27 files changed, 2071 insertions, 1661 deletions

diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 80e8472d618b..d3090187fd76 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -1991,7 +1991,7 @@ struct inode_fs_paths *init_ipath(s32 total_bytes, struct btrfs_root *fs_root,
 
 	ifp = kmalloc(sizeof(*ifp), GFP_NOFS);
 	if (!ifp) {
-		kfree(fspath);
+		vfree(fspath);
 		return ERR_PTR(-ENOMEM);
 	}
 
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index ff61a41ac90b..658c39b70fba 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -743,8 +743,11 @@ out:
 static struct {
 	struct list_head idle_ws;
 	spinlock_t ws_lock;
-	int num_ws;
-	atomic_t alloc_ws;
+	/* Number of free workspaces */
+	int free_ws;
+	/* Total number of allocated workspaces */
+	atomic_t total_ws;
+	/* Waiters for a free workspace */
 	wait_queue_head_t ws_wait;
 } btrfs_comp_ws[BTRFS_COMPRESS_TYPES];
 
@@ -758,16 +761,34 @@ void __init btrfs_init_compress(void)
 	int i;
 
 	for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
+		struct list_head *workspace;
+
 		INIT_LIST_HEAD(&btrfs_comp_ws[i].idle_ws);
 		spin_lock_init(&btrfs_comp_ws[i].ws_lock);
-		atomic_set(&btrfs_comp_ws[i].alloc_ws, 0);
+		atomic_set(&btrfs_comp_ws[i].total_ws, 0);
 		init_waitqueue_head(&btrfs_comp_ws[i].ws_wait);
+
+		/*
+		 * Preallocate one workspace for each compression type so
+		 * we can guarantee forward progress in the worst case
+		 */
+		workspace = btrfs_compress_op[i]->alloc_workspace();
+		if (IS_ERR(workspace)) {
+			printk(KERN_WARNING
+	"BTRFS: cannot preallocate compression workspace, will try later");
+		} else {
+			atomic_set(&btrfs_comp_ws[i].total_ws, 1);
+			btrfs_comp_ws[i].free_ws = 1;
+			list_add(workspace, &btrfs_comp_ws[i].idle_ws);
+		}
 	}
 }
 
 /*
- * this finds an available workspace or allocates a new one
- * ERR_PTR is returned if things go bad.
+ * This finds an available workspace or allocates a new one.
+ * If it's not possible to allocate a new one, waits until there's one.
+ * Preallocation makes a forward progress guarantees and we do not return
+ * errors.
  */
 static struct list_head *find_workspace(int type)
 {
@@ -777,36 +798,58 @@ static struct list_head *find_workspace(int type)
 
 	struct list_head *idle_ws	= &btrfs_comp_ws[idx].idle_ws;
 	spinlock_t *ws_lock		= &btrfs_comp_ws[idx].ws_lock;
-	atomic_t *alloc_ws		= &btrfs_comp_ws[idx].alloc_ws;
+	atomic_t *total_ws		= &btrfs_comp_ws[idx].total_ws;
 	wait_queue_head_t *ws_wait	= &btrfs_comp_ws[idx].ws_wait;
-	int *num_ws			= &btrfs_comp_ws[idx].num_ws;
+	int *free_ws			= &btrfs_comp_ws[idx].free_ws;
 again:
 	spin_lock(ws_lock);
 	if (!list_empty(idle_ws)) {
 		workspace = idle_ws->next;
 		list_del(workspace);
-		(*num_ws)--;
+		(*free_ws)--;
 		spin_unlock(ws_lock);
 		return workspace;
 
 	}
-	if (atomic_read(alloc_ws) > cpus) {
+	if (atomic_read(total_ws) > cpus) {
 		DEFINE_WAIT(wait);
 
 		spin_unlock(ws_lock);
 		prepare_to_wait(ws_wait, &wait, TASK_UNINTERRUPTIBLE);
-		if (atomic_read(alloc_ws) > cpus && !*num_ws)
+		if (atomic_read(total_ws) > cpus && !*free_ws)
 			schedule();
 		finish_wait(ws_wait, &wait);
 		goto again;
 	}
-	atomic_inc(alloc_ws);
+	atomic_inc(total_ws);
 	spin_unlock(ws_lock);
 
 	workspace = btrfs_compress_op[idx]->alloc_workspace();
 	if (IS_ERR(workspace)) {
-		atomic_dec(alloc_ws);
+		atomic_dec(total_ws);
 		wake_up(ws_wait);
+
+		/*
+		 * Do not return the error but go back to waiting. There's a
+		 * workspace preallocated for each type and the compression
+		 * time is bounded so we get to a workspace eventually. This
+		 * makes our caller's life easier.
+		 *
+		 * To prevent silent and low-probability deadlocks (when the
+		 * initial preallocation fails), check if there are any
+		 * workspaces at all.
+		 */
+		if (atomic_read(total_ws) == 0) {
+			static DEFINE_RATELIMIT_STATE(_rs,
+					/* once per minute */ 60 * HZ,
+					/* no burst */ 1);
+
+			if (__ratelimit(&_rs)) {
+				printk(KERN_WARNING
+			    "no compression workspaces, low memory, retrying");
+			}
+		}
+		goto again;
 	}
 	return workspace;
 }
@@ -820,21 +863,21 @@ static void free_workspace(int type, struct list_head *workspace)
 	int idx = type - 1;
 	struct list_head *idle_ws	= &btrfs_comp_ws[idx].idle_ws;
 	spinlock_t *ws_lock		= &btrfs_comp_ws[idx].ws_lock;
-	atomic_t *alloc_ws		= &btrfs_comp_ws[idx].alloc_ws;
+	atomic_t *total_ws		= &btrfs_comp_ws[idx].total_ws;
 	wait_queue_head_t *ws_wait	= &btrfs_comp_ws[idx].ws_wait;
-	int *num_ws			= &btrfs_comp_ws[idx].num_ws;
+	int *free_ws			= &btrfs_comp_ws[idx].free_ws;
 
 	spin_lock(ws_lock);
-	if (*num_ws < num_online_cpus()) {
+	if (*free_ws < num_online_cpus()) {
 		list_add(workspace, idle_ws);
-		(*num_ws)++;
+		(*free_ws)++;
 		spin_unlock(ws_lock);
 		goto wake;
 	}
 	spin_unlock(ws_lock);
 
 	btrfs_compress_op[idx]->free_workspace(workspace);
-	atomic_dec(alloc_ws);
+	atomic_dec(total_ws);
 wake:
 	/*
 	 * Make sure counter is updated before we wake up waiters.
@@ -857,7 +900,7 @@ static void free_workspaces(void)
 			workspace = btrfs_comp_ws[i].idle_ws.next;
 			list_del(workspace);
 			btrfs_compress_op[i]->free_workspace(workspace);
-			atomic_dec(&btrfs_comp_ws[i].alloc_ws);
+			atomic_dec(&btrfs_comp_ws[i].total_ws);
 		}
 	}
 }
@@ -894,8 +937,6 @@ int btrfs_compress_pages(int type, struct address_space *mapping,
 	int ret;
 
 	workspace = find_workspace(type);
-	if (IS_ERR(workspace))
-		return PTR_ERR(workspace);
 
 	ret = btrfs_compress_op[type-1]->compress_pages(workspace, mapping,
 						      start, len, pages,
@@ -930,8 +971,6 @@ static int btrfs_decompress_biovec(int type, struct page **pages_in,
 	int ret;
 
 	workspace = find_workspace(type);
-	if (IS_ERR(workspace))
-		return PTR_ERR(workspace);
 
 	ret = btrfs_compress_op[type-1]->decompress_biovec(workspace, pages_in,
 							 disk_start,
@@ -952,8 +991,6 @@ int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
 	int ret;
 
 	workspace = find_workspace(type);
-	if (IS_ERR(workspace))
-		return PTR_ERR(workspace);
 
 	ret = btrfs_compress_op[type-1]->decompress(workspace, data_in,
 						  dest_page, start_byte,
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index ec7928a27aaa..decd0a3f5d61 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -1011,7 +1011,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 			return ret;
 		if (refs == 0) {
 			ret = -EROFS;
-			btrfs_std_error(root->fs_info, ret, NULL);
+			btrfs_handle_fs_error(root->fs_info, ret, NULL);
 			return ret;
 		}
 	} else {
@@ -1928,7 +1928,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		child = read_node_slot(root, mid, 0);
 		if (!child) {
 			ret = -EROFS;
-			btrfs_std_error(root->fs_info, ret, NULL);
+			btrfs_handle_fs_error(root->fs_info, ret, NULL);
 			goto enospc;
 		}
 
@@ -2031,7 +2031,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		 */
 		if (!left) {
 			ret = -EROFS;
-			btrfs_std_error(root->fs_info, ret, NULL);
+			btrfs_handle_fs_error(root->fs_info, ret, NULL);
 			goto enospc;
 		}
 		wret = balance_node_right(trans, root, mid, left);
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 84a6a5b3384a..e613e48d7122 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -33,6 +33,7 @@
 #include <asm/kmap_types.h>
 #include <linux/pagemap.h>
 #include <linux/btrfs.h>
+#include <linux/btrfs_tree.h>
 #include <linux/workqueue.h>
 #include <linux/security.h>
 #include <linux/sizes.h>
@@ -64,98 +65,6 @@ struct btrfs_ordered_sum;
 
 #define BTRFS_COMPAT_EXTENT_TREE_V0
 
-/* holds pointers to all of the tree roots */
-#define BTRFS_ROOT_TREE_OBJECTID 1ULL
-
-/* stores information about which extents are in use, and reference counts */
-#define BTRFS_EXTENT_TREE_OBJECTID 2ULL
-
-/*
- * chunk tree stores translations from logical -> physical block numbering
- * the super block points to the chunk tree
- */
-#define BTRFS_CHUNK_TREE_OBJECTID 3ULL
-
-/*
- * stores information about which areas of a given device are in use.
- * one per device.  The tree of tree roots points to the device tree
- */
-#define BTRFS_DEV_TREE_OBJECTID 4ULL
-
-/* one per subvolume, storing files and directories */
-#define BTRFS_FS_TREE_OBJECTID 5ULL
-
-/* directory objectid inside the root tree */
-#define BTRFS_ROOT_TREE_DIR_OBJECTID 6ULL
-
-/* holds checksums of all the data extents */
-#define BTRFS_CSUM_TREE_OBJECTID 7ULL
-
-/* holds quota configuration and tracking */
-#define BTRFS_QUOTA_TREE_OBJECTID 8ULL
-
-/* for storing items that use the BTRFS_UUID_KEY* types */
-#define BTRFS_UUID_TREE_OBJECTID 9ULL
-
-/* tracks free space in block groups. */
-#define BTRFS_FREE_SPACE_TREE_OBJECTID 10ULL
-
-/* device stats in the device tree */
-#define BTRFS_DEV_STATS_OBJECTID 0ULL
-
-/* for storing balance parameters in the root tree */
-#define BTRFS_BALANCE_OBJECTID -4ULL
-
-/* orhpan objectid for tracking unlinked/truncated files */
-#define BTRFS_ORPHAN_OBJECTID -5ULL
-
-/* does write ahead logging to speed up fsyncs */
-#define BTRFS_TREE_LOG_OBJECTID -6ULL
-#define BTRFS_TREE_LOG_FIXUP_OBJECTID -7ULL
-
-/* for space balancing */
-#define BTRFS_TREE_RELOC_OBJECTID -8ULL
-#define BTRFS_DATA_RELOC_TREE_OBJECTID -9ULL
-
-/*
- * extent checksums all have this objectid
- * this allows them to share the logging tree
- * for fsyncs
- */
-#define BTRFS_EXTENT_CSUM_OBJECTID -10ULL
-
-/* For storing free space cache */
-#define BTRFS_FREE_SPACE_OBJECTID -11ULL
-
-/*
- * The inode number assigned to the special inode for storing
- * free ino cache
- */
-#define BTRFS_FREE_INO_OBJECTID -12ULL
-
-/* dummy objectid represents multiple objectids */
-#define BTRFS_MULTIPLE_OBJECTIDS -255ULL
-
-/*
- * All files have objectids in this range.
- */
-#define BTRFS_FIRST_FREE_OBJECTID 256ULL
-#define BTRFS_LAST_FREE_OBJECTID -256ULL
-#define BTRFS_FIRST_CHUNK_TREE_OBJECTID 256ULL
-
-
-/*
- * the device items go into the chunk tree.  The key is in the form
- * [ 1 BTRFS_DEV_ITEM_KEY device_id ]
- */
-#define BTRFS_DEV_ITEMS_OBJECTID 1ULL
-
-#define BTRFS_BTREE_INODE_OBJECTID 1
-
-#define BTRFS_EMPTY_SUBVOL_DIR_OBJECTID 2
-
-#define BTRFS_DEV_REPLACE_DEVID 0ULL
-
 /*
  * the max metadata block size.  This limit is somewhat artificial,
  * but the memmove costs go through the roof for larger blocks.
@@ -175,12 +84,6 @@ struct btrfs_ordered_sum;
  */
 #define BTRFS_LINK_MAX 65535U
 
-/* 32 bytes in various csum fields */
-#define BTRFS_CSUM_SIZE 32
-
-/* csum types */
-#define BTRFS_CSUM_TYPE_CRC32	0
-
 static const int btrfs_csum_sizes[] = { 4 };
 
 /* four bytes for CRC32 */
@@ -189,17 +92,6 @@ static const int btrfs_csum_sizes[] = { 4 };
 /* spefic to btrfs_map_block(), therefore not in include/linux/blk_types.h */
 #define REQ_GET_READ_MIRRORS	(1 << 30)
 
-#define BTRFS_FT_UNKNOWN	0
-#define BTRFS_FT_REG_FILE	1
-#define BTRFS_FT_DIR		2
-#define BTRFS_FT_CHRDEV		3
-#define BTRFS_FT_BLKDEV		4
-#define BTRFS_FT_FIFO		5
-#define BTRFS_FT_SOCK		6
-#define BTRFS_FT_SYMLINK	7
-#define BTRFS_FT_XATTR		8
-#define BTRFS_FT_MAX		9
-
 /* ioprio of readahead is set to idle */
 #define BTRFS_IOPRIO_READA (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0))
 
@@ -207,138 +99,10 @@ static const int btrfs_csum_sizes[] = { 4 };
 
 #define BTRFS_MAX_EXTENT_SIZE SZ_128M
 
-/*
- * The key defines the order in the tree, and so it also defines (optimal)
- * block layout.
- *
- * objectid corresponds to the inode number.
- *
- * type tells us things about the object, and is a kind of stream selector.
- * so for a given inode, keys with type of 1 might refer to the inode data,
- * type of 2 may point to file data in the btree and type == 3 may point to
- * extents.
- *
- * offset is the starting byte offset for this key in the stream.
- *
- * btrfs_disk_key is in disk byte order.  struct btrfs_key is always
- * in cpu native order.  Otherwise they are identical and their sizes
- * should be the same (ie both packed)
- */
-struct btrfs_disk_key {
-	__le64 objectid;
-	u8 type;
-	__le64 offset;
-} __attribute__ ((__packed__));
-
-struct btrfs_key {
-	u64 objectid;
-	u8 type;
-	u64 offset;
-} __attribute__ ((__packed__));
-
 struct btrfs_mapping_tree {
 	struct extent_map_tree map_tree;
 };
 
-struct btrfs_dev_item {
-	/* the internal btrfs device id */
-	__le64 devid;
-
-	/* size of the device */
-	__le64 total_bytes;
-
-	/* bytes used */
-	__le64 bytes_used;
-
-	/* optimal io alignment for this device */
-	__le32 io_align;
-
-	/* optimal io width for this device */
-	__le32 io_width;
-
-	/* minimal io size for this device */
-	__le32 sector_size;
-
-	/* type and info about this device */
-	__le64 type;
-
-	/* expected generation for this device */
-	__le64 generation;
-
-	/*
-	 * starting byte of this partition on the device,
-	 * to allow for stripe alignment in the future
-	 */
-	__le64 start_offset;
-
-	/* grouping information for allocation decisions */
-	__le32 dev_group;
-
-	/* seek speed 0-100 where 100 is fastest */
-	u8 seek_speed;
-
-	/* bandwidth 0-100 where 100 is fastest */
-	u8 bandwidth;
-
-	/* btrfs generated uuid for this device */
-	u8 uuid[BTRFS_UUID_SIZE];
-
-	/* uuid of FS who owns this device */
-	u8 fsid[BTRFS_UUID_SIZE];
-} __attribute__ ((__packed__));
-
-struct btrfs_stripe {
-	__le64 devid;
-	__le64 offset;
-	u8 dev_uuid[BTRFS_UUID_SIZE];
-} __attribute__ ((__packed__));
-
-struct btrfs_chunk {
-	/* size of this chunk in bytes */
-	__le64 length;
-
-	/* objectid of the root referencing this chunk */
-	__le64 owner;
-
-	__le64 stripe_len;
-	__le64 type;
-
-	/* optimal io alignment for this chunk */
-	__le32 io_align;
-
-	/* optimal io width for this chunk */
-	__le32 io_width;
-
-	/* minimal io size for this chunk */
-	__le32 sector_size;
-
-	/* 2^16 stripes is quite a lot, a second limit is the size of a single
-	 * item in the btree
-	 */
-	__le16 num_stripes;
-
-	/* sub stripes only matter for raid10 */
-	__le16 sub_stripes;
-	struct btrfs_stripe stripe;
-	/* additional stripes go here */
-} __attribute__ ((__packed__));
-
-#define BTRFS_FREE_SPACE_EXTENT	1
-#define BTRFS_FREE_SPACE_BITMAP	2
-
-struct btrfs_free_space_entry {
-	__le64 offset;
-	__le64 bytes;
-	u8 type;
-} __attribute__ ((__packed__));
-
-struct btrfs_free_space_header {
-	struct btrfs_disk_key location;
-	__le64 generation;
-	__le64 num_entries;
-	__le64 num_bitmaps;
-} __attribute__ ((__packed__));
-
 static inline unsigned long btrfs_chunk_item_size(int num_stripes)
 {
 	BUG_ON(num_stripes == 0);
@@ -346,9 +110,6 @@ static inline unsigned long btrfs_chunk_item_size(int num_stripes)
 		sizeof(struct btrfs_stripe) * (num_stripes - 1);
 }
 
-#define BTRFS_HEADER_FLAG_WRITTEN	(1ULL << 0)
-#define BTRFS_HEADER_FLAG_RELOC		(1ULL << 1)
-
 /*
  * File system states
  */
@@ -357,13 +118,6 @@ static inline unsigned long btrfs_chunk_item_size(int num_stripes)
 #define BTRFS_FS_STATE_TRANS_ABORTED	2
 #define BTRFS_FS_STATE_DEV_REPLACING	3
 
-/* Super block flags */
-/* Errors detected */
-#define BTRFS_SUPER_FLAG_ERROR		(1ULL << 2)
-
-#define BTRFS_SUPER_FLAG_SEEDING	(1ULL << 32)
-#define BTRFS_SUPER_FLAG_METADUMP	(1ULL << 33)
-
 #define BTRFS_BACKREF_REV_MAX		256
 #define BTRFS_BACKREF_REV_SHIFT		56
 #define BTRFS_BACKREF_REV_MASK		(((u64)BTRFS_BACKREF_REV_MAX - 1) << \
@@ -410,7 +164,6 @@ struct btrfs_header {
  * room to translate 14 chunks with 3 stripes each.
  */
 #define BTRFS_SYSTEM_CHUNK_ARRAY_SIZE 2048
-#define BTRFS_LABEL_SIZE 256
 
 /*
  * just in case we somehow lose the roots and are not able to mount,
@@ -507,31 +260,6 @@ struct btrfs_super_block {
  * Compat flags that we support.  If any incompat flags are set other than the
  * ones specified below then we will fail to mount
  */
-#define BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE	(1ULL << 0)
-
-#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)
-/*
- * some patches floated around with a second compression method
- * lets save that incompat here for when they do get in
- * Note we don't actually support it, we're just reserving the
- * number
- */
-#define BTRFS_FEATURE_INCOMPAT_COMPRESS_LZOv2	(1ULL << 4)
-
-/*
- * older kernels tried to do bigger metadata blocks, but the
- * code was pretty buggy.  Lets not let them try anymore.
- */
-#define BTRFS_FEATURE_INCOMPAT_BIG_METADATA	(1ULL << 5)
-
-#define BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF	(1ULL << 6)
-#define BTRFS_FEATURE_INCOMPAT_RAID56		(1ULL << 7)
-#define BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA	(1ULL << 8)
-#define BTRFS_FEATURE_INCOMPAT_NO_HOLES		(1ULL << 9)
-
 #define BTRFS_FEATURE_COMPAT_SUPP		0ULL
 #define BTRFS_FEATURE_COMPAT_SAFE_SET		0ULL
 #define BTRFS_FEATURE_COMPAT_SAFE_CLEAR		0ULL
@@ -624,357 +352,8 @@ struct btrfs_path {
 	unsigned int need_commit_sem:1;
 	unsigned int skip_release_on_error:1;
 };
-
-/*
- * items in the extent btree are used to record the objectid of the
- * owner of the block and the number of references
- */
-
-struct btrfs_extent_item {
-	__le64 refs;
-	__le64 generation;
-	__le64 flags;
-} __attribute__ ((__packed__));
-
-struct btrfs_extent_item_v0 {
-	__le32 refs;
-} __attribute__ ((__packed__));
-
 #define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r) >> 4) - \
 					sizeof(struct btrfs_item))
-
-#define BTRFS_EXTENT_FLAG_DATA		(1ULL << 0)
-#define BTRFS_EXTENT_FLAG_TREE_BLOCK	(1ULL << 1)
-
-/* following flags only apply to tree blocks */
-
-/* use full backrefs for extent pointers in the block */
-#define BTRFS_BLOCK_FLAG_FULL_BACKREF	(1ULL << 8)
-
-/*
- * this flag is only used internally by scrub and may be changed at any time
- * it is only declared here to avoid collisions
- */
-#define BTRFS_EXTENT_FLAG_SUPER		(1ULL << 48)
-
-struct btrfs_tree_block_info {
-	struct btrfs_disk_key key;
-	u8 level;
-} __attribute__ ((__packed__));
-
-struct btrfs_extent_data_ref {
-	__le64 root;
-	__le64 objectid;
-	__le64 offset;
-	__le32 count;
-} __attribute__ ((__packed__));
-
-struct btrfs_shared_data_ref {
-	__le32 count;
-} __attribute__ ((__packed__));
-
-struct btrfs_extent_inline_ref {
-	u8 type;
-	__le64 offset;
-} __attribute__ ((__packed__));
-
-/* old style backrefs item */
-struct btrfs_extent_ref_v0 {
-	__le64 root;
-	__le64 generation;
-	__le64 objectid;
-	__le32 count;
-} __attribute__ ((__packed__));
-
-
-/* dev extents record free space on individual devices.  The owner
- * field points back to the chunk allocation mapping tree that allocated
- * the extent.  The chunk tree uuid field is a way to double check the owner
- */
-struct btrfs_dev_extent {
-	__le64 chunk_tree;
-	__le64 chunk_objectid;
-	__le64 chunk_offset;
-	__le64 length;
-	u8 chunk_tree_uuid[BTRFS_UUID_SIZE];
-} __attribute__ ((__packed__));
-
-struct btrfs_inode_ref {
-	__le64 index;
-	__le16 name_len;
-	/* name goes here */
-} __attribute__ ((__packed__));
-
-struct btrfs_inode_extref {
-	__le64 parent_objectid;
-	__le64 index;
-	__le16 name_len;
-	__u8   name[0];
-	/* name goes here */
-} __attribute__ ((__packed__));
-
-struct btrfs_timespec {
-	__le64 sec;
-	__le32 nsec;
-} __attribute__ ((__packed__));
-
-struct btrfs_inode_item {
-	/* nfs style generation number */
-	__le64 generation;
-	/* transid that last touched this inode */
-	__le64 transid;
-	__le64 size;
-	__le64 nbytes;
-	__le64 block_group;
-	__le32 nlink;
-	__le32 uid;
-	__le32 gid;
-	__le32 mode;
-	__le64 rdev;
-	__le64 flags;
-
-	/* modification sequence number for NFS */
-	__le64 sequence;
-
-	/*
-	 * a little future expansion, for more than this we can
-	 * just grow the inode item and version it
-	 */
-	__le64 reserved[4];
-	struct btrfs_timespec atime;
-	struct btrfs_timespec ctime;
-	struct btrfs_timespec mtime;
-	struct btrfs_timespec otime;
-} __attribute__ ((__packed__));
-
-struct btrfs_dir_log_item {
-	__le64 end;
-} __attribute__ ((__packed__));
-
-struct btrfs_dir_item {
-	struct btrfs_disk_key location;
-	__le64 transid;
-	__le16 data_len;
-	__le16 name_len;
-	u8 type;
-} __attribute__ ((__packed__));
-
-#define BTRFS_ROOT_SUBVOL_RDONLY	(1ULL << 0)
-
-/*
- * Internal in-memory flag that a subvolume has been marked for deletion but
- * still visible as a directory
- */
-#define BTRFS_ROOT_SUBVOL_DEAD		(1ULL << 48)
-
-struct btrfs_root_item {
-	struct btrfs_inode_item inode;
-	__le64 generation;
-	__le64 root_dirid;
-	__le64 bytenr;
-	__le64 byte_limit;
-	__le64 bytes_used;
-	__le64 last_snapshot;
-	__le64 flags;
-	__le32 refs;
-	struct btrfs_disk_key drop_progress;
-	u8 drop_level;
-	u8 level;
-
-	/*
-	 * The following fields appear after subvol_uuids+subvol_times
-	 * were introduced.
-	 */
-
-	/*
-	 * This generation number is used to test if the new fields are valid
-	 * and up to date while reading the root item. Every time the root item
-	 * is written out, the "generation" field is copied into this field. If
-	 * anyone ever mounted the fs with an older kernel, we will have
-	 * mismatching generation values here and thus must invalidate the
-	 * new fields. See btrfs_update_root and btrfs_find_last_root for
-	 * details.
-	 * the offset of generation_v2 is also used as the start for the memset
-	 * when invalidating the fields.
-	 */
-	__le64 generation_v2;
-	u8 uuid[BTRFS_UUID_SIZE];
-	u8 parent_uuid[BTRFS_UUID_SIZE];
-	u8 received_uuid[BTRFS_UUID_SIZE];
-	__le64 ctransid; /* updated when an inode changes */
-	__le64 otransid; /* trans when created */
-	__le64 stransid; /* trans when sent. non-zero for received subvol */
-	__le64 rtransid; /* trans when received. non-zero for received subvol */
-	struct btrfs_timespec ctime;
-	struct btrfs_timespec otime;
-	struct btrfs_timespec stime;
-	struct btrfs_timespec rtime;
-	__le64 reserved[8]; /* for future */
-} __attribute__ ((__packed__));
-
-/*
- * this is used for both forward and backward root refs
- */
-struct btrfs_root_ref {
-	__le64 dirid;
-	__le64 sequence;
-	__le16 name_len;
-} __attribute__ ((__packed__));
-
-struct btrfs_disk_balance_args {
-	/*
-	 * profiles to operate on, single is denoted by
-	 * BTRFS_AVAIL_ALLOC_BIT_SINGLE
-	 */
-	__le64 profiles;
-
-	/*
-	 * usage filter
-	 * BTRFS_BALANCE_ARGS_USAGE with a single value means '0..N'
-	 * BTRFS_BALANCE_ARGS_USAGE_RANGE - range syntax, min..max
-	 */
-	union {
-		__le64 usage;
-		struct {
-			__le32 usage_min;
-			__le32 usage_max;
-		};
-	};
-
-	/* devid filter */
-	__le64 devid;
-
-	/* devid subset filter [pstart..pend) */
-	__le64 pstart;
-	__le64 pend;
-
-	/* btrfs virtual address space subset filter [vstart..vend) */
-	__le64 vstart;
-	__le64 vend;
-
-	/*
-	 * profile to convert to, single is denoted by
-	 * BTRFS_AVAIL_ALLOC_BIT_SINGLE
-	 */
-	__le64 target;
-
-	/* BTRFS_BALANCE_ARGS_* */
-	__le64 flags;
-
-	/*
-	 * BTRFS_BALANCE_ARGS_LIMIT with value 'limit'
-	 * BTRFS_BALANCE_ARGS_LIMIT_RANGE - the extend version can use minimum
-	 * and maximum
-	 */
-	union {
-		__le64 limit;
-		struct {
-			__le32 limit_min;
-			__le32 limit_max;
-		};
-	};
-
-	/*
-	 * Process chunks that cross stripes_min..stripes_max devices,
-	 * BTRFS_BALANCE_ARGS_STRIPES_RANGE
-	 */
-	__le32 stripes_min;
-	__le32 stripes_max;
-
-	__le64 unused[6];
-} __attribute__ ((__packed__));
-
-/*
- * store balance parameters to disk so that balance can be properly
- * resumed after crash or unmount
- */
-struct btrfs_balance_item {
-	/* BTRFS_BALANCE_* */
-	__le64 flags;
-
-	struct btrfs_disk_balance_args data;
-	struct btrfs_disk_balance_args meta;
-	struct btrfs_disk_balance_args sys;
-
-	__le64 unused[4];
-} __attribute__ ((__packed__));
-
-#define BTRFS_FILE_EXTENT_INLINE 0
-#define BTRFS_FILE_EXTENT_REG 1
-#define BTRFS_FILE_EXTENT_PREALLOC 2
-
-struct btrfs_file_extent_item {
-	/*
-	 * transaction id that created this extent
-	 */
-	__le64 generation;
-	/*
-	 * max number of bytes to hold this extent in ram
-	 * when we split a compressed extent we can't know how big
-	 * each of the resulting pieces will be.  So, this is
-	 * an upper limit on the size of the extent in ram instead of
-	 * an exact limit.
-	 */
-	__le64 ram_bytes;
-
-	/*
-	 * 32 bits for the various ways we might encode the data,
-	 * including compression and encryption.  If any of these
-	 * are set to something a given disk format doesn't understand
-	 * it is treated like an incompat flag for reading and writing,
-	 * but not for stat.
-	 */
-	u8 compression;
-	u8 encryption;
-	__le16 other_encoding; /* spare for later use */
-
-	/* are we inline data or a real extent? */
-	u8 type;
-
-	/*
-	 * disk space consumed by the extent, checksum blocks are included
-	 * in these numbers
-	 *
-	 * At this offset in the structure, the inline extent data start.
-	 */
-	__le64 disk_bytenr;
-	__le64 disk_num_bytes;
-	/*
-	 * the logical offset in file blocks (no csums)
-	 * this extent record is for.  This allows a file extent to point
-	 * into the middle of an existing extent on disk, sharing it
-	 * between two snapshots (useful if some bytes in the middle of the
-	 * extent have changed
-	 */
-	__le64 offset;
-	/*
-	 * the logical number of file blocks (no csums included).  This
-	 * always reflects the size uncompressed and without encoding.
-	 */
-	__le64 num_bytes;
-
-} __attribute__ ((__packed__));
-
-struct btrfs_csum_item {
-	u8 csum;
-} __attribute__ ((__packed__));
-
-struct btrfs_dev_stats_item {
-	/*
-	 * grow this item struct at the end for future enhancements and keep
-	 * the existing values unchanged
-	 */
-	__le64 values[BTRFS_DEV_STAT_VALUES_MAX];
-} __attribute__ ((__packed__));
-
-#define BTRFS_DEV_REPLACE_ITEM_CONT_READING_FROM_SRCDEV_MODE_ALWAYS	0
-#define BTRFS_DEV_REPLACE_ITEM_CONT_READING_FROM_SRCDEV_MODE_AVOID	1
-#define BTRFS_DEV_REPLACE_ITEM_STATE_NEVER_STARTED	0
-#define BTRFS_DEV_REPLACE_ITEM_STATE_STARTED		1
-#define BTRFS_DEV_REPLACE_ITEM_STATE_SUSPENDED		2
-#define BTRFS_DEV_REPLACE_ITEM_STATE_FINISHED		3
-#define BTRFS_DEV_REPLACE_ITEM_STATE_CANCELED		4
-
 struct btrfs_dev_replace {
 	u64 replace_state;	/* see #define above */
 	u64 time_started;	/* seconds since 1-Jan-1970 */
@@ -1005,175 +384,6 @@ struct btrfs_dev_replace {
 	struct btrfs_scrub_progress scrub_progress;
 };
 
-struct btrfs_dev_replace_item {
-	/*
-	 * grow this item struct at the end for future enhancements and keep
-	 * the existing values unchanged
-	 */
-	__le64 src_devid;
-	__le64 cursor_left;
-	__le64 cursor_right;
-	__le64 cont_reading_from_srcdev_mode;
-
-	__le64 replace_state;
-	__le64 time_started;
-	__le64 time_stopped;
-	__le64 num_write_errors;
-	__le64 num_uncorrectable_read_errors;
-} __attribute__ ((__packed__));
-
-/* different types of block groups (and chunks) */
-#define BTRFS_BLOCK_GROUP_DATA		(1ULL << 0)
-#define BTRFS_BLOCK_GROUP_SYSTEM	(1ULL << 1)
-#define BTRFS_BLOCK_GROUP_METADATA	(1ULL << 2)
-#define BTRFS_BLOCK_GROUP_RAID0		(1ULL << 3)
-#define BTRFS_BLOCK_GROUP_RAID1		(1ULL << 4)
-#define BTRFS_BLOCK_GROUP_DUP		(1ULL << 5)
-#define BTRFS_BLOCK_GROUP_RAID10	(1ULL << 6)
-#define BTRFS_BLOCK_GROUP_RAID5         (1ULL << 7)
-#define BTRFS_BLOCK_GROUP_RAID6         (1ULL << 8)
-#define BTRFS_BLOCK_GROUP_RESERVED	(BTRFS_AVAIL_ALLOC_BIT_SI