_locked(path->nodes[1]);
 
 	left = read_node_slot(root, path->nodes[1], slot - 1);
+	if (left == NULL)
+		return 1;
+
 	btrfs_tree_lock(left);
 	btrfs_set_lock_blocking(left);
 

fs/btrfs/ctree.h

@@ -295,6 +295,14 @@ static inline unsigned long btrfs_chunk_item_size(int num_stripes)
 #define BTRFS_FSID_SIZE 16
 #define BTRFS_HEADER_FLAG_WRITTEN	(1ULL << 0)
 #define BTRFS_HEADER_FLAG_RELOC		(1ULL << 1)
+
+/*
+ * File system states
+ */
+
+/* Errors detected */
+#define BTRFS_SUPER_FLAG_ERROR		(1ULL << 2)
+
 #define BTRFS_SUPER_FLAG_SEEDING	(1ULL << 32)
 #define BTRFS_SUPER_FLAG_METADUMP	(1ULL << 33)
 
@@ -399,13 +407,15 @@ struct btrfs_super_block {
 #define BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF	(1ULL << 0)
 #define BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL	(1ULL << 1)
 #define BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS	(1ULL << 2)
+#define BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO	(1ULL << 3)
 
 #define BTRFS_FEATURE_COMPAT_SUPP		0ULL
 #define BTRFS_FEATURE_COMPAT_RO_SUPP		0ULL
 #define BTRFS_FEATURE_INCOMPAT_SUPP			\
 	(BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF |		\
 	 BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL |	\
-	 BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS)
+	 BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS |		\
+	 BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO)
 
 /*
  * A leaf is full of items. offset and size tell us where to find
@@ -552,9 +562,11 @@ struct btrfs_timespec {
 } __attribute__ ((__packed__));
 
 enum btrfs_compression_type {
-	BTRFS_COMPRESS_NONE = 0,
-	BTRFS_COMPRESS_ZLIB = 1,
-	BTRFS_COMPRESS_LAST = 2,
+	BTRFS_COMPRESS_NONE  = 0,
+	BTRFS_COMPRESS_ZLIB  = 1,
+	BTRFS_COMPRESS_LZO   = 2,
+	BTRFS_COMPRESS_TYPES = 2,
+	BTRFS_COMPRESS_LAST  = 3,
 };
 
 struct btrfs_inode_item {
@@ -598,6 +610,8 @@ struct btrfs_dir_item {
 	u8 type;
 } __attribute__ ((__packed__));
 
+#define BTRFS_ROOT_SUBVOL_RDONLY	(1ULL << 0)
+
 struct btrfs_root_item {
 	struct btrfs_inode_item inode;
 	__le64 generation;
@@ -896,7 +910,8 @@ struct btrfs_fs_info {
 	 */
 	u64 last_trans_log_full_commit;
 	u64 open_ioctl_trans;
-	unsigned long mount_opt;
+	unsigned long mount_opt:20;
+	unsigned long compress_type:4;
 	u64 max_inline;
 	u64 alloc_start;
 	struct btrfs_transaction *running_transaction;
@@ -1051,6 +1066,9 @@ struct btrfs_fs_info {
 	unsigned metadata_ratio;
 
 	void *bdev_holder;
+
+	/* filesystem state */
+	u64 fs_state;
 };
 
 /*
@@ -1894,6 +1912,11 @@ BTRFS_SETGET_STACK_FUNCS(root_limit, struct btrfs_root_item, byte_limit, 64);
 BTRFS_SETGET_STACK_FUNCS(root_last_snapshot, struct btrfs_root_item,
 			 last_snapshot, 64);
 
+static inline bool btrfs_root_readonly(struct btrfs_root *root)
+{
+	return root->root_item.flags & BTRFS_ROOT_SUBVOL_RDONLY;
+}
+
 /* struct btrfs_super_block */
 
 BTRFS_SETGET_STACK_FUNCS(super_bytenr, struct btrfs_super_block, bytenr, 64);
@@ -2146,6 +2169,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root, u64 group_start);
 u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags);
+u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data);
 void btrfs_set_inode_space_info(struct btrfs_root *root, struct inode *ionde);
 void btrfs_clear_space_info_full(struct btrfs_fs_info *info);
 int btrfs_check_data_free_space(struct inode *inode, u64 bytes);
@@ -2189,6 +2213,12 @@ int btrfs_set_block_group_ro(struct btrfs_root *root,
 int btrfs_set_block_group_rw(struct btrfs_root *root,
 			     struct btrfs_block_group_cache *cache);
 void btrfs_put_block_group_cache(struct btrfs_fs_info *info);
+u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo);
+int btrfs_error_unpin_extent_range(struct btrfs_root *root,
+				   u64 start, u64 end);
+int btrfs_error_discard_extent(struct btrfs_root *root, u64 bytenr,
+			       u64 num_bytes);
+
 /* ctree.c */
 int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key,
 		     int level, int *slot);
@@ -2542,6 +2572,14 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size);
 /* super.c */
 int btrfs_parse_options(struct btrfs_root *root, char *options);
 int btrfs_sync_fs(struct super_block *sb, int wait);
+void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function,
+		     unsigned int line, int errno);
+
+#define btrfs_std_error(fs_info, errno)				\
+do {								\
+	if ((errno))						\
+		__btrfs_std_error((fs_info), __func__, __LINE__, (errno));\
+} while (0)
 
 /* acl.c */
 #ifdef CONFIG_BTRFS_FS_POSIX_ACL

fs/btrfs/disk-io.c

@@ -44,6 +44,20 @@
 static struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
 static void free_fs_root(struct btrfs_root *root);
+static void btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
+				    int read_only);
+static int btrfs_destroy_ordered_operations(struct btrfs_root *root);
+static int btrfs_destroy_ordered_extents(struct btrfs_root *root);
+static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
+				      struct btrfs_root *root);
+static int btrfs_destroy_pending_snapshots(struct btrfs_transaction *t);
+static int btrfs_destroy_delalloc_inodes(struct btrfs_root *root);
+static int btrfs_destroy_marked_extents(struct btrfs_root *root,
+					struct extent_io_tree *dirty_pages,
+					int mark);
+static int btrfs_destroy_pinned_extent(struct btrfs_root *root,
+				       struct extent_io_tree *pinned_extents);
+static int btrfs_cleanup_transaction(struct btrfs_root *root);
 
 /*
  * end_io_wq structs are used to do processing in task context when an IO is
@@ -353,6 +367,10 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
 	WARN_ON(len == 0);
 
 	eb = alloc_extent_buffer(tree, start, len, page, GFP_NOFS);
+	if (eb == NULL) {
+		WARN_ON(1);
+		goto out;
+	}
 	ret = btree_read_extent_buffer_pages(root, eb, start + PAGE_CACHE_SIZE,
 					     btrfs_header_generation(eb));
 	BUG_ON(ret);
@@ -427,6 +445,10 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
 	WARN_ON(len == 0);
 
 	eb = alloc_extent_buffer(tree, start, len, page, GFP_NOFS);
+	if (eb == NULL) {
+		ret = -EIO;
+		goto out;
+	}
 
 	found_start = btrfs_header_bytenr(eb);
 	if (found_start != start) {
@@ -1145,6 +1167,7 @@ struct btrfs_root *btrfs_read_fs_root_no_radix(struct btrfs_root *tree_root,
 	}
 	btrfs_free_path(path);
 	if (ret) {
+		kfree(root);
 		if (ret > 0)
 			ret = -ENOENT;
 		return ERR_PTR(ret);
@@ -1713,8 +1736,10 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 		     fs_info, BTRFS_ROOT_TREE_OBJECTID);
 
 	bh = btrfs_read_dev_super(fs_devices->latest_bdev);
-	if (!bh)
+	if (!bh) {
+		err = -EINVAL;
 		goto fail_iput;
+	}
 
 	memcpy(&fs_info->super_copy, bh->b_data, sizeof(fs_info->super_copy));
 	memcpy(&fs_info->super_for_commit, &fs_info->super_copy,
@@ -1727,6 +1752,11 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	if (!btrfs_super_root(disk_super))
 		goto fail_iput;
 
+	/* check FS state, whether FS is broken. */
+	fs_info->fs_state |= btrfs_super_flags(disk_super);
+
+	btrfs_check_super_valid(fs_info, sb->s_flags & MS_RDONLY);
+
 	ret = btrfs_parse_options(tree_root, options);
 	if (ret) {
 		err = ret;
@@ -1744,10 +1774,10 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	}
 
 	features = btrfs_super_incompat_flags(disk_super);
-	if (!(features & BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF)) {
-		features |= BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF;
-		btrfs_set_super_incompat_flags(disk_super, features);
-	}
+	features |= BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF;
+	if (tree_root->fs_info->compress_type & BTRFS_COMPRESS_LZO)
+		features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
+	btrfs_set_super_incompat_flags(disk_super, features);
 
 	features = btrfs_super_compat_ro_flags(disk_super) &
 		~BTRFS_FEATURE_COMPAT_RO_SUPP;
@@ -1957,7 +1987,9 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 		btrfs_set_opt(fs_info->mount_opt, SSD);
 	}
 
-	if (btrfs_super_log_root(disk_super) != 0) {
+	/* do not make disk changes in broken FS */
+	if (btrfs_super_log_root(disk_super) != 0 &&
+	    !(fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR)) {
 		u64 bytenr = btrfs_super_log_root(disk_super);
 
 		if (fs_devices->rw_devices == 0) {
@@ -2442,8 +2474,28 @@ int close_ctree(struct btrfs_root *root)
 	smp_mb();
 
 	btrfs_put_block_group_cache(fs_info);
+
+	/*
+	 * Here come 2 situations when btrfs is broken to flip readonly:
+	 *
+	 * 1. when btrfs flips readonly somewhere else before
+	 * btrfs_commit_super, sb->s_flags has MS_RDONLY flag,
+	 * and btrfs will skip to write sb directly to keep
+	 * ERROR state on disk.
+	 *
+	 * 2. when btrfs flips readonly just in btrfs_commit_super,
+	 * and in such case, btrfs cannnot write sb via btrfs_commit_super,
+	 * and since fs_state has been set BTRFS_SUPER_FLAG_ERROR flag,
+	 * btrfs will cleanup all FS resources first and write sb then.
+	 */
 	if (!(fs_info->sb->s_flags & MS_RDONLY)) {
-		ret =  btrfs_commit_super(root);
+		ret = btrfs_commit_super(root);
+		if (ret)
+			printk(KERN_ERR "btrfs: commit super ret %d\n", ret);
+	}
+
+	if (fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) {
+		ret = btrfs_error_commit_super(root);
 		if (ret)
 			printk(KERN_ERR "btrfs: commit super ret %d\n", ret);
 	}
@@ -2619,6 +2671,352 @@ out:
 	return 0;
 }
 
+static void btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
+			      int read_only)
+{
+	if (read_only)
+		return;
+
+	if (fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR)
+		printk(KERN_WARNING "warning: mount fs with errors, "
+		       "running btrfsck is recommended\n");
+}
+
+int btrfs_error_commit_super(struct btrfs_root *root)
+{
+	int ret;
+
+	mutex_lock(&root->fs_info->cleaner_mutex);
+	btrfs_run_delayed_iputs(root);
+	mutex_unlock(&root->fs_info->cleaner_mutex);
+
+	down_write(&root->fs_info->cleanup_work_sem);
+	up_write(&root->fs_info->cleanup_work_sem);
+
+	/* cleanup FS via transaction */
+	btrfs_cleanup_transaction(root);
+
+	ret = write_ctree_super(NULL, root, 0);
+
+	return ret;
+}
+
+static int btrfs_destroy_ordered_operations(struct btrfs_root *root)
+{
+	struct btrfs_inode *btrfs_inode;
+	struct list_head splice;
+
+	INIT_LIST_HEAD(&splice);
+
+	mutex_lock(&root->fs_info->ordered_operations_mutex);
+	spin_lock(&root->fs_info->ordered_extent_lock);
+
+	list_splice_init(&root->fs_info->ordered_operations, &splice);
+	while (!list_empty(&splice)) {
+		btrfs_inode = list_entry(splice.next, struct btrfs_inode,
+					 ordered_operations);
+
+		list_del_init(&btrfs_inode->ordered_operations);
+
+		btrfs_invalidate_inodes(btrfs_inode->root);
+	}
+
+	spin_unlock(&root->fs_info->ordered_extent_lock);
+	mutex_unlock(&root->fs_info->ordered_operations_mutex);
+
+	return 0;
+}
+
+static int btrfs_destroy_ordered_extents(struct btrfs_root *root)
+{
+	struct list_head splice;
+	struct btrfs_ordered_extent *ordered;
+	struct inode *inode;
+
+	INIT_LIST_HEAD(&splice);
+
+	spin_lock(&root->fs_info->ordered_extent_lock);
+
+	list_splice_init(&root->fs_info->ordered_extents, &splice);
+	while (!list_empty(&splice)) {
+		ordered = list_entry(splice.next, struct btrfs_ordered_extent,
+				     root_extent_list);
+
+		list_del_init(&ordered->root_extent_list);
+		atomic_inc(&ordered->refs);
+
+		/* the inode may be getting freed (in sys_unlink path). */
+		inode = igrab(ordered->inode);
+
+		spin_unlock(&root->fs_info->ordered_extent_lock);
+		if (inode)
+			iput(inode);
+
+		atomic_set(&ordered->refs, 1);
+		btrfs_put_ordered_extent(ordered);
+
+		spin_lock(&root->fs_info->ordered_extent_lock);
+	}
+
+	spin_unlock(&root->fs_info->ordered_extent_lock);
+
+	return 0;
+}
+
+static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
+				      struct btrfs_root *root)
+{
+	struct rb_node *node;
+	struct btrfs_delayed_ref_root *delayed_refs;
+	struct btrfs_delayed_ref_node *ref;
+	int ret = 0;
+
+	delayed_refs = &trans->delayed_refs;
+
+	spin_lock(&delayed_refs->lock);
+	if (delayed_refs->num_entries == 0) {
+		printk(KERN_INFO "delayed_refs has NO entry\n");
+		return ret;
+	}
+
+	node = rb_first(&delayed_refs->root);
+	while (node) {
+		ref = rb_entry(node, struct btrfs_delayed_ref_node, rb_node);
+		node = rb_next(node);
+
+		ref->in_tree = 0;
+		rb_erase(&ref->rb_node, &delayed_refs->root);
+		delayed_refs->num_entries--;
+
+		atomic_set(&ref->refs, 1);
+		if (btrfs_delayed_ref_is_head(ref)) {
+			struct btrfs_delayed_ref_head *head;
+
+			head = btrfs_delayed_node_to_head(ref);
+			mutex_lock(&head->mutex);
+			kfree(head->extent_op);
+			delayed_refs->num_heads--;
+			if (list_empty(&head->cluster))
+				delayed_refs->num_heads_ready--;
+			list_del_init(&head->cluster);
+			mutex_unlock(&head->mutex);
+		}
+
+		spin_unlock(&delayed_refs->lock);
+		btrfs_put_delayed_ref(ref);
+
+		cond_resched();
+		spin_lock(&delayed_refs->lock);
+	}
+
+	spin_unlock(&delayed_refs->lock);
+
+	return ret;
+}
+
+static int btrfs_destroy_pending_snapshots(struct btrfs_transaction *t)
+{
+	struct btrfs_pending_snapshot *snapshot;
+	struct list_head splice;
+
+	INIT_LIST_HEAD(&splice);
+
+	list_splice_init(&t->pending_snapshots, &splice);
+
+	while (!list_empty(&splice)) {
+		snapshot = list_entry(splice.next,
+				      struct btrfs_pending_snapshot,
+				      list);
+
+		list_del_init(&snapshot->list);
+
+		kfree(snapshot);
+	}
+
+	return 0;
+}
+
+static int btrfs_destroy_delalloc_inodes(struct btrfs_root *root)
+{
+	struct btrfs_inode *btrfs_inode;
+	struct list_head splice;
+
+	INIT_LIST_HEAD(&splice);
+
+	list_splice_init(&root->fs_info->delalloc_inodes, &splice);
+
+	spin_lock(&root->fs_info->delalloc_lock);
+
+	while (!list_empty(&splice)) {
+		btrfs_inode = list_entry(splice.next, struct btrfs_inode,
+				    delalloc_inodes);
+
+		list_del_init(&btrfs_inode->delalloc_inodes);
+
+		btrfs_invalidate_inodes(btrfs_inode->root);
+	}
+
+	spin_unlock(&root->fs_info->delalloc_lock);
+
+	return 0;
+}
+
+static int btrfs_destroy_marked_extents(struct btrfs_root *root,
+					struct extent_io_tree *dirty_pages,
+					int mark)
+{
+	int ret;
+	struct page *page;
+	struct inode *btree_inode = root->fs_info->btree_inode;
+	struct extent_buffer *eb;
+	u64 start = 0;
+	u64 end;
+	u64 offset;
+	unsigned long index;
+
+	while (1) {
+		ret = find_first_extent_bit(dirty_pages, start, &start, &end,
+					    mark);
+		if (ret)
+			break;
+
+		clear_extent_bits(dirty_pages, start, end, mark, GFP_NOFS);
+		while (start <= end) {
+			index = start >> PAGE_CACHE_SHIFT;
+			start = (u64)(index + 1) << PAGE_CACHE_SHIFT;
+			page = find_get_page(btree_inode->i_mapping, index);
+			if (!page)
+				continue;
+			offset = page_offset(page);
+
+			spin_lock(&dirty_pages->buffer_lock);
+			eb = radix_tree_lookup(
+			     &(&BTRFS_I(page->mapping->host)->io_tree)->buffer,
+					       offset >> PAGE_CACHE_SHIFT);
+			spin_unlock(&dirty_pages->buffer_lock);
+			if (eb) {
+				ret = test_and_clear_bit(EXTENT_BUFFER_DIRTY,
+							 &eb->bflags);
+				atomic_set(&eb->refs, 1);
+			}
+			if (PageWriteback(page))
+				end_page_writeback(page);
+
+			lock_page(page);
+			if (PageDirty(page)) {
+				clear_page_dirty_for_io(page);
+				spin_lock_irq(&page->mapping->tree_lock);
+				radix_tree_tag_clear(&page->mapping->page_tree,
+							page_index(page),
+							PAGECACHE_TAG_DIRTY);
+				spin_unlock_irq(&page->mapping->tree_lock);
+			}
+
+			page->mapping->a_ops->invalidatepage(page, 0);
+			unlock_page(page);
+		}
+	}
+
+	return ret;
+}
+
+static int btrfs_destroy_pinned_extent(struct btrfs_root *root,
+				       struct extent_io_tree *pinned_extents)
+{
+	struct extent_io_tree *unpin;
+	u64 start;
+	u64 end;
+	int ret;
+
+	unpin = pinned_extents;
+	while (1) {
+		ret = find_first_extent_bit(unpin, 0, &start, &end,
+					    EXTENT_DIRTY);
+		if (ret)
+			break;
+
+		/* opt_discard */
+		ret = btrfs_error_discard_extent(root, start, end + 1 - start);
+
+		clear_extent_dirty(unpin, start, end, GFP_NOFS);
+		btrfs_error_unpin_extent_range(root, start, end);
+		cond_resched();
+	}
+
+	return 0;
+}
+
+static int btrfs_cleanup_transaction(struct btrfs_root *root)
+{
+	struct btrfs_transaction *t;
+	LIST_HEAD(list);
+
+	WARN_ON(1);
+
+	mutex_lock(&root->fs_info->trans_mutex);
+	mutex_lock(&root->fs_info->transaction_kthread_mutex);
+
+	list_splice_init(&root->fs_info->trans_list, &list);
+	while (!list_empty(&list)) {
+		t = list_entry(list.next, struct btrfs_transaction, list);
+		if (!t)
+			break;
+
+		btrfs_destroy_ordered_operations(root);
+
+		btrfs_destroy_ordered_extents(root);
+
+		btrfs_destroy_delayed_refs(t, root);
+
+		btrfs_block_rsv_release(root,
+					&root->fs_info->trans_block_rsv,
+					t->dirty_pages.dirty_bytes);
+
+		/* FIXME: cleanup wait for commit */
+		t->in_commit = 1;
+		t->blocked = 1;
+		if (waitqueue_active(&root->fs_info->transaction_blocked_wait))
+			wake_up(&root->fs_info->transaction_blocked_wait);
+
+		t->blocked = 0;
+		if (waitqueue_active(&root->fs_info->transaction_wait))
+			wake_up(&root->fs_info->transaction_wait);
+		mutex_unlock(&root->fs_info->trans_mutex);
+
+		mutex_lock(&root->fs_info->trans_mutex);
+		t->commit_done = 1;
+		if (waitqueue_active(&t->commit_wait))
+			wake_up(&t->commit_wait);
+		mutex_unlock(&root->fs_info->trans_mutex);
+
+		mutex_lock(&root->fs_info->trans_mutex);
+
+		btrfs_destroy_pending_snapshots(t);
+
+		btrfs_destroy_delalloc_inodes(root);
+
+		spin_lock(&root->fs_info->new_trans_lock);
+		root->fs_info->running_transaction = NULL;
+		spin_unlock(&root->fs_info->new_trans_lock);
+
+		btrfs_destroy_marked_extents(root, &t->dirty_pages,
+					     EXTENT_DIRTY);
+
+		btrfs_destroy_pinned_extent(root,
+					    root->fs_info->pinned_extents);
+
+		t->use_count = 0;
+		list_del_init(&t->list);
+		memset(t, 0, sizeof(*t));
+		kmem_cache_free(btrfs_transaction_cachep, t);
+	}
+
+	mutex_unlock(&root->fs_info->transaction_kthread_mutex);
+	mutex_unlock(&root->fs_info->trans_mutex);
+
+	return 0;
+}
+
 static struct extent_io_ops btree_extent_io_ops = {
 	.write_cache_pages_lock_hook = btree_lock_page_hook,
 	.readpage_end_io_hook = btree_readpage_end_io_hook,

fs/btrfs/disk-io.h

@@ -52,6 +52,7 @@ int write_ctree_super(struct btrfs_trans_handle *trans,
 		      struct btrfs_root *root, int max_mirrors);
 struct buffer_head *btrfs_read_dev_super(struct block_device *bdev);
 int btrfs_commit_super(struct btrfs_root *root);
+int btrfs_error_commit_super(struct btrfs_root *root);
 struct extent_buffer *btrfs_find_tree_block(struct btrfs_root *root,
 					    u64 bytenr, u32 blocksize);
 struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,

fs/btrfs/extent-tree.c

@@ -3089,7 +3089,7 @@ static u64 get_alloc_profile(struct btrfs_root *root, u64 flags)
 	return btrfs_reduce_alloc_profile(root, flags);
 }
 
-static u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data)
+u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data)
 {
 	u64 flags;
 
@@ -3161,8 +3161,12 @@ alloc:
 					     bytes + 2 * 1024 * 1024,
 					     alloc_target, 0);
 			btrfs_end_transaction(trans, root);
-			if (ret < 0)
-				return ret;
+			if (ret < 0) {
+				if (ret != -ENOSPC)
+					return ret;
+				else
+					goto commit_trans;
+			}
 
 			if (!data_sinfo) {
 				btrfs_set_inode_space_info(root, inode);
@@ -3173,6 +3177,7 @@ alloc:
 		spin_unlock(&data_sinfo->lock);
 
 		/* commit the current transaction and try again */
+commit_trans:
 		if (!committed && !root->fs_info->open_ioctl_trans) {
 			committed = 1;
 			trans = btrfs_join_transaction(root, 1);
@@ -3721,11 +3726,6 @@ int btrfs_block_rsv_check(struct btrfs_trans_handle *trans,
 		return 0;
 	}
 
-	WARN_ON(1);
-	printk(KERN_INFO"block_rsv size %llu reserved %llu freed %llu %llu\n",
-		block_rsv->size, block_rsv->reserved,
-		block_rsv->freed[0], block_rsv->freed[1]);
-
 	return -ENOSPC;
 }
 
@@ -7970,13 +7970,14 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache)
 
 	if (sinfo->bytes_used + sinfo->bytes_reserved + sinfo->bytes_pinned +
 	    sinfo->bytes_may_use + sinfo->bytes_readonly +
-	    cache->reserved_pinned + num_bytes < sinfo->total_bytes) {
+	    cache->reserved_pinned + num_bytes <= sinfo->total_bytes) {
 		sinfo->bytes_readonly += num_bytes;
 		sinfo->bytes_reserved += cache->reserved_pinned;
 		cache->reserved_pinned = 0;
 		cache->ro = 1;
 		ret = 0;
 	}
+
 	spin_unlock(&cache->lock);
 	spin_unlock(&sinfo->lock);
 	return ret;
@@ -8012,6 +8013,62 @@ out:
 	return ret;
 }
 
+/*
+ * helper to account the unused space of all the readonly block group in the
+ * list. takes mirrors into account.
+ */
+static u64 __btrfs_get_ro_block_group_free_space(struct list_head *groups_list)
+{
+	struct btrfs_block_group_cache *block_group;
+	u64 free_bytes = 0;
+	int factor;
+
+	list_for_each_entry(block_group, groups_list, list) {
+		spin_lock(&block_group->lock);
+
+		if (!block_group->ro) {
+			spin_unlock(&block_group->lock);
+			continue;
+		}
+
+		if (block_group->flags & (BTRFS_BLOCK_GROUP_RAID1 |
+					  BTRFS_BLOCK_GROUP_RAID10 |
+					  BTRFS_BLOCK_GROUP_DUP))
+			factor = 2;
+		else
+			factor = 1;
+
+		free_bytes += (block_group->key.offset -
+			       btrfs_block_group_used(&block_group->item)) *
+			       factor;
+
+		spin_unlock(&block_group->lock);
+	}
+
+	return free_bytes;
+}
+
+/*
+ * helper to account the unused space of all the readonly block group in the
+ * space_info. takes mirrors into account.
+ */
+u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo)
+{
+	int i;
+	u64 free_bytes = 0;
+
+	spin_lock(&sinfo->lock);
+
+	for(i = 0; i < BTRFS_NR_RAID_TYPES; i++)
+		if (!list_empty(&sinfo->block_groups[i]))
+			free_bytes += __btrfs_get_ro_block_group_free_space(
+						&sinfo->block_groups[i]);
+
+	spin_unlock(&sinfo->lock);
+
+	return free_bytes;
+}
+
 int btrfs_set_block_group_rw(struct btrfs_root *root,
 			      struct btrfs_block_group_cache *cache)
 {
@@ -8092,7 +8149,7 @@ int btrfs_can_relocate(struct btrfs_root *root, u64 bytenr)
 	mutex_lock(&root->fs_info->chunk_mutex);
 	list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) {
 		u64 min_free = btrfs_block_group_used(&block_group->item);
-		u64 dev_offset, max_avail;
+		u64 dev_offset;
 
 		/*
 		 * check to make sure we can actually find a chunk with enough
@@ -8100,7 +8157,7 @@ int btrfs_can_relocate(struct btrfs_root *root, u64 bytenr)
 		 */
 		if (device->total_bytes > device->bytes_used + min_free) {
 			ret = find_free_dev_extent(NULL, device, min_free,
-						   &dev_offset, &max_avail);
+						   &dev_offset, NULL);
 			if (!ret)
 				break;
 			ret = -1;
@@ -8584,3 +8641,14 @@ out:
 	btrfs_free_path(path);
 	return ret;
 }
+
+int btrfs_error_unpin_extent_range(struct btrfs_root *root, u64 start, u64 end)
+{
+	return unpin_extent_range(root, start, end);
+}
+
+int btrfs_error_discard_extent(struct btrfs_root *root, u64 bytenr,
+			       u64 num_bytes)
+{
+	return btrfs_discard_extent(root, bytenr, num_bytes);
+}

fs/btrfs/extent_io.c

@@ -2028,8 +2028,11 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
 		BUG_ON(extent_map_end(em) <= cur);
 		BUG_ON(end < cur);
 
-		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags))
+		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
 			this_bio_flag = EXTENT_BIO_COMPRESSED;
+			extent_set_compress_type(&this_bio_flag,
+						 em->compress_type);
+		}
 
 		iosize = min(extent_map_end(em) - cur, end - cur + 1);
 		cur_end = min(extent_map_end(em) - 1, end);
@@ -3072,6 +3075,8 @@ static struct extent_buffer *__alloc_extent_buffer(struct extent_io_tree *tree,
 #endif
 
 	eb = kmem_cache_zalloc(extent_buffer_cache, mask);
+	if (eb == NULL)
+		return NULL;
 	eb->start = start;
 	eb->len = len;
 	spin_lock_init(&eb->lock);

fs/btrfs/extent_io.h

@@ -20,8 +20,12 @@
 #define EXTENT_IOBITS (EXTENT_LOCKED | EXTENT_WRITEBACK)
 #define EXTENT_CTLBITS (EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC)
 
-/* flags for bio submission */
+/*
+ * flags for bio submission. The high bits indicate the compression
+ * type for this bio
+ */
 #define EXTENT_BIO_COMPRESSED 1
+#define EXTENT_BIO_FLAG_SHIFT 16
 
 /* these are bit numbers for test/set bit */
 #define EXTENT_BUFFER_UPTODATE 0
@@ -135,6 +139,17 @@ struct extent_buffer {
 	wait_queue_head_t lock_wq;
 };
 
+static inline void extent_set_compress_type(unsigned long *bio_flags,
+					    int compress_type)
+{
+	*bio_flags |= compress_type << EXTENT_BIO_FLAG_SHIFT;
+}
+
+static inline int extent_compress_type(unsigned long bio_flags)
+{
+	return bio_flags >> EXTENT_BIO_FLAG_SHIFT;
+}
+
 struct extent_map_tree;
 
 static inline struct extent_state *extent_state_next(struct extent_state *state)

fs/btrfs/extent_map.c

@@ -3,6 +3,7 @@
 #include <linux/module.h>
 #include <linux/spinlock.h>
 #include <linux/hardirq.h>
+#include "ctree.h"
 #include "extent_map.h"
 
 
@@ -54,6 +55,7 @@ struct extent_map *alloc_extent_map(gfp_t mask)
 		return em;
 	em->in_tree = 0;
 	em->flags = 0;
+	em->compress_type = BTRFS_COMPRESS_NONE;
 	atomic_set(&em->refs, 1);
 	return em;
 }

fs/btrfs/extent_map.h

@@ -26,7 +26,8 @@ struct extent_map {
 	unsigned long flags;
 	struct block_device *bdev;
 	atomic_t refs;
-	int in_tree;
+	unsigned int in_tree:1;
+	unsigned int compress_type:4;
 };
 
 struct extent_map_tree {

fs/btrfs/file.c

@@ -225,6 +225,7 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
 
 			split->bdev = em->bdev;
 			split->flags = flags;
+			split->compress_type = em->compress_type;
 			ret = add_extent_mapping(em_tree, split);
 			BUG_ON(ret);
 			free_extent_map(split);
@@ -239,6 +240,7 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
 			split->len = em->start + em->len - (start + len);
 			split->bdev = em->bdev;
 			split->flags = flags;
+			split->compress_type = em->compress_type;
 
 			if (compressed) {
 				split->block_len = em->block_len;
@@ -891,6 +893,17 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 	if (err)
 		goto out;
 
+	/*
+	 * If BTRFS flips readonly due to some impossible error
+	 * (fs_info->fs_state now has BTRFS_SUPER_FLAG_ERROR),
+	 * although we have opened a file as writable, we have
+	 * to stop this write operation to ensure FS consistency.
+	 */
+	if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) {
+		err = -EROFS;
+		goto out;
+	}
+
 	file_update_time(file);
 	BTRFS_I(inode)->sequence++;
 

fs/btrfs/inode.c

@@ -122,10 +122,10 @@ static noinline int insert_inline_extent(struct btrfs_trans_handle *trans,
 	size_t cur_size = size;
 	size_t datasize;
 	unsigned long offset;
-	int use_compress = 0;
+	int compress_type = BTRFS_COMPRESS_NONE;
 
 	if (compressed_size && compressed_pages) {
-		use_compress = 1;
+		compress_type = root->fs_info->compress_type;
 		cur_size = compressed_size;
 	}
 
@@ -159,7 +159,7 @@ static noinline int insert_inline_extent(struct btrfs_trans_handle *trans,
 	btrfs_set_file_extent_ram_bytes(leaf, ei, size);
 	ptr = btrfs_file_extent_inline_start(ei);
 
-	if (use_compress) {
+	if (compress_type != BTRFS_COMPRESS_NONE) {
 		struct page *cpage;
 		int i = 0;
 		while (compressed_size > 0) {
@@ -176,7 +176,7 @@ static noinline int insert_inline_extent(struct btrfs_trans_handle *trans,
 			compressed_size -= cur_size;
 		}
 		btrfs_set_file_extent_compression(leaf, ei,
-						  BTRFS_COMPRESS_ZLIB);
+						  compress_type);
 	} else {
 		page = find_get_page(inode->i_mapping,
 				     start >> PAGE_CACHE_SHIFT);
@@ -263,6 +263,7 @@ struct async_extent {
 	u64 compressed_size;
 	struct page **pages;
 	unsigned long nr_pages;
+	int compress_type;
 	struct list_head list;
 };
 
@@ -280,7 +281,8 @@ static noinline int add_async_extent(struct async_cow *cow,
 				     u64 start, u64 ram_size,
 				     u64 compressed_size,
 				     struct page **pages,
-				     unsigned long nr_pages)
+				     unsigned long nr_pages,
+				     int compress_type)
 {
 	struct async_extent *async_extent;
 
@@ -290,6 +292,7 @@ static noinline int add_async_extent(struct async_cow *cow,
 	async_extent->compressed_size = compressed_size;
 	async_extent->pages = pages;
 	async_extent->nr_pages = nr_pages;
+	async_extent->compress_type = compress_type;
 	list_add_tail(&async_extent->list, &cow->extents);
 	return 0;
 }
@@ -332,6 +335,7 @@ static noinline int compress_file_range(struct inode *inode,
 	unsigned long max_uncompressed = 128 * 1024;
 	int i;
 	int will_compress;
+	int compress_type = root->fs_info->compress_type;
 
 	actual_end = min_t(u64, isize, end + 1);
 again:
@@ -381,12 +385,16 @@ again:
 		WARN_ON(pages);
 		pages = kzalloc(sizeof(struct page *) * nr_pages, GFP_NOFS);
 
-		ret = btrfs_zlib_compress_pages(inode->i_mapping, start,
-						total_compressed, pages,
-						nr_pages, &nr_pages_ret,
-						&total_in,
-						&total_compressed,
-						max_compressed);
+		if (BTRFS_I(inode)->force_compress)
+			compress_type = BTRFS_I(inode)->force_compress;
+
+		ret = btrfs_compress_pages(compress_type,
+					   inode->i_mapping, start,
+					   total_compressed, pages,
+					   nr_pages, &nr_pages_ret,
+					   &total_in,
+					   &total_compressed,
+					   max_compressed);
 
 		if (!ret) {
 			unsigned long offset = total_compressed &
@@ -493,7 +501,8 @@ again:
 		 * and will submit them to the elevator.
 		 */
 		add_async_extent(async_cow, start, num_bytes,
-				 total_compressed, pages, nr_pages_ret);
+				 total_compressed, pages, nr_pages_ret,
+				 compress_type);
 
 		if (start + num_bytes < end) {
 			start += num_bytes;
@@ -515,7 +524,8 @@ cleanup_and_bail_uncompressed:
 			__set_page_dirty_nobuffers(locked_page);
 			/* unlocked later on in the async handlers */
 		}
-		add_async_extent(async_cow, start, end - start + 1, 0, NULL, 0);
+		add_async_extent(async_cow, start, end - start + 1,
+				 0, NULL, 0, BTRFS_COMPRESS_NONE);
 		*num_added += 1;
 	}
 
@@ -640,6 +650,7 @@ retry:
 		em->block_start = ins.objectid;
 		em->block_len = ins.offset;
 		em->bdev = root->fs_info->fs_devices->latest_bdev;
+		em->compress_type = async_extent->compress_type;
 		set_bit(EXTENT_FLAG_PINNED, &em->flags);
 		set_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
 
@@ -656,11 +667,13 @@ retry:
 						async_extent->ram_size - 1, 0);
 		}
 
-		ret = btrfs_add_ordered_extent(inode, async_extent->start,
-					       ins.objectid,
-					       async_extent->ram_size,
-					       ins.offset,
-					       BTRFS_ORDERED_COMPRESSED);
+		ret = btrfs_add_ordered_extent_compress(inode,
+						async_extent->start,
+						ins.objectid,
+						async_extent->ram_size,
+						ins.offset,
+						BTRFS_ORDERED_COMPRESSED,
+						async_extent->compress_type);
 		BUG_ON(ret);
 
 		/*
@@ -1670,7 +1683,7 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 	struct btrfs_ordered_extent *ordered_extent = NULL;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct extent_state *cached_state = NULL;
-	int compressed = 0;
+	int compress_type = 0;
 	int ret;
 	bool nolock = false;
 
@@ -1711,9 +1724,9 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 	trans->block_rsv = &root->fs_info->delalloc_block_rsv;
 
 	if (test_bit(BTRFS_ORDERED_COMPRESSED, &ordered_extent->flags))
-		compressed = 1;
+		compress_type = ordered_extent->compress_type;
 	if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
-		BUG_ON(compressed);
+		BUG_ON(compress_type);
 		ret = btrfs_mark_extent_written(trans, inode,
 						ordered_extent->file_offset,
 						ordered_extent->file_offset +
@@ -1727,7 +1740,7 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 						ordered_extent->disk_len,
 						ordered_extent->len,
 						ordered_extent->len,
-						compressed, 0, 0,
+						compress_type, 0, 0,
 						BTRFS_FILE_EXTENT_REG);
 		unpin_extent_cache(&BTRFS_I(inode)->extent_tree,
 				   ordered_extent->file_offset,
@@ -1829,6 +1842,8 @@ static int btrfs_io_failed_hook(struct bio *failed_bio,
 		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
 			logical = em->block_start;
 			failrec->bio_flags = EXTENT_BIO_COMPRESSED;
+			extent_set_compress_type(&failrec->bio_flags,
+						 em->compress_type);
 		}
 		failrec->logical = logical;
 		free_extent_map(em);
@@ -3671,8 +3686,12 @@ static int btrfs_setattr_size(struct inode *inode, struct iattr *attr)
 static int btrfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = dentry->d_inode;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int err;
 
+	if (btrfs_root_readonly(root))
+		return -EROFS;
+
 	err = inode_change_ok(inode, attr);
 	if (err)
 		return err;
@@ -4928,8 +4947,10 @@ static noinline int uncompress_inline(struct btrfs_path *path,
 	size_t max_size;
 	unsigned long inline_size;
 	unsigned long ptr;
+	int compress_type;
 
 	WARN_ON(pg_offset != 0);
+	compress_type = btrfs_file_extent_compression(leaf, item);
 	max_size = btrfs_file_extent_ram_bytes(leaf, item);
 	inline_size = btrfs_file_extent_inline_item_len(leaf,
 					btrfs_item_nr(leaf, path->slots[0]));
@@ -4939,8 +4960,8 @@ static noinline int uncompress_inline(struct btrfs_path *path,
 	read_extent_buffer(leaf, tmp, ptr, inline_size);
 
 	max_size = min_t(unsigned long, PAGE_CACHE_SIZE, max_size);
-	ret = btrfs_zlib_decompress(tmp, page, extent_offset,
-				    inline_size, max_size);
+	ret = btrfs_decompress(compress_type, tmp, page,
+			       extent_offset, inline_size, max_size);
 	if (ret) {
 		char *kaddr = kmap_atomic(page, KM_USER0);
 		unsigned long copy_size = min_t(u64,
@@ -4982,7 +5003,7 @@ struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page,
 	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct btrfs_trans_handle *trans = NULL;
-	int compressed;
+	int compress_type;
 
 again:
 	read_lock(&em_tree->lock);
@@ -5041,7 +5062,7 @@ again:
 
 	found_type = btrfs_file_extent_type(leaf, item);
 	extent_start = found_key.offset;
-	compressed = btrfs_file_extent_compression(leaf, item);
+	compress_type = btrfs_file_extent_compression(leaf, item);
 	if (found_type == BTRFS_FILE_EXTENT_REG ||
 	    found_type == BTRFS_FILE_EXTENT_PREALLOC) {
 		extent_end = extent_start +
@@ -5087,8 +5108,9 @@ again:
 			em->block_start = EXTENT_MAP_HOLE;
 			goto insert;
 		}
-		if (compressed) {
+		if (compress_type != BTRFS_COMPRESS_NONE) {
 			set_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
+			em->compress_type = compress_type;
 			em->block_start = bytenr;
 			em->block_len = btrfs_file_extent_disk_num_bytes(leaf,
 									 item);
@@ -5122,12 +5144,14 @@ again:
 		em->len = (copy_size + root->sectorsize - 1) &
 			~((u64)root->sectorsize - 1);
 		em->orig_start = EXTENT_MAP_INLINE;
-		if (compressed)
+		if (compress_type) {
 			set_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
+			em->compress_type = compress_type;
+		}
 		ptr = btrfs_file_extent_inline_start(item) + extent_offset;
 		if (create == 0 && !PageUptodate(page)) {
-			if (btrfs_file_extent_compression(leaf, item) ==
-			    BTRFS_COMPRESS_ZLIB) {
+			if (btrfs_file_extent_compression(leaf, item) !=
+			    BTRFS_COMPRESS_NONE) {
 				ret = uncompress_inline(path, inode, page,
 							pg_offset,
 							extent_offset, item);
@@ -6477,7 +6501,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
 	ei->ordered_data_close = 0;
 	ei->orphan_meta_reserved = 0;
 	ei->dummy_inode = 0;
-	ei->force_compress = 0;
+	ei->force_compress = BTRFS_COMPRESS_NONE;
 
 	inode = &ei->vfs_inode;
 	extent_map_tree_init(&ei->extent_tree, GFP_NOFS);
@@ -7105,6 +7129,10 @@ static int btrfs_set_page_dirty(struct page *page)
 
 static int btrfs_permission(struct inode *inode, int mask, unsigned int flags)
 {
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+
+	if (btrfs_root_readonly(root) && (mask & MAY_WRITE))
+		return -EROFS;
 	if ((BTRFS_I(inode)->flags & BTRFS_INODE_READONLY) && (mask & MAY_WRITE))
 		return -EACCES;
 	return generic_permission(inode, mask, flags, btrfs_check_acl);

fs/btrfs/ioctl.c

@@ -147,6 +147,9 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 	unsigned int flags, oldflags;
 	int ret;
 
+	if (btrfs_root_readonly(root))
+		return -EROFS;
+
 	if (copy_from_user(&flags, arg, sizeof(flags)))
 		return -EFAULT;
 
@@ -360,7 +363,8 @@ fail:
 }
 
 static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
-			   char *name, int namelen, u64 *async_transid)
+			   char *name, int namelen, u64 *async_transid,
+			   bool readonly)
 {
 	struct inode *inode;
 	struct dentry *parent;
@@ -378,6 +382,7 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 	btrfs_init_block_rsv(&pending_snapshot->block_rsv);
 	pending_snapshot->dentry = dentry;
 	pending_snapshot->root = root;
+	pending_snapshot->readonly = readonly;
 
 	trans = btrfs_start_transaction(root->fs_info->extent_root, 5);
 	if (IS_ERR(trans)) {
@@ -509,7 +514,7 @@ static inline int btrfs_may_create(struct inode *dir, struct dentry *child)
 static noinline int btrfs_mksubvol(struct path *parent,
 				   char *name, int namelen,
 				   struct btrfs_root *snap_src,
-				   u64 *async_transid)
+				   u64 *async_transid, bool readonly)
 {
 	struct inode *dir  = parent->dentry->d_inode;
 	struct dentry *dentry;
@@ -541,7 +546,7 @@ static noinline int btrfs_mksubvol(struct path *parent,
 
 	if (snap_src) {
 		error = create_snapshot(snap_src, dentry,
-					name, namelen, async_transid);
+					name, namelen, async_transid, readonly);
 	} else {
 		error = create_subvol(BTRFS_I(dir)->root, dentry,
 				      name, namelen, async_transid);
@@ -638,9 +643,11 @@ static int btrfs_defrag_file(struct file *file,
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct btrfs_ordered_extent *ordered;
 	struct page *page;
+	struct btrfs_super_block *disk_super;
 	unsigned long last_index;
 	unsigned long ra_pages = root->fs_info->bdi.ra_pages;
 	unsigned long total_read = 0;
+	u64 features;
 	u64 page_start;
 	u64 page_end;
 	u64 last_len = 0;
@@ -648,6 +655,14 @@ static int btrfs_defrag_file(struct file *file,
 	u64 defrag_end = 0;
 	unsigned long i;
 	int ret;
+	int compress_type = BTRFS_COMPRESS_ZLIB;
+
+	if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS) {
+		if (range->compress_type > BTRFS_COMPRESS_TYPES)
+			return -EINVAL;
+		if (range->compress_type)
+			compress_type = range->compress_type;
+	}
 
 	if (inode->i_size == 0)
 		return 0;
@@ -683,7 +698,7 @@ static int btrfs_defrag_file(struct file *file,
 		total_read++;
 		mutex_lock(&inode->i_mutex);
 		if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)
-			BTRFS_I(inode)->force_compress = 1;
+			BTRFS_I(inode)->force_compress = compress_type;
 
 		ret  = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE);
 		if (ret)
@@ -781,10 +796,17 @@ loop_unlock:
 		atomic_dec(&root->fs_info->async_submit_draining);
 
 		mutex_lock(&inode->i_mutex);
-		BTRFS_I(inode)->force_compress = 0;
+		BTRFS_I(inode)->force_compress = BTRFS_COMPRESS_NONE;
 		mutex_unlock(&inode->i_mutex);
 	}
 
+	disk_super = &root->fs_info->super_copy;
+	features = btrfs_super_incompat_flags(disk_super);
+	if (range->compress_type == BTRFS_COMPRESS_LZO) {
+		features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
+		btrfs_set_super_incompat_flags(disk_super, features);
+	}
+
 	return 0;
 
 err_reservations:
@@ -901,7 +923,8 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
 						    char *name,
 						    unsigned long fd,
 						    int subvol,
-						    u64 *transid)
+						    u64 *transid,
+						    bool readonly)
 {
 	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
 	struct file *src_file;
@@ -919,7 +942,7 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
 
 	if (subvol) {
 		ret = btrfs_mksubvol(&file->f_path, name, namelen,
-				     NULL, transid);
+				     NULL, transid, readonly);
 	} else {
 		struct inode *src_inode;
 		src_file = fget(fd);
@@ -938,7 +961,7 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
 		}
 		ret = btrfs_mksubvol(&file->f_path, name, namelen,
 				     BTRFS_I(src_inode)->root,
-				     transid);
+				     transid, readonly);
 		fput(src_file);
 	}
 out:
@@ -946,58 +969,139 @@ out:
 }
 
 static noinline int btrfs_ioctl_snap_create(struct file *file,
-					    void __user *arg, int subvol,
-					    int v2)
+					    void __user *arg, int subvol)
 {
-	struct btrfs_ioctl_vol_args *vol_args = NULL;
-	struct btrfs_ioctl_vol_args_v2 *vol_args_v2 = NULL;
-	char *name;
-	u64 fd;
+	struct btrfs_ioctl_vol_args *vol_args;
 	int ret;
 
-	if (v2) {
-		u64 transid = 0;
-		u64 *ptr = NULL;
+	vol_args = memdup_user(arg, sizeof(*vol_args));
+	if (IS_ERR(vol_args))
+		return PTR_ERR(vol_args);
+	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
 
-		vol_args_v2 = memdup_user(arg, sizeof(*vol_args_v2));
-		if (IS_ERR(vol_args_v2))
-			return PTR_ERR(vol_args_v2);
+	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
+					      vol_args->fd, subvol,
+					      NULL, false);
 
-		if (vol_args_v2->flags & ~BTRFS_SUBVOL_CREATE_ASYNC) {
-			ret = -EINVAL;
-			goto out;
-		}
-
-		name = vol_args_v2->name;
-		fd = vol_args_v2->fd;
-		vol_args_v2->name[BTRFS_SUBVOL_NAME_MAX] = '\0';
+	kfree(vol_args);
+	return ret;
+}
 
-		if (vol_args_v2->flags & BTRFS_SUBVOL_CREATE_ASYNC)
-			ptr = &transid;
+static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
+					       void __user *arg, int subvol)
+{
+	struct btrfs_ioctl_vol_args_v2 *vol_args;
+	int ret;
+	u64 transid = 0;
+	u64 *ptr = NULL;
+	bool readonly = false;
 
-		ret = btrfs_ioctl_snap_create_transid(file, name, fd,
-						      subvol, ptr);
+	vol_args = memdup_user(arg, sizeof(*vol_args));
+	if (IS_ERR(vol_args))
+		return PTR_ERR(vol_args);
+	vol_args->name[BTRFS_SUBVOL_NAME_MAX] = '\0';
 
-		if (ret == 0 && ptr &&
-		    copy_to_user(arg +
-				 offsetof(struct btrfs_ioctl_vol_args_v2,
-					  transid), ptr, sizeof(*ptr)))
-			ret = -EFAULT;
-	} else {
-		vol_args = memdup_user(arg, sizeof(*vol_args));
-		if (IS_ERR(vol_args))
-			return PTR_ERR(vol_args);
-		name = vol_args->name;
-		fd = vol_args->fd;
-		vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
-
-		ret = btrfs_ioctl_snap_create_transid(file, name, fd,
-						      subvol, NULL);
+	if (vol_args->flags &
+	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY)) {
+		ret = -EOPNOTSUPP;
+		goto out;
 	}
+
+	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC)
+		ptr = &transid;
+	if (vol_args->flags & BTRFS_SUBVOL_RDONLY)
+		readonly = true;
+
+	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
+					      vol_args->fd, subvol,
+					      ptr, readonly);
+
+	if (ret == 0 && ptr &&
+	    copy_to_user(arg +
+			 offsetof(struct btrfs_ioctl_vol_args_v2,
+				  transid), ptr, sizeof(*ptr)))
+		ret = -EFAULT;
 out:
 	kfree(vol_args);
-	kfree(vol_args_v2);
+	return ret;
+}
 
+static noinline int btrfs_ioctl_subvol_getflags(struct file *file,
+						void __user *arg)
+{
+	struct inode *inode = fdentry(file)->d_inode;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	int ret = 0;
+	u64 flags = 0;
+
+	if (inode->i_ino != BTRFS_FIRST_FREE_OBJECTID)
+		return -EINVAL;
+
+	down_read(&root->fs_info->subvol_sem);
+	if (btrfs_root_readonly(root))
+		flags |= BTRFS_SUBVOL_RDONLY;
+	up_read(&root->fs_info->subvol_sem);
+
+	if (copy_to_user(arg, &flags, sizeof(flags)))
+		ret = -EFAULT;
+
+	return ret;
+}
+
+static noinline int btrfs_ioctl_subvol_setflags(struct file *file,
+					      void __user *arg)
+{
+	struct inode *inode = fdentry(file)->d_inode;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_trans_handle *trans;
+	u64 root_flags;
+	u64 flags;
+	int ret = 0;
+
+	if (root->fs_info->sb->s_flags & MS_RDONLY)
+		return -EROFS;
+
+	if (inode->i_ino != BTRFS_FIRST_FREE_OBJECTID)
+		return -EINVAL;
+
+	if (copy_from_user(&flags, arg, sizeof(flags)))
+		return -EFAULT;
+
+	if (flags & ~BTRFS_SUBVOL_CREATE_ASYNC)
+		return -EINVAL;
+
+	if (flags & ~BTRFS_SUBVOL_RDONLY)
+		return -EOPNOTSUPP;
+
+	down_write(&root->fs_info->subvol_sem);
+
+	/* nothing to do */
+	if (!!(flags & BTRFS_SUBVOL_RDONLY) == btrfs_root_readonly(root))
+		goto out;
+
+	root_flags = btrfs_root_flags(&root->root_item);
+	if (flags & BTRFS_SUBVOL_RDONLY)
+		btrfs_set_root_flags(&root->root_item,
+				     root_flags | BTRFS_ROOT_SUBVOL_RDONLY);
+	else
+		btrfs_set_root_flags(&root->root_item,
+				     root_flags & ~BTRFS_ROOT_SUBVOL_RDONLY);
+
+	trans = btrfs_start_transaction(root, 1);
+	if (IS_ERR(trans)) {
+		ret = PTR_ERR(trans);
+		goto out_reset;
+	}
+
+	ret = btrfs_update_root(trans, root,
+				&root->root_key, &root->root_item);
+
+	btrfs_commit_transaction(trans, root);
+out_reset:
+	if (ret)
+		btrfs_set_root_flags(&root->root_item, root_flags);
+out:
+	up_write(&root->fs_info->subvol_sem);
 	return ret;
 }
 
@@ -1509,6 +1613,9 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
 	struct btrfs_ioctl_defrag_range_args *range;
 	int ret;
 
+	if (btrfs_root_readonly(root))
+		return -EROFS;
+
 	ret = mnt_want_write(file->f_path.mnt);
 	if (ret)
 		return ret;
@@ -1637,6 +1744,9 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 	if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND))
 		return -EINVAL;
 
+	if (btrfs_root_readonly(root))
+		return -EROFS;
+
 	ret = mnt_want_write(file->f_path.mnt);
 	if (ret)
 		return ret;
@@ -1958,6 +2068,10 @@ static long btrfs_ioctl_trans_start(struct file *file)
 	if (file->private_data)
 		goto out;
 
+	ret = -EROFS;
+	if (btrfs_root_readonly(root))
+		goto out;
+
 	ret = mnt_want_write(file->f_path.mnt);
 	if (ret)
 		goto out;
@@ -2257,13 +2371,17 @@ long btrfs_ioctl(struct file *file, unsigned int
 	case FS_IOC_GETVERSION:
 		return btrfs_ioctl_getversion(file, argp);
 	case BTRFS_IOC_SNAP_CREATE:
-		return btrfs_ioctl_snap_create(file, argp, 0, 0);
+		return btrfs_ioctl_snap_create(file, argp, 0);
 	case BTRFS_IOC_SNAP_CREATE_V2:
-		return btrfs_ioctl_snap_create(file, argp, 0, 1);
+		return btrfs_ioctl_snap_create_v2(file, argp, 0);
 	case BTRFS_IOC_SUBVOL_CREATE:
-		return btrfs_ioctl_snap_create(file, argp, 1, 0);
+		return btrfs_ioctl_snap_create(file, argp, 1);
 	case BTRFS_IOC_SNAP_DESTROY:
 		return btrfs_ioctl_snap_destroy(file, argp);
+	case BTRFS_IOC_SUBVOL_GETFLAGS:
+		return btrfs_ioctl_subvol_getflags(file, argp);
+	case BTRFS_IOC_SUBVOL_SETFLAGS:
+		return btrfs_ioctl_subvol_setflags(file, argp);
 	case BTRFS_IOC_DEFAULT_SUBVOL:
 		return btrfs_ioctl_default_subvol(file, argp);
 	case BTRFS_IOC_DEFRAG:

fs/btrfs/ioctl.h

@@ -31,6 +31,7 @@ struct btrfs_ioctl_vol_args {
 };
 
 #define BTRFS_SUBVOL_CREATE_ASYNC	(1ULL << 0)
+#define BTRFS_SUBVOL_RDONLY		(1ULL << 1)
 
 #define BTRFS_SUBVOL_NAME_MAX 4039
 struct btrfs_ioctl_vol_args_v2 {
@@ -133,8 +134,15 @@ struct btrfs_ioctl_defrag_range_args {
 	 */
 	__u32 extent_thresh;
 
+	/*
+	 * which compression method to use if turning on compression
+	 * for this defrag operation.  If unspecified, zlib will
+	 * be used
+	 */
+	__u32 compress_type;
+
 	/* spare for later */
-	__u32 unused[5];
+	__u32 unused[4];
 };
 
 struct btrfs_ioctl_space_info {
@@ -193,4 +201,6 @@ struct btrfs_ioctl_space_args {
 #define BTRFS_IOC_WAIT_SYNC  _IOW(BTRFS_IOCTL_MAGIC, 22, __u64)
 #define BTRFS_IOC_SNAP_CREATE_V2 _IOW(BTRFS_IOCTL_MAGIC, 23, \
 				   struct btrfs_ioctl_vol_args_v2)
+#define BTRFS_IOC_SUBVOL_GETFLAGS _IOW(BTRFS_IOCTL_MAGIC, 25, __u64)
+#define BTRFS_IOC_SUBVOL_SETFLAGS _IOW(BTRFS_IOCTL_MAGIC, 26, __u64)
 #endif

fs/btrfs/lzo.c

@@ -0,0 +1,420 @@
+/*
+ * Copyright (C) 2008 Oracle.  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 v2 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.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+#include <linux/err.h>
+#include <linux/sched.h>
+#include <linux/pagemap.h>
+#include <linux/bio.h>
+#include <linux/lzo.h>
+#include "compression.h"
+
+#define LZO_LEN	4
+
+struct workspace {
+	void *mem;
+	void *buf;	/* where compressed data goes */
+	void *cbuf;	/* where decompressed data goes */
+	struct list_head list;
+};
+
+static void lzo_free_workspace(struct list_head *ws)
+{
+	struct workspace *workspace = list_entry(ws, struct workspace, list);
+
+	vfree(workspace->buf);
+	vfree(workspace->cbuf);
+	vfree(workspace->mem);
+	kfree(workspace);
+}
+
+static struct list_head *lzo_alloc_workspace(void)
+{
+	struct workspace *workspace;
+
+	workspace = kzalloc(sizeof(*workspace), GFP_NOFS);
+	if (!workspace)
+		return ERR_PTR(-ENOMEM);
+
+	workspace->mem = vmalloc(LZO1X_MEM_COMPRESS);
+	workspace->buf = vmalloc(lzo1x_worst_compress(PAGE_CACHE_SIZE));
+	workspace->cbuf = vmalloc(lzo1x_worst_compress(PAGE_CACHE_SIZE));
+	if (!workspace->mem || !workspace->buf || !workspace->cbuf)
+		goto fail;
+
+	INIT_LIST_HEAD(&workspace->list);
+
+	return &workspace->list;
+fail:
+	lzo_free_workspace(&workspace->list);
+	return ERR_PTR(-ENOMEM);
+}
+
+static inline void write_compress_length(char *buf, size_t len)
+{
+	__le32 dlen;
+
+	dlen = cpu_to_le32(len);
+	memcpy(buf, &dlen, LZO_LEN);
+}
+
+static inline size_t read_compress_length(char *buf)
+{
+	__le32 dlen;
+
+	memcpy(&dlen, buf, LZO_LEN);
+	return le32_to_cpu(dlen);
+}
+
+static int lzo_compress_pages(struct list_head *ws,
+			      struct address_space *mapping,
+			      u64 start, unsigned long len,
+			      struct page **pages,
+			      unsigned long nr_dest_pages,
+			      unsigned long *out_pages,
+			      unsigned long *total_in,
+			      unsigned long *total_out,
+			      unsigned long max_out)
+{
+	struct workspace *workspace = list_entry(ws, struct workspace, list);
+	int ret = 0;
+	char *data_in;
+	char *cpage_out;
+	int nr_pages = 0;
+	struct page *in_page = NULL;
+	struct page *out_page = NULL;
+	unsigned long bytes_left;
+
+	size_t in_len;
+	size_t out_len;
+	char *buf;
+	unsigned long tot_in = 0;
+	unsigned long tot_out = 0;
+	unsigned long pg_bytes_left;
+	unsigned long out_offset;
+	unsigned long bytes;
+
+	*out_pages = 0;
+	*total_out = 0;
+	*total_in = 0;
+
+	in_page = find_get_page(mapping, start >> PAGE_CACHE_SHIFT);
+	data_in = kmap(in_page);
+
+	/*
+	 * store the size of all chunks of compressed data in
+	 * the first 4 bytes
+	 */
+	out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+	if (out_page == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	cpage_out = kmap(out_page);
+	out_offset = LZO_LEN;
+	tot_out = LZO_LEN;
+	pages[0] = out_page;
+	nr_pages = 1;
+	pg_bytes_left = PAGE_CACHE_SIZE - LZO_LEN;
+
+	/* compress at most one page of data each time */
+	in_len = min(len, PAGE_CACHE_SIZE);
+	while (tot_in < len) {
+		ret = lzo1x_1_compress(data_in, in_len, workspace->cbuf,
+				       &out_len, workspace->mem);
+		if (ret != LZO_E_OK) {
+			printk(KERN_DEBUG "btrfs deflate in loop returned %d\n",
+			       ret);
+			ret = -1;
+			goto out;
+		}
+
+		/* store the size of this chunk of compressed data */
+		write_compress_length(cpage_out + out_offset, out_len);
+		tot_out += LZO_LEN;
+		out_offset += LZO_LEN;
+		pg_bytes_left -= LZO_LEN;
+
+		tot_in += in_len;
+		tot_out += out_len;
+
+		/* copy bytes from the working buffer into the pages */
+		buf = workspace->cbuf;
+		while (out_len) {
+			bytes = min_t(unsigned long, pg_bytes_left, out_len);
+
+			memcpy(cpage_out + out_offset, buf, bytes);
+
+			out_len -= bytes;
+			pg_bytes_left -= bytes;
+			buf += bytes;
+			out_offset += bytes;
+
+			/*
+			 * we need another page for writing out.
+			 *
+			 * Note if there's less than 4 bytes left, we just
+			 * skip to a new page.
+			 */
+			if ((out_len == 0 && pg_bytes_left < LZO_LEN) ||
+			    pg_bytes_left == 0) {
+				if (pg_bytes_left) {
+					memset(cpage_out + out_offset, 0,
+					       pg_bytes_left);
+					tot_out += pg_bytes_left;
+				}
+
+				/* we're done, don't allocate new page */
+				if (out_len == 0 && tot_in >= len)
+					break;
+
+				kunmap(out_page);
+				if (nr_pages == nr_dest_pages) {
+					out_page = NULL;
+					ret = -1;
+					goto out;
+				}
+
+				out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+				if (out_page == NULL) {
+					ret = -ENOMEM;
+					goto out;
+				}
+				cpage_out = kmap(out_page);
+				pages[nr_pages++] = out_page;
+
+				pg_bytes_left = PAGE_CACHE_SIZE;
+				out_offset = 0;
+			}
+		}
+
+		/* we're making it bigger, give up */
+		if (tot_in > 8192 && tot_in < tot_out)
+			goto out;
+
+		/* we're all done */
+		if (tot_in >= len)
+			break;
+
+		if (tot_out > max_out)
+			break;
+
+		bytes_left = len - tot_in;
+		kunmap(in_page);
+		page_cache_release(in_page);
+
+		start += PAGE_CACHE_SIZE;
+		in_page = find_get_page(mapping, start >> PAGE_CACHE_SHIFT);
+		data_in = kmap(in_page);
+		in_len = min(bytes_left, PAGE_CACHE_SIZE);
+	}
+
+	if (tot_out > tot_in)
+		goto out;
+
+	/* store the size of all chunks of compressed data */
+	cpage_out = kmap(pages[0]);
+	write_compress_length(cpage_out, tot_out);
+
+	kunmap(pages[0]);
+
+	ret = 0;
+	*total_out = tot_out;
+	*total_in = tot_in;
+out:
+	*out_pages = nr_pages;
+	if (out_page)
+		kunmap(out_page);
+
+	if (in_page) {
+		kunmap(in_page);
+		page_cache_release(in_page);
+	}
+
+	return ret;
+}
+
+static int lzo_decompress_biovec(struct list_head *ws,
+				 struct page **pages_in,
+				 u64 disk_start,
+				 struct bio_vec *bvec,
+				 int vcnt,
+				 size_t srclen)
+{
+	struct workspace *workspace = list_entry(ws, struct workspace, list);
+	int ret = 0, ret2;
+	char *data_in;
+	unsigned long page_in_index = 0;
+	unsigned long page_out_index = 0;
+	unsigned long total_pages_in = (srclen + PAGE_CACHE_SIZE - 1) /
+					PAGE_CACHE_SIZE;
+	unsigned long buf_start;
+	unsigned long buf_offset = 0;
+	unsigned long bytes;
+	unsigned long working_bytes;
+	unsigned long pg_offset;
+
+	size_t in_len;
+	size_t out_len;
+	unsigned long in_offset;
+	unsigned long in_page_bytes_left;
+	unsigned long tot_in;
+	unsigned long tot_out;
+	unsigned long tot_len;
+	char *buf;
+
+	data_in = kmap(pages_in[0]);
+	tot_len = read_compress_length(data_in);
+
+	tot_in = LZO_LEN;
+	in_offset = LZO_LEN;
+	tot_len = min_t(size_t, srclen, tot_len);
+	in_page_bytes_left = PAGE_CACHE_SIZE - LZO_LEN;
+
+	tot_out = 0;
+	pg_offset = 0;
+
+	while (tot_in < tot_len) {
+		in_len = read_compress_length(data_in + in_offset);
+		in_page_bytes_left -= LZO_LEN;
+		in_offset += LZO_LEN;
+		tot_in += LZO_LEN;
+
+		tot_in += in_len;
+		working_bytes = in_len;
+
+		/* fast path: avoid using the working buffer */
+		if (in_page_bytes_left >= in_len) {
+			buf = data_in + in_offset;
+			bytes = in_len;
+			goto cont;
+		}
+
+		/* copy bytes from the pages into the working buffer */
+		buf = workspace->cbuf;
+		buf_offset = 0;
+		while (working_bytes) {
+			bytes = min(working_bytes, in_page_bytes_left);
+
+			memcpy(buf + buf_offset, data_in + in_offset, bytes);
+			buf_offset += bytes;
+cont:
+			working_bytes -= bytes;
+			in_page_bytes_left -= bytes;
+			in_offset += bytes;
+
+			/* check if we need to pick another page */
+			if ((working_bytes == 0 && in_page_bytes_left < LZO_LEN)
+			    || in_page_bytes_left == 0) {
+				tot_in += in_page_bytes_left;
+
+				if (working_bytes == 0 && tot_in >= tot_len)
+					break;
+
+				kunmap(pages_in[page_in_index]);
+				page_in_index++;
+				if (page_in_index >= total_pages_in) {
+					ret = -1;
+					data_in = NULL;
+					goto done;
+				}
+				data_in = kmap(pages_in[page_in_index]);
+
+				in_page_bytes_left = PAGE_CACHE_SIZE;
+				in_offset = 0;
+			}
+		}
+
+		out_len = lzo1x_worst_compress(PAGE_CACHE_SIZE);
+		ret = lzo1x_decompress_safe(buf, in_len, workspace->buf,
+					    &out_len);
+		if (ret != LZO_E_OK) {
+			printk(KERN_WARNING "btrfs decompress failed\n");
+			ret = -1;
+			break;
+		}
+
+		buf_start = tot_out;
+		tot_out += out_len;
+
+		ret2 = btrfs_decompress_buf2page(workspace->buf, buf_start,
+						 tot_out, disk_start,
+						 bvec, vcnt,
+						 &page_out_index, &pg_offset);
+		if (ret2 == 0)
+			break;
+	}
+done:
+	if (data_in)
+		kunmap(pages_in[page_in_index]);
+	return ret;
+}
+
+static int lzo_decompress(struct list_head *ws, unsigned char *data_in,
+			  struct page *dest_page,
+			  unsigned long start_byte,
+			  size_t srclen, size_t destlen)
+{
+	struct workspace *workspace = list_entry(ws, struct workspace, list);
+	size_t in_len;
+	size_t out_len;
+	size_t tot_len;
+	int ret = 0;
+	char *kaddr;
+	unsigned long bytes;
+
+	BUG_ON(srclen < LZO_LEN);
+
+	tot_len = read_compress_length(data_in);
+	data_in += LZO_LEN;
+
+	in_len = read_compress_length(data_in);
+	data_in += LZO_LEN;
+
+	out_len = PAGE_CACHE_SIZE;
+	ret = lzo1x_decompress_safe(data_in, in_len, workspace->buf, &out_len);
+	if (ret != LZO_E_OK) {
+		printk(KERN_WARNING "btrfs decompress failed!\n");
+		ret = -1;
+		goto out;
+	}
+
+	if (out_len < start_byte) {
+		ret = -1;
+		goto out;
+	}
+
+	bytes = min_t(unsigned long, destlen, out_len - start_byte);
+
+	kaddr = kmap_atomic(dest_page, KM_USER0);
+	memcpy(kaddr, workspace->buf + start_byte, bytes);
+	kunmap_atomic(kaddr, KM_USER0);
+out:
+	return ret;
+}
+
+struct btrfs_compress_op btrfs_lzo_compress = {
+	.alloc_workspace	= lzo_alloc_workspace,
+	.free_workspace		= lzo_free_workspace,
+	.compress_pages		= lzo_compress_pages,
+	.decompress_biovec	= lzo_decompress_biovec,
+	.decompress		= lzo_decompress,
+};

fs/btrfs/ordered-data.c

@@ -172,7 +172,7 @@ static inline struct rb_node *tree_search(struct btrfs_ordered_inode_tree *tree,
  */
 static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 				      u64 start, u64 len, u64 disk_len,
-				      int type, int dio)
+				      int type, int dio, int compress_type)
 {
 	struct btrfs_ordered_inode_tree *tree;
 	struct rb_node *node;
@@ -189,6 +189,7 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 	entry->disk_len = disk_len;
 	entry->bytes_left = len;
 	entry->inode = inode;
+	entry->compress_type = compress_type;
 	if (type != BTRFS_ORDERED_IO_DONE && type != BTRFS_ORDERED_COMPLETE)
 		set_bit(type, &entry->flags);
 
@@ -220,14 +221,25 @@ int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 			     u64 start, u64 len, u64 disk_len, int type)
 {
 	return __btrfs_add_ordered_extent(inode, file_offset, start, len,
-					  disk_len, type, 0);
+					  disk_len, type, 0,
+					  BTRFS_COMPRESS_NONE);
 }
 
 int btrfs_add_ordered_extent_dio(struct inode *inode, u64 file_offset,
 				 u64 start, u64 len, u64 disk_len, int type)
 {
 	return __btrfs_add_ordered_extent(inode, file_offset, start, len,
-					  disk_len, type, 1);
+					  disk_len, type, 1,
+					  BTRFS_COMPRESS_NONE);
+}
+
+int btrfs_add_ordered_extent_compress(struct inode *inode, u64 file_offset,
+				      u64 start, u64 len, u64 disk_len,
+				      int type, int compress_type)
+{
+	return __btrfs_add_ordered_extent(inode, file_offset, start, len,
+					  disk_len, type, 0,
+					  compress_type);
 }
 
 /*

fs/btrfs/ordered-data.h

@@ -68,7 +68,7 @@ struct btrfs_ordered_sum {
 
 #define BTRFS_ORDERED_NOCOW 2 /* set when we want to write in place */
 
-#define BTRFS_ORDERED_COMPRESSED 3 /* writing a compressed extent */
+#define BTRFS_ORDERED_COMPRESSED 3 /* writing a zlib compressed extent */
 
 #define BTRFS_ORDERED_PREALLOC 4 /* set when writing to prealloced extent */
 
@@ -93,6 +93,9 @@ struct btrfs_ordered_extent {
 	/* flags (described above) */
 	unsigned long flags;
 
+	/* compression algorithm */
+	int compress_type;
+
 	/* reference count */
 	atomic_t refs;
 
@@ -148,6 +151,9 @@ int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 			     u64 start, u64 len, u64 disk_len, int type);
 int btrfs_add_ordered_extent_dio(struct inode *inode, u64 file_offset,
 				 u64 start, u64 len, u64 disk_len, int type);
+int btrfs_add_ordered_extent_compress(struct inode *inode, u64 file_offset,
+				      u64 start, u64 len, u64 disk_len,
+				      int type, int compress_type);
 int btrfs_add_ordered_sum(struct inode *inode,
 			  struct btrfs_ordered_extent *entry,
 			  struct btrfs_ordered_sum *sum);

fs/btrfs/super.c

@@ -54,6 +54,90 @@
 
 static const struct super_operations btrfs_super_ops;
 
+static const char *btrfs_decode_error(struct btrfs_fs_info *fs_info, int errno,
+				      char nbuf[16])
+{
+	char *errstr = NULL;
+
+	switch (errno) {
+	case -EIO:
+		errstr = "IO failure";
+		break;
+	case -ENOMEM:
+		errstr = "Out of memory";
+		break;
+	case -EROFS:
+		errstr = "Readonly filesystem";
+		break;
+	default:
+		if (nbuf) {
+			if (snprintf(nbuf, 16, "error %d", -errno) >= 0)
+				errstr = nbuf;
+		}
+		break;
+	}
+
+	return errstr;
+}
+
+static void __save_error_info(struct btrfs_fs_info *fs_info)
+{
+	/*
+	 * today we only save the error info into ram.  Long term we'll
+	 * also send it down to the disk
+	 */
+	fs_info->fs_state = BTRFS_SUPER_FLAG_ERROR;
+}
+
+/* NOTE:
+ *	We move write_super stuff at umount in order to avoid deadlock
+ *	for umount hold all lock.
+ */
+static void save_error_info(struct btrfs_fs_info *fs_info)
+{
+	__save_error_info(fs_info);
+}
+
+/* btrfs handle error by forcing the filesystem readonly */
+static void btrfs_handle_error(struct btrfs_fs_info *fs_info)
+{
+	struct super_block *sb = fs_info->sb;
+
+	if (sb->s_flags & MS_RDONLY)
+		return;
+
+	if (fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) {
+		sb->s_flags |= MS_RDONLY;
+		printk(KERN_INFO "btrfs is forced readonly\n");
+	}
+}
+
+/*
+ * __btrfs_std_error decodes expected errors from the caller and
+ * invokes the approciate error response.
+ */
+void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function,
+		     unsigned int line, int errno)
+{
+	struct super_block *sb = fs_info->sb;
+	char nbuf[16];
+	const char *errstr;
+
+	/*
+	 * Special case: if the error is EROFS, and we're already
+	 * under MS_RDONLY, then it is safe here.
+	 */
+	if (errno == -EROFS && (sb->s_flags & MS_RDONLY))
+		return;
+
+	errstr = btrfs_decode_error(fs_info, errno, nbuf);
+	printk(KERN_CRIT "BTRFS error (device %s) in %s:%d: %s\n",
+		sb->s_id, function, line, errstr);
+	save_error_info(fs_info);
+
+	btrfs_handle_error(fs_info);
+}
+
 static void btrfs_put_super(struct super_block *sb)
 {
 	struct btrfs_root *root = btrfs_sb(sb);
@@ -69,9 +153,9 @@ enum {
 	Opt_degraded, Opt_subvol, Opt_subvolid, Opt_device, Opt_nodatasum,
 	Opt_nodatacow, Opt_max_inline, Opt_alloc_start, Opt_nobarrier, Opt_ssd,
 	Opt_nossd, Opt_ssd_spread, Opt_thread_pool, Opt_noacl, Opt_compress,
-	Opt_compress_force, Opt_notreelog, Opt_ratio, Opt_flushoncommit,
-	Opt_discard, Opt_space_cache, Opt_clear_cache, Opt_err,
-	Opt_user_subvol_rm_allowed,
+	Opt_compress_type, Opt_compress_force, Opt_compress_force_type,
+	Opt_notreelog, Opt_ratio, Opt_flushoncommit, Opt_discard,
+	Opt_space_cache, Opt_clear_cache, Opt_user_subvol_rm_allowed, Opt_err,
 };
 
 static match_table_t tokens = {
@@ -86,7 +170,9 @@ static match_table_t tokens = {
 	{Opt_alloc_start, "alloc_start=%s"},
 	{Opt_thread_pool, "thread_pool=%d"},
 	{Opt_compress, "compress"},
+	{Opt_compress_type, "compress=%s"},
 	{Opt_compress_force, "compress-force"},
+	{Opt_compress_force_type, "compress-force=%s"},
 	{Opt_ssd, "ssd"},
 	{Opt_ssd_spread, "ssd_spread"},
 	{Opt_nossd, "nossd"},
@@ -112,6 +198,8 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 	char *p, *num, *orig;
 	int intarg;
 	int ret = 0;
+	char *compress_type;
+	bool compress_force = false;
 
 	if (!options)
 		return 0;
@@ -154,14 +242,32 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 			btrfs_set_opt(info->mount_opt, NODATACOW);
 			btrfs_set_opt(info->mount_opt, NODATASUM);
 			break;
-		case Opt_compress:
-			printk(KERN_INFO "btrfs: use compression\n");
-			btrfs_set_opt(info->mount_opt, COMPRESS);
-			break;
 		case Opt_compress_force:
-			printk(KERN_INFO "btrfs: forcing compression\n");
-			btrfs_set_opt(info->mount_opt, FORCE_COMPRESS);
+		case Opt_compress_force_type:
+			compress_force = true;
+		case Opt_compress:
+		case Opt_compress_type:
+			if (token == Opt_compress ||
+			    token == Opt_compress_force ||
+			    strcmp(args[0].from, "zlib") == 0) {
+				compress_type = "zlib";
+				info->compress_type = BTRFS_COMPRESS_ZLIB;
+			} else if (strcmp(args[0].from, "lzo") == 0) {
+				compress_type = "lzo";
+				info->compress_type = BTRFS_COMPRESS_LZO;
+			} else {
+				ret = -EINVAL;
+				goto out;
+			}
+
 			btrfs_set_opt(info->mount_opt, COMPRESS);
+			if (compress_force) {
+				btrfs_set_opt(info->mount_opt, FORCE_COMPRESS);
+				pr_info("btrfs: force %s compression\n",
+					compress_type);
+			} else
+				pr_info("btrfs: use %s compression\n",
+					compress_type);
 			break;
 		case Opt_ssd:
 			printk(KERN_INFO "btrfs: use ssd allocation scheme\n");
@@ -753,6 +859,127 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 	return 0;
 }
 
+/*
+ * The helper to calc the free space on the devices that can be used to store
+ * file data.
+ */
+static int btrfs_calc_avail_data_space(struct btrfs_root *root, u64 *free_bytes)
+{
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct btrfs_device_info *devices_info;
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+	struct btrfs_device *device;
+	u64 skip_space;
+	u64 type;
+	u64 avail_space;
+	u64 used_space;
+	u64 min_stripe_size;
+	int min_stripes = 1;
+	int i = 0, nr_devices;
+	int ret;
+
+	nr_devices = fs_info->fs_devices->rw_devices;
+	BUG_ON(!nr_devices);
+
+	devices_info = kmalloc(sizeof(*devices_info) * nr_devices,
+			       GFP_NOFS);
+	if (!devices_info)
+		return -ENOMEM;
+
+	/* calc min stripe number for data space alloction */
+	type = btrfs_get_alloc_profile(root, 1);
+	if (type & BTRFS_BLOCK_GROUP_RAID0)
+		min_stripes = 2;
+	else if (type & BTRFS_BLOCK_GROUP_RAID1)
+		min_stripes = 2;
+	else if (type & BTRFS_BLOCK_GROUP_RAID10)
+		min_stripes = 4;
+
+	if (type & BTRFS_BLOCK_GROUP_DUP)
+		min_stripe_size = 2 * BTRFS_STRIPE_LEN;
+	else
+		min_stripe_size = BTRFS_STRIPE_LEN;
+
+	list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) {
+		if (!device->in_fs_metadata)
+			continue;
+
+		avail_space = device->total_bytes - device->bytes_used;
+
+		/* align with stripe_len */
+		do_div(avail_space, BTRFS_STRIPE_LEN);
+		avail_space *= BTRFS_STRIPE_LEN;
+
+		/*
+		 * In order to avoid overwritting the superblock on the drive,
+		 * btrfs starts at an offset of at least 1MB when doing chunk
+		 * allocation.
+		 */
+		skip_space = 1024 * 1024;
+
+		/* user can set the offset in fs_info->alloc_start. */
+		if (fs_info->alloc_start + BTRFS_STRIPE_LEN <=
+		    device->total_bytes)
+			skip_space = max(fs_info->alloc_start, skip_space);
+
+		/*
+		 * btrfs can not use the free space in [0, skip_space - 1],
+		 * we must subtract it from the total. In order to implement
+		 * it, we account the used space in this range first.
+		 */
+		ret = btrfs_account_dev_extents_size(device, 0, skip_space - 1,
+						     &used_space);
+		if (ret) {
+			kfree(devices_info);
+			return ret;
+		}
+
+		/* calc the free space in [0, skip_space - 1] */
+		skip_space -= used_space;
+
+		/*
+		 * we can use the free space in [0, skip_space - 1], subtract
+		 * it from the total.
+		 */
+		if (avail_space && avail_space >= skip_space)
+			avail_space -= skip_space;
+		else
+			avail_space = 0;
+
+		if (avail_space < min_stripe_size)
+			continue;
+
+		devices_info[i].dev = device;
+		devices_info[i].max_avail = avail_space;
+
+		i++;
+	}
+
+	nr_devices = i;
+
+	btrfs_descending_sort_devices(devices_info, nr_devices);
+
+	i = nr_devices - 1;
+	avail_space = 0;
+	while (nr_devices >= min_stripes) {
+		if (devices_info[i].max_avail >= min_stripe_size) {
+			int j;
+			u64 alloc_size;
+
+			avail_space += devices_info[i].max_avail * min_stripes;
+			alloc_size = devices_info[i].max_avail;
+			for (j = i + 1 - min_stripes; j <= i; j++)
+				devices_info[j].max_avail -= alloc_size;
+		}
+		i--;
+		nr_devices--;
+	}
+
+	kfree(devices_info);
+	*free_bytes = avail_space;
+	return 0;
+}
+
 static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct btrfs_root *root = btrfs_sb(dentry->d_sb);
@@ -760,17 +987,21 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	struct list_head *head = &root->fs_info->space_info;
 	struct btrfs_space_info *found;
 	u64 total_used = 0;
-	u64 total_used_data = 0;
+	u64 total_free_data = 0;
 	int bits = dentry->d_sb->s_blocksize_bits;
 	__be32 *fsid = (__be32 *)root->fs_info->fsid;
+	int ret;
 
+	/* holding chunk_muext to avoid allocating new chunks */
+	mutex_lock(&root->fs_info->chunk_mutex);
 	rcu_read_lock();
 	list_for_each_entry_rcu(found, head, list) {
-		if (found->flags & (BTRFS_BLOCK_GROUP_METADATA |
-				    BTRFS_BLOCK_GROUP_SYSTEM))
-			total_used_data += found->disk_total;
-		else
-			total_used_data += found->disk_used;
+		if (found->flags & BTRFS_BLOCK_GROUP_DATA) {
+			total_free_data += found->disk_total - found->disk_used;
+			total_free_data -=
+				btrfs_account_ro_block_groups_free_space(found);
+		}
+
 		total_used += found->disk_used;
 	}
 	rcu_read_unlock();
@@ -778,9 +1009,17 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_namelen = BTRFS_NAME_LEN;
 	buf->f_blocks = btrfs_super_total_bytes(disk_super) >> bits;
 	buf->f_bfree = buf->f_blocks - (total_used >> bits);
-	buf->f_bavail = buf->f_blocks - (total_used_data >> bits);
 	buf->f_bsize = dentry->d_sb->s_blocksize;
 	buf->f_type = BTRFS_SUPER_MAGIC;
+	buf->f_bavail = total_free_data;
+	ret = btrfs_calc_avail_data_space(root, &total_free_data);
+	if (ret) {
+		mutex_unlock(&root->fs_info->chunk_mutex);
+		return ret;
+	}
+	buf->f_bavail += total_free_data;
+	buf->f_bavail = buf->f_bavail >> bits;
+	mutex_unlock(&root->fs_info->chunk_mutex);
 
 	/* We treat it as constant endianness (it doesn't matter _which_)
 	   because we want the fsid to come out the same whether mounted
@@ -897,10 +1136,14 @@ static int __init init_btrfs_fs(void)
 	if (err)
 		return err;
 
-	err = btrfs_init_cachep();
+	err = btrfs_init_compress();
 	if (err)
 		goto free_sysfs;
 
+	err = btrfs_init_cachep();
+	if (err)
+		goto free_compress;
+
 	err = extent_io_init();
 	if (err)
 		goto free_cachep;
@@ -928,6 +1171,8 @@ free_extent_io:
 	extent_io_exit();
 free_cachep:
 	btrfs_destroy_cachep();
+free_compress:
+	btrfs_exit_compress();
 free_sysfs:
 	btrfs_exit_sysfs();
 	return err;
@@ -942,7 +1187,7 @@ static void __exit exit_btrfs_fs(void)
 	unregister_filesystem(&btrfs_fs_type);
 	btrfs_exit_sysfs();
 	btrfs_cleanup_fs_uuids();
-	btrfs_zlib_exit();
+	btrfs_exit_compress();
 }
 
 module_init(init_btrfs_fs)

fs/btrfs/transaction.c

@@ -181,6 +181,9 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root,
 	struct btrfs_trans_handle *h;
 	struct btrfs_transaction *cur_trans;
 	int ret;
+
+	if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR)
+		return ERR_PTR(-EROFS);
 again:
 	h = kmem_cache_alloc(btrfs_trans_handle_cachep, GFP_NOFS);
 	if (!h)
@@ -910,6 +913,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	u64 to_reserve = 0;
 	u64 index = 0;
 	u64 objectid;
+	u64 root_flags;
 
 	new_root_item = kmalloc(sizeof(*new_root_item), GFP_NOFS);
 	if (!new_root_item) {
@@ -967,6 +971,13 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	btrfs_set_root_last_snapshot(&root->root_item, trans->transid);
 	memcpy(new_root_item, &root->root_item, sizeof(*new_root_item));
 
+	root_flags = btrfs_root_flags(new_root_item);
+	if (pending->readonly)
+		root_flags |= BTRFS_ROOT_SUBVOL_RDONLY;
+	else
+		root_flags &= ~BTRFS_ROOT_SUBVOL_RDONLY;
+	btrfs_set_root_flags(new_root_item, root_flags);
+
 	old = btrfs_lock_root_node(root);
 	btrfs_cow_block(trans, root, old, NULL, 0, &old);
 	btrfs_set_lock_blocking(old);

fs/btrfs/transaction.h

@@ -62,6 +62,7 @@ struct btrfs_pending_snapshot {
 	struct btrfs_block_rsv block_rsv;
 	/* extra metadata reseration for relocation */
 	int error;
+	bool readonly;
 	struct list_head list;
 };
 

fs/btrfs/volumes.c

@@ -22,6 +22,7 @@
 #include <linux/blkdev.h>
 #include <linux/random.h>
 #include <linux/iocontext.h>
+#include <linux/capability.h>
 #include <asm/div64.h>
 #include "compat.h"
 #include "ctree.h"
@@ -600,8 +601,10 @@ static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		set_blocksize(bdev, 4096);
 
 		bh = btrfs_read_dev_super(bdev);
-		if (!bh)
+		if (!bh) {
+			ret = -EINVAL;
 			goto error_close;
+		}
 
 		disk_super = (struct btrfs_super_block *)bh->b_data;
 		devid = btrfs_stack_device_id(&disk_super->dev_item);
@@ -703,7 +706,7 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 		goto error_close;
 	bh = btrfs_read_dev_super(bdev);
 	if (!bh) {
-		ret = -EIO;
+		ret = -EINVAL;
 		goto error_close;
 	}
 	disk_super = (struct btrfs_super_block *)bh->b_data;
@@ -729,59 +732,167 @@ error:
 	return ret;
 }
 
+/* helper to account the used device space in the range */
+int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
+				   u64 end, u64 *length)
+{
+	struct btrfs_key key;
+	struct btrfs_root *root = device->dev_root;
+	struct btrfs_dev_extent *dev_extent;
+	struct btrfs_path *path;
+	u64 extent_end;
+	int ret;
+	int slot;
+	struct extent_buffer *l;
+
+	*length = 0;
+
+	if (start >= device->total_bytes)
+		return 0;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+	path->reada = 2;
+
+	key.objectid = device->devid;
+	key.offset = start;
+	key.type = BTRFS_DEV_EXTENT_KEY;
+
+	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	if (ret < 0)
+		goto out;
+	if (ret > 0) {
+		ret = btrfs_previous_item(root, path, key.objectid, key.type);
+		if (ret < 0)
+			goto out;
+	}
+
+	while (1) {
+		l = path->nodes[0];
+		slot = path->slots[0];
+		if (slot >= btrfs_header_nritems(l)) {
+			ret = btrfs_next_leaf(root, path);
+			if (ret == 0)
+				continue;
+			if (ret < 0)
+				goto out;
+
+			break;
+		}
+		btrfs_item_key_to_cpu(l, &key, slot);
+
+		if (key.objectid < device->devid)
+			goto next;
+
+		if (key.objectid > device->devid)
+			break;
+
+		if (btrfs_key_type(&key) != BTRFS_DEV_EXTENT_KEY)
+			goto next;
+
+		dev_extent = btrfs_item_ptr(l, slot, struct btrfs_dev_extent);
+		extent_end = key.offset + btrfs_dev_extent_length(l,
+								  dev_extent);
+		if (key.offset <= start && extent_end > end) {
+			*length = end - start + 1;
+			break;
+		} else if (key.offset <= start && extent_end > start)
+			*length += extent_end - start;
+		else if (key.offset > start && extent_end <= end)
+			*length += extent_end - key.offset;
+		else if (key.offset > start && key.offset <= end) {
+			*length += end - key.offset + 1;
+			break;
+		} else if (key.offset > end)
+			break;
+
+next:
+		path->slots[0]++;
+	}
+	ret = 0;
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
 /*
+ * find_free_dev_extent - find free space in the specified device
+ * @trans:	transaction handler
+ * @device:	the device which we search the free space in
+ * @num_bytes:	the size of the free space that we need
+ * @start:	store the start of the free space.
+ * @len:	the size of the free space. that we find, or the size of the max
+ * 		free space if we don't find suitable free space
+ *
  * this uses a pretty simple search, the expectation is that it is
  * called very infrequently and that a given device has a small number
  * of extents
+ *
+ * @start is used to store the start of the free space if we find. But if we
+ * don't find suitable free space, it will be used to store the start position
+ * of the max free space.
+ *
+ * @len is used to store the size of the free space that we find.
+ * But if we don't find suitable free space, it is used to store the size of
+ * the max free space.
  */
 int find_free_dev_extent(struct btrfs_trans_handle *trans,
 			 struct btrfs_device *device, u64 num_bytes,
-			 u64 *start, u64 *max_avail)
+			 u64 *start, u64 *len)
 {
 	struct btrfs_key key;
 	struct btrfs_root *root = device->dev_root;
-	struct btrfs_dev_extent *dev_extent = NULL;
+	struct btrfs_dev_extent *dev_extent;
 	struct btrfs_path *path;
-	u64 hole_size = 0;
-	u64 last_byte = 0;
-	u64 search_start = 0;
+	u64 hole_size;
+	u64 max_hole_start;
+	u64 max_hole_size;
+	u64 extent_end;
+	u64 search_start;
 	u64 search_end = device->total_bytes;
 	int ret;
-	int slot = 0;
-	int start_found;
+	int slot;
 	struct extent_buffer *l;
 
-	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
-	path->reada = 2;
-	start_found = 0;
-
 	/* FIXME use last free of some kind */
 
 	/* we don't want to overwrite the superblock on the drive,
 	 * so we make sure to start at an offset of at least 1MB
 	 */
-	search_start = max((u64)1024 * 1024, search_start);
+	search_start = 1024 * 1024;
 
-	if (root->fs_info->alloc_start + num_bytes <= device->total_bytes)
+	if (root->fs_info->alloc_start + num_bytes <= search_end)
 		search_start = max(root->fs_info->alloc_start, search_start);
 
+	max_hole_start = search_start;
+	max_hole_size = 0;
+
+	if (search_start >= search_end) {
+		ret = -ENOSPC;
+		goto error;
+	}
+
+	path = btrfs_alloc_path();
+	if (!path) {
+		ret = -ENOMEM;
+		goto error;
+	}
+	path->reada = 2;
+
 	key.objectid = device->devid;
 	key.offset = search_start;
 	key.type = BTRFS_DEV_EXTENT_KEY;
+
 	ret = btrfs_search_slot(trans, root, &key, path, 0, 0);
 	if (ret < 0)
-		goto error;
+		goto out;
 	if (ret > 0) {
 		ret = btrfs_previous_item(root, path, key.objectid, key.type);
 		if (ret < 0)
-			goto error;
-		if (ret > 0)
-			start_found = 1;
+			goto out;
 	}
-	l = path->nodes[0];
-	btrfs_item_key_to_cpu(l, &key, path->slots[0]);
+
 	while (1) {
 		l = path->nodes[0];
 		slot = path->slots[0];
@@ -790,24 +901,9 @@ int find_free_dev_extent(struct btrfs_trans_handle *trans,
 			if (ret == 0)
 				continue;
 			if (ret < 0)
-				goto error;
-no_more_items:
-			if (!start_found) {
-				if (search_start >= search_end) {
-					ret = -ENOSPC;
-					goto error;
-				}
-				*start = search_start;
-				start_found = 1;
-				goto check_pending;
-			}
-			*start = last_byte > search_start ?
-				last_byte : search_start;
-			if (search_end <= *start) {
-				ret = -ENOSPC;
-				goto error;
-			}
-			goto check_pending;
+				goto out;
+
+			break;
 		}
 		btrfs_item_key_to_cpu(l, &key, slot);
 
@@ -815,48 +911,62 @@ no_more_items:
 			goto next;
 
 		if (key.objectid > device->devid)
-			goto no_more_items;
+			break;
 
-		if (key.offset >= search_start && key.offset > last_byte &&
-		    start_found) {
-			if (last_byte < search_start)
-				last_byte = search_start;
-			hole_size = key.offset - last_byte;
+		if (btrfs_key_type(&key) != BTRFS_DEV_EXTENT_KEY)
+			goto next;
 
-			if (hole_size > *max_avail)
-				*max_avail = hole_size;
+		if (key.offset > search_start) {
+			hole_size = key.offset - search_start;
 
-			if (key.offset > last_byte &&
-			    hole_size >= num_bytes) {
-				*start = last_byte;
-				goto check_pending;
+			if (hole_size > max_hole_size) {
+				max_hole_start = search_start;
+				max_hole_size = hole_size;
+			}
+
+			/*
+			 * If this free space is greater than which we need,
+			 * it must be the max free space that we have found
+			 * until now, so max_hole_start must point to the start
+			 * of this free space and the length of this free space
+			 * is stored in max_hole_size. Thus, we return
+			 * max_hole_start and max_hole_size and go back to the
+			 * caller.
+			 */
+			if (hole_size >= num_bytes) {
+				ret = 0;
+				goto out;
 			}
 		}
-		if (btrfs_key_type(&key) != BTRFS_DEV_EXTENT_KEY)
-			goto next;
 
-		start_found = 1;
 		dev_extent = btrfs_item_ptr(l, slot, struct btrfs_dev_extent);
-		last_byte = key.offset + btrfs_dev_extent_length(l, dev_extent);
+		extent_end = key.offset + btrfs_dev_extent_length(l,
+								  dev_extent);
+		if (extent_end > search_start)
+			search_start = extent_end;
 next:
 		path->slots[0]++;
 		cond_resched();
 	}
-check_pending:
-	/* we have to make sure we didn't find an extent that has already
-	 * been allocated by the map tree or the original allocation
-	 */
-	BUG_ON(*start < search_start);
 
-	if (*start + num_bytes > search_end) {
-		ret = -ENOSPC;
-		goto error;
+	hole_size = search_end- search_start;
+	if (hole_size > max_hole_size) {
+		max_hole_start = search_start;
+		max_hole_size = hole_size;
 	}
-	/* check for pending inserts here */
-	ret = 0;
 
-error:
+	/* See above. */
+	if (hole_size < num_bytes)
+		ret = -ENOSPC;
+	else
+		ret = 0;
+
+out:
 	btrfs_free_path(path);
+error:
+	*start = max_hole_start;
+	if (len)
+		*len = max_hole_size;
 	return ret;
 }
 
@@ -1196,7 +1306,7 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
 		set_blocksize(bdev, 4096);
 		bh = btrfs_read_dev_super(bdev);
 		if (!bh) {
-			ret = -EIO;
+			ret = -EINVAL;
 			goto error_close;
 		}
 		disk_super = (struct btrfs_super_block *)bh->b_data;
@@ -1916,6 +2026,9 @@ int btrfs_balance(struct btrfs_root *dev_root)
 	if (dev_root->fs_info->sb->s_flags & MS_RDONLY)
 		return -EROFS;
 
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
 	mutex_lock(&dev_root->fs_info->volume_mutex);
 	dev_root = dev_root->fs_info->dev_root;
 
@@ -2154,66 +2267,67 @@ static noinline u64 chunk_bytes_by_type(u64 type, u64 calc_size,
 		return calc_size * num_stripes;
 }
 
-static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *extent_root,
-			       struct map_lookup **map_ret,
-			       u64 *num_bytes, u64 *stripe_size,
-			       u64 start, u64 type)
+/* Used to sort the devices by max_avail(descending sort) */
+int btrfs_cmp_device_free_bytes(const void *dev_info1, const void *dev_info2)
 {
-	struct btrfs_fs_info *info = extent_root->fs_info;
-	struct btrfs_device *device = NULL;
-	struct btrfs_fs_devices *fs_devices = info->fs_devices;
-	struct list_head *cur;
-	struct map_lookup *map = NULL;
-	struct extent_map_tree *em_tree;
-	struct extent_map *em;
-	struct list_head private_devs;
-	int min_stripe_size = 1 * 1024 * 1024;
-	u64 calc_size = 1024 * 1024 * 1024;
-	u64 max_chunk_size = calc_size;
-	u64 min_free;
-	u64 avail;
-	u64 max_avail = 0;
-	u64 dev_offset;
-	int num_stripes = 1;
-	int min_stripes = 1;
-	int sub_stripes = 0;
-	int looped = 0;
-	int ret;
-	int index;
-	int stripe_len = 64 * 1024;
+	if (((struct btrfs_device_info *)dev_info1)->max_avail >
+	    ((struct btrfs_device_info *)dev_info2)->max_avail)
+		return -1;
+	else if (((struct btrfs_device_info *)dev_info1)->max_avail <
+		 ((struct btrfs_device_info *)dev_info2)->max_avail)
+		return 1;
+	else
+		return 0;
+}
 
-	if ((type & BTRFS_BLOCK_GROUP_RAID1) &&
-	    (type & BTRFS_BLOCK_GROUP_DUP)) {
-		WARN_ON(1);
-		type &= ~BTRFS_BLOCK_GROUP_DUP;
-	}
-	if (list_empty(&fs_devices->alloc_list))
-		return -ENOSPC;
+static int __btrfs_calc_nstripes(struct btrfs_fs_devices *fs_devices, u64 type,
+				 int *num_stripes, int *min_stripes,
+				 int *sub_stripes)
+{
+	*num_stripes = 1;
+	*min_stripes = 1;
+	*sub_stripes = 0;
 
 	if (type & (BTRFS_BLOCK_GROUP_RAID0)) {
-		num_stripes = fs_devices->rw_devices;
-		min_stripes = 2;
+		*num_stripes = fs_devices->rw_devices;
+		*min_stripes = 2;
 	}
 	if (type & (BTRFS_BLOCK_GROUP_DUP)) {
-		num_stripes = 2;
-		min_stripes = 2;
+		*num_stripes = 2;
+		*min_stripes = 2;
 	}
 	if (type & (BTRFS_BLOCK_GROUP_RAID1)) {
 		if (fs_devices->rw_devices < 2)
 			return -ENOSPC;
-		num_stripes = 2;
-		min_stripes = 2;
+		*num_stripes = 2;
+		*min_stripes = 2;
 	}
 	if (type & (BTRFS_BLOCK_GROUP_RAID10)) {
-		num_stripes = fs_devices->rw_devices;
-		if (num_stripes < 4)
+		*num_stripes = fs_devices->rw_devices;
+		if (*num_stripes < 4)
 			return -ENOSPC;
-		num_stripes &= ~(u32)1;
-		sub_stripes = 2;
-		min_stripes = 4;
+		*num_stripes &= ~(u32)1;
+		*sub_stripes = 2;
+		*min_stripes = 4;
 	}
 
+	return 0;
+}
+
+static u64 __btrfs_calc_stripe_size(struct btrfs_fs_devices *fs_devices,
+				    u64 proposed_size, u64 type,
+				    int num_stripes, int small_stripe)
+{
+	int min_stripe_size = 1 * 1024 * 1024;
+	u64 calc_size = proposed_size;
+	u64 max_chunk_size = calc_size;
+	int ncopies = 1;
+
+	if (type & (BTRFS_BLOCK_GROUP_RAID1 |
+		    BTRFS_BLOCK_GROUP_DUP |
+		    BTRFS_BLOCK_GROUP_RAID10))
+		ncopies = 2;
+
 	if (type & BTRFS_BLOCK_GROUP_DATA) {
 		max_chunk_size = 10 * calc_size;
 		min_stripe_size = 64 * 1024 * 1024;
@@ -2230,51 +2344,209 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	max_chunk_size = min(div_factor(fs_devices->total_rw_bytes, 1),
 			     max_chunk_size);
 
-again:
-	max_avail = 0;
-	if (!map || map->num_stripes != num_stripes) {
-		kfree(map);
-		map = kmalloc(map_lookup_size(num_stripes), GFP_NOFS);
-		if (!map)
-			return -ENOMEM;
-		map->num_stripes = num_stripes;
-	}
-
-	if (calc_size * num_stripes > max_chunk_size) {
-		calc_size = max_chunk_size;
+	if (calc_size * num_stripes > max_chunk_size * ncopies) {
+		calc_size = max_chunk_size * ncopies;
 		do_div(calc_size, num_stripes);
-		do_div(calc_size, stripe_len);
-		calc_size *= stripe_len;
+		do_div(calc_size, BTRFS_STRIPE_LEN);
+		calc_size *= BTRFS_STRIPE_LEN;
 	}
 
 	/* we don't want tiny stripes */
-	if (!looped)
+	if (!small_stripe)
 		calc_size = max_t(u64, min_stripe_size, calc_size);
 
 	/*
-	 * we're about to do_div by the stripe_len so lets make sure
+	 * we're about to do_div by the BTRFS_STRIPE_LEN so lets make sure
 	 * we end up with something bigger than a stripe
 	 */
-	calc_size = max_t(u64, calc_size, stripe_len * 4);
+	calc_size = max_t(u64, calc_size, BTRFS_STRIPE_LEN);
+
+	do_div(calc_size, BTRFS_STRIPE_LEN);
+	calc_size *= BTRFS_STRIPE_LEN;
+
+	return calc_size;
+}
+
+static struct map_lookup *__shrink_map_lookup_stripes(struct map_lookup *map,
+						      int num_stripes)
+{
+	struct map_lookup *new;
+	size_t len = map_lookup_size(num_stripes);
+
+	BUG_ON(map->num_stripes < num_stripes);
+
+	if (map->num_stripes == num_stripes)
+		return map;
+
+	new = kmalloc(len, GFP_NOFS);
+	if (!new) {
+		/* just change map->num_stripes */
+		map->num_stripes = num_stripes;
+		return map;
+	}
+
+	memcpy(new, map, len);
+	new->num_stripes = num_stripes;
+	kfree(map);
+	return new;
+}
+
+/*
+ * helper to allocate device space from btrfs_device_info, in which we stored
+ * max free space information of every device. It is used when we can not
+ * allocate chunks by default size.
+ *
+ * By this helper, we can allocate a new chunk as larger as possible.
+ */
+static int __btrfs_alloc_tiny_space(struct btrfs_trans_handle *trans,
+				    struct btrfs_fs_devices *fs_devices,
+				    struct btrfs_device_info *devices,
+				    int nr_device, u64 type,
+				    struct map_lookup **map_lookup,
+				    int min_stripes, u64 *stripe_size)
+{
+	int i, index, sort_again = 0;
+	int min_devices = min_stripes;
+	u64 max_avail, min_free;
+	struct map_lookup *map = *map_lookup;
+	int ret;
+
+	if (nr_device < min_stripes)
+		return -ENOSPC;
+
+	btrfs_descending_sort_devices(devices, nr_device);
+
+	max_avail = devices[0].max_avail;
+	if (!max_avail)
+		return -ENOSPC;
+
+	for (i = 0; i < nr_device; i++) {
+		/*
+		 * if dev_offset = 0, it means the free space of this device
+		 * is less than what we need, and we didn't search max avail
+		 * extent on this device, so do it now.
+		 */
+		if (!devices[i].dev_offset) {
+			ret = find_free_dev_extent(trans, devices[i].dev,
+						   max_avail,
+						   &devices[i].dev_offset,
+						   &devices[i].max_avail);
+			if (ret != 0 && ret != -ENOSPC)
+				return ret;
+			sort_again = 1;
+		}
+	}
+
+	/* we update the max avail free extent of each devices, sort again */
+	if (sort_again)
+		btrfs_descending_sort_devices(devices, nr_device);
+
+	if (type & BTRFS_BLOCK_GROUP_DUP)
+		min_devices = 1;
+
+	if (!devices[min_devices - 1].max_avail)
+		return -ENOSPC;
+
+	max_avail = devices[min_devices - 1].max_avail;
+	if (type & BTRFS_BLOCK_GROUP_DUP)
+		do_div(max_avail, 2);
+
+	max_avail = __btrfs_calc_stripe_size(fs_devices, max_avail, type,
+					     min_stripes, 1);
+	if (type & BTRFS_BLOCK_GROUP_DUP)
+		min_free = max_avail * 2;
+	else
+		min_free = max_avail;
+
+	if (min_free > devices[min_devices - 1].max_avail)
+		return -ENOSPC;
+
+	map = __shrink_map_lookup_stripes(map, min_stripes);
+	*stripe_size = max_avail;
+
+	index = 0;
+	for (i = 0; i < min_stripes; i++) {
+		map->stripes[i].dev = devices[index].dev;
+		map->stripes[i].physical = devices[index].dev_offset;
+		if (type & BTRFS_BLOCK_GROUP_DUP) {
+			i++;
+			map->stripes[i].dev = devices[index].dev;
+			map->stripes[i].physical = devices[index].dev_offset +
+						   max_avail;
+		}
+		index++;
+	}
+	*map_lookup = map;
+
+	return 0;
+}
 
-	do_div(calc_size, stripe_len);
-	calc_size *= stripe_len;
+static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
+			       struct btrfs_root *extent_root,
+			       struct map_lookup **map_ret,
+			       u64 *num_bytes, u64 *stripe_size,
+			       u64 start, u64 type)
+{
+	struct btrfs_fs_info *info = extent_root->fs_info;
+	struct btrfs_device *device = NULL;
+	struct btrfs_fs_devices *fs_devices = info->fs_devices;
+	struct list_head *cur;
+	struct map_lookup *map;
+	struct extent_map_tree *em_tree;
+	struct extent_map *em;
+	struct btrfs_device_info *devices_info;
+	struct list_head private_devs;
+	u64 calc_size = 1024 * 1024 * 1024;
+	u64 min_free;
+	u64 avail;
+	u64 dev_offset;
+	int num_stripes;
+	int min_stripes;
+	int sub_stripes;
+	int min_devices;	/* the min number of devices we need */
+	int i;
+	int ret;
+	int index;
+
+	if ((type & BTRFS_BLOCK_GROUP_RAID1) &&
+	    (type & BTRFS_BLOCK_GROUP_DUP)) {
+		WARN_ON(1);
+		type &= ~BTRFS_BLOCK_GROUP_DUP;
+	}
+	if (list_empty(&fs_devices->alloc_list))
+		return -ENOSPC;
+
+	ret = __btrfs_calc_nstripes(fs_devices, type, &num_stripes,
+				    &min_stripes, &sub_stripes);
+	if (ret)
+		return ret;
+
+	devices_info = kzalloc(sizeof(*devices_info) * fs_devices->rw_devices,
+			       GFP_NOFS);
+	if (!devices_info)
+		return -ENOMEM;
+
+	map = kmalloc(map_lookup_size(num_stripes), GFP_NOFS);
+	if (!map) {
+		ret = -ENOMEM;
+		goto error;
+	}
+	map->num_stripes = num_stripes;
 
 	cur = fs_devices->alloc_list.next;
 	index = 0;
+	i = 0;
 
-	if (type & BTRFS_BLOCK_GROUP_DUP)
+	calc_size = __btrfs_calc_stripe_size(fs_devices, calc_size, type,
+					     num_stripes, 0);
+
+	if (type & BTRFS_BLOCK_GROUP_DUP) {
 		min_free = calc_size * 2;
-	else
+		min_devices = 1;
+	} else {
 		min_free = calc_size;
-
-	/*
-	 * we add 1MB because we never use the first 1MB of the device, unless
-	 * we've looped, then we are likely allocating the maximum amount of
-	 * space left already
-	 */
-	if (!looped)
-		min_free += 1024 * 1024;
+		min_devices = min_stripes;
+	}
 
 	INIT_LIST_HEAD(&private_devs);
 	while (index < num_stripes) {
@@ -2287,27 +2559,39 @@ again:
 		cur = cur->next;
 
 		if (device->in_fs_metadata && avail >= min_free) {
-			ret = find_free_dev_extent(trans, device,
-						   min_free, &dev_offset,
-						   &max_avail);
+			ret = find_free_dev_extent(trans, device, min_free,
+						   &devices_info[i].dev_offset,
+						   &devices_info[i].max_avail);
 			if (ret == 0) {
 				list_move_tail(&device->dev_alloc_list,
 					       &private_devs);
 				map->stripes[index].dev = device;
-				map->stripes[index].physical = dev_offset;
+				map->stripes[index].physical =
+						devices_info[i].dev_offset;
 				index++;
 				if (type & BTRFS_BLOCK_GROUP_DUP) {
 					map->stripes[index].dev = device;
 					map->stripes[index].physical =
-						dev_offset + calc_size;
+						devices_info[i].dev_offset +
+						calc_size;
 					index++;
 				}
-			}
-		} else if (device->in_fs_metadata && avail > max_avail)
-			max_avail = avail;
+			} else if (ret != -ENOSPC)
+				goto error;
+
+			devices_info[i].dev = device;
+			i++;
+		} else if (device->in_fs_metadata &&
+			   avail >= BTRFS_STRIPE_LEN) {
+			devices_info[i].dev = device;
+			devices_info[i].max_avail = avail;
+			i++;
+		}
+
 		if (cur == &fs_devices->alloc_list)
 			break;
 	}
+
 	list_splice(&private_devs, &fs_devices->alloc_list);
 	if (index < num_stripes) {
 		if (index >= min_stripes) {
@@ -2316,34 +2600,36 @@ again:
 				num_stripes /= sub_stripes;
 				num_stripes *= sub_stripes;
 			}
-			looped = 1;
-			goto again;
-		}
-		if (!looped && max_avail > 0) {
-			looped = 1;
-			calc_size = max_avail;
-			goto again;
+
+			map = __shrink_map_lookup_stripes(map, num_stripes);
+		} else if (i >= min_devices) {
+			ret = __btrfs_alloc_tiny_space(trans, fs_devices,
+						       devices_info, i, type,
+						       &map, min_stripes,
+						       &calc_size);
+			if (ret)
+				goto error;
+		} else {
+			ret = -ENOSPC;
+			goto error;
 		}
-		kfree(map);
-		return -ENOSPC;
 	}
 	map->sector_size = extent_root->sectorsize;
-	map->stripe_len = stripe_len;
-	map->io_align = stripe_len;
-	map->io_width = stripe_len;
+	map->stripe_len = BTRFS_STRIPE_LEN;
+	map->io_align = BTRFS_STRIPE_LEN;
+	map->io_width = BTRFS_STRIPE_LEN;
 	map->type = type;
-	map->num_stripes = num_stripes;
 	map->sub_stripes = sub_stripes;
 
 	*map_ret = map;
 	*stripe_size = calc_size;
 	*num_bytes = chunk_bytes_by_type(type, calc_size,
-					 num_stripes, sub_stripes);
+					 map->num_stripes, sub_stripes);
 
 	em = alloc_extent_map(GFP_NOFS);
 	if (!em) {
-		kfree(map);
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto error;
 	}
 	em->bdev = (struct block_device *)map;
 	em->start = start;
@@ -2376,7 +2662,13 @@ again:
 		index++;
 	}
 
+	kfree(devices_info);
 	return 0;
+
+error:
+	kfree(map);
+	kfree(devices_info);
+	return ret;
 }
 
 static int __finish_chunk_alloc(struct btrfs_trans_handle *trans,

fs/btrfs/volumes.h

@@ -20,8 +20,11 @@
 #define __BTRFS_VOLUMES_
 
 #include <linux/bio.h>
+#include <linux/sort.h>
 #include "async-thread.h"
 
+#define BTRFS_STRIPE_LEN	(64 * 1024)
+
 struct buffer_head;
 struct btrfs_pending_bios {
 	struct bio *head;
@@ -136,6 +139,30 @@ struct btrfs_multi_bio {
 	struct btrfs_bio_stripe stripes[];
 };
 
+struct btrfs_device_info {
+	struct btrfs_device *dev;
+	u64 dev_offset;
+	u64 max_avail;
+};
+
+/* Used to sort the devices by max_avail(descending sort) */
+int btrfs_cmp_device_free_bytes(const void *dev_info1, const void *dev_info2);
+
+/*
+ * sort the devices by max_avail, in which max free extent size of each device
+ * is stored.(Descending Sort)
+ */
+static inline void btrfs_descending_sort_devices(
+					struct btrfs_device_info *devices,
+					size_t nr_devices)
+{
+	sort(devices, nr_devices, sizeof(struct btrfs_device_info),
+	     btrfs_cmp_device_free_bytes, NULL);
+}
+
+int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
+				   u64 end, u64 *length);
+
 #define btrfs_multi_bio_size(n) (sizeof(struct btrfs_multi_bio) + \
 			    (sizeof(struct btrfs_bio_stripe) * (n)))
 

fs/btrfs/xattr.c

@@ -316,6 +316,15 @@ ssize_t btrfs_getxattr(struct dentry *dentry, const char *name,
 int btrfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 		   size_t size, int flags)
 {
+	struct btrfs_root *root = BTRFS_I(dentry->d_inode)->root;
+
+	/*
+	 * The permission on security.* and system.* is not checked
+	 * in permission().
+	 */
+	if (btrfs_root_readonly(root))
+		return -EROFS;
+
 	/*
 	 * If this is a request for a synthetic attribute in the system.*
 	 * namespace use the generic infrastructure to resolve a handler
@@ -336,6 +345,15 @@ int btrfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 
 int btrfs_removexattr(struct dentry *dentry, const char *name)
 {
+	struct btrfs_root *root = BTRFS_I(dentry->d_inode)->root;
+
+	/*
+	 * The permission on security.* and system.* is not checked
+	 * in permission().
+	 */
+	if (btrfs_root_readonly(root))
+		return -EROFS;
+
 	/*
 	 * If this is a request for a synthetic attribute in the system.*
 	 * namespace use the generic infrastructure to resolve a handler

fs/btrfs/zlib.c

@@ -32,15 +32,6 @@
 #include <linux/bio.h>
 #include "compression.h"
 
-/* Plan: call deflate() with avail_in == *sourcelen,
-	avail_out = *dstlen - 12 and flush == Z_FINISH.
-	If it doesn't manage to finish,	call it again with
-	avail_in == 0 and avail_out set to the remaining 12
-	bytes for it to clean up.
-   Q: Is 12 bytes sufficient?
-*/
-#define STREAM_END_SPACE 12
-
 struct workspace {
 	z_stream inf_strm;
 	z_stream def_strm;
@@ -48,152 +39,51 @@ struct workspace {
 	struct list_head list;
 };
 
-static LIST_HEAD(idle_workspace);
-static DEFINE_SPINLOCK(workspace_lock);
-static unsigned long num_workspace;
-static atomic_t alloc_workspace = ATOMIC_INIT(0);
-static DECLARE_WAIT_QUEUE_HEAD(workspace_wait);
+static void zlib_free_workspace(struct list_head *ws)
+{
+	struct workspace *workspace = list_entry(ws, struct workspace, list);
 
-/*
- * this finds an available zlib workspace or allocates a new one
- * NULL or an ERR_PTR is returned if things go bad.
- */
-static struct workspace *find_zlib_workspace(void)
+	vfree(workspace->def_strm.workspace);
+	vfree(workspace->inf_strm.workspace);
+	kfree(workspace->buf);
+	kfree(workspace);
+}
+
+static struct list_head *zlib_alloc_workspace(void)
 {
 	struct workspace *workspace;
-	int ret;
-	int cpus = num_online_cpus();
-
-again:
-	spin_lock(&workspace_lock);
-	if (!list_empty(&idle_workspace)) {
-		workspace = list_entry(idle_workspace.next, struct workspace,
-				       list);
-		list_del(&workspace->list);
-		num_workspace--;
-		spin_unlock(&workspace_lock);
-		return workspace;
 
-	}
-	spin_unlock(&workspace_lock);
-	if (atomic_read(&alloc_workspace) > cpus) {
-		DEFINE_WAIT(wait);
-		prepare_to_wait(&workspace_wait, &wait, TASK_UNINTERRUPTIBLE);
-		if (atomic_read(&alloc_workspace) > cpus)
-			schedule();
-		finish_wait(&workspace_wait, &wait);
-		goto again;
-	}
-	atomic_inc(&alloc_workspace);
 	workspace = kzalloc(sizeof(*workspace), GFP_NOFS);
-	if (!workspace) {
-		ret = -ENOMEM;
-		goto fail;
-	}
+	if (!workspace)
+		return ERR_PTR(-ENOMEM);
 
 	workspace->def_strm.workspace = vmalloc(zlib_deflate_workspacesize());
-	if (!workspace->def_strm.workspace) {
-		ret = -ENOMEM;
-		goto fail;
-	}
 	workspace->inf_strm.workspace = vmalloc(zlib_inflate_workspacesize());
-	if (!workspace->inf_strm.workspace) {
-		ret = -ENOMEM;
-		goto fail_inflate;
-	}
 	workspace->buf = kmalloc(PAGE_CACHE_SIZE, GFP_NOFS);
-	if (!workspace->buf) {
-		ret = -ENOMEM;
-		goto fail_kmalloc;
-	}
-	return workspace;
-
-fail_kmalloc:
-	vfree(workspace->inf_strm.workspace);
-fail_inflate:
-	vfree(workspace->def_strm.workspace);
-fail:
-	kfree(workspace);
-	atomic_dec(&alloc_workspace);
-	wake_up(&workspace_wait);
-	return ERR_PTR(ret);
-}
-
-/*
- * put a workspace struct back on the list or free it if we have enough
- * idle ones sitting around
- */
-static int free_workspace(struct workspace *workspace)
-{
-	spin_lock(&workspace_lock);
-	if (num_workspace < num_online_cpus()) {
-		list_add_tail(&workspace->list, &idle_workspace);
-		num_workspace++;
-		spin_unlock(&workspace_lock);
-		if (waitqueue_active(&workspace_wait))
-			wake_up(&workspace_wait);
-		return 0;
-	}
-	spin_unlock(&workspace_lock);
-	vfree(workspace->def_strm.workspace);
-	vfree(workspace->inf_strm.workspace);
-	kfree(workspace->buf);
-	kfree(workspace);
+	if (!workspace->def_strm.workspace ||
+	    !workspace->inf_strm.workspace || !workspace->buf)
+		goto fail;
 
-	atomic_dec(&alloc_workspace);
-	if (waitqueue_active(&workspace_wait))
-		wake_up(&workspace_wait);
-	return 0;
-}
+	INIT_LIST_HEAD(&workspace->list);
 
-/*
- * cleanup function for module exit
- */
-static void free_workspaces(void)
-{
-	struct workspace *workspace;
-	while (!list_empty(&idle_workspace)) {
-		workspace = list_entry(idle_workspace.next, struct workspace,
-				       list);
-		list_del(&workspace->list);
-		vfree(workspace->def_strm.workspace);
-		vfree(workspace->inf_strm.workspace);
-		kfree(workspace->buf);
-		kfree(workspace);
-		atomic_dec(&alloc_workspace);
-	}
+	return &workspace->list;
+fail:
+	zlib_free_workspace(&workspace->list);
+	return ERR_PTR(-ENOMEM);
 }
 
-/*
- * given an address space and start/len, compress the bytes.
- *
- * pages are allocated to hold the compressed result and stored
- * in 'pages'
- *
- * out_pages is used to return the number of pages allocated.  There
- * may be pages allocated even if we return an error
- *
- * total_in is used to return the number of bytes actually read.  It
- * may be smaller then len if we had to exit early because we
- * ran out of room in the pages array or because we cross the
- * max_out threshold.
- *
- * total_out is used to return the total number of compressed bytes
- *
- * max_out tells us the max number of bytes that we're allowed to
- * stuff into pages
- */
-int btrfs_zlib_compress_pages(struct address_space *mapping,
-			      u64 start, unsigned long len,
-			      struct page **pages,
-			      unsigned long nr_dest_pages,
-			      unsigned long *out_pages,
-			      unsigned long *total_in,
-			      unsigned long *total_out,
-			      unsigned long max_out)
+static int zlib_compress_pages(struct list_head *ws,
+			       struct address_space *mapping,
+			       u64 start, unsigned long len,
+			       struct page **pages,
+			       unsigned long nr_dest_pages,
+			       unsigned long *out_pages,
+			       unsigned long *total_in,
+			       unsigned long *total_out,
+			       unsigned long max_out)
 {
+	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	int ret;
-	struct workspace *workspace;
 	char *data_in;
 	char *cpage_out;
 	int nr_pages = 0;
@@ -205,10 +95,6 @@ int btrfs_zlib_compress_pages(struct address_space *mapping,
 	*total_out = 0;
 	*total_in = 0;
 
-	workspace = find_zlib_workspace();
-	if (IS_ERR(workspace))
-		return -1;
-
 	if (Z_OK != zlib_deflateInit(&workspace->def_strm, 3)) {
 		printk(KERN_WARNING "deflateInit failed\n");
 		ret = -1;
@@ -222,6 +108,10 @@ int btrfs_zlib_compress_pages(struct address_space *mapping,
 	data_in = kmap(in_page);
 
 	out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+	if (out_page == NULL) {
+		ret = -1;
+		goto out;
+	}
 	cpage_out = kmap(out_page);
 	pages[0] = out_page;
 	nr_pages = 1;
@@ -260,6 +150,10 @@ int btrfs_zlib_compress_pages(struct address_space *mapping,
 				goto out;
 			}
 			out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+			if (out_page == NULL) {
+				ret = -1;
+				goto out;
+			}
 			cpage_out = kmap(out_page);
 			pages[nr_pages] = out_page;
 			nr_pages++;
@@ -314,55 +208,26 @@ out:
 		kunmap(in_page);
 		page_cache_release(in_page);
 	}
-	free_workspace(workspace);
 	return ret;
 }
 
-/*
- * pages_in is an array of pages with compressed data.
- *
- * disk_start is the starting logical offset of this array in the file
- *
- * bvec is a bio_vec of pages from the file that we want to decompress into
- *
- * vcnt is the count of pages in the biovec
- *
- * srclen is the number of bytes in pages_in
- *
- * The basic idea is that we have a bio that was created by readpages.
- * The pages in the bio are for the uncompressed data, and they may not
- * be contiguous.  They all correspond to the range of bytes covered by
- * the compressed extent.
- */
-int btrfs_zlib_decompress_biovec(struct page **pages_in,
-			      u64 disk_start,
-			      struct bio_vec *bvec,
-			      int vcnt,
-			      size_t srclen)
+static int zlib_decompress_biovec(struct list_head *ws, struct page **pages_in,
+				  u64 disk_start,
+				  struct bio_vec *bvec,
+				  int vcnt,
+				  size_t srclen)
 {
-	int ret = 0;
+	struct workspace *workspace = list_entry(ws, struct workspace, list);
+	int ret = 0, ret2;
 	int wbits = MAX_WBITS;
-	struct workspace *workspace;
 	char *data_in;
 	size_t total_out = 0;
-	unsigned long page_bytes_left;
 	unsigned long page_in_index = 0;
 	unsigned long page_out_index = 0;
-	struct page *page_out;
 	unsigned long total_pages_in = (srclen + PAGE_CACHE_SIZE - 1) /
 					PAGE_CACHE_SIZE;
 	unsigned long buf_start;
-	unsigned long buf_offset;
-	unsigned long bytes;
-	unsigned long working_bytes;
 	unsigned long pg_offset;
-	unsigned long start_byte;
-	unsigned long current_buf_start;
-	char *kaddr;
-
-	workspace = find_zlib_workspace();
-	if (IS_ERR(workspace))
-		return -ENOMEM;
 
 	data_in = kmap(pages_in[page_in_index]);
 	workspace->inf_strm.next_in = data_in;
@@ -372,8 +237,6 @@ int btrfs_zlib_decompress_biovec(struct page **pages_in,
 	workspace->inf_strm.total_out = 0;
 	workspace->inf_strm.next_out = workspace->buf;
 	workspace->inf_strm.avail_out = PAGE_CACHE_SIZE;
-	page_out = bvec[page_out_index].bv_page;
-	page_bytes_left = PAGE_CACHE_SIZE;
 	pg_offset = 0;
 
 	/* If it's deflate, and it's got no preset dictionary, then
@@ -389,107 +252,29 @@ int btrfs_zlib_decompress_biovec(struct page **pages_in,
 
 	if (Z_OK != zlib_inflateInit2(&workspace->inf_strm, wbits)) {
 		printk(KERN_WARNING "inflateInit failed\n");
-		ret = -1;
-		goto out;
+		return -1;
 	}
 	while (workspace->inf_strm.total_in < srclen) {
 		ret = zlib_inflate(&workspace->inf_strm, Z_NO_FLUSH);
 		if (ret != Z_OK && ret != Z_STREAM_END)
 			break;
-		/*
-		 * buf start is the byte offset we're of the start of
-		 * our workspace buffer
-		 */
-		buf_start = total_out;
 
-		/* total_out is the last byte of the workspace buffer */
+		buf_start = total_out;
 		total_out = workspace->inf_strm.total_out;
 
-		working_bytes = total_out - buf_start;
-
-		/*
-		 * start byte is the first byte of the page we're currently
-		 * copying into relative to the start of the compressed data.
-		 */
-		start_byte = page_offset(page_out) - disk_start;
-
-		if (working_bytes == 0) {
-			/* we didn't make progress in this inflate
-			 * call, we're done
-			 */
-			if (ret != Z_STREAM_END)
-				ret = -1;
+		/* we didn't make progress in this inflate call, we're done */
+		if (buf_start == total_out)
 			break;
-		}
 
-		/* we haven't yet hit data corresponding to this page */
-		if (total_out <= start_byte)
-			goto next;
-
-		/*
-		 * the start of the data we care about is offset into
-		 * the middle of our working buffer
-		 */
-		if (total_out > start_byte && buf_start < start_byte) {
-			buf_offset = start_byte - buf_start;
-			working_bytes -= buf_offset;
-		} else {
-			buf_offset = 0;
-		}
-		current_buf_start = buf_start;
-
-		/* copy bytes from the working buffer into the pages */
-		while (working_bytes > 0) {
-			bytes = min(PAGE_CACHE_SIZE - pg_offset,
-				    PAGE_CACHE_SIZE - buf_offset);
-			bytes = min(bytes, working_bytes);
-			kaddr = kmap_atomic(page_out, KM_USER0);
-			memcpy(kaddr + pg_offset, workspace->buf + buf_offset,
-			       bytes);
-			kunmap_atomic(kaddr, KM_USER0);
-			flush_dcache_page(page_out);
-
-			pg_offset += bytes;
-			page_bytes_left -= bytes;
-			buf_offset += bytes;
-			working_bytes -= bytes;
-			current_buf_start += bytes;
-
-			/* check if we need to pick another page */
-			if (page_bytes_left == 0) {
-				page_out_index++;
-				if (page_out_index >= vcnt) {
-					ret = 0;
-					goto done;
-				}
-
-				page_out = bvec[page_out_index].bv_page;
-				pg_offset = 0;
-				page_bytes_left = PAGE_CACHE_SIZE;
-				start_byte = page_offset(page_out) - disk_start;
-
-				/*
-				 * make sure our new page is covered by this
-				 * working buffer
-				 */
-				if (total_out <= start_byte)
-					goto next;
-
-				/* the next page in the biovec might not
-				 * be adjacent to the last page, but it
-				 * might still be found inside this working
-				 * buffer.  bump our offset pointer
-				 */
-				if (total_out > start_byte &&
-				    current_buf_start < start_byte) {
-					buf_offset = start_byte - buf_start;
-					working_bytes = total_out - start_byte;
-					current_buf_start = buf_start +
-						buf_offset;
-				}
-			}
+		ret2 = btrfs_decompress_buf2page(workspace->buf, buf_start,
+						 total_out, disk_start,
+						 bvec, vcnt,
+						 &page_out_index, &pg_offset);
+		if (ret2 == 0) {
+			ret = 0;
+			goto done;
 		}
-next:
+
 		workspace->inf_strm.next_out = workspace->buf;
 		workspace->inf_strm.avail_out = PAGE_CACHE_SIZE;
 
@@ -516,35 +301,21 @@ done:
 	zlib_inflateEnd(&workspace->inf_strm);
 	if (data_in)
 		kunmap(pages_in[page_in_index]);
-out:
-	free_workspace(workspace);
 	return ret;
 }
 
-/*
- * a less complex decompression routine.  Our compressed data fits in a
- * single page, and we want to read a single page out of it.
- * start_byte tells us the offset into the compressed data we're interested in
- */
-int btrfs_zlib_decompress(unsigned char *data_in,
-			  struct page *dest_page,
-			  unsigned long start_byte,
-			  size_t srclen, size_t destlen)
+static int zlib_decompress(struct list_head *ws, unsigned char *data_in,
+			   struct page *dest_page,
+			   unsigned long start_byte,
+			   size_t srclen, size_t destlen)
 {
+	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	int ret = 0;
 	int wbits = MAX_WBITS;
-	struct workspace *workspace;
 	unsigned long bytes_left = destlen;
 	unsigned long total_out = 0;
 	char *kaddr;
 
-	if (destlen > PAGE_CACHE_SIZE)
-		return -ENOMEM;
-
-	workspace = find_zlib_workspace();
-	if (IS_ERR(workspace))
-		return -ENOMEM;
-
 	workspace->inf_strm.next_in = data_in;
 	workspace->inf_strm.avail_in = srclen;
 	workspace->inf_strm.total_in = 0;
@@ -565,8 +336,7 @@ int btrfs_zlib_decompress(unsigned char *data_in,
 
 	if (Z_OK != zlib_inflateInit2(&workspace->inf_strm, wbits)) {
 		printk(KERN_WARNING "inflateInit failed\n");
-		ret = -1;
-		goto out;
+		return -1;
 	}
 
 	while (bytes_left > 0) {
@@ -616,12 +386,13 @@ next:
 		ret = 0;
 
 	zlib_inflateEnd(&workspace->inf_strm);
-out:
-	free_workspace(workspace);
 	return ret;
 }
 
-void btrfs_zlib_exit(void)
-{
-    free_workspaces();
-}
+struct btrfs_compress_op btrfs_zlib_compress = {
+	.alloc_workspace	= zlib_alloc_workspace,
+	.free_workspace		= zlib_free_workspace,
+	.compress_pages		= zlib_compress_pages,
+	.decompress_biovec	= zlib_decompress_biovec,
+	.decompress		= zlib_decompress,
+};