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

Pull btrfs updates from Chris Mason:
 "This pull is mostly cleanups and fixes:

   - The raid5/6 cleanups from Zhao Lei fixup some long standing warts
     in the code and add improvements on top of the scrubbing support
     from 3.19.

   - Josef has round one of our ENOSPC fixes coming from large btrfs
     clusters here at FB.

   - Dave Sterba continues a long series of cleanups (thanks Dave), and
     Filipe continues hammering on corner cases in fsync and others

  This all was held up a little trying to track down a use-after-free in
  btrfs raid5/6.  It's not clear yet if this is just made easier to
  trigger with this pull or if its a new bug from the raid5/6 cleanups.
  Dave Sterba is the only one to trigger it so far, but he has a
  consistent way to reproduce, so we'll get it nailed shortly"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (68 commits)
  Btrfs: don't remove extents and xattrs when logging new names
  Btrfs: fix fsync data loss after adding hard link to inode
  Btrfs: fix BUG_ON in btrfs_orphan_add() when delete unused block group
  Btrfs: account for large extents with enospc
  Btrfs: don't set and clear delalloc for O_DIRECT writes
  Btrfs: only adjust outstanding_extents when we do a short write
  btrfs: Fix out-of-space bug
  Btrfs: scrub, fix sleep in atomic context
  Btrfs: fix scheduler warning when syncing log
  Btrfs: Remove unnecessary placeholder in btrfs_err_code
  btrfs: cleanup init for list in free-space-cache
  btrfs: delete chunk allocation attemp when setting block group ro
  btrfs: clear bio reference after submit_one_bio()
  Btrfs: fix scrub race leading to use-after-free
  Btrfs: add missing cleanup on sysfs init failure
  Btrfs: fix race between transaction commit and empty block group removal
  btrfs: add more checks to btrfs_read_sys_array
  btrfs: cleanup, rename a few variables in btrfs_read_sys_array
  btrfs: add checks for sys_chunk_array sizes
  btrfs: more superblock checks, lower bounds on devices and sectorsize/nodesize
  ...

fs/btrfs/backref.c

@@ -1246,25 +1246,6 @@ int btrfs_check_shared(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-/*
- * this makes the path point to (inum INODE_ITEM ioff)
- */
-int inode_item_info(u64 inum, u64 ioff, struct btrfs_root *fs_root,
-			struct btrfs_path *path)
-{
-	struct btrfs_key key;
-	return btrfs_find_item(fs_root, path, inum, ioff,
-			BTRFS_INODE_ITEM_KEY, &key);
-}
-
-static int inode_ref_info(u64 inum, u64 ioff, struct btrfs_root *fs_root,
-				struct btrfs_path *path,
-				struct btrfs_key *found_key)
-{
-	return btrfs_find_item(fs_root, path, inum, ioff,
-			BTRFS_INODE_REF_KEY, found_key);
-}
-
 int btrfs_find_one_extref(struct btrfs_root *root, u64 inode_objectid,
 			  u64 start_off, struct btrfs_path *path,
 			  struct btrfs_inode_extref **ret_extref,
@@ -1374,7 +1355,8 @@ char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
 			btrfs_tree_read_unlock_blocking(eb);
 			free_extent_buffer(eb);
 		}
-		ret = inode_ref_info(parent, 0, fs_root, path, &found_key);
+		ret = btrfs_find_item(fs_root, path, parent, 0,
+				BTRFS_INODE_REF_KEY, &found_key);
 		if (ret > 0)
 			ret = -ENOENT;
 		if (ret)
@@ -1727,8 +1709,10 @@ static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root,
 	struct btrfs_key found_key;
 
 	while (!ret) {
-		ret = inode_ref_info(inum, parent ? parent+1 : 0, fs_root, path,
-				     &found_key);
+		ret = btrfs_find_item(fs_root, path, inum,
+				parent ? parent + 1 : 0, BTRFS_INODE_REF_KEY,
+				&found_key);
+
 		if (ret < 0)
 			break;
 		if (ret) {

fs/btrfs/backref.h

@@ -32,9 +32,6 @@ struct inode_fs_paths {
 typedef int (iterate_extent_inodes_t)(u64 inum, u64 offset, u64 root,
 		void *ctx);
 
-int inode_item_info(u64 inum, u64 ioff, struct btrfs_root *fs_root,
-			struct btrfs_path *path);
-
 int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical,
 			struct btrfs_path *path, struct btrfs_key *found_key,
 			u64 *flags);

fs/btrfs/btrfs_inode.h

@@ -185,6 +185,9 @@ struct btrfs_inode {
 
 	struct btrfs_delayed_node *delayed_node;
 
+	/* File creation time. */
+	struct timespec i_otime;
+
 	struct inode vfs_inode;
 };
 

fs/btrfs/ctree.c

@@ -213,11 +213,19 @@ static struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
  */
 static void add_root_to_dirty_list(struct btrfs_root *root)
 {
+	if (test_bit(BTRFS_ROOT_DIRTY, &root->state) ||
+	    !test_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state))
+		return;
+
 	spin_lock(&root->fs_info->trans_lock);
-	if (test_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state) &&
-	    list_empty(&root->dirty_list)) {
-		list_add(&root->dirty_list,
-			 &root->fs_info->dirty_cowonly_roots);
+	if (!test_and_set_bit(BTRFS_ROOT_DIRTY, &root->state)) {
+		/* Want the extent tree to be the last on the list */
+		if (root->objectid == BTRFS_EXTENT_TREE_OBJECTID)
+			list_move_tail(&root->dirty_list,
+				       &root->fs_info->dirty_cowonly_roots);
+		else
+			list_move(&root->dirty_list,
+				  &root->fs_info->dirty_cowonly_roots);
 	}
 	spin_unlock(&root->fs_info->trans_lock);
 }
@@ -1363,8 +1371,7 @@ tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
 
 	if (tm->op == MOD_LOG_KEY_REMOVE_WHILE_FREEING) {
 		BUG_ON(tm->slot != 0);
-		eb_rewin = alloc_dummy_extent_buffer(eb->start,
-						fs_info->tree_root->nodesize);
+		eb_rewin = alloc_dummy_extent_buffer(fs_info, eb->start);
 		if (!eb_rewin) {
 			btrfs_tree_read_unlock_blocking(eb);
 			free_extent_buffer(eb);
@@ -1444,7 +1451,7 @@ get_old_root(struct btrfs_root *root, u64 time_seq)
 	} else if (old_root) {
 		btrfs_tree_read_unlock(eb_root);
 		free_extent_buffer(eb_root);
-		eb = alloc_dummy_extent_buffer(logical, root->nodesize);
+		eb = alloc_dummy_extent_buffer(root->fs_info, logical);
 	} else {
 		btrfs_set_lock_blocking_rw(eb_root, BTRFS_READ_LOCK);
 		eb = btrfs_clone_extent_buffer(eb_root);
@@ -2282,7 +2289,7 @@ static void reada_for_search(struct btrfs_root *root,
 		if ((search <= target && target - search <= 65536) ||
 		    (search > target && search - target <= 65536)) {
 			gen = btrfs_node_ptr_generation(node, nr);
-			readahead_tree_block(root, search, blocksize);
+			readahead_tree_block(root, search);
 			nread += blocksize;
 		}
 		nscan++;
@@ -2301,7 +2308,6 @@ static noinline void reada_for_balance(struct btrfs_root *root,
 	u64 gen;
 	u64 block1 = 0;
 	u64 block2 = 0;
-	int blocksize;
 
 	parent = path->nodes[level + 1];
 	if (!parent)
@@ -2309,7 +2315,6 @@ static noinline void reada_for_balance(struct btrfs_root *root,
 
 	nritems = btrfs_header_nritems(parent);
 	slot = path->slots[level + 1];
-	blocksize = root->nodesize;
 
 	if (slot > 0) {
 		block1 = btrfs_node_blockptr(parent, slot - 1);
@@ -2334,9 +2339,9 @@ static noinline void reada_for_balance(struct btrfs_root *root,
 	}
 
 	if (block1)
-		readahead_tree_block(root, block1, blocksize);
+		readahead_tree_block(root, block1);
 	if (block2)
-		readahead_tree_block(root, block2, blocksize);
+		readahead_tree_block(root, block2);
 }
 
 
@@ -2609,32 +2614,24 @@ static int key_search(struct extent_buffer *b, struct btrfs_key *key,
 	return 0;
 }
 
-int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *found_path,
+int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
 		u64 iobjectid, u64 ioff, u8 key_type,
 		struct btrfs_key *found_key)
 {
 	int ret;
 	struct btrfs_key key;
 	struct extent_buffer *eb;
-	struct btrfs_path *path;
+
+	ASSERT(path);
+	ASSERT(found_key);
 
 	key.type = key_type;
 	key.objectid = iobjectid;
 	key.offset = ioff;
 
-	if (found_path == NULL) {
-		path = btrfs_alloc_path();
-		if (!path)
-			return -ENOMEM;
-	} else
-		path = found_path;
-
 	ret = btrfs_search_slot(NULL, fs_root, &key, path, 0, 0);
-	if ((ret < 0) || (found_key == NULL)) {
-		if (path != found_path)
-			btrfs_free_path(path);
+	if (ret < 0)
 		return ret;
-	}
 
 	eb = path->nodes[0];
 	if (ret && path->slots[0] >= btrfs_header_nritems(eb)) {
@@ -3383,7 +3380,7 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans,
 	add_root_to_dirty_list(root);
 	extent_buffer_get(c);
 	path->nodes[level] = c;
-	path->locks[level] = BTRFS_WRITE_LOCK;
+	path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING;
 	path->slots[level] = 0;
 	return 0;
 }
@@ -4356,13 +4353,15 @@ static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
 	path->search_for_split = 1;
 	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
 	path->search_for_split = 0;
+	if (ret > 0)
+		ret = -EAGAIN;
 	if (ret < 0)
 		goto err;
 
 	ret = -EAGAIN;
 	leaf = path->nodes[0];
-	/* if our item isn't there or got smaller, return now */
-	if (ret > 0 || item_size != btrfs_item_size_nr(leaf, path->slots[0]))
+	/* if our item isn't there, return now */
+	if (item_size != btrfs_item_size_nr(leaf, path->slots[0]))
 		goto err;
 
 	/* the leaf has  changed, it now has room.  return now */

fs/btrfs/ctree.h

@@ -198,6 +198,8 @@ static int btrfs_csum_sizes[] = { 4, 0 };
 
 #define BTRFS_DIRTY_METADATA_THRESH	(32 * 1024 * 1024)
 
+#define BTRFS_MAX_EXTENT_SIZE (128 * 1024 * 1024)
+
 /*
  * The key defines the order in the tree, and so it also defines (optimal)
  * block layout.
@@ -1020,6 +1022,9 @@ enum btrfs_raid_types {
 					 BTRFS_BLOCK_GROUP_RAID6 |   \
 					 BTRFS_BLOCK_GROUP_DUP |     \
 					 BTRFS_BLOCK_GROUP_RAID10)
+#define BTRFS_BLOCK_GROUP_RAID56_MASK	(BTRFS_BLOCK_GROUP_RAID5 |   \
+					 BTRFS_BLOCK_GROUP_RAID6)
+
 /*
  * We need a bit for restriper to be able to tell when chunks of type
  * SINGLE are available.  This "extended" profile format is used in
@@ -1239,7 +1244,6 @@ enum btrfs_disk_cache_state {
 	BTRFS_DC_ERROR		= 1,
 	BTRFS_DC_CLEAR		= 2,
 	BTRFS_DC_SETUP		= 3,
-	BTRFS_DC_NEED_WRITE	= 4,
 };
 
 struct btrfs_caching_control {
@@ -1277,7 +1281,6 @@ struct btrfs_block_group_cache {
 	unsigned long full_stripe_len;
 
 	unsigned int ro:1;
-	unsigned int dirty:1;
 	unsigned int iref:1;
 	unsigned int has_caching_ctl:1;
 	unsigned int removed:1;
@@ -1315,6 +1318,9 @@ struct btrfs_block_group_cache {
 	struct list_head ro_list;
 
 	atomic_t trimming;
+
+	/* For dirty block groups */
+	struct list_head dirty_list;
 };
 
 /* delayed seq elem */
@@ -1741,6 +1747,7 @@ struct btrfs_fs_info {
 
 	spinlock_t unused_bgs_lock;
 	struct list_head unused_bgs;
+	struct mutex unused_bg_unpin_mutex;
 
 	/* For btrfs to record security options */
 	struct security_mnt_opts security_opts;
@@ -1776,6 +1783,7 @@ struct btrfs_subvolume_writers {
 #define BTRFS_ROOT_DEFRAG_RUNNING	6
 #define BTRFS_ROOT_FORCE_COW		7
 #define BTRFS_ROOT_MULTI_LOG_TASKS	8
+#define BTRFS_ROOT_DIRTY		9
 
 /*
  * in ram representation of the tree.  extent_root is used for all allocations
@@ -1794,8 +1802,6 @@ struct btrfs_root {
 	struct btrfs_fs_info *fs_info;
 	struct extent_io_tree dirty_log_pages;
 
-	struct kobject root_kobj;
-	struct completion kobj_unregister;
 	struct mutex objectid_mutex;
 
 	spinlock_t accounting_lock;
@@ -2465,31 +2471,6 @@ BTRFS_SETGET_STACK_FUNCS(stack_inode_gid, struct btrfs_inode_item, gid, 32);
 BTRFS_SETGET_STACK_FUNCS(stack_inode_mode, struct btrfs_inode_item, mode, 32);
 BTRFS_SETGET_STACK_FUNCS(stack_inode_rdev, struct btrfs_inode_item, rdev, 64);
 BTRFS_SETGET_STACK_FUNCS(stack_inode_flags, struct btrfs_inode_item, flags, 64);
-
-static inline struct btrfs_timespec *
-btrfs_inode_atime(struct btrfs_inode_item *inode_item)
-{
-	unsigned long ptr = (unsigned long)inode_item;
-	ptr += offsetof(struct btrfs_inode_item, atime);
-	return (struct btrfs_timespec *)ptr;
-}
-
-static inline struct btrfs_timespec *
-btrfs_inode_mtime(struct btrfs_inode_item *inode_item)
-{
-	unsigned long ptr = (unsigned long)inode_item;
-	ptr += offsetof(struct btrfs_inode_item, mtime);
-	return (struct btrfs_timespec *)ptr;
-}
-
-static inline struct btrfs_timespec *
-btrfs_inode_ctime(struct btrfs_inode_item *inode_item)
-{
-	unsigned long ptr = (unsigned long)inode_item;
-	ptr += offsetof(struct btrfs_inode_item, ctime);
-	return (struct btrfs_timespec *)ptr;
-}
-
 BTRFS_SETGET_FUNCS(timespec_sec, struct btrfs_timespec, sec, 64);
 BTRFS_SETGET_FUNCS(timespec_nsec, struct btrfs_timespec, nsec, 32);
 BTRFS_SETGET_STACK_FUNCS(stack_timespec_sec, struct btrfs_timespec, sec, 64);

fs/btrfs/delayed-inode.c

@@ -1755,27 +1755,31 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans,
 	btrfs_set_stack_inode_flags(inode_item, BTRFS_I(inode)->flags);
 	btrfs_set_stack_inode_block_group(inode_item, 0);
 
-	btrfs_set_stack_timespec_sec(btrfs_inode_atime(inode_item),
+	btrfs_set_stack_timespec_sec(&inode_item->atime,
 				     inode->i_atime.tv_sec);
-	btrfs_set_stack_timespec_nsec(btrfs_inode_atime(inode_item),
+	btrfs_set_stack_timespec_nsec(&inode_item->atime,
 				      inode->i_atime.tv_nsec);
 
-	btrfs_set_stack_timespec_sec(btrfs_inode_mtime(inode_item),
+	btrfs_set_stack_timespec_sec(&inode_item->mtime,
 				     inode->i_mtime.tv_sec);
-	btrfs_set_stack_timespec_nsec(btrfs_inode_mtime(inode_item),
+	btrfs_set_stack_timespec_nsec(&inode_item->mtime,
 				      inode->i_mtime.tv_nsec);
 
-	btrfs_set_stack_timespec_sec(btrfs_inode_ctime(inode_item),
+	btrfs_set_stack_timespec_sec(&inode_item->ctime,
 				     inode->i_ctime.tv_sec);
-	btrfs_set_stack_timespec_nsec(btrfs_inode_ctime(inode_item),
+	btrfs_set_stack_timespec_nsec(&inode_item->ctime,
 				      inode->i_ctime.tv_nsec);
+
+	btrfs_set_stack_timespec_sec(&inode_item->otime,
+				     BTRFS_I(inode)->i_otime.tv_sec);
+	btrfs_set_stack_timespec_nsec(&inode_item->otime,
+				     BTRFS_I(inode)->i_otime.tv_nsec);
 }
 
 int btrfs_fill_inode(struct inode *inode, u32 *rdev)
 {
 	struct btrfs_delayed_node *delayed_node;
 	struct btrfs_inode_item *inode_item;
-	struct btrfs_timespec *tspec;
 
 	delayed_node = btrfs_get_delayed_node(inode);
 	if (!delayed_node)
@@ -1802,17 +1806,19 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
 	*rdev = btrfs_stack_inode_rdev(inode_item);
 	BTRFS_I(inode)->flags = btrfs_stack_inode_flags(inode_item);
 
-	tspec = btrfs_inode_atime(inode_item);
-	inode->i_atime.tv_sec = btrfs_stack_timespec_sec(tspec);
-	inode->i_atime.tv_nsec = btrfs_stack_timespec_nsec(tspec);
+	inode->i_atime.tv_sec = btrfs_stack_timespec_sec(&inode_item->atime);
+	inode->i_atime.tv_nsec = btrfs_stack_timespec_nsec(&inode_item->atime);
+
+	inode->i_mtime.tv_sec = btrfs_stack_timespec_sec(&inode_item->mtime);
+	inode->i_mtime.tv_nsec = btrfs_stack_timespec_nsec(&inode_item->mtime);
 
-	tspec = btrfs_inode_mtime(inode_item);
-	inode->i_mtime.tv_sec = btrfs_stack_timespec_sec(tspec);
-	inode->i_mtime.tv_nsec = btrfs_stack_timespec_nsec(tspec);
+	inode->i_ctime.tv_sec = btrfs_stack_timespec_sec(&inode_item->ctime);
+	inode->i_ctime.tv_nsec = btrfs_stack_timespec_nsec(&inode_item->ctime);
 
-	tspec = btrfs_inode_ctime(inode_item);
-	inode->i_ctime.tv_sec = btrfs_stack_timespec_sec(tspec);
-	inode->i_ctime.tv_nsec = btrfs_stack_timespec_nsec(tspec);
+	BTRFS_I(inode)->i_otime.tv_sec =
+		btrfs_stack_timespec_sec(&inode_item->otime);
+	BTRFS_I(inode)->i_otime.tv_nsec =
+		btrfs_stack_timespec_nsec(&inode_item->otime);
 
 	inode->i_generation = BTRFS_I(inode)->generation;
 	BTRFS_I(inode)->index_cnt = (u64)-1;

fs/btrfs/dev-replace.c

@@ -440,18 +440,9 @@ int btrfs_dev_replace_start(struct btrfs_root *root,
  */
 static void btrfs_rm_dev_replace_blocked(struct btrfs_fs_info *fs_info)
 {
-	s64 writers;
-	DEFINE_WAIT(wait);
-
 	set_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state);
-	do {
-		prepare_to_wait(&fs_info->replace_wait, &wait,
-				TASK_UNINTERRUPTIBLE);
-		writers = percpu_counter_sum(&fs_info->bio_counter);
-		if (writers)
-			schedule();
-		finish_wait(&fs_info->replace_wait, &wait);
-	} while (writers);
+	wait_event(fs_info->replace_wait, !percpu_counter_sum(
+		   &fs_info->bio_counter));
 }
 
 /*
@@ -932,15 +923,15 @@ void btrfs_bio_counter_sub(struct btrfs_fs_info *fs_info, s64 amount)
 
 void btrfs_bio_counter_inc_blocked(struct btrfs_fs_info *fs_info)
 {
-	DEFINE_WAIT(wait);
-again:
-	percpu_counter_inc(&fs_info->bio_counter);
-	if (test_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state)) {
+	while (1) {
+		percpu_counter_inc(&fs_info->bio_counter);
+		if (likely(!test_bit(BTRFS_FS_STATE_DEV_REPLACING,
+				     &fs_info->fs_state)))
+			break;
+
 		btrfs_bio_counter_dec(fs_info);
 		wait_event(fs_info->replace_wait,
 			   !test_bit(BTRFS_FS_STATE_DEV_REPLACING,
 				     &fs_info->fs_state));
-		goto again;
 	}
-
 }

fs/btrfs/disk-io.h

@@ -46,11 +46,11 @@ struct btrfs_fs_devices;
 
 struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr,
 				      u64 parent_transid);
-void readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize);
-int reada_tree_block_flagged(struct btrfs_root *root, u64 bytenr, u32 blocksize,
+void readahead_tree_block(struct btrfs_root *root, u64 bytenr);
+int reada_tree_block_flagged(struct btrfs_root *root, u64 bytenr,
 			 int mirror_num, struct extent_buffer **eb);
 struct extent_buffer *btrfs_find_create_tree_block(struct btrfs_root *root,
-						   u64 bytenr, u32 blocksize);
+						   u64 bytenr);
 void clean_tree_block(struct btrfs_trans_handle *trans,
 		      struct btrfs_root *root, struct extent_buffer *buf);
 int open_ctree(struct super_block *sb,

fs/btrfs/extent_io.h

@@ -4,22 +4,22 @@
 #include <linux/rbtree.h>
 
 /* bits for the extent state */
-#define EXTENT_DIRTY 1
-#define EXTENT_WRITEBACK (1 << 1)
-#define EXTENT_UPTODATE (1 << 2)
-#define EXTENT_LOCKED (1 << 3)
-#define EXTENT_NEW (1 << 4)
-#define EXTENT_DELALLOC (1 << 5)
-#define EXTENT_DEFRAG (1 << 6)
-#define EXTENT_BOUNDARY (1 << 9)
-#define EXTENT_NODATASUM (1 << 10)
-#define EXTENT_DO_ACCOUNTING (1 << 11)
-#define EXTENT_FIRST_DELALLOC (1 << 12)
-#define EXTENT_NEED_WAIT (1 << 13)
-#define EXTENT_DAMAGED (1 << 14)
-#define EXTENT_NORESERVE (1 << 15)
-#define EXTENT_IOBITS (EXTENT_LOCKED | EXTENT_WRITEBACK)
-#define EXTENT_CTLBITS (EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC)
+#define EXTENT_DIRTY		(1U << 0)
+#define EXTENT_WRITEBACK	(1U << 1)
+#define EXTENT_UPTODATE		(1U << 2)
+#define EXTENT_LOCKED		(1U << 3)
+#define EXTENT_NEW		(1U << 4)
+#define EXTENT_DELALLOC		(1U << 5)
+#define EXTENT_DEFRAG		(1U << 6)
+#define EXTENT_BOUNDARY		(1U << 9)
+#define EXTENT_NODATASUM	(1U << 10)
+#define EXTENT_DO_ACCOUNTING	(1U << 11)
+#define EXTENT_FIRST_DELALLOC	(1U << 12)
+#define EXTENT_NEED_WAIT	(1U << 13)
+#define EXTENT_DAMAGED		(1U << 14)
+#define EXTENT_NORESERVE	(1U << 15)
+#define EXTENT_IOBITS		(EXTENT_LOCKED | EXTENT_WRITEBACK)
+#define EXTENT_CTLBITS		(EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC)
 
 /*
  * flags for bio submission. The high bits indicate the compression
@@ -81,9 +81,9 @@ struct extent_io_ops {
 	int (*writepage_end_io_hook)(struct page *page, u64 start, u64 end,
 				      struct extent_state *state, int uptodate);
 	void (*set_bit_hook)(struct inode *inode, struct extent_state *state,
-			     unsigned long *bits);
+			     unsigned *bits);
 	void (*clear_bit_hook)(struct inode *inode, struct extent_state *state,
-			       unsigned long *bits);
+			       unsigned *bits);
 	void (*merge_extent_hook)(struct inode *inode,
 				  struct extent_state *new,
 				  struct extent_state *other);
@@ -108,7 +108,7 @@ struct extent_state {
 	/* ADD NEW ELEMENTS AFTER THIS */
 	wait_queue_head_t wq;
 	atomic_t refs;
-	unsigned long state;
+	unsigned state;
 
 	/* for use by the FS */
 	u64 private;
@@ -188,7 +188,7 @@ int try_release_extent_mapping(struct extent_map_tree *map,
 int try_release_extent_buffer(struct page *page);
 int lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
 int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
-		     unsigned long bits, struct extent_state **cached);
+		     unsigned bits, struct extent_state **cached);
 int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end);
 int unlock_extent_cached(struct extent_io_tree *tree, u64 start, u64 end,
 			 struct extent_state **cached, gfp_t mask);
@@ -202,21 +202,21 @@ void extent_io_exit(void);
 
 u64 count_range_bits(struct extent_io_tree *tree,
 		     u64 *start, u64 search_end,
-		     u64 max_bytes, unsigned long bits, int contig);
+		     u64 max_bytes, unsigned bits, int contig);
 
 void free_extent_state(struct extent_state *state);
 int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		   unsigned long bits, int filled,
+		   unsigned bits, int filled,
 		   struct extent_state *cached_state);
 int clear_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
-		      unsigned long bits, gfp_t mask);
+		      unsigned bits, gfp_t mask);
 int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		     unsigned long bits, int wake, int delete,
+		     unsigned bits, int wake, int delete,
 		     struct extent_state **cached, gfp_t mask);
 int set_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
-		    unsigned long bits, gfp_t mask);
+		    unsigned bits, gfp_t mask);
 int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		   unsigned long bits, u64 *failed_start,
+		   unsigned bits, u64 *failed_start,
 		   struct extent_state **cached_state, gfp_t mask);
 int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
 			struct extent_state **cached_state, gfp_t mask);
@@ -229,14 +229,14 @@ int set_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
 int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
 		       gfp_t mask);
 int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		       unsigned long bits, unsigned long clear_bits,
+		       unsigned bits, unsigned clear_bits,
 		       struct extent_state **cached_state, gfp_t mask);
 int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end,
 			struct extent_state **cached_state, gfp_t mask);
 int set_extent_defrag(struct extent_io_tree *tree, u64 start, u64 end,
 		      struct extent_state **cached_state, gfp_t mask);
 int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
-			  u64 *start_ret, u64 *end_ret, unsigned long bits,
+			  u64 *start_ret, u64 *end_ret, unsigned bits,
 			  struct extent_state **cached_state);
 int extent_invalidatepage(struct extent_io_tree *tree,
 			  struct page *page, unsigned long offset);
@@ -262,8 +262,9 @@ int get_state_private(struct extent_io_tree *tree, u64 start, u64 *private);
 void set_page_extent_mapped(struct page *page);
 
 struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
-					  u64 start, unsigned long len);
-struct extent_buffer *alloc_dummy_extent_buffer(u64 start, unsigned long len);
+					  u64 start);
+struct extent_buffer *alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
+		u64 start);
 struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src);
 struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
 					 u64 start);
@@ -322,7 +323,7 @@ int extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
 int extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
 int extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
 				 struct page *locked_page,
-				 unsigned long bits_to_clear,
+				 unsigned bits_to_clear,
 				 unsigned long page_ops);
 struct bio *
 btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
@@ -377,5 +378,5 @@ noinline u64 find_lock_delalloc_range(struct inode *inode,
 				      u64 *end, u64 max_bytes);
 #endif
 struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
-					       u64 start, unsigned long len);
+					       u64 start);
 #endif

fs/btrfs/free-space-cache.c

@@ -651,15 +651,13 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
 	struct io_ctl io_ctl;
 	struct btrfs_key key;
 	struct btrfs_free_space *e, *n;
-	struct list_head bitmaps;
+	LIST_HEAD(bitmaps);
 	u64 num_entries;
 	u64 num_bitmaps;
 	u64 generation;
 	u8 type;
 	int ret = 0;
 
-	INIT_LIST_HEAD(&bitmaps);
-
 	/* Nothing in the space cache, goodbye */
 	if (!i_size_read(inode))
 		return 0;
@@ -1243,6 +1241,7 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	struct inode *inode;
 	int ret = 0;
+	enum btrfs_disk_cache_state dcs = BTRFS_DC_WRITTEN;
 
 	root = root->fs_info->tree_root;
 
@@ -1266,9 +1265,7 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 	ret = __btrfs_write_out_cache(root, inode, ctl, block_group, trans,
 				      path, block_group->key.objectid);
 	if (ret) {
-		spin_lock(&block_group->lock);
-		block_group->disk_cache_state = BTRFS_DC_ERROR;
-		spin_unlock(&block_group->lock);
+		dcs = BTRFS_DC_ERROR;
 		ret = 0;
 #ifdef DEBUG
 		btrfs_err(root->fs_info,
@@ -1277,6 +1274,9 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 #endif
 	}
 
+	spin_lock(&block_group->lock);
+	block_group->disk_cache_state = dcs;
+	spin_unlock(&block_group->lock);
 	iput(inode);
 	return ret;
 }
@@ -2903,7 +2903,6 @@ int btrfs_find_space_cluster(struct btrfs_root *root,
 	trace_btrfs_find_cluster(block_group, offset, bytes, empty_size,
 				 min_bytes);
 
-	INIT_LIST_HEAD(&bitmaps);
 	ret = setup_cluster_no_bitmap(block_group, cluster, &bitmaps, offset,
 				      bytes + empty_size,
 				      cont1_bytes, min_bytes);

fs/btrfs/inode-item.c

@@ -344,6 +344,7 @@ int btrfs_insert_inode_ref(struct btrfs_trans_handle *trans,
 		return -ENOMEM;
 
 	path->leave_spinning = 1;
+	path->skip_release_on_error = 1;
 	ret = btrfs_insert_empty_item(trans, root, path, &key,
 				      ins_len);
 	if (ret == -EEXIST) {
@@ -362,8 +363,12 @@ int btrfs_insert_inode_ref(struct btrfs_trans_handle *trans,
 		ptr = (unsigned long)(ref + 1);
 		ret = 0;
 	} else if (ret < 0) {
-		if (ret == -EOVERFLOW)
-			ret = -EMLINK;
+		if (ret == -EOVERFLOW) {
+			if (find_name_in_backref(path, name, name_len, &ref))
+				ret = -EEXIST;
+			else
+				ret = -EMLINK;
+		}
 		goto out;
 	} else {
 		ref = btrfs_item_ptr(path->nodes[0], path->slots[0],

fs/btrfs/qgroup.c

@@ -1431,9 +1431,8 @@ static int qgroup_excl_accounting(struct btrfs_fs_info *fs_info,
 		qgroup = u64_to_ptr(unode->aux);
 		qgroup->rfer += sign * oper->num_bytes;
 		qgroup->rfer_cmpr += sign * oper->num_bytes;
+		WARN_ON(sign < 0 && qgroup->excl < oper->num_bytes);
 		qgroup->excl += sign * oper->num_bytes;
-		if (sign < 0)
-			WARN_ON(qgroup->excl < oper->num_bytes);
 		qgroup->excl_cmpr += sign * oper->num_bytes;
 		qgroup_dirty(fs_info, qgroup);
 

fs/btrfs/raid56.c

@@ -58,15 +58,6 @@
  */
 #define RBIO_CACHE_READY_BIT	3
 
-/*
- * bbio and raid_map is managed by the caller, so we shouldn't free
- * them here. And besides that, all rbios with this flag should not
- * be cached, because we need raid_map to check the rbios' stripe
- * is the same or not, but it is very likely that the caller has
- * free raid_map, so don't cache those rbios.
- */
-#define RBIO_HOLD_BBIO_MAP_BIT	4
-
 #define RBIO_CACHE_SIZE 1024
 
 enum btrfs_rbio_ops {
@@ -79,13 +70,6 @@ struct btrfs_raid_bio {
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_bio *bbio;
 
-	/*
-	 * logical block numbers for the start of each stripe
-	 * The last one or two are p/q.  These are sorted,
-	 * so raid_map[0] is the start of our full stripe
-	 */
-	u64 *raid_map;
-
 	/* while we're doing rmw on a stripe
 	 * we put it into a hash table so we can
 	 * lock the stripe and merge more rbios
@@ -303,7 +287,7 @@ static void cache_rbio_pages(struct btrfs_raid_bio *rbio)
  */
 static int rbio_bucket(struct btrfs_raid_bio *rbio)
 {
-	u64 num = rbio->raid_map[0];
+	u64 num = rbio->bbio->raid_map[0];
 
 	/*
 	 * we shift down quite a bit.  We're using byte
@@ -606,8 +590,8 @@ static int rbio_can_merge(struct btrfs_raid_bio *last,
 	    test_bit(RBIO_CACHE_BIT, &cur->flags))
 		return 0;
 
-	if (last->raid_map[0] !=
-	    cur->raid_map[0])
+	if (last->bbio->raid_map[0] !=
+	    cur->bbio->raid_map[0])
 		return 0;
 
 	/* we can't merge with different operations */
@@ -689,7 +673,7 @@ static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
 	spin_lock_irqsave(&h->lock, flags);
 	list_for_each_entry(cur, &h->hash_list, hash_list) {
 		walk++;
-		if (cur->raid_map[0] == rbio->raid_map[0]) {
+		if (cur->bbio->raid_map[0] == rbio->bbio->raid_map[0]) {
 			spin_lock(&cur->bio_list_lock);
 
 			/* can we steal this cached rbio's pages? */
@@ -841,21 +825,6 @@ static noinline void unlock_stripe(struct btrfs_raid_bio *rbio)
 		remove_rbio_from_cache(rbio);
 }
 
-static inline void
-__free_bbio_and_raid_map(struct btrfs_bio *bbio, u64 *raid_map, int need)
-{
-	if (need) {
-		kfree(raid_map);
-		kfree(bbio);
-	}
-}
-
-static inline void free_bbio_and_raid_map(struct btrfs_raid_bio *rbio)
-{
-	__free_bbio_and_raid_map(rbio->bbio, rbio->raid_map,
-			!test_bit(RBIO_HOLD_BBIO_MAP_BIT, &rbio->flags));
-}
-
 static void __free_raid_bio(struct btrfs_raid_bio *rbio)
 {
 	int i;
@@ -875,8 +844,7 @@ static void __free_raid_bio(struct btrfs_raid_bio *rbio)
 		}
 	}
 
-	free_bbio_and_raid_map(rbio);
-
+	btrfs_put_bbio(rbio->bbio);
 	kfree(rbio);
 }
 
@@ -985,8 +953,7 @@ static unsigned long rbio_nr_pages(unsigned long stripe_len, int nr_stripes)
  * this does not allocate any pages for rbio->pages.
  */
 static struct btrfs_raid_bio *alloc_rbio(struct btrfs_root *root,
-			  struct btrfs_bio *bbio, u64 *raid_map,
-			  u64 stripe_len)
+			  struct btrfs_bio *bbio, u64 stripe_len)
 {
 	struct btrfs_raid_bio *rbio;
 	int nr_data = 0;
@@ -1007,7 +974,6 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_root *root,
 	INIT_LIST_HEAD(&rbio->stripe_cache);
 	INIT_LIST_HEAD(&rbio->hash_list);
 	rbio->bbio = bbio;
-	rbio->raid_map = raid_map;
 	rbio->fs_info = root->fs_info;
 	rbio->stripe_len = stripe_len;
 	rbio->nr_pages = num_pages;
@@ -1028,10 +994,12 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_root *root,
 	rbio->bio_pages = p + sizeof(struct page *) * num_pages;
 	rbio->dbitmap = p + sizeof(struct page *) * num_pages * 2;
 
-	if (raid_map[real_stripes - 1] == RAID6_Q_STRIPE)
+	if (bbio->map_type & BTRFS_BLOCK_GROUP_RAID5)
+		nr_data = real_stripes - 1;
+	else if (bbio->map_type & BTRFS_BLOCK_GROUP_RAID6)
 		nr_data = real_stripes - 2;
 	else
-		nr_data = real_stripes - 1;
+		BUG();
 
 	rbio->nr_data = nr_data;
 	return rbio;
@@ -1182,7 +1150,7 @@ static void index_rbio_pages(struct btrfs_raid_bio *rbio)
 	spin_lock_irq(&rbio->bio_list_lock);
 	bio_list_for_each(bio, &rbio->bio_list) {
 		start = (u64)bio->bi_iter.bi_sector << 9;
-		stripe_offset = start - rbio->raid_map[0];
+		stripe_offset = start - rbio->bbio->raid_map[0];
 		page_index = stripe_offset >> PAGE_CACHE_SHIFT;
 
 		for (i = 0; i < bio->bi_vcnt; i++) {
@@ -1402,7 +1370,7 @@ static int find_logical_bio_stripe(struct btrfs_raid_bio *rbio,
 	logical <<= 9;
 
 	for (i = 0; i < rbio->nr_data; i++) {
-		stripe_start = rbio->raid_map[i];
+		stripe_start = rbio->bbio->raid_map[i];
 		if (logical >= stripe_start &&
 		    logical < stripe_start + rbio->stripe_len) {
 			return i;
@@ -1776,17 +1744,16 @@ static void btrfs_raid_unplug(struct blk_plug_cb *cb, bool from_schedule)
  * our main entry point for writes from the rest of the FS.
  */
 int raid56_parity_write(struct btrfs_root *root, struct bio *bio,
-			struct btrfs_bio *bbio, u64 *raid_map,
-			u64 stripe_len)
+			struct btrfs_bio *bbio, u64 stripe_len)
 {
 	struct btrfs_raid_bio *rbio;
 	struct btrfs_plug_cb *plug = NULL;
 	struct blk_plug_cb *cb;
 	int ret;
 
-	rbio = alloc_rbio(root, bbio, raid_map, stripe_len);
+	rbio = alloc_rbio(root, bbio, stripe_len);
 	if (IS_ERR(rbio)) {
-		__free_bbio_and_raid_map(bbio, raid_map, 1);
+		btrfs_put_bbio(bbio);
 		return PTR_ERR(rbio);
 	}
 	bio_list_add(&rbio->bio_list, bio);
@@ -1885,9 +1852,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 		}
 
 		/* all raid6 handling here */
-		if (rbio->raid_map[rbio->real_stripes - 1] ==
-		    RAID6_Q_STRIPE) {
-
+		if (rbio->bbio->map_type & BTRFS_BLOCK_GROUP_RAID6) {
 			/*
 			 * single failure, rebuild from parity raid5
 			 * style
@@ -1922,8 +1887,9 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 			 * here due to a crc mismatch and we can't give them the
 			 * data they want
 			 */
-			if (rbio->raid_map[failb] == RAID6_Q_STRIPE) {
-				if (rbio->raid_map[faila] == RAID5_P_STRIPE) {
+			if (rbio->bbio->raid_map[failb] == RAID6_Q_STRIPE) {
+				if (rbio->bbio->raid_map[faila] ==
+				    RAID5_P_STRIPE) {
 					err = -EIO;
 					goto cleanup;
 				}
@@ -1934,7 +1900,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 				goto pstripe;
 			}
 
-			if (rbio->raid_map[failb] == RAID5_P_STRIPE) {
+			if (rbio->bbio->raid_map[failb] == RAID5_P_STRIPE) {
 				raid6_datap_recov(rbio->real_stripes,
 						  PAGE_SIZE, faila, pointers);
 			} else {
@@ -2001,8 +1967,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 
 cleanup_io:
 	if (rbio->operation == BTRFS_RBIO_READ_REBUILD) {
-		if (err == 0 &&
-		    !test_bit(RBIO_HOLD_BBIO_MAP_BIT, &rbio->flags))
+		if (err == 0)
 			cache_rbio_pages(rbio);
 		else
 			clear_bit(RBIO_CACHE_READY_BIT, &rbio->flags);
@@ -2156,15 +2121,16 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
  * of the drive.
  */
 int raid56_parity_recover(struct btrfs_root *root, struct bio *bio,
-			  struct btrfs_bio *bbio, u64 *raid_map,
-			  u64 stripe_len, int mirror_num, int generic_io)
+			  struct btrfs_bio *bbio, u64 stripe_len,
+			  int mirror_num, int generic_io)
 {
 	struct btrfs_raid_bio *rbio;
 	int ret;
 
-	rbio = alloc_rbio(root, bbio, raid_map, stripe_len);
+	rbio = alloc_rbio(root, bbio, stripe_len);
 	if (IS_ERR(rbio)) {
-		__free_bbio_and_raid_map(bbio, raid_map, generic_io);
+		if (generic_io)
+			btrfs_put_bbio(bbio);
 		return PTR_ERR(rbio);
 	}
 
@@ -2175,7 +2141,8 @@ int raid56_parity_recover(struct btrfs_root *root, struct bio *bio,
 	rbio->faila = find_logical_bio_stripe(rbio, bio);
 	if (rbio->faila == -1) {
 		BUG();
-		__free_bbio_and_raid_map(bbio, raid_map, generic_io);
+		if (generic_io)
+			btrfs_put_bbio(bbio);
 		kfree(rbio);
 		return -EIO;
 	}
@@ -2184,7 +2151,7 @@ int raid56_parity_recover(struct btrfs_root *root, struct bio *bio,
 		btrfs_bio_counter_inc_noblocked(root->fs_info);
 		rbio->generic_bio_cnt = 1;
 	} else {
-		set_bit(RBIO_HOLD_BBIO_MAP_BIT, &rbio->flags);
+		btrfs_get_bbio(bbio);
 	}
 
 	/*
@@ -2240,14 +2207,14 @@ static void read_rebuild_work(struct btrfs_work *work)
 
 struct btrfs_raid_bio *
 raid56_parity_alloc_scrub_rbio(struct btrfs_root *root, struct bio *bio,
-			       struct btrfs_bio *bbio, u64 *raid_map,
-			       u64 stripe_len, struct btrfs_device *scrub_dev,
+			       struct btrfs_bio *bbio, u64 stripe_len,
+			       struct btrfs_device *scrub_dev,
 			       unsigned long *dbitmap, int stripe_nsectors)
 {
 	struct btrfs_raid_bio *rbio;
 	int i;
 
-	rbio = alloc_rbio(root, bbio, raid_map, stripe_len);
+	rbio = alloc_rbio(root, bbio, stripe_len);
 	if (IS_ERR(rbio))
 		return NULL;
 	bio_list_add(&rbio->bio_list, bio);
@@ -2279,10 +2246,10 @@ void raid56_parity_add_scrub_pages(struct btrfs_raid_bio *rbio,
 	int stripe_offset;
 	int index;
 
-	ASSERT(logical >= rbio->raid_map[0]);
-	ASSERT(logical + PAGE_SIZE <= rbio->raid_map[0] +
+	ASSERT(logical >= rbio->bbio->raid_map[0]);
+	ASSERT(logical + PAGE_SIZE <= rbio->bbio->raid_map[0] +
 				rbio->stripe_len * rbio->nr_data);
-	stripe_offset = (int)(logical - rbio->raid_map[0]);
+	stripe_offset = (int)(logical - rbio->bbio->raid_map[0]);
 	index = stripe_offset >> PAGE_CACHE_SHIFT;
 	rbio->bio_pages[index] = page;
 }

fs/btrfs/raid56.h

@@ -43,16 +43,15 @@ struct btrfs_raid_bio;
 struct btrfs_device;
 
 int raid56_parity_recover(struct btrfs_root *root, struct bio *bio,
-			  struct btrfs_bio *bbio, u64 *raid_map,
-			  u64 stripe_len, int mirror_num, int generic_io);
+			  struct btrfs_bio *bbio, u64 stripe_len,
+			  int mirror_num, int generic_io);
 int raid56_parity_write(struct btrfs_root *root, struct bio *bio,
-			       struct btrfs_bio *bbio, u64 *raid_map,
-			       u64 stripe_len);
+			       struct btrfs_bio *bbio, u64 stripe_len);
 
 struct btrfs_raid_bio *
 raid56_parity_alloc_scrub_rbio(struct btrfs_root *root, struct bio *bio,
-			       struct btrfs_bio *bbio, u64 *raid_map,
-			       u64 stripe_len, struct btrfs_device *scrub_dev,
+			       struct btrfs_bio *bbio, u64 stripe_len,
+			       struct btrfs_device *scrub_dev,
 			       unsigned long *dbitmap, int stripe_nsectors);
 void raid56_parity_add_scrub_pages(struct btrfs_raid_bio *rbio,
 				   struct page *page, u64 logical);

fs/btrfs/reada.c

@@ -66,7 +66,6 @@ struct reada_extctl {
 struct reada_extent {
 	u64			logical;
 	struct btrfs_key	top;
-	u32			blocksize;
 	int			err;
 	struct list_head	extctl;
 	int 			refcnt;
@@ -349,7 +348,6 @@ static struct reada_extent *reada_find_extent(struct btrfs_root *root,
 
 	blocksize = root->nodesize;
 	re->logical = logical;
-	re->blocksize = blocksize;
 	re->top = *top;
 	INIT_LIST_HEAD(&re->extctl);
 	spin_lock_init(&re->lock);
@@ -463,7 +461,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_root *root,
 	spin_unlock(&fs_info->reada_lock);
 	btrfs_dev_replace_unlock(&fs_info->dev_replace);
 
-	kfree(bbio);
+	btrfs_put_bbio(bbio);
 	return re;
 
 error:
@@ -488,7 +486,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_root *root,
 		kref_put(&zone->refcnt, reada_zone_release);
 		spin_unlock(&fs_info->reada_lock);
 	}
-	kfree(bbio);
+	btrfs_put_bbio(bbio);
 	kfree(re);
 	return re_exist;
 }
@@ -660,7 +658,6 @@ static int reada_start_machine_dev(struct btrfs_fs_info *fs_info,
 	int mirror_num = 0;
 	struct extent_buffer *eb = NULL;
 	u64 logical;
-	u32 blocksize;
 	int ret;
 	int i;
 	int need_kick = 0;
@@ -694,7 +691,7 @@ static int reada_start_machine_dev(struct btrfs_fs_info *fs_info,
 		spin_unlock(&fs_info->reada_lock);
 		return 0;
 	}
-	dev->reada_next = re->logical + re->blocksize;
+	dev->reada_next = re->logical + fs_info->tree_root->nodesize;
 	re->refcnt++;
 
 	spin_unlock(&fs_info->reada_lock);
@@ -709,7 +706,6 @@ static int reada_start_machine_dev(struct btrfs_fs_info *fs_info,
 		}
 	}
 	logical = re->logical;
-	blocksize = re->blocksize;
 
 	spin_lock(&re->lock);
 	if (re->scheduled_for == NULL) {
@@ -724,8 +720,8 @@ static int reada_start_machine_dev(struct btrfs_fs_info *fs_info,
 		return 0;
 
 	atomic_inc(&dev->reada_in_flight);
-	ret = reada_tree_block_flagged(fs_info->extent_root, logical, blocksize,
-			 mirror_num, &eb);
+	ret = reada_tree_block_flagged(fs_info->extent_root, logical,
+			mirror_num, &eb);
 	if (ret)
 		__readahead_hook(fs_info->extent_root, NULL, logical, ret);
 	else if (eb)
@@ -851,7 +847,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all)
 				break;
 			printk(KERN_DEBUG
 				"  re: logical %llu size %u empty %d for %lld",
-				re->logical, re->blocksize,
+				re->logical, fs_info->tree_root->nodesize,
 				list_empty(&re->extctl), re->scheduled_for ?
 				re->scheduled_for->devid : -1);
 
@@ -886,7 +882,8 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all)
 		}
 		printk(KERN_DEBUG
 			"re: logical %llu size %u list empty %d for %lld",
-			re->logical, re->blocksize, list_empty(&re->extctl),
+			re->logical, fs_info->tree_root->nodesize,
+			list_empty(&re->extctl),
 			re->scheduled_for ? re->scheduled_for->devid : -1);
 		for (i = 0; i < re->nzones; ++i) {
 			printk(KERN_CONT " zone %llu-%llu devs",

fs/btrfs/relocation.c

@@ -2855,9 +2855,10 @@ static void update_processed_blocks(struct reloc_control *rc,
 	}
 }
 
-static int tree_block_processed(u64 bytenr, u32 blocksize,
-				struct reloc_control *rc)
+static int tree_block_processed(u64 bytenr, struct reloc_control *rc)
 {
+	u32 blocksize = rc->extent_root->nodesize;
+
 	if (test_range_bit(&rc->processed_blocks, bytenr,
 			   bytenr + blocksize - 1, EXTENT_DIRTY, 1, NULL))
 		return 1;
@@ -2965,8 +2966,7 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
 	while (rb_node) {
 		block = rb_entry(rb_node, struct tree_block, rb_node);
 		if (!block->key_ready)
-			readahead_tree_block(rc->extent_root, block->bytenr,
-					block->key.objectid);
+			readahead_tree_block(rc->extent_root, block->bytenr);
 		rb_node = rb_next(rb_node);
 	}
 
@@ -3353,7 +3353,7 @@ static int __add_tree_block(struct reloc_control *rc,
 	bool skinny = btrfs_fs_incompat(rc->extent_root->fs_info,
 					SKINNY_METADATA);
 
-	if (tree_block_processed(bytenr, blocksize, rc))
+	if (tree_block_processed(bytenr, rc))
 		return 0;
 
 	if (tree_search(blocks, bytenr))
@@ -3611,7 +3611,7 @@ static int find_data_references(struct reloc_control *rc,
 		if (added)
 			goto next;
 
-		if (!tree_block_processed(leaf->start, leaf->len, rc)) {
+		if (!tree_block_processed(leaf->start, rc)) {
 			block = kmalloc(sizeof(*block), GFP_NOFS);
 			if (!block) {
 				err = -ENOMEM;

fs/btrfs/send.c

@@ -2471,12 +2471,9 @@ verbose_printk("btrfs: send_utimes %llu\n", ino);
 	if (ret < 0)
 		goto out;
 	TLV_PUT_PATH(sctx, BTRFS_SEND_A_PATH, p);
-	TLV_PUT_BTRFS_TIMESPEC(sctx, BTRFS_SEND_A_ATIME, eb,
-			btrfs_inode_atime(ii));
-	TLV_PUT_BTRFS_TIMESPEC(sctx, BTRFS_SEND_A_MTIME, eb,
-			btrfs_inode_mtime(ii));
-	TLV_PUT_BTRFS_TIMESPEC(sctx, BTRFS_SEND_A_CTIME, eb,
-			btrfs_inode_ctime(ii));
+	TLV_PUT_BTRFS_TIMESPEC(sctx, BTRFS_SEND_A_ATIME, eb, &ii->atime);
+	TLV_PUT_BTRFS_TIMESPEC(sctx, BTRFS_SEND_A_MTIME, eb, &ii->mtime);
+	TLV_PUT_BTRFS_TIMESPEC(sctx, BTRFS_SEND_A_CTIME, eb, &ii->ctime);
 	/* TODO Add otime support when the otime patches get into upstream */
 
 	ret = send_cmd(sctx);

fs/btrfs/super.c

@@ -1958,11 +1958,6 @@ static int btrfs_freeze(struct super_block *sb)
 	return btrfs_commit_transaction(trans, root);
 }
 
-static int btrfs_unfreeze(struct super_block *sb)
-{
-	return 0;
-}
-
 static int btrfs_show_devname(struct seq_file *m, struct dentry *root)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(root->d_sb);
@@ -2011,7 +2006,6 @@ static const struct super_operations btrfs_super_ops = {
 	.statfs		= btrfs_statfs,
 	.remount_fs	= btrfs_remount,
 	.freeze_fs	= btrfs_freeze,
-	.unfreeze_fs	= btrfs_unfreeze,
 };
 
 static const struct file_operations btrfs_ctl_fops = {

fs/btrfs/sysfs.c

@@ -733,10 +733,18 @@ int btrfs_init_sysfs(void)
 
 	ret = btrfs_init_debugfs();
 	if (ret)
-		return ret;
+		goto out1;
 
 	init_feature_attrs();
 	ret = sysfs_create_group(&btrfs_kset->kobj, &btrfs_feature_attr_group);
+	if (ret)
+		goto out2;
+
+	return 0;
+out2:
+	debugfs_remove_recursive(btrfs_debugfs_root_dentry);
+out1:
+	kset_unregister(btrfs_kset);
 
 	return ret;
 }

fs/btrfs/tests/extent-buffer-tests.c

@@ -53,7 +53,7 @@ static int test_btrfs_split_item(void)
 		return -ENOMEM;
 	}
 
-	path->nodes[0] = eb = alloc_dummy_extent_buffer(0, 4096);
+	path->nodes[0] = eb = alloc_dummy_extent_buffer(NULL, 4096);
 	if (!eb) {
 		test_msg("Could not allocate dummy buffer\n");
 		ret = -ENOMEM;

fs/btrfs/tests/extent-io-tests.c

@@ -258,8 +258,7 @@ static int test_find_delalloc(void)
 	}
 	ret = 0;
 out_bits:
-	clear_extent_bits(&tmp, 0, total_dirty - 1,
-			  (unsigned long)-1, GFP_NOFS);
+	clear_extent_bits(&tmp, 0, total_dirty - 1, (unsigned)-1, GFP_NOFS);
 out:
 	if (locked_page)
 		page_cache_release(locked_page);

fs/btrfs/tests/inode-tests.c

@@ -255,7 +255,7 @@ static noinline int test_btrfs_get_extent(void)
 		goto out;
 	}
 
-	root->node = alloc_dummy_extent_buffer(0, 4096);
+	root->node = alloc_dummy_extent_buffer(NULL, 4096);
 	if (!root->node) {
 		test_msg("Couldn't allocate dummy buffer\n");
 		goto out;
@@ -843,7 +843,7 @@ static int test_hole_first(void)
 		goto out;
 	}
 
-	root->node = alloc_dummy_extent_buffer(0, 4096);
+	root->node = alloc_dummy_extent_buffer(NULL, 4096);
 	if (!root->node) {
 		test_msg("Couldn't allocate dummy buffer\n");
 		goto out;

fs/btrfs/tests/qgroup-tests.c

@@ -404,12 +404,22 @@ int btrfs_test_qgroups(void)
 		ret = -ENOMEM;
 		goto out;
 	}
+	/* We are using this root as our extent root */
+	root->fs_info->extent_root = root;
+
+	/*
+	 * Some of the paths we test assume we have a filled out fs_info, so we
+	 * just need to add the root in there so we don't panic.
+	 */
+	root->fs_info->tree_root = root;
+	root->fs_info->quota_root = root;
+	root->fs_info->quota_enabled = 1;
 
 	/*
 	 * Can't use bytenr 0, some things freak out
 	 * *cough*backref walking code*cough*
 	 */
-	root->node = alloc_test_extent_buffer(root->fs_info, 4096, 4096);
+	root->node = alloc_test_extent_buffer(root->fs_info, 4096);
 	if (!root->node) {
 		test_msg("Couldn't allocate dummy buffer\n");
 		ret = -ENOMEM;
@@ -448,17 +458,6 @@ int btrfs_test_qgroups(void)
 		goto out;
 	}
 
-	/* We are using this root as our extent root */
-	root->fs_info->extent_root = root;
-
-	/*
-	 * Some of the paths we test assume we have a filled out fs_info, so we
-	 * just need to addt he root in there so we don't panic.
-	 */
-	root->fs_info->tree_root = root;
-	root->fs_info->quota_root = root;
-	root->fs_info->quota_enabled = 1;
-
 	test_msg("Running qgroup tests\n");
 	ret = test_no_shared_qgroup(root);
 	if (ret)

fs/btrfs/transaction.c

@@ -220,6 +220,7 @@ static noinline int join_transaction(struct btrfs_root *root, unsigned int type)
 	 * commit the transaction.
 	 */
 	atomic_set(&cur_trans->use_count, 2);
+	cur_trans->have_free_bgs = 0;
 	cur_trans->start_time = get_seconds();
 
 	cur_trans->delayed_refs.href_root = RB_ROOT;
@@ -248,6 +249,8 @@ static noinline int join_transaction(struct btrfs_root *root, unsigned int type)
 	INIT_LIST_HEAD(&cur_trans->pending_chunks);
 	INIT_LIST_HEAD(&cur_trans->switch_commits);
 	INIT_LIST_HEAD(&cur_trans->pending_ordered);
+	INIT_LIST_HEAD(&cur_trans->dirty_bgs);
+	spin_lock_init(&cur_trans->dirty_bgs_lock);
 	list_add_tail(&cur_trans->list, &fs_info->trans_list);
 	extent_io_tree_init(&cur_trans->dirty_pages,
 			     fs_info->btree_inode->i_mapping);
@@ -1020,6 +1023,7 @@ static int update_cowonly_root(struct btrfs_trans_handle *trans,
 	u64 old_root_bytenr;
 	u64 old_root_used;
 	struct btrfs_root *tree_root = root->fs_info->tree_root;
+	bool extent_root = (root->objectid == BTRFS_EXTENT_TREE_OBJECTID);
 
 	old_root_used = btrfs_root_used(&root->root_item);
 	btrfs_write_dirty_block_groups(trans, root);
@@ -1027,7 +1031,9 @@ static int update_cowonly_root(struct btrfs_trans_handle *trans,
 	while (1) {
 		old_root_bytenr = btrfs_root_bytenr(&root->root_item);
 		if (old_root_bytenr == root->node->start &&
-		    old_root_used == btrfs_root_used(&root->root_item))
+		    old_root_used == btrfs_root_used(&root->root_item) &&
+		    (!extent_root ||
+		     list_empty(&trans->transaction->dirty_bgs)))
 			break;
 
 		btrfs_set_root_node(&root->root_item, root->node);
@@ -1038,7 +1044,15 @@ static int update_cowonly_root(struct btrfs_trans_handle *trans,
 			return ret;
 
 		old_root_used = btrfs_root_used(&root->root_item);
-		ret = btrfs_write_dirty_block_groups(trans, root);
+		if (extent_root) {
+			ret = btrfs_write_dirty_block_groups(trans, root);
+			if (ret)
+				return ret;
+		}
+		ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
+		if (ret)
+			return ret;
+		ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
 		if (ret)
 			return ret;
 	}
@@ -1061,10 +1075,6 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans,
 	struct extent_buffer *eb;
 	int ret;
 
-	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
-	if (ret)
-		return ret;
-
 	eb = btrfs_lock_root_node(fs_info->tree_root);
 	ret = btrfs_cow_block(trans, fs_info->tree_root, eb, NULL,
 			      0, &eb);
@@ -1097,6 +1107,7 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans,
 		next = fs_info->dirty_cowonly_roots.next;
 		list_del_init(next);
 		root = list_entry(next, struct btrfs_root, dirty_list);
+		clear_bit(BTRFS_ROOT_DIRTY, &root->state);
 
 		if (root != fs_info->extent_root)
 			list_add_tail(&root->dirty_list,
@@ -1983,6 +1994,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 	switch_commit_roots(cur_trans, root->fs_info);
 
 	assert_qgroups_uptodate(trans);
+	ASSERT(list_empty(&cur_trans->dirty_bgs));
 	update_super_roots(root);
 
 	btrfs_set_super_log_root(root->fs_info->super_copy, 0);
@@ -2026,6 +2038,9 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 	btrfs_finish_extent_commit(trans, root);
 
+	if (cur_trans->have_free_bgs)
+		btrfs_clear_space_info_full(root->fs_info);
+
 	root->fs_info->last_trans_committed = cur_trans->transid;
 	/*
 	 * We needn't acquire the lock here because there is no other task

fs/btrfs/transaction.h

@@ -47,6 +47,11 @@ struct btrfs_transaction {
 	atomic_t num_writers;
 	atomic_t use_count;
 
+	/*
+	 * true if there is free bgs operations in this transaction
+	 */
+	int have_free_bgs;
+
 	/* Be protected by fs_info->trans_lock when we want to change it. */
 	enum btrfs_trans_state state;
 	struct list_head list;
@@ -58,6 +63,8 @@ struct btrfs_transaction {
 	struct list_head pending_chunks;
 	struct list_head pending_ordered;
 	struct list_head switch_commits;
+	struct list_head dirty_bgs;
+	spinlock_t dirty_bgs_lock;
 	struct btrfs_delayed_ref_root delayed_refs;
 	int aborted;
 };

fs/btrfs/volumes.h

@@ -295,8 +295,10 @@ typedef void (btrfs_bio_end_io_t) (struct btrfs_bio *bio, int err);
 #define BTRFS_BIO_ORIG_BIO_SUBMITTED	(1 << 0)
 
 struct btrfs_bio {
+	atomic_t refs;
 	atomic_t stripes_pending;
 	struct btrfs_fs_info *fs_info;
+	u64 map_type; /* get from map_lookup->type */
 	bio_end_io_t *end_io;
 	struct bio *orig_bio;
 	unsigned long flags;
@@ -307,6 +309,12 @@ struct btrfs_bio {
 	int mirror_num;
 	int num_tgtdevs;
 	int *tgtdev_map;
+	/*
+	 * logical block numbers for the start of each stripe
+	 * The last one or two are p/q.  These are sorted,
+	 * so raid_map[0] is the start of our full stripe
+	 */
+	u64 *raid_map;
 	struct btrfs_bio_stripe stripes[];
 };
 
@@ -388,19 +396,15 @@ struct btrfs_balance_control {
 
 int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
 				   u64 end, u64 *length);
-
-#define btrfs_bio_size(total_stripes, real_stripes)		\
-	(sizeof(struct btrfs_bio) +				\
-	 (sizeof(struct btrfs_bio_stripe) * (total_stripes)) +	\
-	 (sizeof(int) * (real_stripes)))
-
+void btrfs_get_bbio(struct btrfs_bio *bbio);
+void btrfs_put_bbio(struct btrfs_bio *bbio);
 int btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		    u64 logical, u64 *length,
 		    struct btrfs_bio **bbio_ret, int mirror_num);
 int btrfs_map_sblock(struct btrfs_fs_info *fs_info, int rw,
 		     u64 logical, u64 *length,
 		     struct btrfs_bio **bbio_ret, int mirror_num,
-		     u64 **raid_map_ret);
+		     int need_raid_map);
 int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
 		     u64 chunk_start, u64 physical, u64 devid,
 		     u64 **logical, int *naddrs, int *stripe_len);

include/uapi/linux/btrfs.h

@@ -495,8 +495,7 @@ struct btrfs_ioctl_send_args {
 
 /* Error codes as returned by the kernel */
 enum btrfs_err_code {
-	notused,
-	BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET,
+	BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET = 1,
 	BTRFS_ERROR_DEV_RAID10_MIN_NOT_MET,
 	BTRFS_ERROR_DEV_RAID5_MIN_NOT_MET,
 	BTRFS_ERROR_DEV_RAID6_MIN_NOT_MET,