Merge tag 'for-4.18-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "User visible features:

   - added support for the ioctl FS_IOC_FSGETXATTR, per-inode flags,
     successor of GET/SETFLAGS; now supports only existing flags:
     append, immutable, noatime, nodump, sync

   - 3 new unprivileged ioctls to allow users to enumerate subvolumes

   - dedupe syscall implementation does not restrict the range to 16MiB,
     though it still splits the whole range to 16MiB chunks

   - on user demand, rmdir() is able to delete an empty subvolume,
     export the capability in sysfs

   - fix inode number types in tracepoints, other cleanups

   - send: improved speed when dealing with a large removed directory,
     measurements show decrease from 2000 minutes to 2 minutes on a
     directory with 2 million entries

   - pre-commit check of superblock to detect a mysterious in-memory
     corruption

   - log message updates

  Other changes:

   - orphan inode cleanup improved, does no keep long-standing
     reservations that could lead up to early ENOSPC in some cases

   - slight improvement of handling snapshotted NOCOW files by avoiding
     some unnecessary tree searches

   - avoid OOM when dealing with many unmergeable small extents at flush
     time

   - speedup conversion of free space tree representations from/to
     bitmap/tree

   - code refactoring, deletion, cleanups:
      + delayed refs
      + delayed iput
      + redundant argument removals
      + memory barrier cleanups
      + remove a redundant mutex supposedly excluding several ioctls to
        run in parallel

   - new tracepoints for blockgroup manipulation

   - more sanity checks of compressed headers"

* tag 'for-4.18-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (183 commits)
  btrfs: Add unprivileged version of ino_lookup ioctl
  btrfs: Add unprivileged ioctl which returns subvolume's ROOT_REF
  btrfs: Add unprivileged ioctl which returns subvolume information
  Btrfs: clean up error handling in btrfs_truncate()
  btrfs: Factor out write portion of btrfs_get_blocks_direct
  btrfs: Factor out read portion of btrfs_get_blocks_direct
  btrfs: return ENOMEM if path allocation fails in btrfs_cross_ref_exist
  btrfs: raid56: Remove VLA usage
  btrfs: return error value if create_io_em failed in cow_file_range
  btrfs: drop useless member qgroup_reserved of btrfs_pending_snapshot
  btrfs: drop unused parameter qgroup_reserved
  btrfs: balance dirty metadata pages in btrfs_finish_ordered_io
  btrfs: lift some btrfs_cross_ref_exist checks in nocow path
  btrfs: Remove fs_info argument from btrfs_uuid_tree_rem
  btrfs: Remove fs_info argument from btrfs_uuid_tree_add
  Btrfs: remove unused check of skip_locking
  Btrfs: remove always true check in unlock_up
  Btrfs: grab write lock directly if write_lock_level is the max level
  Btrfs: move get root out of btrfs_search_slot to a helper
  Btrfs: use more straightforward extent_buffer_uptodate check
  ...

fs/btrfs/btrfs_inode.h

@@ -19,17 +19,17 @@
  * ordered operations list so that we make sure to flush out any
  * new data the application may have written before commit.
  */
-#define BTRFS_INODE_ORDERED_DATA_CLOSE		0
-#define BTRFS_INODE_ORPHAN_META_RESERVED	1
-#define BTRFS_INODE_DUMMY			2
-#define BTRFS_INODE_IN_DEFRAG			3
-#define BTRFS_INODE_HAS_ORPHAN_ITEM		4
-#define BTRFS_INODE_HAS_ASYNC_EXTENT		5
-#define BTRFS_INODE_NEEDS_FULL_SYNC		6
-#define BTRFS_INODE_COPY_EVERYTHING		7
-#define BTRFS_INODE_IN_DELALLOC_LIST		8
-#define BTRFS_INODE_READDIO_NEED_LOCK		9
-#define BTRFS_INODE_HAS_PROPS		        10
+enum {
+	BTRFS_INODE_ORDERED_DATA_CLOSE = 0,
+	BTRFS_INODE_DUMMY,
+	BTRFS_INODE_IN_DEFRAG,
+	BTRFS_INODE_HAS_ASYNC_EXTENT,
+	BTRFS_INODE_NEEDS_FULL_SYNC,
+	BTRFS_INODE_COPY_EVERYTHING,
+	BTRFS_INODE_IN_DELALLOC_LIST,
+	BTRFS_INODE_READDIO_NEED_LOCK,
+	BTRFS_INODE_HAS_PROPS,
+};
 
 /* in memory btrfs inode */
 struct btrfs_inode {

fs/btrfs/compression.c

@@ -990,12 +990,7 @@ static void __free_workspace(int type, struct list_head *workspace,
 		btrfs_compress_op[idx]->free_workspace(workspace);
 	atomic_dec(total_ws);
 wake:
-	/*
-	 * Make sure counter is updated before we wake up waiters.
-	 */
-	smp_mb();
-	if (waitqueue_active(ws_wait))
-		wake_up(ws_wait);
+	cond_wake_up(ws_wait);
 }
 
 static void free_workspace(int type, struct list_head *ws)

fs/btrfs/compression.h

@@ -6,6 +6,8 @@
 #ifndef BTRFS_COMPRESSION_H
 #define BTRFS_COMPRESSION_H
 
+#include <linux/sizes.h>
+
 /*
  * We want to make sure that amount of RAM required to uncompress an extent is
  * reasonable, so we limit the total size in ram of a compressed extent to

fs/btrfs/ctree.c

@@ -2330,7 +2330,7 @@ static noinline void unlock_up(struct btrfs_path *path, int level,
 			no_skips = 1;
 
 		t = path->nodes[i];
-		if (i >= lowest_unlock && i > skip_level && path->locks[i]) {
+		if (i >= lowest_unlock && i > skip_level) {
 			btrfs_tree_unlock_rw(t, path->locks[i]);
 			path->locks[i] = 0;
 			if (write_lock_level &&
@@ -2432,7 +2432,6 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
 	btrfs_unlock_up_safe(p, level + 1);
 	btrfs_set_path_blocking(p);
 
-	free_extent_buffer(tmp);
 	if (p->reada != READA_NONE)
 		reada_for_search(fs_info, p, level, slot, key->objectid);
 
@@ -2446,7 +2445,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
 		 * and give up so that our caller doesn't loop forever
 		 * on our EAGAINs.
 		 */
-		if (!btrfs_buffer_uptodate(tmp, 0, 0))
+		if (!extent_buffer_uptodate(tmp))
 			ret = -EIO;
 		free_extent_buffer(tmp);
 	} else {
@@ -2599,6 +2598,78 @@ int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
 	return 0;
 }
 
+static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
+							struct btrfs_path *p,
+							int write_lock_level)
+{
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct extent_buffer *b;
+	int root_lock;
+	int level = 0;
+
+	/* We try very hard to do read locks on the root */
+	root_lock = BTRFS_READ_LOCK;
+
+	if (p->search_commit_root) {
+		/* The commit roots are read only so we always do read locks */
+		if (p->need_commit_sem)
+			down_read(&fs_info->commit_root_sem);
+		b = root->commit_root;
+		extent_buffer_get(b);
+		level = btrfs_header_level(b);
+		if (p->need_commit_sem)
+			up_read(&fs_info->commit_root_sem);
+		/*
+		 * Ensure that all callers have set skip_locking when
+		 * p->search_commit_root = 1.
+		 */
+		ASSERT(p->skip_locking == 1);
+
+		goto out;
+	}
+
+	if (p->skip_locking) {
+		b = btrfs_root_node(root);
+		level = btrfs_header_level(b);
+		goto out;
+	}
+
+	/*
+	 * If the level is set to maximum, we can skip trying to get the read
+	 * lock.
+	 */
+	if (write_lock_level < BTRFS_MAX_LEVEL) {
+		/*
+		 * We don't know the level of the root node until we actually
+		 * have it read locked
+		 */
+		b = btrfs_read_lock_root_node(root);
+		level = btrfs_header_level(b);
+		if (level > write_lock_level)
+			goto out;
+
+		/* Whoops, must trade for write lock */
+		btrfs_tree_read_unlock(b);
+		free_extent_buffer(b);
+	}
+
+	b = btrfs_lock_root_node(root);
+	root_lock = BTRFS_WRITE_LOCK;
+
+	/* The level might have changed, check again */
+	level = btrfs_header_level(b);
+
+out:
+	p->nodes[level] = b;
+	if (!p->skip_locking)
+		p->locks[level] = root_lock;
+	/*
+	 * Callers are responsible for dropping b's references.
+	 */
+	return b;
+}
+
+
 /*
  * btrfs_search_slot - look for a key in a tree and perform necessary
  * modifications to preserve tree invariants.
@@ -2635,7 +2706,6 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	int err;
 	int level;
 	int lowest_unlock = 1;
-	int root_lock;
 	/* everything at write_lock_level or lower must be write locked */
 	int write_lock_level = 0;
 	u8 lowest_level = 0;
@@ -2673,50 +2743,7 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 
 again:
 	prev_cmp = -1;
-	/*
-	 * we try very hard to do read locks on the root
-	 */
-	root_lock = BTRFS_READ_LOCK;
-	level = 0;
-	if (p->search_commit_root) {
-		/*
-		 * the commit roots are read only
-		 * so we always do read locks
-		 */
-		if (p->need_commit_sem)
-			down_read(&fs_info->commit_root_sem);
-		b = root->commit_root;
-		extent_buffer_get(b);
-		level = btrfs_header_level(b);
-		if (p->need_commit_sem)
-			up_read(&fs_info->commit_root_sem);
-		if (!p->skip_locking)
-			btrfs_tree_read_lock(b);
-	} else {
-		if (p->skip_locking) {
-			b = btrfs_root_node(root);
-			level = btrfs_header_level(b);
-		} else {
-			/* we don't know the level of the root node
-			 * until we actually have it read locked
-			 */
-			b = btrfs_read_lock_root_node(root);
-			level = btrfs_header_level(b);
-			if (level <= write_lock_level) {
-				/* whoops, must trade for write lock */
-				btrfs_tree_read_unlock(b);
-				free_extent_buffer(b);
-				b = btrfs_lock_root_node(root);
-				root_lock = BTRFS_WRITE_LOCK;
-
-				/* the level might have changed, check again */
-				level = btrfs_header_level(b);
-			}
-		}
-	}
-	p->nodes[level] = b;
-	if (!p->skip_locking)
-		p->locks[level] = root_lock;
+	b = btrfs_search_slot_get_root(root, p, write_lock_level);
 
 	while (b) {
 		level = btrfs_header_level(b);

fs/btrfs/ctree.h

@@ -739,6 +739,12 @@ struct btrfs_delayed_root;
  */
 #define BTRFS_FS_NEED_ASYNC_COMMIT		17
 
+/*
+ * Indicate that balance has been set up from the ioctl and is in the main
+ * phase. The fs_info::balance_ctl is initialized.
+ */
+#define BTRFS_FS_BALANCE_RUNNING		18
+
 struct btrfs_fs_info {
 	u8 fsid[BTRFS_FSID_SIZE];
 	u8 chunk_tree_uuid[BTRFS_UUID_SIZE];
@@ -838,7 +844,6 @@ struct btrfs_fs_info {
 	struct mutex transaction_kthread_mutex;
 	struct mutex cleaner_mutex;
 	struct mutex chunk_mutex;
-	struct mutex volume_mutex;
 
 	/*
 	 * this is taken to make sure we don't set block groups ro after
@@ -1004,7 +1009,6 @@ struct btrfs_fs_info {
 	/* restriper state */
 	spinlock_t balance_lock;
 	struct mutex balance_mutex;
-	atomic_t balance_running;
 	atomic_t balance_pause_req;
 	atomic_t balance_cancel_req;
 	struct btrfs_balance_control *balance_ctl;
@@ -1219,9 +1223,6 @@ struct btrfs_root {
 	spinlock_t log_extents_lock[2];
 	struct list_head logged_list[2];
 
-	spinlock_t orphan_lock;
-	atomic_t orphan_inodes;
-	struct btrfs_block_rsv *orphan_block_rsv;
 	int orphan_cleanup_state;
 
 	spinlock_t inode_lock;
@@ -2764,13 +2765,9 @@ void btrfs_delalloc_release_space(struct inode *inode,
 void btrfs_free_reserved_data_space_noquota(struct inode *inode, u64 start,
 					    u64 len);
 void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans);
-int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
-				  struct btrfs_inode *inode);
-void btrfs_orphan_release_metadata(struct btrfs_inode *inode);
 int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 				     struct btrfs_block_rsv *rsv,
-				     int nitems,
-				     u64 *qgroup_reserved, bool use_global_rsv);
+				     int nitems, bool use_global_rsv);
 void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info,
 				      struct btrfs_block_rsv *rsv);
 void btrfs_delalloc_release_extents(struct btrfs_inode *inode, u64 num_bytes,
@@ -2828,7 +2825,7 @@ void btrfs_wait_for_snapshot_creation(struct btrfs_root *root);
 void check_system_chunk(struct btrfs_trans_handle *trans,
 			struct btrfs_fs_info *fs_info, const u64 type);
 u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
-		       struct btrfs_fs_info *info, u64 start, u64 end);
+		       u64 start, u64 end);
 
 /* ctree.c */
 int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
@@ -3042,11 +3039,9 @@ void btrfs_update_root_times(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root);
 
 /* uuid-tree.c */
-int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, u8 *uuid, u8 type,
+int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 			u64 subid);
-int btrfs_uuid_tree_rem(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, u8 *uuid, u8 type,
+int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 			u64 subid);
 int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info,
 			    int (*check_func)(struct btrfs_fs_info *, u8 *, u8,
@@ -3163,18 +3158,6 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 				     struct extent_map *em);
 
 /* inode.c */
-struct btrfs_delalloc_work {
-	struct inode *inode;
-	int delay_iput;
-	struct completion completion;
-	struct list_head list;
-	struct btrfs_work work;
-};
-
-struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
-						    int delay_iput);
-void btrfs_wait_and_free_delalloc_work(struct btrfs_delalloc_work *work);
-
 struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
 		struct page *page, size_t pg_offset, u64 start,
 		u64 len, int create);
@@ -3193,10 +3176,7 @@ int btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 int btrfs_add_link(struct btrfs_trans_handle *trans,
 		   struct btrfs_inode *parent_inode, struct btrfs_inode *inode,
 		   const char *name, int name_len, int add_backref, u64 index);
-int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
-			struct btrfs_root *root,
-			struct inode *dir, u64 objectid,
-			const char *name, int name_len);
+int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry);
 int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len,
 			int front);
 int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
@@ -3204,9 +3184,8 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 			       struct inode *inode, u64 new_size,
 			       u32 min_type);
 
-int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput);
-int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput,
-			       int nr);
+int btrfs_start_delalloc_inodes(struct btrfs_root *root);
+int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr);
 int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
 			      unsigned int extra_bits,
 			      struct extent_state **cached_state, int dedupe);
@@ -3240,10 +3219,7 @@ int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
 int btrfs_orphan_add(struct btrfs_trans_handle *trans,
 		struct btrfs_inode *inode);
 int btrfs_orphan_cleanup(struct btrfs_root *root);
-void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans,
-			      struct btrfs_root *root);
 int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size);
-void btrfs_invalidate_inodes(struct btrfs_root *root);
 void btrfs_add_delayed_iput(struct inode *inode);
 void btrfs_run_delayed_iputs(struct btrfs_fs_info *fs_info);
 int btrfs_prealloc_file_range(struct inode *inode, int mode,
@@ -3262,14 +3238,14 @@ void btrfs_test_inode_set_ops(struct inode *inode);
 long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 int btrfs_ioctl_get_supported_features(void __user *arg);
-void btrfs_update_iflags(struct inode *inode);
+void btrfs_sync_inode_flags_to_i_flags(struct inode *inode);
 int btrfs_is_empty_uuid(u8 *uuid);
 int btrfs_defrag_file(struct inode *inode, struct file *file,
 		      struct btrfs_ioctl_defrag_range_args *range,
 		      u64 newer_than, unsigned long max_pages);
 void btrfs_get_block_group_info(struct list_head *groups_list,
 				struct btrfs_ioctl_space_info *space);
-void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock,
+void btrfs_update_ioctl_balance_args(struct btrfs_fs_info *fs_info,
 			       struct btrfs_ioctl_balance_args *bargs);
 ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen,
 			   struct file *dst_file, u64 dst_loff);
@@ -3767,4 +3743,26 @@ static inline int btrfs_is_testing(struct btrfs_fs_info *fs_info)
 	return 0;
 }
 
+static inline void cond_wake_up(struct wait_queue_head *wq)
+{
+	/*
+	 * This implies a full smp_mb barrier, see comments for
+	 * waitqueue_active why.
+	 */
+	if (wq_has_sleeper(wq))
+		wake_up(wq);
+}
+
+static inline void cond_wake_up_nomb(struct wait_queue_head *wq)
+{
+	/*
+	 * Special case for conditional wakeup where the barrier required for
+	 * waitqueue_active is implied by some of the preceding code. Eg. one
+	 * of such atomic operations (atomic_dec_and_return, ...), or a
+	 * unlock/lock sequence, etc.
+	 */
+	if (waitqueue_active(wq))
+		wake_up(wq);
+}
+
 #endif

fs/btrfs/delayed-inode.c

@@ -460,13 +460,10 @@ static void finish_one_item(struct btrfs_delayed_root *delayed_root)
 {
 	int seq = atomic_inc_return(&delayed_root->items_seq);
 
-	/*
-	 * atomic_dec_return implies a barrier for waitqueue_active
-	 */
+	/* atomic_dec_return implies a barrier */
 	if ((atomic_dec_return(&delayed_root->items) <
-	    BTRFS_DELAYED_BACKGROUND || seq % BTRFS_DELAYED_BATCH == 0) &&
-	    waitqueue_active(&delayed_root->wait))
-		wake_up(&delayed_root->wait);
+	    BTRFS_DELAYED_BACKGROUND || seq % BTRFS_DELAYED_BATCH == 0))
+		cond_wake_up_nomb(&delayed_root->wait);
 }
 
 static void __btrfs_remove_delayed_item(struct btrfs_delayed_item *delayed_item)

fs/btrfs/delayed-ref.c

@@ -286,10 +286,10 @@ static bool merge_ref(struct btrfs_trans_handle *trans,
 }
 
 void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
-			      struct btrfs_fs_info *fs_info,
 			      struct btrfs_delayed_ref_root *delayed_refs,
 			      struct btrfs_delayed_ref_head *head)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_delayed_ref_node *ref;
 	struct rb_node *node;
 	u64 seq = 0;
@@ -323,9 +323,7 @@ again:
 	}
 }
 
-int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
-			    struct btrfs_delayed_ref_root *delayed_refs,
-			    u64 seq)
+int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq)
 {
 	struct seq_list *elem;
 	int ret = 0;
@@ -336,10 +334,9 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
 					struct seq_list, list);
 		if (seq >= elem->seq) {
 			btrfs_debug(fs_info,
-				"holding back delayed_ref %#x.%x, lowest is %#x.%x (%p)",
+				"holding back delayed_ref %#x.%x, lowest is %#x.%x",
 				(u32)(seq >> 32), (u32)seq,
-				(u32)(elem->seq >> 32), (u32)elem->seq,
-				delayed_refs);
+				(u32)(elem->seq >> 32), (u32)elem->seq);
 			ret = 1;
 		}
 	}
@@ -529,33 +526,20 @@ update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs,
 	spin_unlock(&existing->lock);
 }
 
-/*
- * helper function to actually insert a head node into the rbtree.
- * this does all the dirty work in terms of maintaining the correct
- * overall modification count.
- */
-static noinline struct btrfs_delayed_ref_head *
-add_delayed_ref_head(struct btrfs_fs_info *fs_info,
-		     struct btrfs_trans_handle *trans,
-		     struct btrfs_delayed_ref_head *head_ref,
-		     struct btrfs_qgroup_extent_record *qrecord,
-		     u64 bytenr, u64 num_bytes, u64 ref_root, u64 reserved,
-		     int action, int is_data, int is_system,
-		     int *qrecord_inserted_ret,
-		     int *old_ref_mod, int *new_ref_mod)
-
+static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref,
+				  struct btrfs_qgroup_extent_record *qrecord,
+				  u64 bytenr, u64 num_bytes, u64 ref_root,
+				  u64 reserved, int action, bool is_data,
+				  bool is_system)
 {
-	struct btrfs_delayed_ref_head *existing;
-	struct btrfs_delayed_ref_root *delayed_refs;
 	int count_mod = 1;
 	int must_insert_reserved = 0;
-	int qrecord_inserted = 0;
 
 	/* If reserved is provided, it must be a data extent. */
 	BUG_ON(!is_data && reserved);
 
 	/*
-	 * the head node stores the sum of all the mods, so dropping a ref
+	 * The head node stores the sum of all the mods, so dropping a ref
 	 * should drop the sum in the head node by one.
 	 */
 	if (action == BTRFS_UPDATE_DELAYED_HEAD)
@@ -564,12 +548,11 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 		count_mod = -1;
 
 	/*
-	 * BTRFS_ADD_DELAYED_EXTENT means that we need to update
-	 * the reserved accounting when the extent is finally added, or
-	 * if a later modification deletes the delayed ref without ever
-	 * inserting the extent into the extent allocation tree.
-	 * ref->must_insert_reserved is the flag used to record
-	 * that accounting mods are required.
+	 * BTRFS_ADD_DELAYED_EXTENT means that we need to update the reserved
+	 * accounting when the extent is finally added, or if a later
+	 * modification deletes the delayed ref without ever inserting the
+	 * extent into the extent allocation tree.  ref->must_insert_reserved
+	 * is the flag used to record that accounting mods are required.
 	 *
 	 * Once we record must_insert_reserved, switch the action to
 	 * BTRFS_ADD_DELAYED_REF because other special casing is not required.
@@ -579,8 +562,6 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 	else
 		must_insert_reserved = 0;
 
-	delayed_refs = &trans->transaction->delayed_refs;
-
 	refcount_set(&head_ref->refs, 1);
 	head_ref->bytenr = bytenr;
 	head_ref->num_bytes = num_bytes;
@@ -598,7 +579,6 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 	spin_lock_init(&head_ref->lock);
 	mutex_init(&head_ref->mutex);
 
-	/* Record qgroup extent info if provided */
 	if (qrecord) {
 		if (ref_root && reserved) {
 			head_ref->qgroup_ref_root = ref_root;
@@ -608,20 +588,44 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 		qrecord->bytenr = bytenr;
 		qrecord->num_bytes = num_bytes;
 		qrecord->old_roots = NULL;
+	}
+}
+
+/*
+ * helper function to actually insert a head node into the rbtree.
+ * this does all the dirty work in terms of maintaining the correct
+ * overall modification count.
+ */
+static noinline struct btrfs_delayed_ref_head *
+add_delayed_ref_head(struct btrfs_trans_handle *trans,
+		     struct btrfs_delayed_ref_head *head_ref,
+		     struct btrfs_qgroup_extent_record *qrecord,
+		     int action, int *qrecord_inserted_ret,
+		     int *old_ref_mod, int *new_ref_mod)
+{
+	struct btrfs_delayed_ref_head *existing;
+	struct btrfs_delayed_ref_root *delayed_refs;
+	int qrecord_inserted = 0;
 
-		if(btrfs_qgroup_trace_extent_nolock(fs_info,
+	delayed_refs = &trans->transaction->delayed_refs;
+
+	/* Record qgroup extent info if provided */
+	if (qrecord) {
+		if (btrfs_qgroup_trace_extent_nolock(trans->fs_info,
 					delayed_refs, qrecord))
 			kfree(qrecord);
 		else
 			qrecord_inserted = 1;
 	}
 
-	trace_add_delayed_ref_head(fs_info, head_ref, action);
+	trace_add_delayed_ref_head(trans->fs_info, head_ref, action);
 
 	existing = htree_insert(&delayed_refs->href_root,
 				&head_ref->href_node);
 	if (existing) {
-		WARN_ON(ref_root && reserved && existing->qgroup_ref_root
+		WARN_ON(qrecord && head_ref->qgroup_ref_root
+			&& head_ref->qgroup_reserved
+			&& existing->qgroup_ref_root
 			&& existing->qgroup_reserved);
 		update_existing_head_ref(delayed_refs, existing, head_ref,
 					 old_ref_mod);
@@ -634,8 +638,8 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 	} else {
 		if (old_ref_mod)
 			*old_ref_mod = 0;
-		if (is_data && count_mod < 0)
-			delayed_refs->pending_csums += num_bytes;
+		if (head_ref->is_data && head_ref->ref_mod < 0)
+			delayed_refs->pending_csums += head_ref->num_bytes;
 		delayed_refs->num_heads++;
 		delayed_refs->num_heads_ready++;
 		atomic_inc(&delayed_refs->num_entries);
@@ -645,90 +649,48 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 		*qrecord_inserted_ret = qrecord_inserted;
 	if (new_ref_mod)
 		*new_ref_mod = head_ref->total_ref_mod;
-	return head_ref;
-}
-
-/*
- * helper to insert a delayed tree ref into the rbtree.
- */
-static noinline void
-add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
-		     struct btrfs_trans_handle *trans,
-		     struct btrfs_delayed_ref_head *head_ref,
-		     struct btrfs_delayed_ref_node *ref, u64 bytenr,
-		     u64 num_bytes, u64 parent, u64 ref_root, int level,
-		     int action)
-{
-	struct btrfs_delayed_tree_ref *full_ref;
-	struct btrfs_delayed_ref_root *delayed_refs;
-	u64 seq = 0;
-	int ret;
-
-	if (action == BTRFS_ADD_DELAYED_EXTENT)
-		action = BTRFS_ADD_DELAYED_REF;
 
-	if (is_fstree(ref_root))
-		seq = atomic64_read(&fs_info->tree_mod_seq);
-	delayed_refs = &trans->transaction->delayed_refs;
-
-	/* first set the basic ref node struct up */
-	refcount_set(&ref->refs, 1);
-	ref->bytenr = bytenr;
-	ref->num_bytes = num_bytes;
-	ref->ref_mod = 1;
-	ref->action = action;
-	ref->is_head = 0;
-	ref->in_tree = 1;
-	ref->seq = seq;
-	RB_CLEAR_NODE(&ref->ref_node);
-	INIT_LIST_HEAD(&ref->add_list);
-
-	full_ref = btrfs_delayed_node_to_tree_ref(ref);
-	full_ref->parent = parent;
-	full_ref->root = ref_root;
-	if (parent)
-		ref->type = BTRFS_SHARED_BLOCK_REF_KEY;
-	else
-		ref->type = BTRFS_TREE_BLOCK_REF_KEY;
-	full_ref->level = level;
-
-	trace_add_delayed_tree_ref(fs_info, ref, full_ref, action);
-
-	ret = insert_delayed_ref(trans, delayed_refs, head_ref, ref);
-
-	/*
-	 * XXX: memory should be freed at the same level allocated.
-	 * But bad practice is anywhere... Follow it now. Need cleanup.
-	 */
-	if (ret > 0)
-		kmem_cache_free(btrfs_delayed_tree_ref_cachep, full_ref);
+	return head_ref;
 }
 
 /*
- * helper to insert a delayed data ref into the rbtree.
+ * init_delayed_ref_common - Initialize the structure which represents a
+ *			     modification to a an extent.
+ *
+ * @fs_info:    Internal to the mounted filesystem mount structure.
+ *
+ * @ref:	The structure which is going to be initialized.
+ *
+ * @bytenr:	The logical address of the extent for which a modification is
+ *		going to be recorded.
+ *
+ * @num_bytes:  Size of the extent whose modification is being recorded.
+ *
+ * @ref_root:	The id of the root where this modification has originated, this
+ *		can be either one of the well-known metadata trees or the
+ *		subvolume id which references this extent.
+ *
+ * @action:	Can be one of BTRFS_ADD_DELAYED_REF/BTRFS_DROP_DELAYED_REF or
+ *		BTRFS_ADD_DELAYED_EXTENT
+ *
+ * @ref_type:	Holds the type of the extent which is being recorded, can be
+ *		one of BTRFS_SHARED_BLOCK_REF_KEY/BTRFS_TREE_BLOCK_REF_KEY
+ *		when recording a metadata extent or BTRFS_SHARED_DATA_REF_KEY/
+ *		BTRFS_EXTENT_DATA_REF_KEY when recording data extent
  */
-static noinline void
-add_delayed_data_ref(struct btrfs_fs_info *fs_info,
-		     struct btrfs_trans_handle *trans,
-		     struct btrfs_delayed_ref_head *head_ref,
-		     struct btrfs_delayed_ref_node *ref, u64 bytenr,
-		     u64 num_bytes, u64 parent, u64 ref_root, u64 owner,
-		     u64 offset, int action)
+static void init_delayed_ref_common(struct btrfs_fs_info *fs_info,
+				    struct btrfs_delayed_ref_node *ref,
+				    u64 bytenr, u64 num_bytes, u64 ref_root,
+				    int action, u8 ref_type)
 {
-	struct btrfs_delayed_data_ref *full_ref;
-	struct btrfs_delayed_ref_root *delayed_refs;
 	u64 seq = 0;
-	int ret;
 
 	if (action == BTRFS_ADD_DELAYED_EXTENT)
 		action = BTRFS_ADD_DELAYED_REF;
 
-	delayed_refs = &trans->transaction->delayed_refs;
-
 	if (is_fstree(ref_root))
 		seq = atomic64_read(&fs_info->tree_mod_seq);
 
-	/* first set the basic ref node struct up */
 	refcount_set(&ref->refs, 1);
 	ref->bytenr = bytenr;
 	ref->num_bytes = num_bytes;
@@ -737,25 +699,9 @@ add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 	ref->is_head = 0;
 	ref->in_tree = 1;
 	ref->seq = seq;
+	ref->type = ref_type;
 	RB_CLEAR_NODE(&ref->ref_node);
 	INIT_LIST_HEAD(&ref->add_list);
-
-	full_ref = btrfs_delayed_node_to_data_ref(ref);
-	full_ref->parent = parent;
-	full_ref->root = ref_root;
-	if (parent)
-		ref->type = BTRFS_SHARED_DATA_REF_KEY;
-	else
-		ref->type = BTRFS_EXTENT_DATA_REF_KEY;
-
-	full_ref->objectid = owner;
-	full_ref->offset = offset;
-
-	trace_add_delayed_data_ref(fs_info, ref, full_ref, action);
-
-	ret = insert_delayed_ref(trans, delayed_refs, head_ref, ref);
-	if (ret > 0)
-		kmem_cache_free(btrfs_delayed_data_ref_cachep, full_ref);
 }
 
 /*
@@ -775,13 +721,25 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 	struct btrfs_delayed_ref_root *delayed_refs;
 	struct btrfs_qgroup_extent_record *record = NULL;
 	int qrecord_inserted;
-	int is_system = (ref_root == BTRFS_CHUNK_TREE_OBJECTID);
+	bool is_system = (ref_root == BTRFS_CHUNK_TREE_OBJECTID);
+	int ret;
+	u8 ref_type;
 
 	BUG_ON(extent_op && extent_op->is_data);
 	ref = kmem_cache_alloc(btrfs_delayed_tree_ref_cachep, GFP_NOFS);
 	if (!ref)
 		return -ENOMEM;
 
+	if (parent)
+		ref_type = BTRFS_SHARED_BLOCK_REF_KEY;
+	else
+		ref_type = BTRFS_TREE_BLOCK_REF_KEY;
+	init_delayed_ref_common(fs_info, &ref->node, bytenr, num_bytes,
+				ref_root, action, ref_type);
+	ref->root = ref_root;
+	ref->parent = parent;
+	ref->level = level;
+
 	head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS);
 	if (!head_ref)
 		goto free_ref;
@@ -793,6 +751,8 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 			goto free_head_ref;
 	}
 
+	init_delayed_ref_head(head_ref, record, bytenr, num_bytes,
+			      ref_root, 0, action, false, is_system);
 	head_ref->extent_op = extent_op;
 
 	delayed_refs = &trans->transaction->delayed_refs;
@@ -802,15 +762,19 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 	 * insert both the head node and the new ref without dropping
 	 * the spin lock
 	 */
-	head_ref = add_delayed_ref_head(fs_info, trans, head_ref, record,
-					bytenr, num_bytes, 0, 0, action, 0,
-					is_system, &qrecord_inserted,
+	head_ref = add_delayed_ref_head(trans, head_ref, record,
+					action, &qrecord_inserted,
 					old_ref_mod, new_ref_mod);
 
-	add_delayed_tree_ref(fs_info, trans, head_ref, &ref->node, bytenr,
-			     num_bytes, parent, ref_root, level, action);
+	ret = insert_delayed_ref(trans, delayed_refs, head_ref, &ref->node);
 	spin_unlock(&delayed_refs->lock);
 
+	trace_add_delayed_tree_ref(fs_info, &ref->node, ref,
+				   action == BTRFS_ADD_DELAYED_EXTENT ?
+				   BTRFS_ADD_DELAYED_REF : action);
+	if (ret > 0)
+		kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref);
+
 	if (qrecord_inserted)
 		btrfs_qgroup_trace_extent_post(fs_info, record);
 
@@ -839,11 +803,25 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 	struct btrfs_delayed_ref_root *delayed_refs;
 	struct btrfs_qgroup_extent_record *record = NULL;
 	int qrecord_inserted;
+	int ret;
+	u8 ref_type;
 
 	ref = kmem_cache_alloc(btrfs_delayed_data_ref_cachep, GFP_NOFS);
 	if (!ref)
 		return -ENOMEM;
 
+	if (parent)
+	        ref_type = BTRFS_SHARED_DATA_REF_KEY;
+	else
+	        ref_type = BTRFS_EXTENT_DATA_REF_KEY;
+	init_delayed_ref_common(fs_info, &ref->node, bytenr, num_bytes,
+				ref_root, action, ref_type);
+	ref->root = ref_root;
+	ref->parent = parent;
+	ref->objectid = owner;
+	ref->offset = offset;
+
+
 	head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS);
 	if (!head_ref) {
 		kmem_cache_free(btrfs_delayed_data_ref_cachep, ref);
@@ -861,6 +839,8 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 		}
 	}
 
+	init_delayed_ref_head(head_ref, record, bytenr, num_bytes, ref_root,
+			      reserved, action, true, false);
 	head_ref->extent_op = NULL;
 
 	delayed_refs = &trans->transaction->delayed_refs;
@@ -870,16 +850,20 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 	 * insert both the head node and the new ref without dropping
 	 * the spin lock
 	 */
-	head_ref = add_delayed_ref_head(fs_info, trans, head_ref, record,
-					bytenr, num_bytes, ref_root, reserved,
-					action, 1, 0, &qrecord_inserted,
+	head_ref = add_delayed_ref_head(trans, head_ref, record,
+					action, &qrecord_inserted,
 					old_ref_mod, new_ref_mod);
 
-	add_delayed_data_ref(fs_info, trans, head_ref, &ref->node, bytenr,
-				   num_bytes, parent, ref_root, owner, offset,
-				   action);
+	ret = insert_delayed_ref(trans, delayed_refs, head_ref, &ref->node);
 	spin_unlock(&delayed_refs->lock);
 
+	trace_add_delayed_data_ref(trans->fs_info, &ref->node, ref,
+				   action == BTRFS_ADD_DELAYED_EXTENT ?
+				   BTRFS_ADD_DELAYED_REF : action);
+	if (ret > 0)
+		kmem_cache_free(btrfs_delayed_data_ref_cachep, ref);
+
+
 	if (qrecord_inserted)
 		return btrfs_qgroup_trace_extent_post(fs_info, record);
 	return 0;
@@ -897,19 +881,16 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
 	if (!head_ref)
 		return -ENOMEM;
 
+	init_delayed_ref_head(head_ref, NULL, bytenr, num_bytes, 0, 0,
+			      BTRFS_UPDATE_DELAYED_HEAD, extent_op->is_data,
+			      false);
 	head_ref->extent_op = extent_op;
 
 	delayed_refs = &trans->transaction->delayed_refs;
 	spin_lock(&delayed_refs->lock);
 
-	/*
-	 * extent_ops just modify the flags of an extent and they don't result
-	 * in ref count changes, hence it's safe to pass false/0 for is_system
-	 * argument
-	 */
-	add_delayed_ref_head(fs_info, trans, head_ref, NULL, bytenr,
-			     num_bytes, 0, 0, BTRFS_UPDATE_DELAYED_HEAD,
-			     extent_op->is_data, 0, NULL, NULL, NULL);
+	add_delayed_ref_head(trans, head_ref, NULL, BTRFS_UPDATE_DELAYED_HEAD,
+			     NULL, NULL, NULL);
 
 	spin_unlock(&delayed_refs->lock);
 	return 0;

fs/btrfs/delayed-ref.h

@@ -251,7 +251,6 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
 				u64 bytenr, u64 num_bytes,
 				struct btrfs_delayed_extent_op *extent_op);
 void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
-			      struct btrfs_fs_info *fs_info,
 			      struct btrfs_delayed_ref_root *delayed_refs,
 			      struct btrfs_delayed_ref_head *head);
 
@@ -269,9 +268,7 @@ static inline void btrfs_delayed_ref_unlock(struct btrfs_delayed_ref_head *head)
 struct btrfs_delayed_ref_head *
 btrfs_select_ref_head(struct btrfs_trans_handle *trans);
 
-int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
-			    struct btrfs_delayed_ref_root *delayed_refs,
-			    u64 seq);
+int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq);
 
 /*
  * helper functions to cast a node into its container

fs/btrfs/dev-replace.c

@@ -33,8 +33,6 @@ static void btrfs_dev_replace_update_device_in_mapping_tree(
 						struct btrfs_device *srcdev,
 						struct btrfs_device *tgtdev);
 static int btrfs_dev_replace_kthread(void *data);
-static int btrfs_dev_replace_continue_on_mount(struct btrfs_fs_info *fs_info);
-
 
 int btrfs_init_dev_replace(struct btrfs_fs_info *fs_info)
 {
@@ -178,6 +176,105 @@ out:
 	return ret;
 }
 
+/*
+ * Initialize a new device for device replace target from a given source dev
+ * and path.
+ *
+ * Return 0 and new device in @device_out, otherwise return < 0
+ */
+static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
+				  const char *device_path,
+				  struct btrfs_device *srcdev,
+				  struct btrfs_device **device_out)
+{
+	struct btrfs_device *device;
+	struct block_device *bdev;
+	struct list_head *devices;
+	struct rcu_string *name;
+	u64 devid = BTRFS_DEV_REPLACE_DEVID;
+	int ret = 0;
+
+	*device_out = NULL;
+	if (fs_info->fs_devices->seeding) {
+		btrfs_err(fs_info, "the filesystem is a seed filesystem!");
+		return -EINVAL;
+	}
+
+	bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL,
+				  fs_info->bdev_holder);
+	if (IS_ERR(bdev)) {
+		btrfs_err(fs_info, "target device %s is invalid!", device_path);
+		return PTR_ERR(bdev);
+	}
+
+	filemap_write_and_wait(bdev->bd_inode->i_mapping);
+
+	devices = &fs_info->fs_devices->devices;
+	list_for_each_entry(device, devices, dev_list) {
+		if (device->bdev == bdev) {
+			btrfs_err(fs_info,
+				  "target device is in the filesystem!");
+			ret = -EEXIST;
+			goto error;
+		}
+	}
+
+
+	if (i_size_read(bdev->bd_inode) <
+	    btrfs_device_get_total_bytes(srcdev)) {
+		btrfs_err(fs_info,
+			  "target device is smaller than source device!");
+		ret = -EINVAL;
+		goto error;
+	}
+
+
+	device = btrfs_alloc_device(NULL, &devid, NULL);
+	if (IS_ERR(device)) {
+		ret = PTR_ERR(device);
+		goto error;
+	}
+
+	name = rcu_string_strdup(device_path, GFP_KERNEL);
+	if (!name) {
+		btrfs_free_device(device);
+		ret = -ENOMEM;
+		goto error;
+	}
+	rcu_assign_pointer(device->name, name);
+
+	mutex_lock(&fs_info->fs_devices->device_list_mutex);
+	set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
+	device->generation = 0;
+	device->io_width = fs_info->sectorsize;
+	device->io_align = fs_info->sectorsize;
+	device->sector_size = fs_info->sectorsize;
+	device->total_bytes = btrfs_device_get_total_bytes(srcdev);
+	device->disk_total_bytes = btrfs_device_get_disk_total_bytes(srcdev);
+	device->bytes_used = btrfs_device_get_bytes_used(srcdev);
+	device->commit_total_bytes = srcdev->commit_total_bytes;
+	device->commit_bytes_used = device->bytes_used;
+	device->fs_info = fs_info;
+	device->bdev = bdev;
+	set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
+	set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
+	device->mode = FMODE_EXCL;
+	device->dev_stats_valid = 1;
+	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
+	device->fs_devices = fs_info->fs_devices;
+	list_add(&device->dev_list, &fs_info->fs_devices->devices);
+	fs_info->fs_devices->num_devices++;
+	fs_info->fs_devices->open_devices++;
+	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+
+	*device_out = device;
+	return 0;
+
+error:
+	blkdev_put(bdev, FMODE_EXCL);
+	return ret;
+}
+
 /*
  * called from commit_transaction. Writes changed device replace state to
  * disk.
@@ -317,18 +414,13 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	struct btrfs_device *tgt_device = NULL;
 	struct btrfs_device *src_device = NULL;
 
-	/* the disk copy procedure reuses the scrub code */
-	mutex_lock(&fs_info->volume_mutex);
 	ret = btrfs_find_device_by_devspec(fs_info, srcdevid,
 					    srcdev_name, &src_device);
-	if (ret) {
-		mutex_unlock(&fs_info->volume_mutex);
+	if (ret)
 		return ret;
-	}
 
 	ret = btrfs_init_dev_replace_tgtdev(fs_info, tgtdev_name,
 					    src_device, &tgt_device);
-	mutex_unlock(&fs_info->volume_mutex);
 	if (ret)
 		return ret;
 
@@ -360,7 +452,6 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	dev_replace->cont_reading_from_srcdev_mode = read_src;
 	WARN_ON(!src_device);
 	dev_replace->srcdev = src_device;
-	WARN_ON(!tgt_device);
 	dev_replace->tgtdev = tgt_device;
 
 	btrfs_info_in_rcu(fs_info,
@@ -503,7 +594,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	 * flush all outstanding I/O and inode extent mappings before the
 	 * copy operation is declared as being finished
 	 */
-	ret = btrfs_start_delalloc_roots(fs_info, 0, -1);
+	ret = btrfs_start_delalloc_roots(fs_info, -1);
 	if (ret) {
 		mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
 		return ret;
@@ -518,7 +609,6 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	ret = btrfs_commit_transaction(trans);
 	WARN_ON(ret);
 
-	mutex_lock(&uuid_mutex);
 	/* keep away write_all_supers() during the finishing procedure */
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
 	mutex_lock(&fs_info->chunk_mutex);
@@ -545,7 +635,6 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 		btrfs_dev_replace_write_unlock(dev_replace);
 		mutex_unlock(&fs_info->chunk_mutex);
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-		mutex_unlock(&uuid_mutex);
 		btrfs_rm_dev_replace_blocked(fs_info);
 		if (tgt_device)
 			btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
@@ -596,7 +685,6 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	 */
 	mutex_unlock(&fs_info->chunk_mutex);
 	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-	mutex_unlock(&uuid_mutex);
 
 	/* replace the sysfs entry */
 	btrfs_sysfs_rm_device_link(fs_info->fs_devices, src_device);
@@ -800,7 +888,17 @@ int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info)
 	}
 	btrfs_dev_replace_write_unlock(dev_replace);
 
-	WARN_ON(test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags));
+	/*
+	 * This could collide with a paused balance, but the exclusive op logic
+	 * should never allow both to start and pause. We don't want to allow
+	 * dev-replace to start anyway.
+	 */
+	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
+		btrfs_info(fs_info,
+		"cannot resume dev-replace, other exclusive operation running");
+		return 0;
+	}
+
 	task = kthread_run(btrfs_dev_replace_kthread, fs_info, "btrfs-devrepl");
 	return PTR_ERR_OR_ZERO(task);
 }
@@ -810,6 +908,7 @@ static int btrfs_dev_replace_kthread(void *data)
 	struct btrfs_fs_info *fs_info = data;
 	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
 	u64 progress;
+	int ret;
 
 	progress = btrfs_dev_replace_progress(fs_info);
 	progress = div_u64(progress, 10);
@@ -820,23 +919,14 @@ static int btrfs_dev_replace_kthread(void *data)
 		btrfs_dev_name(dev_replace->tgtdev),
 		(unsigned int)progress);
 
-	btrfs_dev_replace_continue_on_mount(fs_info);
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
-
-	return 0;
-}
-
-static int btrfs_dev_replace_continue_on_mount(struct btrfs_fs_info *fs_info)
-{
-	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
-	int ret;
-
 	ret = btrfs_scrub_dev(fs_info, dev_replace->srcdev->devid,
 			      dev_replace->committed_cursor_left,
 			      btrfs_device_get_total_bytes(dev_replace->srcdev),
 			      &dev_replace->scrub_progress, 0, 1);
 	ret = btrfs_dev_replace_finishing(fs_info, ret);
 	WARN_ON(ret);
+
+	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
 	return 0;
 }
 
@@ -916,9 +1006,9 @@ void btrfs_dev_replace_clear_lock_blocking(
 	ASSERT(atomic_read(&dev_replace->read_locks) > 0);
 	ASSERT(atomic_read(&dev_replace->blocking_readers) > 0);
 	read_lock(&dev_replace->lock);
-	if (atomic_dec_and_test(&dev_replace->blocking_readers) &&
-	    waitqueue_active(&dev_replace->read_lock_wq))
-		wake_up(&dev_replace->read_lock_wq);
+	/* Barrier implied by atomic_dec_and_test */
+	if (atomic_dec_and_test(&dev_replace->blocking_readers))
+		cond_wake_up_nomb(&dev_replace->read_lock_wq);
 }
 
 void btrfs_bio_counter_inc_noblocked(struct btrfs_fs_info *fs_info)
@@ -929,9 +1019,7 @@ void btrfs_bio_counter_inc_noblocked(struct btrfs_fs_info *fs_info)
 void btrfs_bio_counter_sub(struct btrfs_fs_info *fs_info, s64 amount)
 {
 	percpu_counter_sub(&fs_info->bio_counter, amount);
-
-	if (waitqueue_active(&fs_info->replace_wait))
-		wake_up(&fs_info->replace_wait);
+	cond_wake_up_nomb(&fs_info->replace_wait);
 }
 
 void btrfs_bio_counter_inc_blocked(struct btrfs_fs_info *fs_info)

fs/btrfs/disk-io.c

@@ -55,7 +55,6 @@
 static const 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 int btrfs_check_super_valid(struct btrfs_fs_info *fs_info);
 static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
 static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 				      struct btrfs_fs_info *fs_info);
@@ -416,7 +415,7 @@ static int btrfs_check_super_csum(struct btrfs_fs_info *fs_info,
 
 static int verify_level_key(struct btrfs_fs_info *fs_info,
 			    struct extent_buffer *eb, int level,
-			    struct btrfs_key *first_key)
+			    struct btrfs_key *first_key, u64 parent_transid)
 {
 	int found_level;
 	struct btrfs_key found_key;
@@ -454,10 +453,11 @@ static int verify_level_key(struct btrfs_fs_info *fs_info,
 	if (ret) {
 		WARN_ON(1);
 		btrfs_err(fs_info,
-"tree first key mismatch detected, bytenr=%llu key expected=(%llu, %u, %llu) has=(%llu, %u, %llu)",
-			  eb->start, first_key->objectid, first_key->type,
-			  first_key->offset, found_key.objectid,
-			  found_key.type, found_key.offset);
+"tree first key mismatch detected, bytenr=%llu parent_transid=%llu key expected=(%llu,%u,%llu) has=(%llu,%u,%llu)",
+			  eb->start, parent_transid, first_key->objectid,
+			  first_key->type, first_key->offset,
+			  found_key.objectid, found_key.type,
+			  found_key.offset);
 	}
 #endif
 	return ret;
@@ -493,7 +493,7 @@ static int btree_read_extent_buffer_pages(struct btrfs_fs_info *fs_info,
 						   parent_transid, 0))
 				ret = -EIO;
 			else if (verify_level_key(fs_info, eb, level,
-						  first_key))
+						  first_key, parent_transid))
 				ret = -EUCLEAN;
 			else
 				break;
@@ -1185,7 +1185,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	root->inode_tree = RB_ROOT;
 	INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
 	root->block_rsv = NULL;
-	root->orphan_block_rsv = NULL;
 
 	INIT_LIST_HEAD(&root->dirty_list);
 	INIT_LIST_HEAD(&root->root_list);
@@ -1195,7 +1194,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	INIT_LIST_HEAD(&root->ordered_root);
 	INIT_LIST_HEAD(&root->logged_list[0]);
 	INIT_LIST_HEAD(&root->logged_list[1]);
-	spin_lock_init(&root->orphan_lock);
 	spin_lock_init(&root->inode_lock);
 	spin_lock_init(&root->delalloc_lock);
 	spin_lock_init(&root->ordered_extent_lock);
@@ -1216,7 +1214,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	atomic_set(&root->log_commit[1], 0);
 	atomic_set(&root->log_writers, 0);
 	atomic_set(&root->log_batch, 0);
-	atomic_set(&root->orphan_inodes, 0);
 	refcount_set(&root->refs, 1);
 	atomic_set(&root->will_be_snapshotted, 0);
 	root->log_transid = 0;
@@ -2164,7 +2161,6 @@ static void btrfs_init_balance(struct btrfs_fs_info *fs_info)
 {
 	spin_lock_init(&fs_info->balance_lock);
 	mutex_init(&fs_info->balance_mutex);
-	atomic_set(&fs_info->balance_running, 0);
 	atomic_set(&fs_info->balance_pause_req, 0);
 	atomic_set(&fs_info->balance_cancel_req, 0);
 	fs_info->balance_ctl = NULL;
@@ -2442,6 +2438,211 @@ out:
 	return ret;
 }
 
+/*
+ * Real super block validation
+ * NOTE: super csum type and incompat features will not be checked here.
+ *
+ * @sb:		super block to check
+ * @mirror_num:	the super block number to check its bytenr:
+ * 		0	the primary (1st) sb
+ * 		1, 2	2nd and 3rd backup copy
+ * 	       -1	skip bytenr check
+ */
+static int validate_super(struct btrfs_fs_info *fs_info,
+			    struct btrfs_super_block *sb, int mirror_num)
+{
+	u64 nodesize = btrfs_super_nodesize(sb);
+	u64 sectorsize = btrfs_super_sectorsize(sb);
+	int ret = 0;
+
+	if (btrfs_super_magic(sb) != BTRFS_MAGIC) {
+		btrfs_err(fs_info, "no valid FS found");
+		ret = -EINVAL;
+	}
+	if (btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP) {
+		btrfs_err(fs_info, "unrecognized or unsupported super flag: %llu",
+				btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP);
+		ret = -EINVAL;
+	}
+	if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) {
+		btrfs_err(fs_info, "tree_root level too big: %d >= %d",
+				btrfs_super_root_level(sb), BTRFS_MAX_LEVEL);
+		ret = -EINVAL;
+	}
+	if (btrfs_super_chunk_root_level(sb) >= BTRFS_MAX_LEVEL) {
+		btrfs_err(fs_info, "chunk_root level too big: %d >= %d",
+				btrfs_super_chunk_root_level(sb), BTRFS_MAX_LEVEL);
+		ret = -EINVAL;
+	}
+	if (btrfs_super_log_root_level(sb) >= BTRFS_MAX_LEVEL) {
+		btrfs_err(fs_info, "log_root level too big: %d >= %d",
+				btrfs_super_log_root_level(sb), BTRFS_MAX_LEVEL);
+		ret = -EINVAL;
+	}
+
+	/*
+	 * Check sectorsize and nodesize first, other check will need it.
+	 * Check all possible sectorsize(4K, 8K, 16K, 32K, 64K) here.
+	 */
+	if (!is_power_of_2(sectorsize) || sectorsize < 4096 ||
+	    sectorsize > BTRFS_MAX_METADATA_BLOCKSIZE) {
+		btrfs_err(fs_info, "invalid sectorsize %llu", sectorsize);
+		ret = -EINVAL;
+	}
+	/* Only PAGE SIZE is supported yet */
+	if (sectorsize != PAGE_SIZE) {
+		btrfs_err(fs_info,
+			"sectorsize %llu not supported yet, only support %lu",
+			sectorsize, PAGE_SIZE);
+		ret = -EINVAL;
+	}
+	if (!is_power_of_2(nodesize) || nodesize < sectorsize ||
+	    nodesize > BTRFS_MAX_METADATA_BLOCKSIZE) {
+		btrfs_err(fs_info, "invalid nodesize %llu", nodesize);
+		ret = -EINVAL;
+	}
+	if (nodesize != le32_to_cpu(sb->__unused_leafsize)) {
+		btrfs_err(fs_info, "invalid leafsize %u, should be %llu",
+			  le32_to_cpu(sb->__unused_leafsize), nodesize);
+		ret = -EINVAL;
+	}
+
+	/* Root alignment check */
+	if (!IS_ALIGNED(btrfs_super_root(sb), sectorsize)) {
+		btrfs_warn(fs_info, "tree_root block unaligned: %llu",
+			   btrfs_super_root(sb));
+		ret = -EINVAL;
+	}
+	if (!IS_ALIGNED(btrfs_super_chunk_root(sb), sectorsize)) {
+		btrfs_warn(fs_info, "chunk_root block unaligned: %llu",
+			   btrfs_super_chunk_root(sb));
+		ret = -EINVAL;
+	}
+	if (!IS_ALIGNED(btrfs_super_log_root(sb), sectorsize)) {
+		btrfs_warn(fs_info, "log_root block unaligned: %llu",
+			   btrfs_super_log_root(sb));
+		ret = -EINVAL;
+	}
+
+	if (memcmp(fs_info->fsid, sb->dev_item.fsid, BTRFS_FSID_SIZE) != 0) {
+		btrfs_err(fs_info,
+			   "dev_item UUID does not match fsid: %pU != %pU",
+			   fs_info->fsid, sb->dev_item.fsid);
+		ret = -EINVAL;
+	}
+
+	/*
+	 * Hint to catch really bogus numbers, bitflips or so, more exact checks are
+	 * done later
+	 */
+	if (btrfs_super_bytes_used(sb) < 6 * btrfs_super_nodesize(sb)) {
+		btrfs_err(fs_info, "bytes_used is too small %llu",
+			  btrfs_super_bytes_used(sb));
+		ret = -EINVAL;
+	}
+	if (!is_power_of_2(btrfs_super_stripesize(sb))) {
+		btrfs_err(fs_info, "invalid stripesize %u",
+			  btrfs_super_stripesize(sb));
+		ret = -EINVAL;
+	}
+	if (btrfs_super_num_devices(sb) > (1UL << 31))
+		btrfs_warn(fs_info, "suspicious number of devices: %llu",
+			   btrfs_super_num_devices(sb));
+	if (btrfs_super_num_devices(sb) == 0) {
+		btrfs_err(fs_info, "number of devices is 0");
+		ret = -EINVAL;
+	}
+
+	if (mirror_num >= 0 &&
+	    btrfs_super_bytenr(sb) != btrfs_sb_offset(mirror_num)) {
+		btrfs_err(fs_info, "super offset mismatch %llu != %u",
+			  btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET);
+		ret = -EINVAL;
+	}
+
+	/*
+	 * Obvious sys_chunk_array corruptions, it must hold at least one key
+	 * and one chunk
+	 */
+	if (btrfs_super_sys_array_size(sb) > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) {
+		btrfs_err(fs_info, "system chunk array too big %u > %u",
+			  btrfs_super_sys_array_size(sb),
+			  BTRFS_SYSTEM_CHUNK_ARRAY_SIZE);
+		ret = -EINVAL;
+	}
+	if (btrfs_super_sys_array_size(sb) < sizeof(struct btrfs_disk_key)
+			+ sizeof(struct btrfs_chunk)) {
+		btrfs_err(fs_info, "system chunk array too small %u < %zu",
+			  btrfs_super_sys_array_size(sb),
+			  sizeof(struct btrfs_disk_key)
+			  + sizeof(struct btrfs_chunk));
+		ret = -EINVAL;
+	}
+
+	/*
+	 * The generation is a global counter, we'll trust it more than the others
+	 * but it's still possible that it's the one that's wrong.
+	 */
+	if (btrfs_super_generation(sb) < btrfs_super_chunk_root_generation(sb))
+		btrfs_warn(fs_info,
+			"suspicious: generation < chunk_root_generation: %llu < %llu",
+			btrfs_super_generation(sb),
+			btrfs_super_chunk_root_generation(sb));
+	if (btrfs_super_generation(sb) < btrfs_super_cache_generation(sb)
+	    && btrfs_super_cache_generation(sb) != (u64)-1)
+		btrfs_warn(fs_info,
+			"suspicious: generation < cache_generation: %llu < %llu",
+			btrfs_super_generation(sb),
+			btrfs_super_cache_generation(sb));
+
+	return ret;
+}
+
+/*
+ * Validation of super block at mount time.
+ * Some checks already done early at mount time, like csum type and incompat
+ * flags will be skipped.
+ */
+static int btrfs_validate_mount_super(struct btrfs_fs_info *fs_info)
+{
+	return validate_super(fs_info, fs_info->super_copy, 0);
+}
+
+/*
+ * Validation of super block at write time.
+ * Some checks like bytenr check will be skipped as their values will be
+ * overwritten soon.
+ * Extra checks like csum type and incompat flags will be done here.
+ */
+static int btrfs_validate_write_super(struct btrfs_fs_info *fs_info,
+				      struct btrfs_super_block *sb)
+{
+	int ret;
+
+	ret = validate_super(fs_info, sb, -1);
+	if (ret < 0)
+		goto out;
+	if (btrfs_super_csum_type(sb) != BTRFS_CSUM_TYPE_CRC32) {
+		ret = -EUCLEAN;
+		btrfs_err(fs_info, "invalid csum type, has %u want %u",
+			  btrfs_super_csum_type(sb), BTRFS_CSUM_TYPE_CRC32);
+		goto out;
+	}
+	if (btrfs_super_incompat_flags(sb) & ~BTRFS_FEATURE_INCOMPAT_SUPP) {
+		ret = -EUCLEAN;
+		btrfs_err(fs_info,
+		"invalid incompat flags, has 0x%llx valid mask 0x%llx",
+			  btrfs_super_incompat_flags(sb),
+			  (unsigned long long)BTRFS_FEATURE_INCOMPAT_SUPP);
+		goto out;
+	}
+out:
+	if (ret < 0)
+		btrfs_err(fs_info,
+		"super block corruption detected before writing it to disk");
+	return ret;
+}
+
 int open_ctree(struct super_block *sb,
 	       struct btrfs_fs_devices *fs_devices,
 	       char *options)
@@ -2601,7 +2802,6 @@ int open_ctree(struct super_block *sb,
 	mutex_init(&fs_info->chunk_mutex);
 	mutex_init(&fs_info->transaction_kthread_mutex);
 	mutex_init(&fs_info->cleaner_mutex);
-	mutex_init(&fs_info->volume_mutex);
 	mutex_init(&fs_info->ro_block_group_mutex);
 	init_rwsem(&fs_info->commit_root_sem);
 	init_rwsem(&fs_info->cleanup_work_sem);
@@ -2668,7 +2868,7 @@ int open_ctree(struct super_block *sb,
 
 	memcpy(fs_info->fsid, fs_info->super_copy->fsid, BTRFS_FSID_SIZE);
 
-	ret = btrfs_check_super_valid(fs_info);
+	ret = btrfs_validate_mount_super(fs_info);
 	if (ret) {
 		btrfs_err(fs_info, "superblock contains fatal errors");
 		err = -EINVAL;
@@ -3523,7 +3723,7 @@ int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags)
 	for (raid_type = 0; raid_type < BTRFS_NR_RAID_TYPES; raid_type++) {
 		if (raid_type == BTRFS_RAID_SINGLE)
 			continue;
-		if (!(flags & btrfs_raid_group[raid_type]))
+		if (!(flags & btrfs_raid_array[raid_type].bg_flag))
 			continue;
 		min_tolerated = min(min_tolerated,
 				    btrfs_raid_array[raid_type].
@@ -3603,6 +3803,14 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
 		flags = btrfs_super_flags(sb);
 		btrfs_set_super_flags(sb, flags | BTRFS_HEADER_FLAG_WRITTEN);
 
+		ret = btrfs_validate_write_super(fs_info, sb);
+		if (ret < 0) {
+			mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+			btrfs_handle_fs_error(fs_info, -EUCLEAN,
+				"unexpected superblock corruption detected");
+			return -EUCLEAN;
+		}
+
 		ret = write_dev_supers(dev, sb, max_mirrors);
 		if (ret)
 			total_errors++;
@@ -3674,8 +3882,6 @@ static void free_fs_root(struct btrfs_root *root)
 {
 	iput(root->ino_cache_inode);
 	WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
-	btrfs_free_block_rsv(root->fs_info, root->orphan_block_rsv);
-	root->orphan_block_rsv = NULL;
 	if (root->anon_dev)
 		free_anon_bdev(root->anon_dev);
 	if (root->subv_writers)
@@ -3766,7 +3972,6 @@ int btrfs_commit_super(struct btrfs_fs_info *fs_info)
 
 void close_ctree(struct btrfs_fs_info *fs_info)
 {
-	struct btrfs_root *root = fs_info->tree_root;
 	int ret;
 
 	set_bit(BTRFS_FS_CLOSING_START, &fs_info->flags);
@@ -3862,9 +4067,6 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 	btrfs_free_stripe_hash_table(fs_info);
 	btrfs_free_ref_cache(fs_info);
 
-	__btrfs_free_block_rsv(root->orphan_block_rsv);
-	root->orphan_block_rsv = NULL;
-
 	while (!list_empty(&fs_info->pinned_chunks)) {
 		struct extent_map *em;
 
@@ -3975,155 +4177,6 @@ int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid, int level,
 					      level, first_key);
 }
 
-static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info)
-{
-	struct btrfs_super_block *sb = fs_info->super_copy;
-	u64 nodesize = btrfs_super_nodesize(sb);
-	u64 sectorsize = btrfs_super_sectorsize(sb);
-	int ret = 0;
-
-	if (btrfs_super_magic(sb) != BTRFS_MAGIC) {
-		btrfs_err(fs_info, "no valid FS found");
-		ret = -EINVAL;
-	}
-	if (btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP) {
-		btrfs_err(fs_info, "unrecognized or unsupported super flag: %llu",
-				btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP);
-		ret = -EINVAL;
-	}
-	if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) {
-		btrfs_err(fs_info, "tree_root level too big: %d >= %d",
-				btrfs_super_root_level(sb), BTRFS_MAX_LEVEL);
-		ret = -EINVAL;
-	}
-	if (btrfs_super_chunk_root_level(sb) >= BTRFS_MAX_LEVEL) {
-		btrfs_err(fs_info, "chunk_root level too big: %d >= %d",
-				btrfs_super_chunk_root_level(sb), BTRFS_MAX_LEVEL);
-		ret = -EINVAL;
-	}
-	if (btrfs_super_log_root_level(sb) >= BTRFS_MAX_LEVEL) {
-		btrfs_err(fs_info, "log_root level too big: %d >= %d",
-				btrfs_super_log_root_level(sb), BTRFS_MAX_LEVEL);
-		ret = -EINVAL;
-	}
-
-	/*
-	 * Check sectorsize and nodesize first, other check will need it.
-	 * Check all possible sectorsize(4K, 8K, 16K, 32K, 64K) here.
-	 */
-	if (!is_power_of_2(sectorsize) || sectorsize < 4096 ||
-	    sectorsize > BTRFS_MAX_METADATA_BLOCKSIZE) {
-		btrfs_err(fs_info, "invalid sectorsize %llu", sectorsize);
-		ret = -EINVAL;
-	}
-	/* Only PAGE SIZE is supported yet */
-	if (sectorsize != PAGE_SIZE) {
-		btrfs_err(fs_info,
-			"sectorsize %llu not supported yet, only support %lu",
-			sectorsize, PAGE_SIZE);
-		ret = -EINVAL;
-	}
-	if (!is_power_of_2(nodesize) || nodesize < sectorsize ||
-	    nodesize > BTRFS_MAX_METADATA_BLOCKSIZE) {
-		btrfs_err(fs_info, "invalid nodesize %llu", nodesize);
-		ret = -EINVAL;
-	}
-	if (nodesize != le32_to_cpu(sb->__unused_leafsize)) {
-		btrfs_err(fs_info, "invalid leafsize %u, should be %llu",
-			  le32_to_cpu(sb->__unused_leafsize), nodesize);
-		ret = -EINVAL;
-	}
-
-	/* Root alignment check */
-	if (!IS_ALIGNED(btrfs_super_root(sb), sectorsize)) {
-		btrfs_warn(fs_info, "tree_root block unaligned: %llu",
-			   btrfs_super_root(sb));
-		ret = -EINVAL;
-	}
-	if (!IS_ALIGNED(btrfs_super_chunk_root(sb), sectorsize)) {
-		btrfs_warn(fs_info, "chunk_root block unaligned: %llu",
-			   btrfs_super_chunk_root(sb));
-		ret = -EINVAL;
-	}
-	if (!IS_ALIGNED(btrfs_super_log_root(sb), sectorsize)) {
-		btrfs_warn(fs_info, "log_root block unaligned: %llu",
-			   btrfs_super_log_root(sb));
-		ret = -EINVAL;
-	}
-
-	if (memcmp(fs_info->fsid, sb->dev_item.fsid, BTRFS_FSID_SIZE) != 0) {
-		btrfs_err(fs_info,
-			   "dev_item UUID does not match fsid: %pU != %pU",
-			   fs_info->fsid, sb->dev_item.fsid);
-		ret = -EINVAL;
-	}
-
-	/*
-	 * Hint to catch really bogus numbers, bitflips or so, more exact checks are
-	 * done later
-	 */
-	if (btrfs_super_bytes_used(sb) < 6 * btrfs_super_nodesize(sb)) {
-		btrfs_err(fs_info, "bytes_used is too small %llu",
-			  btrfs_super_bytes_used(sb));
-		ret = -EINVAL;
-	}
-	if (!is_power_of_2(btrfs_super_stripesize(sb))) {
-		btrfs_err(fs_info, "invalid stripesize %u",
-			  btrfs_super_stripesize(sb));
-		ret = -EINVAL;
-	}
-	if (btrfs_super_num_devices(sb) > (1UL << 31))
-		btrfs_warn(fs_info, "suspicious number of devices: %llu",
-			   btrfs_super_num_devices(sb));
-	if (btrfs_super_num_devices(sb) == 0) {
-		btrfs_err(fs_info, "number of devices is 0");
-		ret = -EINVAL;
-	}
-
-	if (btrfs_super_bytenr(sb) != BTRFS_SUPER_INFO_OFFSET) {
-		btrfs_err(fs_info, "super offset mismatch %llu != %u",
-			  btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET);
-		ret = -EINVAL;
-	}
-
-	/*
-	 * Obvious sys_chunk_array corruptions, it must hold at least one key
-	 * and one chunk
-	 */
-	if (btrfs_super_sys_array_size(sb) > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) {
-		btrfs_err(fs_info, "system chunk array too big %u > %u",
-			  btrfs_super_sys_array_size(sb),
-			  BTRFS_SYSTEM_CHUNK_ARRAY_SIZE);
-		ret = -EINVAL;
-	}
-	if (btrfs_super_sys_array_size(sb) < sizeof(struct btrfs_disk_key)
-			+ sizeof(struct btrfs_chunk)) {
-		btrfs_err(fs_info, "system chunk array too small %u < %zu",
-			  btrfs_super_sys_array_size(sb),
-			  sizeof(struct btrfs_disk_key)
-			  + sizeof(struct btrfs_chunk));
-		ret = -EINVAL;
-	}
-
-	/*
-	 * The generation is a global counter, we'll trust it more than the others
-	 * but it's still possible that it's the one that's wrong.
-	 */
-	if (btrfs_super_generation(sb) < btrfs_super_chunk_root_generation(sb))
-		btrfs_warn(fs_info,
-			"suspicious: generation < chunk_root_generation: %llu < %llu",
-			btrfs_super_generation(sb),
-			btrfs_super_chunk_root_generation(sb));
-	if (btrfs_super_generation(sb) < btrfs_super_cache_generation(sb)
-	    && btrfs_super_cache_generation(sb) != (u64)-1)
-		btrfs_warn(fs_info,
-			"suspicious: generation < cache_generation: %llu < %llu",
-			btrfs_super_generation(sb),
-			btrfs_super_cache_generation(sb));
-
-	return ret;
-}
-
 static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info)
 {
 	/* cleanup FS via transaction */

fs/btrfs/extent-tree.c

@@ -66,10 +66,8 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
 				      u64 flags, u64 owner, u64 offset,
 				      struct btrfs_key *ins, int ref_mod);
 static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
-				     struct btrfs_fs_info *fs_info,
-				     u64 parent, u64 root_objectid,
-				     u64 flags, struct btrfs_disk_key *key,
-				     int level, struct btrfs_key *ins);
+				     struct btrfs_delayed_ref_node *node,
+				     struct btrfs_delayed_extent_op *extent_op);
 static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 			  struct btrfs_fs_info *fs_info, u64 flags,
 			  int force);
@@ -256,7 +254,7 @@ static int exclude_super_stripes(struct btrfs_fs_info *fs_info,
 	for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
 		bytenr = btrfs_sb_offset(i);
 		ret = btrfs_rmap_block(fs_info, cache->key.objectid,
-				       bytenr, 0, &logical, &nr, &stripe_len);
+				       bytenr, &logical, &nr, &stripe_len);
 		if (ret)
 			return ret;
 
@@ -343,8 +341,9 @@ static void fragment_free_space(struct btrfs_block_group_cache *block_group)
  * since their free space will be released as soon as the transaction commits.
  */
 u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
-		       struct btrfs_fs_info *info, u64 start, u64 end)
+		       u64 start, u64 end)
 {
+	struct btrfs_fs_info *info = block_group->fs_info;
 	u64 extent_start, extent_end, size, total_added = 0;
 	int ret;
 
@@ -489,8 +488,7 @@ next:
 
 		if (key.type == BTRFS_EXTENT_ITEM_KEY ||
 		    key.type == BTRFS_METADATA_ITEM_KEY) {
-			total_found += add_new_free_space(block_group,
-							  fs_info, last,
+			total_found += add_new_free_space(block_group, last,
 							  key.objectid);
 			if (key.type == BTRFS_METADATA_ITEM_KEY)
 				last = key.objectid +
@@ -508,7 +506,7 @@ next:
 	}
 	ret = 0;
 
-	total_found += add_new_free_space(block_group, fs_info, last,
+	total_found += add_new_free_space(block_group, last,
 					  block_group->key.objectid +
 					  block_group->key.offset);
 	caching_ctl->progress = (u64)-1;
@@ -744,12 +742,12 @@ static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
 }
 
 static void add_pinned_bytes(struct btrfs_fs_info *fs_info, s64 num_bytes,
-			     u64 owner, u64 root_objectid)
+			     bool metadata, u64 root_objectid)
 {
 	struct btrfs_space_info *space_info;
 	u64 flags;
 
-	if (owner < BTRFS_FIRST_FREE_OBJECTID) {
+	if (metadata) {
 		if (root_objectid == BTRFS_CHUNK_TREE_OBJECTID)
 			flags = BTRFS_BLOCK_GROUP_SYSTEM;
 		else
@@ -2200,8 +2198,11 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 						 &old_ref_mod, &new_ref_mod);
 	}
 
-	if (ret == 0 && old_ref_mod < 0 && new_ref_mod >= 0)
-		add_pinned_bytes(fs_info, -num_bytes, owner, root_objectid);
+	if (ret == 0 && old_ref_mod < 0 && new_ref_mod >= 0) {
+		bool metadata = owner < BTRFS_FIRST_FREE_OBJECTID;
+
+		add_pinned_bytes(fs_info, -num_bytes, metadata, root_objectid);
+	}
 
 	return ret;
 }
@@ -2428,10 +2429,8 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
 {
 	int ret = 0;
 	struct btrfs_delayed_tree_ref *ref;
-	struct btrfs_key ins;
 	u64 parent = 0;
 	u64 ref_root = 0;
-	bool skinny_metadata = btrfs_fs_incompat(fs_info, SKINNY_METADATA);
 
 	ref = btrfs_delayed_node_to_tree_ref(node);
 	trace_run_delayed_tree_ref(fs_info, node, ref, node->action);
@@ -2440,15 +2439,6 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
 		parent = ref->parent;
 	ref_root = ref->root;
 
-	ins.objectid = node->bytenr;
-	if (skinny_metadata) {
-		ins.offset = ref->level;
-		ins.type = BTRFS_METADATA_ITEM_KEY;
-	} else {
-		ins.offset = node->num_bytes;
-		ins.type = BTRFS_EXTENT_ITEM_KEY;
-	}
-
 	if (node->ref_mod != 1) {
 		btrfs_err(fs_info,
 	"btree block(%llu) has %d references rather than 1: action %d ref_root %llu parent %llu",
@@ -2458,11 +2448,7 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
 	}
 	if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) {
 		BUG_ON(!extent_op || !extent_op->update_flags);
-		ret = alloc_reserved_tree_block(trans, fs_info,
-						parent, ref_root,
-						extent_op->flags_to_set,
-						&extent_op->key,
-						ref->level, &ins);
+		ret = alloc_reserved_tree_block(trans, node, extent_op);
 	} else if (node->action == BTRFS_ADD_DELAYED_REF) {
 		ret = __btrfs_inc_extent_ref(trans, fs_info, node,
 					     parent, ref_root,
@@ -2594,8 +2580,8 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans,
 	delayed_refs->num_heads--;
 	rb_erase(&head->href_node, &delayed_refs->href_root);
 	RB_CLEAR_NODE(&head->href_node);
-	spin_unlock(&delayed_refs->lock);
 	spin_unlock(&head->lock);
+	spin_unlock(&delayed_refs->lock);
 	atomic_dec(&delayed_refs->num_entries);
 
 	trace_run_delayed_ref_head(fs_info, head, 0);
@@ -2700,17 +2686,12 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 		 * insert_inline_extent_backref()).
 		 */
 		spin_lock(&locked_ref->lock);
-		btrfs_merge_delayed_refs(trans, fs_info, delayed_refs,
-					 locked_ref);
+		btrfs_merge_delayed_refs(trans, delayed_refs, locked_ref);
 
-		/*
-		 * locked_ref is the head node, so we have to go one
-		 * node back for any delayed ref updates
-		 */
 		ref = select_delayed_ref(locked_ref);
 
 		if (ref && ref->seq &&
-		    btrfs_check_delayed_seq(fs_info, delayed_refs, ref->seq)) {
+		    btrfs_check_delayed_seq(fs_info, ref->seq)) {
 			spin_unlock(&locked_ref->lock);
 			unselect_delayed_ref_head(delayed_refs, locked_ref);
 			locked_ref = NULL;
@@ -3291,7 +3272,7 @@ int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset,
 
 	path = btrfs_alloc_path();
 	if (!path)
-		return -ENOENT;
+		return -ENOMEM;
 
 	do {
 		ret = check_committed_ref(root, path, objectid,
@@ -4026,8 +4007,7 @@ static const char *alloc_name(u64 flags)
 	};
 }
 
-static int create_space_info(struct btrfs_fs_info *info, u64 flags,
-			     struct btrfs_space_info **new)
+static int create_space_info(struct btrfs_fs_info *info, u64 flags)
 {
 
 	struct btrfs_space_info *space_info;
@@ -4065,7 +4045,6 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags,
 		return ret;
 	}
 
-	*new = space_info;
 	list_add_rcu(&space_info->list, &info->space_info);
 	if (flags & BTRFS_BLOCK_GROUP_DATA)
 		info->data_sinfo = space_info;
@@ -4122,7 +4101,7 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
  * returns target flags in extended format or 0 if restripe for this
  * chunk_type is not in progress
  *
- * should be called with either volume_mutex or balance_lock held
+ * should be called with balance_lock held
  */
 static u64 get_restripe_target(struct btrfs_fs_info *fs_info, u64 flags)
 {
@@ -4178,7 +4157,7 @@ static u64 btrfs_reduce_alloc_profile(struct btrfs_fs_info *fs_info, u64 flags)
 	/* First, mask out the RAID levels which aren't possible */
 	for (raid_type = 0; raid_type < BTRFS_NR_RAID_TYPES; raid_type++) {
 		if (num_devices >= btrfs_raid_array[raid_type].devs_min)
-			allowed |= btrfs_raid_group[raid_type];
+			allowed |= btrfs_raid_array[raid_type].bg_flag;
 	}
 	allowed &= flags;
 
@@ -4341,7 +4320,7 @@ commit_trans:
 			need_commit--;
 
 			if (need_commit > 0) {
-				btrfs_start_delalloc_roots(fs_info, 0, -1);
+				btrfs_start_delalloc_roots(fs_info, -1);
 				btrfs_wait_ordered_roots(fs_info, U64_MAX, 0,
 							 (u64)-1);
 			}
@@ -4678,12 +4657,14 @@ again:
 	trans->allocating_chunk = false;
 
 	spin_lock(&space_info->lock);
-	if (ret < 0 && ret != -ENOSPC)
-		goto out;
-	if (ret)
-		space_info->full = 1;
-	else
+	if (ret < 0) {
+		if (ret == -ENOSPC)
+			space_info->full = 1;
+		else
+			goto out;
+	} else {
 		ret = 1;
+	}
 
 	space_info->force_alloc = CHUNK_ALLOC_NO_FORCE;
 out:
@@ -4792,7 +4773,7 @@ static void btrfs_writeback_inodes_sb_nr(struct btrfs_fs_info *fs_info,
 		 * the filesystem is readonly(all dirty pages are written to
 		 * the disk).
 		 */
-		btrfs_start_delalloc_roots(fs_info, 0, nr_items);
+		btrfs_start_delalloc_roots(fs_info, nr_items);
 		if (!current->journal_info)
 			btrfs_wait_ordered_roots(fs_info, nr_items, 0, (u64)-1);
 	}
@@ -5949,44 +5930,6 @@ void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans)
 	trans->chunk_bytes_reserved = 0;
 }
 
-/* Can only return 0 or -ENOSPC */
-int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
-				  struct btrfs_inode *inode)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
-	struct btrfs_root *root = inode->root;
-	/*
-	 * We always use trans->block_rsv here as we will have reserved space
-	 * for our orphan when starting the transaction, using get_block_rsv()
-	 * here will sometimes make us choose the wrong block rsv as we could be
-	 * doing a reloc inode for a non refcounted root.
-	 */
-	struct btrfs_block_rsv *src_rsv = trans->block_rsv;
-	struct btrfs_block_rsv *dst_rsv = root->orphan_block_rsv;
-
-	/*
-	 * We need to hold space in order to delete our orphan item once we've
-	 * added it, so this takes the reservation so we can release it later
-	 * when we are truly done with the orphan item.
-	 */
-	u64 num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
-
-	trace_btrfs_space_reservation(fs_info, "orphan", btrfs_ino(inode),
-			num_bytes, 1);
-	return btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
-}
-
-void btrfs_orphan_release_metadata(struct btrfs_inode *inode)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
-	struct btrfs_root *root = inode->root;
-	u64 num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
-
-	trace_btrfs_space_reservation(fs_info, "orphan", btrfs_ino(inode),
-			num_bytes, 0);
-	btrfs_block_rsv_release(fs_info, root->orphan_block_rsv, num_bytes);
-}
-
 /*
  * btrfs_subvolume_reserve_metadata() - reserve space for subvolume operation
  * root: the root of the parent directory
@@ -6004,7 +5947,6 @@ void btrfs_orphan_release_metadata(struct btrfs_inode *inode)
 int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 				     struct btrfs_block_rsv *rsv,
 				     int items,
-				     u64 *qgroup_reserved,
 				     bool use_global_rsv)
 {
 	u64 num_bytes;
@@ -6022,8 +5964,6 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 		num_bytes = 0;
 	}
 
-	*qgroup_reserved = num_bytes;
-
 	num_bytes = btrfs_calc_trans_metadata_size(fs_info, items);
 	rsv->space_info = __find_space_info(fs_info,
 					    BTRFS_BLOCK_GROUP_METADATA);
@@ -6033,8 +5973,8 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 	if (ret == -ENOSPC && use_global_rsv)
 		ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes, 1);
 
-	if (ret && *qgroup_reserved)
-		btrfs_qgroup_free_meta_prealloc(root, *qgroup_reserved);
+	if (ret && num_bytes)
+		btrfs_qgroup_free_meta_prealloc(root, num_bytes);
 
 	return ret;
 }
@@ -6354,6 +6294,7 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 			spin_lock(&info->unused_bgs_lock);
 			if (list_empty(&cache->bg_list)) {
 				btrfs_get_block_group(cache);
+				trace_btrfs_add_unused_block_group(cache);
 				list_add_tail(&cache->bg_list,
 					      &info->unused_bgs);
 			}
@@ -6511,6 +6452,7 @@ int btrfs_exclude_logged_extents(struct btrfs_fs_info *fs_info,
 	struct btrfs_key key;
 	int found_type;
 	int i;
+	int ret = 0;
 
 	if (!btrfs_fs_incompat(fs_info, MIXED_GROUPS))
 		return 0;
@@ -6527,10 +6469,12 @@ int btrfs_exclude_logged_extents(struct btrfs_fs_info *fs_info,
 			continue;
 		key.objectid = btrfs_file_extent_disk_bytenr(eb, item);
 		key.offset = btrfs_file_extent_disk_num_bytes(eb, item);
-		__exclude_logged_extent(fs_info, key.objectid, key.offset);
+		ret = __exclude_logged_extent(fs_info, key.objectid, key.offset);
+		if (ret)
+			break;
 	}
 
-	return 0;
+	return ret;
 }
 
 static void
@@ -7122,7 +7066,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 			}
 		}
 
-		ret = add_to_free_space_tree(trans, info, bytenr, num_bytes);
+		ret = add_to_free_space_tree(trans, bytenr, num_bytes);
 		if (ret) {
 			btrfs_abort_transaction(trans, ret);
 			goto out;
@@ -7266,7 +7210,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 	}
 out:
 	if (pin)
-		add_pinned_bytes(fs_info, buf->len, btrfs_header_level(buf),
+		add_pinned_bytes(fs_info, buf->len, true,
 				 root->root_key.objectid);
 
 	if (last_ref) {
@@ -7320,8 +7264,11 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans,
 						 &old_ref_mod, &new_ref_mod);
 	}
 
-	if (ret == 0 && old_ref_mod >= 0 && new_ref_mod < 0)
-		add_pinned_bytes(fs_info, num_bytes, owner, root_objectid);
+	if (ret == 0 && old_ref_mod >= 0 && new_ref_mod < 0) {
+		bool metadata = owner < BTRFS_FIRST_FREE_OBJECTID;
+
+		add_pinned_bytes(fs_info, num_bytes, metadata, root_objectid);
+	}
 
 	return ret;
 }
@@ -7373,24 +7320,6 @@ wait_block_group_cache_done(struct btrfs_block_group_cache *cache)
 	return ret;
 }
 
-static const char *btrfs_raid_type_names[BTRFS_NR_RAID_TYPES] = {
-	[BTRFS_RAID_RAID10]	= "raid10",
-	[BTRFS_RAID_RAID1]	= "raid1",
-	[BTRFS_RAID_DUP]	= "dup",
-	[BTRFS_RAID_RAID0]	= "raid0",
-	[BTRFS_RAID_SINGLE]	= "single",
-	[BTRFS_RAID_RAID5]	= "raid5",
-	[BTRFS_RAID_RAID6]	= "raid6",
-};
-
-static const char *get_raid_name(enum btrfs_raid_types type)
-{
-	if (type >= BTRFS_NR_RAID_TYPES)
-		return NULL;
-
-	return btrfs_raid_type_names[type];
-}
-
 enum btrfs_loop_type {
 	LOOP_CACHING_NOWAIT = 0,
 	LOOP_CACHING_WAIT = 1,
@@ -7662,7 +7591,7 @@ have_block_group:
 			if (offset) {
 				/* we have a block, we're done */
 				spin_unlock(&last_ptr->refill_lock);
-				trace_btrfs_reserve_extent_cluster(fs_info,
+				trace_btrfs_reserve_extent_cluster(
 						used_block_group,
 						search_start, num_bytes);
 				if (used_block_group != block_group) {
@@ -7735,7 +7664,7 @@ refill_cluster:
 				if (offset) {
 					/* we found one, proceed */
 					spin_unlock(&last_ptr->refill_lock);
-					trace_btrfs_reserve_extent_cluster(fs_info,
+					trace_btrfs_reserve_extent_cluster(
 						block_group, search_start,
 						num_bytes);
 					goto checks;
@@ -7835,8 +7764,7 @@ checks:
 		ins->objectid = search_start;
 		ins->offset = num_bytes;
 
-		trace_btrfs_reserve_extent(fs_info, block_group,
-					   search_start, num_bytes);
+		trace_btrfs_reserve_extent(block_group, search_start, num_bytes);
 		btrfs_release_block_group(block_group, delalloc);
 		break;
 loop:
@@ -8184,8 +8112,7 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(path->nodes[0]);
 	btrfs_free_path(path);
 
-	ret = remove_from_free_space_tree(trans, fs_info, ins->objectid,
-					  ins->offset);
+	ret = remove_from_free_space_tree(trans, ins->objectid, ins->offset);
 	if (ret)
 		return ret;
 
@@ -8200,37 +8127,52 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
 }
 
 static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
-				     struct btrfs_fs_info *fs_info,
-				     u64 parent, u64 root_objectid,
-				     u64 flags, struct btrfs_disk_key *key,
-				     int level, struct btrfs_key *ins)
+				     struct btrfs_delayed_ref_node *node,
+				     struct btrfs_delayed_extent_op *extent_op)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret;
 	struct btrfs_extent_item *extent_item;
+	struct btrfs_key extent_key;
 	struct btrfs_tree_block_info *block_info;
 	struct btrfs_extent_inline_ref *iref;
 	struct btrfs_path *path;
 	struct extent_buffer *leaf;
+	struct btrfs_delayed_tree_ref *ref;
 	u32 size = sizeof(*extent_item) + sizeof(*iref);
-	u64 num_bytes = ins->offset;
+	u64 num_bytes;
+	u64 flags = extent_op->flags_to_set;
 	bool skinny_metadata = btrfs_fs_incompat(fs_info, SKINNY_METADATA);
 
-	if (!skinny_metadata)
+	ref = btrfs_delayed_node_to_tree_ref(node);
+
+	extent_key.objectid = node->bytenr;
+	if (skinny_metadata) {
+		extent_key.offset = ref->level;
+		extent_key.type = BTRFS_METADATA_ITEM_KEY;
+		num_bytes = fs_info->nodesize;
+	} else {
+		extent_key.offset = node->num_bytes;
+		extent_key.type = BTRFS_EXTENT_ITEM_KEY;
 		size += sizeof(*block_info);
+		num_bytes = node->num_bytes;
+	}
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		btrfs_free_and_pin_reserved_extent(fs_info, ins->objectid,
+		btrfs_free_and_pin_reserved_extent(fs_info,
+						   extent_key.objectid,
 						   fs_info->nodesize);
 		return -ENOMEM;
 	}
 
 	path->leave_spinning = 1;
 	ret = btrfs_insert_empty_item(trans, fs_info->extent_root, path,
-				      ins, size);
+				      &extent_key, size);
 	if (ret) {
 		btrfs_free_path(path);
-		btrfs_free_and_pin_reserved_extent(fs_info, ins->objectid,
+		btrfs_free_and_pin_reserved_extent(fs_info,
+						   extent_key.objectid,
 						   fs_info->nodesize);
 		return ret;
 	}
@@ -8245,42 +8187,41 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
 
 	if (skinny_metadata) {
 		iref = (struct btrfs_extent_inline_ref *)(extent_item + 1);
-		num_bytes = fs_info->nodesize;
 	} else {
 		block_info = (struct btrfs_tree_block_info *)(extent_item + 1);
-		btrfs_set_tree_block_key(leaf, block_info, key);
-		btrfs_set_tree_block_level(leaf, block_info, level);
+		btrfs_set_tree_block_key(leaf, block_info, &extent_op->key);
+		btrfs_set_tree_block_level(leaf, block_info, ref->level);
 		iref = (struct btrfs_extent_inline_ref *)(block_info + 1);
 	}
 
-	if (parent > 0) {
+	if (node->type == BTRFS_SHARED_BLOCK_REF_KEY) {
 		BUG_ON(!(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF));
 		btrfs_set_extent_inline_ref_type(leaf, iref,
 						 BTRFS_SHARED_BLOCK_REF_KEY);
-		btrfs_set_extent_inline_ref_offset(leaf, iref, parent);
+		btrfs_set_extent_inline_ref_offset(leaf, iref, ref->parent);
 	} else {
 		btrfs_set_extent_inline_ref_type(leaf, iref,
 						 BTRFS_TREE_BLOCK_REF_KEY);
-		btrfs_set_extent_inline_ref_offset(leaf, iref, root_objectid);
+		btrfs_set_extent_inline_ref_offset(leaf, iref, ref->root);
 	}
 
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_free_path(path);
 
-	ret = remove_from_free_space_tree(trans, fs_info, ins->objectid,
+	ret = remove_from_free_space_tree(trans, extent_key.objectid,
 					  num_bytes);
 	if (ret)
 		return ret;
 
-	ret = update_block_group(trans, fs_info, ins->objectid,
+	ret = update_block_group(trans, fs_info, extent_key.objectid,
 				 fs_info->nodesize, 1);
 	if (ret) { /* -ENOENT, logic error */
 		btrfs_err(fs_info, "update block group failed for %llu %llu",
-			ins->objectid, ins->offset);
+			extent_key.objectid, extent_key.offset);
 		BUG();
 	}
 
-	trace_btrfs_reserved_extent_alloc(fs_info, ins->objectid,
+	trace_btrfs_reserved_extent_alloc(fs_info, extent_key.objectid,
 					  fs_info->nodesize);
 	return ret;
 }
@@ -10173,8 +10114,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 		} else if (btrfs_block_group_used(&cache->item) == 0) {
 			cache->last_byte_to_unpin = (u64)-1;
 			cache->cached = BTRFS_CACHE_FINISHED;
-			add_new_free_space(cache, info,
-					   found_key.objectid,
+			add_new_free_space(cache, found_key.objectid,
 					   found_key.objectid +
 					   found_key.offset);
 			free_excluded_extents(info, cache);
@@ -10204,6 +10144,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 			/* Should always be true but just in case. */
 			if (list_empty(&cache->bg_list)) {
 				btrfs_get_block_group(cache);
+				trace_btrfs_add_unused_block_group(cache);
 				list_add_tail(&cache->bg_list,
 					      &info->unused_bgs);
 			}
@@ -10269,7 +10210,7 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 					       key.offset);
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
-		add_block_group_free_space(trans, fs_info, block_group);
+		add_block_group_free_space(trans, block_group);
 		/* already aborted the transaction if it failed. */
 next:
 		list_del_init(&block_group->bg_list);
@@ -10310,7 +10251,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 		return ret;
 	}
 
-	add_new_free_space(cache, fs_info, chunk_offset, chunk_offset + size);
+	add_new_free_space(cache, chunk_offset, chunk_offset + size);
 
 	free_excluded_extents(fs_info, cache);
 
@@ -10391,6 +10332,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	BUG_ON(!block_group);
 	BUG_ON(!block_group->ro);
 
+	trace_btrfs_remove_block_group(block_group);
 	/*
 	 * Free the reserved super bytes from this block group before
 	 * remove it.
@@ -10648,7 +10590,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 
 	mutex_unlock(&fs_info->chunk_mutex);
 
-	ret = remove_block_group_free_space(trans, fs_info, block_group);
+	ret = remove_block_group_free_space(trans, block_group);
 	if (ret)
 		goto out;
 
@@ -10755,6 +10697,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 			 * the ro check in case balance is currently acting on
 			 * this block group.
 			 */
+			trace_btrfs_skip_unused_block_group(block_group);
 			spin_unlock(&block_group->lock);
 			up_write(&space_info->groups_sem);
 			goto next;
@@ -10877,7 +10820,6 @@ next:
 
 int btrfs_init_space_info(struct btrfs_fs_info *fs_info)
 {
-	struct btrfs_space_info *space_info;
 	struct btrfs_super_block *disk_super;
 	u64 features;
 	u64 flags;
@@ -10893,21 +10835,21 @@ int btrfs_init_space_info(struct btrfs_fs_info *fs_info)
 		mixed = 1;
 
 	flags = BTRFS_BLOCK_GROUP_SYSTEM;
-	ret = create_space_info(fs_info, flags, &space_info);
+	ret = create_space_info(fs_info, flags);
 	if (ret)
 		goto out;
 
 	if (mixed) {
 		flags = BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA;
-		ret = create_space_info(fs_info, flags, &space_info);
+		ret = create_space_info(fs_info, flags);
 	} else {
 		flags = BTRFS_BLOCK_GROUP_METADATA;
-		ret = create_space_info(fs_info, flags, &space_info);
+		ret = create_space_info(fs_info, flags);
 		if (ret)
 			goto out;
 
 		flags = BTRFS_BLOCK_GROUP_DATA;
-		ret = create_space_info(fs_info, flags, &space_info);
+		ret = create_space_info(fs_info, flags);
 	}
 out:
 	return ret;
@@ -11092,12 +11034,7 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
 void btrfs_end_write_no_snapshotting(struct btrfs_root *root)
 {
 	percpu_counter_dec(&root->subv_writers->counter);
-	/*
-	 * Make sure counter is updated before we wake up waiters.
-	 */
-	smp_mb();
-	if (waitqueue_active(&root->subv_writers->wait))
-		wake_up(&root->subv_writers->wait);
+	cond_wake_up(&root->subv_writers->wait);
 }
 
 int btrfs_start_write_no_snapshotting(struct btrfs_root *root)

fs/btrfs/extent_io.c

@@ -4106,14 +4106,13 @@ int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 	return ret;
 }
 
-int extent_writepages(struct extent_io_tree *tree,
-		      struct address_space *mapping,
+int extent_writepages(struct address_space *mapping,
 		      struct writeback_control *wbc)
 {
 	int ret = 0;
 	struct extent_page_data epd = {
 		.bio = NULL,
-		.tree = tree,
+		.tree = &BTRFS_I(mapping->host)->io_tree,
 		.extent_locked = 0,
 		.sync_io = wbc->sync_mode == WB_SYNC_ALL,
 	};
@@ -4123,9 +4122,8 @@ int extent_writepages(struct extent_io_tree *tree,
 	return ret;
 }
 
-int extent_readpages(struct extent_io_tree *tree,
-		     struct address_space *mapping,
-		     struct list_head *pages, unsigned nr_pages)
+int extent_readpages(struct address_space *mapping, struct list_head *pages,
+		     unsigned nr_pages)
 {
 	struct bio *bio = NULL;
 	unsigned page_idx;
@@ -4133,6 +4131,7 @@ int extent_readpages(struct extent_io_tree *tree,
 	struct page *pagepool[16];
 	struct page *page;
 	struct extent_map *em_cached = NULL;
+	struct extent_io_tree *tree = &BTRFS_I(mapping->host)->io_tree;
 	int nr = 0;
 	u64 prev_em_start = (u64)-1;
 
@@ -4199,8 +4198,7 @@ int extent_invalidatepage(struct extent_io_tree *tree,
  * are locked or under IO and drops the related state bits if it is safe
  * to drop the page.
  */
-static int try_release_extent_state(struct extent_map_tree *map,
-				    struct extent_io_tree *tree,
+static int try_release_extent_state(struct extent_io_tree *tree,
 				    struct page *page, gfp_t mask)
 {
 	u64 start = page_offset(page);
@@ -4235,13 +4233,13 @@ static int try_release_extent_state(struct extent_map_tree *map,
  * in the range corresponding to the page, both state records and extent
  * map records are removed
  */
-int try_release_extent_mapping(struct extent_map_tree *map,
-			       struct extent_io_tree *tree, struct page *page,
-			       gfp_t mask)
+int try_release_extent_mapping(struct page *page, gfp_t mask)
 {
 	struct extent_map *em;
 	u64 start = page_offset(page);
 	u64 end = start + PAGE_SIZE - 1;
+	struct extent_io_tree *tree = &BTRFS_I(page->mapping->host)->io_tree;
+	struct extent_map_tree *map = &BTRFS_I(page->mapping->host)->extent_tree;
 
 	if (gfpflags_allow_blocking(mask) &&
 	    page->mapping->host->i_size > SZ_16M) {
@@ -4275,7 +4273,7 @@ int try_release_extent_mapping(struct extent_map_tree *map,
 			free_extent_map(em);
 		}
 	}
-	return try_release_extent_state(map, tree, page, mask);
+	return try_release_extent_state(tree, page, mask);
 }
 
 /*
@@ -5617,46 +5615,6 @@ void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
 	}
 }
 
-void le_bitmap_set(u8 *map, unsigned int start, int len)
-{
-	u8 *p = map + BIT_BYTE(start);
-	const unsigned int size = start + len;
-	int bits_to_set = BITS_PER_BYTE - (start % BITS_PER_BYTE);
-	u8 mask_to_set = BITMAP_FIRST_BYTE_MASK(start);
-
-	while (len - bits_to_set >= 0) {
-		*p |= mask_to_set;
-		len -= bits_to_set;
-		bits_to_set = BITS_PER_BYTE;
-		mask_to_set = ~0;
-		p++;
-	}
-	if (len) {
-		mask_to_set &= BITMAP_LAST_BYTE_MASK(size);
-		*p |= mask_to_set;
-	}
-}
-
-void le_bitmap_clear(u8 *map, unsigned int start, int len)
-{
-	u8 *p = map + BIT_BYTE(start);
-	const unsigned int size = start + len;
-	int bits_to_clear = BITS_PER_BYTE - (start % BITS_PER_BYTE);
-	u8 mask_to_clear = BITMAP_FIRST_BYTE_MASK(start);
-
-	while (len - bits_to_clear >= 0) {
-		*p &= ~mask_to_clear;
-		len -= bits_to_clear;
-		bits_to_clear = BITS_PER_BYTE;
-		mask_to_clear = ~0;
-		p++;
-	}
-	if (len) {
-		mask_to_clear &= BITMAP_LAST_BYTE_MASK(size);
-		*p &= ~mask_to_clear;
-	}
-}
-
 /*
  * eb_bitmap_offset() - calculate the page and offset of the byte containing the
  * given bit number

fs/btrfs/extent_io.h

@@ -79,14 +79,6 @@
 #define BITMAP_LAST_BYTE_MASK(nbits) \
 	(BYTE_MASK >> (-(nbits) & (BITS_PER_BYTE - 1)))
 
-static inline int le_test_bit(int nr, const u8 *addr)
-{
-	return 1U & (addr[BIT_BYTE(nr)] >> (nr & (BITS_PER_BYTE-1)));
-}
-
-void le_bitmap_set(u8 *map, unsigned int start, int len);
-void le_bitmap_clear(u8 *map, unsigned int start, int len);
-
 struct extent_state;
 struct btrfs_root;
 struct btrfs_inode;
@@ -278,9 +270,7 @@ typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode,
 					  int create);
 
 void extent_io_tree_init(struct extent_io_tree *tree, void *private_data);
-int try_release_extent_mapping(struct extent_map_tree *map,
-			       struct extent_io_tree *tree, struct page *page,
-			       gfp_t mask);
+int try_release_extent_mapping(struct page *page, gfp_t mask);
 int try_release_extent_buffer(struct page *page);
 int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
 		     struct extent_state **cached);
@@ -421,14 +411,12 @@ int extent_invalidatepage(struct extent_io_tree *tree,
 int extent_write_full_page(struct page *page, struct writeback_control *wbc);
 int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 			      int mode);
-int extent_writepages(struct extent_io_tree *tree,
-		      struct address_space *mapping,
+int extent_writepages(struct address_space *mapping,
 		      struct writeback_control *wbc);
 int btree_write_cache_pages(struct address_space *mapping,
 			    struct writeback_control *wbc);
-int extent_readpages(struct extent_io_tree *tree,
-		     struct address_space *mapping,
-		     struct list_head *pages, unsigned nr_pages);
+int extent_readpages(struct address_space *mapping, struct list_head *pages,
+		     unsigned nr_pages);
 int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		__u64 start, __u64 len);
 void set_page_extent_mapped(struct page *page);

fs/btrfs/extent_map.c

@@ -518,6 +518,7 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree,
 
 /**
  * btrfs_add_extent_mapping - add extent mapping into em_tree
+ * @fs_info - used for tracepoint
  * @em_tree - the extent tree into which we want to insert the extent mapping
  * @em_in   - extent we are inserting
  * @start   - start of the logical range btrfs_get_extent() is requesting
@@ -535,7 +536,8 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree,
  * Return 0 on success, otherwise -EEXIST.
  *
  */
-int btrfs_add_extent_mapping(struct extent_map_tree *em_tree,
+int btrfs_add_extent_mapping(struct btrfs_fs_info *fs_info,
+			     struct extent_map_tree *em_tree,
 			     struct extent_map **em_in, u64 start, u64 len)
 {
 	int ret;
@@ -553,7 +555,7 @@ int btrfs_add_extent_mapping(struct extent_map_tree *em_tree,
 
 		existing = search_extent_mapping(em_tree, start, len);
 
-		trace_btrfs_handle_em_exist(existing, em, start, len);
+		trace_btrfs_handle_em_exist(fs_info, existing, em, start, len);
 
 		/*
 		 * existing will always be non-NULL, since there must be

fs/btrfs/extent_map.h

@@ -92,7 +92,8 @@ int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len, u64 gen
 void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em);
 struct extent_map *search_extent_mapping(struct extent_map_tree *tree,
 					 u64 start, u64 len);
-int btrfs_add_extent_mapping(struct extent_map_tree *em_tree,
+int btrfs_add_extent_mapping(struct btrfs_fs_info *fs_info,
+			     struct extent_map_tree *em_tree,
 			     struct extent_map **em_in, u64 start, u64 len);
 
 #endif

fs/btrfs/free-space-cache.c

@@ -253,10 +253,8 @@ int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans,
 	truncate_pagecache(inode, 0);
 
 	/*
-	 * We don't need an orphan item because truncating the free space cache
-	 * will never be split across transactions.
-	 * We don't need to check for -EAGAIN because we're a free space
-	 * cache inode
+	 * We skip the throttling logic for free space cache inodes, so we don't
+	 * need to check for -EAGAIN.
 	 */
 	ret = btrfs_truncate_inode_items(trans, root, inode,
 					 0, BTRFS_EXTENT_DATA_KEY);

fs/btrfs/free-space-tree.c

@@ -12,7 +12,6 @@
 #include "transaction.h"
 
 static int __add_block_group_free_space(struct btrfs_trans_handle *trans,
-					struct btrfs_fs_info *fs_info,
 					struct btrfs_block_group_cache *block_group,
 					struct btrfs_path *path);
 
@@ -45,11 +44,10 @@ void set_free_space_tree_thresholds(struct btrfs_block_group_cache *cache)
 }
 
 static int add_new_free_space_info(struct btrfs_trans_handle *trans,
-				   struct btrfs_fs_info *fs_info,
 				   struct btrfs_block_group_cache *block_group,
 				   struct btrfs_path *path)
 {
-	struct btrfs_root *root = fs_info->free_space_root;
+	struct btrfs_root *root = trans->fs_info->free_space_root;
 	struct btrfs_free_space_info *info;
 	struct btrfs_key key;
 	struct extent_buffer *leaf;
@@ -138,10 +136,11 @@ static inline u32 free_space_bitmap_size(u64 size, u32 sectorsize)
 	return DIV_ROUND_UP((u32)div_u64(size, sectorsize), BITS_PER_BYTE);
 }
 
-static u8 *alloc_bitmap(u32 bitmap_size)
+static unsigned long *alloc_bitmap(u32 bitmap_size)
 {
-	u8 *ret;
+	unsigned long *ret;
 	unsigned int nofs_flag;
+	u32 bitmap_rounded_size = round_up(bitmap_size, sizeof(unsigned long));
 
 	/*
 	 * GFP_NOFS doesn't work with kvmalloc(), but we really can't recurse
@@ -152,21 +151,42 @@ static u8 *alloc_bitmap(u32 bitmap_size)
 	 * know that recursion is unsafe.
 	 */
 	nofs_flag = memalloc_nofs_save();
-	ret = kvzalloc(bitmap_size, GFP_KERNEL);
+	ret = kvzalloc(bitmap_rounded_size, GFP_KERNEL);
 	memalloc_nofs_restore(nofs_flag);
 	return ret;
 }
 
+static void le_bitmap_set(unsigned long *map, unsigned int start, int len)
+{
+	u8 *p = ((u8 *)map) + BIT_BYTE(start);
+	const unsigned int size = start + len;
+	int bits_to_set = BITS_PER_BYTE - (start % BITS_PER_BYTE);
+	u8 mask_to_set = BITMAP_FIRST_BYTE_MASK(start);
+
+	while (len - bits_to_set >= 0) {
+		*p |= mask_to_set;
+		len -= bits_to_set;
+		bits_to_set = BITS_PER_BYTE;
+		mask_to_set = ~0;
+		p++;
+	}
+	if (len) {
+		mask_to_set &= BITMAP_LAST_BYTE_MASK(size);
+		*p |= mask_to_set;
+	}
+}
+
 int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *block_group,
 				  struct btrfs_path *path)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root = fs_info->free_space_root;
 	struct btrfs_free_space_info *info;
 	struct btrfs_key key, found_key;
 	struct extent_buffer *leaf;
-	u8 *bitmap, *bitmap_cursor;
+	unsigned long *bitmap;
+	char *bitmap_cursor;
 	u64 start, end;
 	u64 bitmap_range, i;
 	u32 bitmap_size, flags, expected_extent_count;
@@ -255,7 +275,7 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	bitmap_cursor = bitmap;
+	bitmap_cursor = (char *)bitmap;
 	bitmap_range = fs_info->sectorsize * BTRFS_FREE_SPACE_BITMAP_BITS;
 	i = start;
 	while (i < end) {
@@ -296,21 +316,18 @@ out:
 }
 
 int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *block_group,
 				  struct btrfs_path *path)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root = fs_info->free_space_root;
 	struct btrfs_free_space_info *info;
 	struct btrfs_key key, found_key;
 	struct extent_buffer *leaf;
-	u8 *bitmap;
+	unsigned long *bitmap;
 	u64 start, end;
-	/* Initialize to silence GCC. */
-	u64 extent_start = 0;
-	u64 offset;
 	u32 bitmap_size, flags, expected_extent_count;
-	int prev_bit = 0, bit, bitnr;
+	unsigned long nrbits, start_bit, end_bit;
 	u32 extent_count = 0;
 	int done = 0, nr;
 	int ret;
@@ -348,7 +365,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 				break;
 			} else if (found_key.type == BTRFS_FREE_SPACE_BITMAP_KEY) {
 				unsigned long ptr;
-				u8 *bitmap_cursor;
+				char *bitmap_cursor;
 				u32 bitmap_pos, data_size;
 
 				ASSERT(found_key.objectid >= start);
@@ -358,7 +375,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 				bitmap_pos = div_u64(found_key.objectid - start,
 						     fs_info->sectorsize *
 						     BITS_PER_BYTE);
-				bitmap_cursor = bitmap + bitmap_pos;
+				bitmap_cursor = ((char *)bitmap) + bitmap_pos;
 				data_size = free_space_bitmap_size(found_key.offset,
 								   fs_info->sectorsize);
 
@@ -392,32 +409,16 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_release_path(path);
 
-	offset = start;
-	bitnr = 0;
-	while (offset < end) {
-		bit = !!le_test_bit(bitnr, bitmap);
-		if (prev_bit == 0 && bit == 1) {
-			extent_start = offset;
-		} else if (prev_bit == 1 && bit == 0) {
-			key.objectid = extent_start;
-			key.type = BTRFS_FREE_SPACE_EXTENT_KEY;
-			key.offset = offset - extent_start;
-
-			ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
-			if (ret)
-				goto out;
-			btrfs_release_path(path);
+	nrbits = div_u64(block_group->key.offset, block_group->fs_info->sectorsize);
+	start_bit = find_next_bit_le(bitmap, nrbits, 0);
 
-			extent_count++;
-		}
-		prev_bit = bit;
-		offset += fs_info->sectorsize;
-		bitnr++;
-	}
-	if (prev_bit == 1) {
-		key.objectid = extent_start;
+	while (start_bit < nrbits) {
+		end_bit = find_next_zero_bit_le(bitmap, nrbits, start_bit);
+		ASSERT(start_bit < end_bit);
+
+		key.objectid = start + start_bit * block_group->fs_info->sectorsize;
 		key.type = BTRFS_FREE_SPACE_EXTENT_KEY;
-		key.offset = end - extent_start;
+		key.offset = (end_bit - start_bit) * block_group->fs_info->sectorsize;
 
 		ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
 		if (ret)
@@ -425,6 +426,8 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 		btrfs_release_path(path);
 
 		extent_count++;
+
+		start_bit = find_next_bit_le(bitmap, nrbits, end_bit);
 	}
 
 	if (extent_count != expected_extent_count) {
@@ -446,7 +449,6 @@ out:
 }
 
 static int update_free_space_extent_count(struct btrfs_trans_handle *trans,
-					  struct btrfs_fs_info *fs_info,
 					  struct btrfs_block_group_cache *block_group,
 					  struct btrfs_path *path,
 					  int new_extents)
@@ -459,7 +461,8 @@ static int update_free_space_extent_count(struct btrfs_trans_handle *trans,
 	if (new_extents == 0)
 		return 0;
 
-	info = search_free_space_info(trans, fs_info, block_group, path, 1);
+	info = search_free_space_info(trans, trans->fs_info, block_group, path,
+				      1);
 	if (IS_ERR(info)) {
 		ret = PTR_ERR(info);
 		goto out;
@@ -474,12 +477,10 @@ static int update_free_space_extent_count(struct btrfs_trans_handle *trans,
 
 	if (!(flags & BTRFS_FREE_SPACE_USING_BITMAPS) &&
 	    extent_count > block_group->bitmap_high_thresh) {
-		ret = convert_free_space_to_bitmaps(trans, fs_info, block_group,
-						    path);
+		ret = convert_free_space_to_bitmaps(trans, block_group, path);
 	} else if ((flags & BTRFS_FREE_SPACE_USING_BITMAPS) &&
 		   extent_count < block_group->bitmap_low_thresh) {
-		ret = convert_free_space_to_extents(trans, fs_info, block_group,
-						    path);
+		ret = convert_free_space_to_extents(trans, block_group, path);
 	}
 
 out:
@@ -576,12 +577,11 @@ static int free_space_next_bitmap(struct btrfs_trans_handle *trans,
  * the bitmap.
  */
 static int modify_free_space_bitmap(struct btrfs_trans_handle *trans,
-				    struct btrfs_fs_info *fs_info,
 				    struct btrfs_block_group_cache *block_group,
 				    struct btrfs_path *path,
 				    u64 start, u64 size, int remove)
 {
-	struct btrfs_root *root = fs_info->free_space_root;
+	struct btrfs_root *root = block_group->fs_info->free_space_root;
 	struct btrfs_key key;
 	u64 end = start + size;
 	u64 cur_start, cur_size;
@@ -682,7 +682,7 @@ static int modify_free_space_bitmap(struct btrfs_trans_handle *trans,
 	}
 
 	btrfs_release_path(path);
-	ret = update_free_space_extent_count(trans, fs_info, block_group, path,
+	ret = update_free_space_extent_count(trans, block_group, path,
 					     new_extents);
 
 out:
@@ -690,12 +690,11 @@ out:
 }
 
 static int remove_free_space_extent(struct btrfs_trans_handle *trans,
-				    struct btrfs_fs_info *fs_info,
 				    struct btrfs_block_group_cache *block_group,
 				    struct btrfs_path *path,
 				    u64 start, u64 size)
 {
-	struct btrfs_root *root = fs_info->free_space_root;
+	struct btrfs_root *root = trans->fs_info->free_space_root;
 	struct btrfs_key key;
 	u64 found_start, found_end;
 	u64 end = start + size;
@@ -769,7 +768,7 @@ static int remove_free_space_extent(struct btrfs_trans_handle *trans,
 	}
 
 	btrfs_release_path(path);
-	ret = update_free_space_extent_count(trans, fs_info, block_group, path,
+	ret = update_free_space_extent_count(trans, block_group, path,
 					     new_extents);
 
 out:
@@ -777,7 +776,6 @@ out:
 }
 
 int __remove_from_free_space_tree(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *block_group,
 				  struct btrfs_path *path, u64 start, u64 size)
 {
@@ -786,36 +784,35 @@ int __remove_from_free_space_tree(struct btrfs_trans_handle *trans,
 	int ret;
 
 	if (block_group->needs_free_space) {
-		ret = __add_block_group_free_space(trans, fs_info, block_group,
-						   path);
+		ret = __add_block_group_free_space(trans, block_group, path);
 		if (ret)
 			return ret;
 	}
 
-	info = search_free_space_info(NULL, fs_info, block_group, path, 0);
+	info = search_free_space_info(NULL, trans->fs_info, block_group, path,
+				      0);
 	if (IS_ERR(info))
 		return PTR_ERR(info);
 	flags = btrfs_free_space_flags(path->nodes[0], info);
 	btrfs_release_path(path);
 
 	if (flags & BTRFS_FREE_SPACE_USING_BITMAPS) {
-		return modify_free_space_bitmap(trans, fs_info, block_group,
-						path, start, size, 1);
+		return modify_free_space_bitmap(trans, block_group, path,
+						start, size, 1);
 	} else {
-		return remove_free_space_extent(trans, fs_info, block_group,
-						path, start, size);
+		return remove_free_space_extent(trans, block_group, path,
+						start, size);
 	}
 }
 
 int remove_from_free_space_tree(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *fs_info,
 				u64 start, u64 size)
 {
 	struct btrfs_block_group_cache *block_group;
 	struct btrfs_path *path;
 	int ret;
 
-	if (!btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE))
+	if (!btrfs_fs_compat_ro(trans->fs_info, FREE_SPACE_TREE))
 		return 0;
 
 	path = btrfs_alloc_path();
@@ -824,7 +821,7 @@ int remove_from_free_space_tree(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	block_group = btrfs_lookup_block_group(fs_info, start);
+	block_group = btrfs_lookup_block_group(trans->fs_info, start);
 	if (!block_group) {
 		ASSERT(0);
 		ret = -ENOENT;
@@ -832,8 +829,8 @@ int remove_from_free_space_tree(struct btrfs_trans_handle *trans,
 	}
 
 	mutex_lock(&block_group->free_space_lock);
-	ret = __remove_from_free_space_tree(trans, fs_info, block_group, path,
-					    start, size);
+	ret = __remove_from_free_space_tree(trans, block_group, path, start,
+					    size);
 	mutex_unlock(&block_group->free_space_lock);
 
 	btrfs_put_block_group(block_group);
@@ -845,12 +842,11 @@ out:
 }
 
 static int add_free_space_extent(struct btrfs_trans_handle *trans,
-				 struct btrfs_fs_info *fs_info,
 				 struct btrfs_block_group_cache *block_group,
 				 struct btrfs_path *path,
 				 u64 start, u64 size)
 {
-	struct btrfs_root *root = fs_info->free_space_root;
+	struct btrfs_root *root = trans->fs_info->free_space_root;
 	struct btrfs_key key, new_key;
 	u64 found_start, found_end;
 	u64 end = start + size;
@@ -965,7 +961,7 @@ insert:
 		goto out;
 
 	btrfs_release_path(path);
-	ret = update_free_space_extent_count(trans, fs_info, block_group, path,
+	ret = update_free_space_extent_count(trans, block_group, path,
 					     new_extents);
 
 out:
@@ -973,17 +969,16 @@ out:
 }
 
 int __add_to_free_space_tree(struct btrfs_trans_handle *trans,
-			     struct btrfs_fs_info *fs_info,
 			     struct btrfs_block_group_cache *block_group,
 			     struct btrfs_path *path, u64 start, u64 size)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_free_space_info *info;
 	u32 flags;
 	int ret;
 
 	if (block_group->needs_free_space) {
-		ret = __add_block_group_free_space(trans, fs_info, block_group,
-						   path);
+		ret = __add_block_group_free_space(trans, block_group, path);
 		if (ret)
 			return ret;
 	}
@@ -995,23 +990,22 @@ int __add_to_free_space_tree(struct btrfs_trans_handle *trans,
 	btrfs_release_path(path);
 
 	if (flags & BTRFS_FREE_SPACE_USING_BITMAPS) {
-		return modify_free_space_bitmap(trans, fs_info, block_group,
-						path, start, size, 0);
+		return modify_free_space_bitmap(trans, block_group, path,
+						start, size, 0);
 	} else {
-		return add_free_space_extent(trans, fs_info, block_group, path,
-					     start, size);
+		return add_free_space_extent(trans, block_group, path, start,
+					     size);
 	}
 }
 
 int add_to_free_space_tree(struct btrfs_trans_handle *trans,
-			   struct btrfs_fs_info *fs_info,
 			   u64 start, u64 size)
 {
 	struct btrfs_block_group_cache *block_group;
 	struct btrfs_path *path;
 	int ret;
 
-	if (!btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE))
+	if (!btrfs_fs_compat_ro(trans->fs_info, FREE_SPACE_TREE))
 		return 0;
 
 	path = btrfs_alloc_path();
@@ -1020,7 +1014,7 @@ int add_to_free_space_tree(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	block_group = btrfs_lookup_block_group(fs_info, start);
+	block_group = btrfs_lookup_block_group(trans->fs_info, start);
 	if (!block_group) {
 		ASSERT(0);
 		ret = -ENOENT;
@@ -1028,8 +1022,7 @@ int add_to_free_space_tree(struct btrfs_trans_handle *trans,
 	}
 
 	mutex_lock(&block_group->free_space_lock);
-	ret = __add_to_free_space_tree(trans, fs_info, block_group, path, start,
-				       size);
+	ret = __add_to_free_space_tree(trans, block_group, path, start, size);
 	mutex_unlock(&block_group->free_space_lock);
 
 	btrfs_put_block_group(block_group);
@@ -1046,10 +1039,9 @@ out:
  * through the normal add/remove hooks.
  */
 static int populate_free_space_tree(struct btrfs_trans_handle *trans,
-				    struct btrfs_fs_info *fs_info,
 				    struct btrfs_block_group_cache *block_group)
 {
-	struct btrfs_root *extent_root = fs_info->extent_root;
+	struct btrfs_root *extent_root = trans->fs_info->extent_root;
 	struct btrfs_path *path, *path2;
 	struct btrfs_key key;
 	u64 start, end;
@@ -1066,7 +1058,7 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans,
 		return -ENOMEM;
 	}
 
-	ret = add_new_free_space_info(trans, fs_info, block_group, path2);
+	ret = add_new_free_space_info(trans, block_group, path2);
 	if (ret)
 		goto out;
 
@@ -1099,7 +1091,7 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans,
 				break;
 
 			if (start < key.objectid) {
-				ret = __add_to_free_space_tree(trans, fs_info,
+				ret = __add_to_free_space_tree(trans,
 							       block_group,
 							       path2, start,
 							       key.objectid -
@@ -1109,7 +1101,7 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans,
 			}
 			start = key.objectid;
 			if (key.type == BTRFS_METADATA_ITEM_KEY)
-				start += fs_info->nodesize;
+				start += trans->fs_info->nodesize;
 			else
 				start += key.offset;
 		} else if (key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) {
@@ -1124,8 +1116,8 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans,
 			break;
 	}
 	if (start < end) {
-		ret = __add_to_free_space_tree(trans, fs_info, block_group,
-					       path2, start, end - start);
+		ret = __add_to_free_space_tree(trans, block_group, path2,
+					       start, end - start);
 		if (ret)
 			goto out_locked;
 	}
@@ -1165,7 +1157,7 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
 	while (node) {
 		block_group = rb_entry(node, struct btrfs_block_group_cache,
 				       cache_node);
-		ret = populate_free_space_tree(trans, fs_info, block_group);
+		ret = populate_free_space_tree(trans, block_group);
 		if (ret)
 			goto abort;
 		node = rb_next(node);
@@ -1269,7 +1261,6 @@ abort:
 }
 
 static int __add_block_group_free_space(struct btrfs_trans_handle *trans,
-					struct btrfs_fs_info *fs_info,
 					struct btrfs_block_group_cache *block_group,
 					struct btrfs_path *path)
 {
@@ -1277,19 +1268,19 @@ static int __add_block_group_free_space(struct btrfs_trans_handle *trans,
 
 	block_group->needs_free_space = 0;
 
-	ret = add_new_free_space_info(trans, fs_info, block_group, path);
+	ret = add_new_free_space_info(trans, block_group, path);
 	if (ret)
 		return ret;
 
-	return __add_to_free_space_tree(trans, fs_info, block_group, path,
+	return __add_to_free_space_tree(trans, block_group, path,
 					block_group->key.objectid,
 					block_group->key.offset);
 }
 
 int add_block_group_free_space(struct btrfs_trans_handle *trans,
-			       struct btrfs_fs_info *fs_info,
 			       struct btrfs_block_group_cache *block_group)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_path *path = NULL;
 	int ret = 0;
 
@@ -1306,7 +1297,7 @@ int add_block_group_free_space(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	ret = __add_block_group_free_space(trans, fs_info, block_group, path);
+	ret = __add_block_group_free_space(trans, block_group, path);
 
 out:
 	btrfs_free_path(path);
@@ -1317,10 +1308,9 @@ out:
 }
 
 int remove_block_group_free_space(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *block_group)
 {
-	struct btrfs_root *root = fs_info->free_space_root;
+	struct btrfs_root *root = trans->fs_info->free_space_root;
 	struct btrfs_path *path;
 	struct btrfs_key key, found_key;
 	struct extent_buffer *leaf;
@@ -1328,7 +1318,7 @@ int remove_block_group_free_space(struct btrfs_trans_handle *trans,
 	int done = 0, nr;
 	int ret;
 
-	if (!btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE))
+	if (!btrfs_fs_compat_ro(trans->fs_info, FREE_SPACE_TREE))
 		return 0;
 
 	if (block_group->needs_free_space) {
@@ -1439,7 +1429,6 @@ static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl,
 				extent_start = offset;
 			} else if (prev_bit == 1 && bit == 0) {
 				total_found += add_new_free_space(block_group,
-								  fs_info,
 								  extent_start,
 								  offset);
 				if (total_found > CACHING_CTL_WAKE_UP) {
@@ -1453,8 +1442,8 @@ static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl,
 		}
 	}
 	if (prev_bit == 1) {
-		total_found += add_new_free_space(block_group, fs_info,
-						  extent_start, end);
+		total_found += add_new_free_space(block_group, extent_start,
+						  end);
 		extent_count++;
 	}
 
@@ -1511,8 +1500,7 @@ static int load_free_space_extents(struct btrfs_caching_control *caching_ctl,
 
 		caching_ctl->progress = key.objectid;
 
-		total_found += add_new_free_space(block_group, fs_info,
-						  key.objectid,
+		total_found += add_new_free_space(block_group, key.objectid,
 						  key.objectid + key.offset);
 		if (total_found > CACHING_CTL_WAKE_UP) {
 			total_found = 0;

fs/btrfs/free-space-tree.h

@@ -19,16 +19,12 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info);
 int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info);
 int load_free_space_tree(struct btrfs_caching_control *caching_ctl);
 int add_block_group_free_space(struct btrfs_trans_handle *trans,
-			       struct btrfs_fs_info *fs_info,
 			       struct btrfs_block_group_cache *block_group);
 int remove_block_group_free_space(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *block_group);
 int add_to_free_space_tree(struct btrfs_trans_handle *trans,
-			   struct btrfs_fs_info *fs_info,
 			   u64 start, u64 size);
 int remove_from_free_space_tree(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *fs_info,
 				u64 start, u64 size);
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
@@ -38,19 +34,15 @@ search_free_space_info(struct btrfs_trans_handle *trans,
 		       struct btrfs_block_group_cache *block_group,
 		       struct btrfs_path *path, int cow);
 int __add_to_free_space_tree(struct btrfs_trans_handle *trans,
-			     struct btrfs_fs_info *fs_info,
 			     struct btrfs_block_group_cache *block_group,
 			     struct btrfs_path *path, u64 start, u64 size);
 int __remove_from_free_space_tree(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *block_group,
 				  struct btrfs_path *path, u64 start, u64 size);
 int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *block_group,
 				  struct btrfs_path *path);
 int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
-				  struct btrfs_fs_info *fs_info,
 				  struct btrfs_block_group_cache *block_group,
 				  struct btrfs_path *path);
 int free_space_test_bit(struct btrfs_block_group_cache *block_group,

fs/btrfs/inode.c

@@ -1018,8 +1018,10 @@ static noinline int cow_file_range(struct inode *inode,
 				  ram_size, /* ram_bytes */
 				  BTRFS_COMPRESS_NONE, /* compress_type */
 				  BTRFS_ORDERED_REGULAR /* type */);
-		if (IS_ERR(em))
+		if (IS_ERR(em)) {
+			ret = PTR_ERR(em);
 			goto out_reserve;
+		}
 		free_extent_map(em);
 
 		ret = btrfs_add_ordered_extent(inode, start, ins.objectid,
@@ -1156,13 +1158,10 @@ static noinline void async_cow_submit(struct btrfs_work *work)
 	nr_pages = (async_cow->end - async_cow->start + PAGE_SIZE) >>
 		PAGE_SHIFT;
 
-	/*
-	 * atomic_sub_return implies a barrier for waitqueue_active
-	 */
+	/* atomic_sub_return implies a barrier */
 	if (atomic_sub_return(nr_pages, &fs_info->async_delalloc_pages) <
-	    5 * SZ_1M &&
-	    waitqueue_active(&fs_info->async_submit_wait))
-		wake_up(&fs_info->async_submit_wait);
+	    5 * SZ_1M)
+		cond_wake_up_nomb(&fs_info->async_submit_wait);
 
 	if (async_cow->inode)
 		submit_compressed_extents(async_cow->inode, async_cow);
@@ -1373,6 +1372,13 @@ next_slot:
 			    btrfs_file_extent_encryption(leaf, fi) ||
 			    btrfs_file_extent_other_encoding(leaf, fi))
 				goto out_check;
+			/*
+			 * Do the same check as in btrfs_cross_ref_exist but
+			 * without the unnecessary search.
+			 */
+			if (btrfs_file_extent_generation(leaf, fi) <=
+			    btrfs_root_last_snapshot(&root->root_item))
+				goto out_check;
 			if (extent_type == BTRFS_FILE_EXTENT_REG && !force)
 				goto out_check;
 			if (btrfs_extent_readonly(fs_info, disk_bytenr))
@@ -1754,6 +1760,7 @@ void __btrfs_del_delalloc_inode(struct btrfs_root *root,
 			  &inode->runtime_flags);
 		root->nr_delalloc_inodes--;
 		if (!root->nr_delalloc_inodes) {
+			ASSERT(list_empty(&root->delalloc_inodes));
 			spin_lock(&fs_info->delalloc_root_lock);
 			BUG_ON(list_empty(&root->delalloc_root));
 			list_del_init(&root->delalloc_root);
@@ -3158,6 +3165,9 @@ out:
 	/* once for the tree */
 	btrfs_put_ordered_extent(ordered_extent);
 
+	/* Try to release some metadata so we don't get an OOM but don't wait */
+	btrfs_btree_balance_dirty_nodelay(fs_info);
+
 	return ret;
 }
 
@@ -3300,177 +3310,31 @@ void btrfs_run_delayed_iputs(struct btrfs_fs_info *fs_info)
 }
 
 /*
- * This is called in transaction commit time. If there are no orphan
- * files in the subvolume, it removes orphan item and frees block_rsv
- * structure.
- */
-void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans,
-			      struct btrfs_root *root)
-{
-	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct btrfs_block_rsv *block_rsv;
-	int ret;
-
-	if (atomic_read(&root->orphan_inodes) ||
-	    root->orphan_cleanup_state != ORPHAN_CLEANUP_DONE)
-		return;
-
-	spin_lock(&root->orphan_lock);
-	if (atomic_read(&root->orphan_inodes)) {
-		spin_unlock(&root->orphan_lock);
-		return;
-	}
-
-	if (root->orphan_cleanup_state != ORPHAN_CLEANUP_DONE) {
-		spin_unlock(&root->orphan_lock);
-		return;
-	}
-
-	block_rsv = root->orphan_block_rsv;
-	root->orphan_block_rsv = NULL;
-	spin_unlock(&root->orphan_lock);
-
-	if (test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state) &&
-	    btrfs_root_refs(&root->root_item) > 0) {
-		ret = btrfs_del_orphan_item(trans, fs_info->tree_root,
-					    root->root_key.objectid);
-		if (ret)
-			btrfs_abort_transaction(trans, ret);
-		else
-			clear_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED,
-				  &root->state);
-	}
-
-	if (block_rsv) {
-		WARN_ON(block_rsv->size > 0);
-		btrfs_free_block_rsv(fs_info, block_rsv);
-	}
-}
-
-/*
- * This creates an orphan entry for the given inode in case something goes
- * wrong in the middle of an unlink/truncate.
- *
- * NOTE: caller of this function should reserve 5 units of metadata for
- *	 this function.
+ * This creates an orphan entry for the given inode in case something goes wrong
+ * in the middle of an unlink.
  */
 int btrfs_orphan_add(struct btrfs_trans_handle *trans,
-		struct btrfs_inode *inode)
+		     struct btrfs_inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
-	struct btrfs_root *root = inode->root;
-	struct btrfs_block_rsv *block_rsv = NULL;
-	int reserve = 0;
-	bool insert = false;
 	int ret;
 
-	if (!root->orphan_block_rsv) {
-		block_rsv = btrfs_alloc_block_rsv(fs_info,
-						  BTRFS_BLOCK_RSV_TEMP);
-		if (!block_rsv)
-			return -ENOMEM;
-	}
-
-	if (!test_and_set_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
-			      &inode->runtime_flags))
-		insert = true;
-
-	if (!test_and_set_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
-			      &inode->runtime_flags))
-		reserve = 1;
-
-	spin_lock(&root->orphan_lock);
-	/* If someone has created ->orphan_block_rsv, be happy to use it. */
-	if (!root->orphan_block_rsv) {
-		root->orphan_block_rsv = block_rsv;
-	} else if (block_rsv) {
-		btrfs_free_block_rsv(fs_info, block_rsv);
-		block_rsv = NULL;
-	}
-
-	if (insert)
-		atomic_inc(&root->orphan_inodes);
-	spin_unlock(&root->orphan_lock);
-
-	/* grab metadata reservation from transaction handle */
-	if (reserve) {
-		ret = btrfs_orphan_reserve_metadata(trans, inode);
-		ASSERT(!ret);
-		if (ret) {
-			/*
-			 * dec doesn't need spin_lock as ->orphan_block_rsv
-			 * would be released only if ->orphan_inodes is
-			 * zero.
-			 */
-			atomic_dec(&root->orphan_inodes);
-			clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
-				  &inode->runtime_flags);
-			if (insert)
-				clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
-					  &inode->runtime_flags);
-			return ret;
-		}
-	}
-
-	/* insert an orphan item to track this unlinked/truncated file */
-	if (insert) {
-		ret = btrfs_insert_orphan_item(trans, root, btrfs_ino(inode));
-		if (ret) {
-			if (reserve) {
-				clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
-					  &inode->runtime_flags);
-				btrfs_orphan_release_metadata(inode);
-			}
-			/*
-			 * btrfs_orphan_commit_root may race with us and set
-			 * ->orphan_block_rsv to zero, in order to avoid that,
-			 * decrease ->orphan_inodes after everything is done.
-			 */
-			atomic_dec(&root->orphan_inodes);
-			if (ret != -EEXIST) {
-				clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
-					  &inode->runtime_flags);
-				btrfs_abort_transaction(trans, ret);
-				return ret;
-			}
-		}
-		ret = 0;
+	ret = btrfs_insert_orphan_item(trans, inode->root, btrfs_ino(inode));
+	if (ret && ret != -EEXIST) {
+		btrfs_abort_transaction(trans, ret);
+		return ret;
 	}
 
 	return 0;
 }
 
 /*
- * We have done the truncate/delete so we can go ahead and remove the orphan
- * item for this particular inode.
+ * We have done the delete so we can go ahead and remove the orphan item for
+ * this particular inode.
  */
 static int btrfs_orphan_del(struct btrfs_trans_handle *trans,
 			    struct btrfs_inode *inode)
 {
-	struct btrfs_root *root = inode->root;
-	int delete_item = 0;
-	int ret = 0;
-
-	if (test_and_clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
-			       &inode->runtime_flags))
-		delete_item = 1;
-
-	if (delete_item && trans)
-		ret = btrfs_del_orphan_item(trans, root, btrfs_ino(inode));
-
-	if (test_and_clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
-			       &inode->runtime_flags))
-		btrfs_orphan_release_metadata(inode);
-
-	/*
-	 * btrfs_orphan_commit_root may race with us and set ->orphan_block_rsv
-	 * to zero, in order to avoid that, decrease ->orphan_inodes after
-	 * everything is done.
-	 */
-	if (delete_item)
-		atomic_dec(&root->orphan_inodes);
-
-	return ret;
+	return btrfs_del_orphan_item(trans, inode->root, btrfs_ino(inode));
 }
 
 /*
@@ -3486,7 +3350,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 	struct btrfs_trans_handle *trans;
 	struct inode *inode;
 	u64 last_objectid = 0;
-	int ret = 0, nr_unlink = 0, nr_truncate = 0;
+	int ret = 0, nr_unlink = 0;
 
 	if (cmpxchg(&root->orphan_cleanup_state, 0, ORPHAN_CLEANUP_STARTED))
 		return 0;
@@ -3586,12 +3450,31 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 				key.offset = found_key.objectid - 1;
 				continue;
 			}
+
 		}
+
 		/*
-		 * Inode is already gone but the orphan item is still there,
-		 * kill the orphan item.
+		 * If we have an inode with links, there are a couple of
+		 * possibilities. Old kernels (before v3.12) used to create an
+		 * orphan item for truncate indicating that there were possibly
+		 * extent items past i_size that needed to be deleted. In v3.12,
+		 * truncate was changed to update i_size in sync with the extent
+		 * items, but the (useless) orphan item was still created. Since
+		 * v4.18, we don't create the orphan item for truncate at all.
+		 *
+		 * So, this item could mean that we need to do a truncate, but
+		 * only if this filesystem was last used on a pre-v3.12 kernel
+		 * and was not cleanly unmounted. The odds of that are quite
+		 * slim, and it's a pain to do the truncate now, so just delete
+		 * the orphan item.
+		 *
+		 * It's also possible that this orphan item was supposed to be
+		 * deleted but wasn't. The inode number may have been reused,
+		 * but either way, we can delete the orphan item.
 		 */
-		if (ret == -ENOENT) {
+		if (ret == -ENOENT || inode->i_nlink) {
+			if (!ret)
+				iput(inode);
 			trans = btrfs_start_transaction(root, 1);
 			if (IS_ERR(trans)) {
 				ret = PTR_ERR(trans);
@@ -3607,42 +3490,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 			continue;
 		}
 
-		/*
-		 * add this inode to the orphan list so btrfs_orphan_del does
-		 * the proper thing when we hit it
-		 */
-		set_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
-			&BTRFS_I(inode)->runtime_flags);
-		atomic_inc(&root->orphan_inodes);
-
-		/* if we have links, this was a truncate, lets do that */
-		if (inode->i_nlink) {
-			if (WARN_ON(!S_ISREG(inode->i_mode))) {
-				iput(inode);
-				continue;
-			}
-			nr_truncate++;
-
-			/* 1 for the orphan item deletion. */
-			trans = btrfs_start_transaction(root, 1);
-			if (IS_ERR(trans)) {
-				iput(inode);
-				ret = PTR_ERR(trans);
-				goto out;
-			}
-			ret = btrfs_orphan_add(trans, BTRFS_I(inode));
-			btrfs_end_transaction(trans);
-			if (ret) {
-				iput(inode);
-				goto out;
-			}
-
-			ret = btrfs_truncate(inode, false);
-			if (ret)
-				btrfs_orphan_del(NULL, BTRFS_I(inode));
-		} else {
-			nr_unlink++;
-		}
+		nr_unlink++;
 
 		/* this will do delete_inode and everything for us */
 		iput(inode);
@@ -3654,12 +3502,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 
 	root->orphan_cleanup_state = ORPHAN_CLEANUP_DONE;
 
-	if (root->orphan_block_rsv)
-		btrfs_block_rsv_release(fs_info, root->orphan_block_rsv,
-					(u64)-1);
-
-	if (root->orphan_block_rsv ||
-	    test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state)) {
+	if (test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state)) {
 		trans = btrfs_join_transaction(root);
 		if (!IS_ERR(trans))
 			btrfs_end_transaction(trans);
@@ -3667,8 +3510,6 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 
 	if (nr_unlink)
 		btrfs_debug(fs_info, "unlinked %d orphans", nr_unlink);
-	if (nr_truncate)
-		btrfs_debug(fs_info, "truncated %d orphans", nr_truncate);
 
 out:
 	if (ret)
@@ -3931,7 +3772,7 @@ cache_acl:
 		break;
 	}
 
-	btrfs_update_iflags(inode);
+	btrfs_sync_inode_flags_to_i_flags(inode);
 	return 0;
 
 make_bad:
@@ -4245,7 +4086,7 @@ out:
 	return ret;
 }
 
-int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
+static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 			struct btrfs_root *root,
 			struct inode *dir, u64 objectid,
 			const char *name, int name_len)
@@ -4326,6 +4167,262 @@ out:
 	return ret;
 }
 
+/*
+ * Helper to check if the subvolume references other subvolumes or if it's
+ * default.
+ */
+static noinline int may_destroy_subvol(struct btrfs_root *root)
+{
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct btrfs_path *path;
+	struct btrfs_dir_item *di;
+	struct btrfs_key key;
+	u64 dir_id;
+	int ret;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	/* Make sure this root isn't set as the default subvol */
+	dir_id = btrfs_super_root_dir(fs_info->super_copy);
+	di = btrfs_lookup_dir_item(NULL, fs_info->tree_root, path,
+				   dir_id, "default", 7, 0);
+	if (di && !IS_ERR(di)) {
+		btrfs_dir_item_key_to_cpu(path->nodes[0], di, &key);
+		if (key.objectid == root->root_key.objectid) {
+			ret = -EPERM;
+			btrfs_err(fs_info,
+				  "deleting default subvolume %llu is not allowed",
+				  key.objectid);
+			goto out;
+		}
+		btrfs_release_path(path);
+	}
+
+	key.objectid = root->root_key.objectid;
+	key.type = BTRFS_ROOT_REF_KEY;
+	key.offset = (u64)-1;
+
+	ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0);
+	if (ret < 0)
+		goto out;
+	BUG_ON(ret == 0);
+
+	ret = 0;
+	if (path->slots[0] > 0) {
+		path->slots[0]--;
+		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+		if (key.objectid == root->root_key.objectid &&
+		    key.type == BTRFS_ROOT_REF_KEY)
+			ret = -ENOTEMPTY;
+	}
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
+/* Delete all dentries for inodes belonging to the root */
+static void btrfs_prune_dentries(struct btrfs_root *root)
+{
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct rb_node *node;
+	struct rb_node *prev;
+	struct btrfs_inode *entry;
+	struct inode *inode;
+	u64 objectid = 0;
+
+	if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
+		WARN_ON(btrfs_root_refs(&root->root_item) != 0);
+
+	spin_lock(&root->inode_lock);
+again:
+	node = root->inode_tree.rb_node;
+	prev = NULL;
+	while (node) {
+		prev = node;
+		entry = rb_entry(node, struct btrfs_inode, rb_node);
+
+		if (objectid < btrfs_ino(BTRFS_I(&entry->vfs_inode)))
+			node = node->rb_left;
+		else if (objectid > btrfs_ino(BTRFS_I(&entry->vfs_inode)))
+			node = node->rb_right;
+		else
+			break;
+	}
+	if (!node) {
+		while (prev) {
+			entry = rb_entry(prev, struct btrfs_inode, rb_node);
+			if (objectid <= btrfs_ino(BTRFS_I(&entry->vfs_inode))) {
+				node = prev;
+				break;
+			}
+			prev = rb_next(prev);
+		}
+	}
+	while (node) {
+		entry = rb_entry(node, struct btrfs_inode, rb_node);
+		objectid = btrfs_ino(BTRFS_I(&entry->vfs_inode)) + 1;
+		inode = igrab(&entry->vfs_inode);
+		if (inode) {
+			spin_unlock(&root->inode_lock);
+			if (atomic_read(&inode->i_count) > 1)
+				d_prune_aliases(inode);
+			/*
+			 * btrfs_drop_inode will have it removed from the inode
+			 * cache when its usage count hits zero.
+			 */
+			iput(inode);
+			cond_resched();
+			spin_lock(&root->inode_lock);
+			goto again;
+		}
+
+		if (cond_resched_lock(&root->inode_lock))
+			goto again;
+
+		node = rb_next(node);
+	}
+	spin_unlock(&root->inode_lock);
+}
+
+int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry)
+{
+	struct btrfs_fs_info *fs_info = btrfs_sb(dentry->d_sb);
+	struct btrfs_root *root = BTRFS_I(dir)->root;
+	struct inode *inode = d_inode(dentry);
+	struct btrfs_root *dest = BTRFS_I(inode)->root;
+	struct btrfs_trans_handle *trans;
+	struct btrfs_block_rsv block_rsv;
+	u64 root_flags;
+	int ret;
+	int err;
+
+	/*
+	 * Don't allow to delete a subvolume with send in progress. This is
+	 * inside the inode lock so the error handling that has to drop the bit
+	 * again is not run concurrently.
+	 */
+	spin_lock(&dest->root_item_lock);
+	root_flags = btrfs_root_flags(&dest->root_item);
+	if (dest->send_in_progress == 0) {
+		btrfs_set_root_flags(&dest->root_item,
+				root_flags | BTRFS_ROOT_SUBVOL_DEAD);
+		spin_unlock(&dest->root_item_lock);
+	} else {
+		spin_unlock(&dest->root_item_lock);
+		btrfs_warn(fs_info,
+			   "attempt to delete subvolume %llu during send",
+			   dest->root_key.objectid);
+		return -EPERM;
+	}
+
+	down_write(&fs_info->subvol_sem);
+
+	err = may_destroy_subvol(dest);
+	if (err)
+		goto out_up_write;
+
+	btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP);
+	/*
+	 * One for dir inode,
+	 * two for dir entries,
+	 * two for root ref/backref.
+	 */
+	err = btrfs_subvolume_reserve_metadata(root, &block_rsv, 5, true);
+	if (err)
+		goto out_up_write;
+
+	trans = btrfs_start_transaction(root, 0);
+	if (IS_ERR(trans)) {
+		err = PTR_ERR(trans);
+		goto out_release;
+	}
+	trans->block_rsv = &block_rsv;
+	trans->bytes_reserved = block_rsv.size;
+
+	btrfs_record_snapshot_destroy(trans, BTRFS_I(dir));
+
+	ret = btrfs_unlink_subvol(trans, root, dir,
+				dest->root_key.objectid,
+				dentry->d_name.name,
+				dentry->d_name.len);
+	if (ret) {
+		err = ret;
+		btrfs_abort_transaction(trans, ret);
+		goto out_end_trans;
+	}
+
+	btrfs_record_root_in_trans(trans, dest);
+
+	memset(&dest->root_item.drop_progress, 0,
+		sizeof(dest->root_item.drop_progress));
+	dest->root_item.drop_level = 0;
+	btrfs_set_root_refs(&dest->root_item, 0);
+
+	if (!test_and_set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &dest->state)) {
+		ret = btrfs_insert_orphan_item(trans,
+					fs_info->tree_root,
+					dest->root_key.objectid);
+		if (ret) {
+			btrfs_abort_transaction(trans, ret);
+			err = ret;
+			goto out_end_trans;
+		}
+	}
+
+	ret = btrfs_uuid_tree_remove(trans, dest->root_item.uuid,
+				  BTRFS_UUID_KEY_SUBVOL,
+				  dest->root_key.objectid);
+	if (ret && ret != -ENOENT) {
+		btrfs_abort_transaction(trans, ret);
+		err = ret;
+		goto out_end_trans;
+	}
+	if (!btrfs_is_empty_uuid(dest->root_item.received_uuid)) {
+		ret = btrfs_uuid_tree_remove(trans,
+					  dest->root_item.received_uuid,
+					  BTRFS_UUID_KEY_RECEIVED_SUBVOL,
+					  dest->root_key.objectid);
+		if (ret && ret != -ENOENT) {
+			btrfs_abort_transaction(trans, ret);
+			err = ret;
+			goto out_end_trans;
+		}
+	}
+
+out_end_trans:
+	trans->block_rsv = NULL;
+	trans->bytes_reserved = 0;
+	ret = btrfs_end_transaction(trans);
+	if (ret && !err)
+		err = ret;
+	inode->i_flags |= S_DEAD;
+out_release:
+	btrfs_subvolume_release_metadata(fs_info, &block_rsv);
+out_up_write:
+	up_write(&fs_info->subvol_sem);
+	if (err) {
+		spin_lock(&dest->root_item_lock);
+		root_flags = btrfs_root_flags(&dest->root_item);
+		btrfs_set_root_flags(&dest->root_item,
+				root_flags & ~BTRFS_ROOT_SUBVOL_DEAD);
+		spin_unlock(&dest->root_item_lock);
+	} else {
+		d_invalidate(dentry);
+		btrfs_prune_dentries(dest);
+		ASSERT(dest->send_in_progress == 0);
+
+		/* the last ref */
+		if (dest->ino_cache_inode) {
+			iput(dest->ino_cache_inode);
+			dest->ino_cache_inode = NULL;
+		}
+	}
+
+	return err;
+}
+
 static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 {
 	struct inode *inode = d_inode(dentry);
@@ -4337,7 +4434,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 	if (inode->i_size > BTRFS_EMPTY_DIR_SIZE)
 		return -ENOTEMPTY;
 	if (btrfs_ino(BTRFS_I(inode)) == BTRFS_FIRST_FREE_OBJECTID)
-		return -EPERM;
+		return btrfs_delete_subvolume(dir, dentry);
 
 	trans = __unlink_start_trans(dir);
 	if (IS_ERR(trans))
@@ -4449,7 +4546,6 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 	int pending_del_slot = 0;
 	int extent_type = -1;
 	int ret;
-	int err = 0;
 	u64 ino = btrfs_ino(BTRFS_I(inode));
 	u64 bytes_deleted = 0;
 	bool be_nice = false;
@@ -4501,22 +4597,19 @@ search_again:
 	 * up a huge file in a single leaf.  Most of the time that
 	 * bytes_deleted is > 0, it will be huge by the time we get here
 	 */
-	if (be_nice && bytes_deleted > SZ_32M) {
-		if (btrfs_should_end_transaction(trans)) {
-			err = -EAGAIN;
-			goto error;
-		}
+	if (be_nice && bytes_deleted > SZ_32M &&
+	    btrfs_should_end_transaction(trans)) {
+		ret = -EAGAIN;
+		goto out;
 	}
 
-
 	path->leave_spinning = 1;
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
-	if (ret < 0) {
-		err = ret;
+	if (ret < 0)
 		goto out;
-	}
 
 	if (ret > 0) {
+		ret = 0;
 		/* there are no items in the tree for us to truncate, we're
 		 * done
 		 */
@@ -4627,7 +4720,7 @@ search_again:
 				 * We have to bail so the last_size is set to
 				 * just before this extent.
 				 */
-				err = NEED_TRUNCATE_BLOCK;
+				ret = NEED_TRUNCATE_BLOCK;
 				break;
 			}
 
@@ -4666,7 +4759,10 @@ delete:
 						extent_num_bytes, 0,
 						btrfs_header_owner(leaf),
 						ino, extent_offset);
-			BUG_ON(ret);
+			if (ret) {
+				btrfs_abort_transaction(trans, ret);
+				break;
+			}
 			if (btrfs_should_throttle_delayed_refs(trans, fs_info))
 				btrfs_async_run_delayed_refs(fs_info,
 					trans->delayed_ref_updates * 2,
@@ -4694,7 +4790,7 @@ delete:
 						pending_del_nr);
 				if (ret) {
 					btrfs_abort_transaction(trans, ret);
-					goto error;
+					break;
 				}
 				pending_del_nr = 0;
 			}
@@ -4705,8 +4801,8 @@ delete:
 					trans->delayed_ref_updates = 0;
 					ret = btrfs_run_delayed_refs(trans,
 								   updates * 2);
-					if (ret && !err)
-						err = ret;
+					if (ret)
+						break;
 				}
 			}
 			/*
@@ -4714,8 +4810,8 @@ delete:
 			 * and let the transaction restart
 			 */
 			if (should_end) {
-				err = -EAGAIN;
-				goto error;
+				ret = -EAGAIN;
+				break;
 			}
 			goto search_again;
 		} else {
@@ -4723,32 +4819,37 @@ delete:
 		}
 	}
 out:
-	if (pending_del_nr) {
-		ret = btrfs_del_items(trans, root, path, pending_del_slot,
+	if (ret >= 0 && pending_del_nr) {
+		int err;
+
+		err = btrfs_del_items(trans, root, path, pending_del_slot,
 				      pending_del_nr);
-		if (ret)
-			btrfs_abort_transaction(trans, ret);
+		if (err) {
+			btrfs_abort_transaction(trans, err);
+			ret = err;
+		}
 	}
-error:
 	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
 		ASSERT(last_size >= new_size);
-		if (!err && last_size > new_size)
+		if (!ret && last_size > new_size)
 			last_size = new_size;
 		btrfs_ordered_update_i_size(inode, last_size, NULL);
 	}
 
 	btrfs_free_path(path);
 
-	if (be_nice && bytes_deleted > SZ_32M) {
+	if (be_nice && bytes_deleted > SZ_32M && (ret >= 0 || ret == -EAGAIN)) {
 		unsigned long updates = trans->delayed_ref_updates;
+		int err;
+
 		if (updates) {
 			trans->delayed_ref_updates = 0;
-			ret = btrfs_run_delayed_refs(trans, updates * 2);
-			if (ret && !err)
-				err = ret;
+			err = btrfs_run_delayed_refs(trans, updates * 2);
+			if (err)
+				ret = err;
 		}
 	}
-	return err;
+	return ret;
 }
 
 /*
@@ -5090,30 +5191,6 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 			set_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
 				&BTRFS_I(inode)->runtime_flags);
 
-		/*
-		 * 1 for the orphan item we're going to add
-		 * 1 for the orphan item deletion.
-		 */
-		trans = btrfs_start_transaction(root, 2);
-		if (IS_ERR(trans))
-			return PTR_ERR(trans);
-
-		/*
-		 * We need to do this in case we fail at _any_ point during the
-		 * actual truncate.  Once we do the truncate_setsize we could
-		 * invalidate pages which forces any outstanding ordered io to
-		 * be instantly completed which will give us extents that need
-		 * to be truncated.  If we fail to get an orphan inode down we
-		 * could have left over extents that were never meant to live,
-		 * so we need to guarantee from this point on that everything
-		 * will be consistent.
-		 */
-		ret = btrfs_orphan_add(trans, BTRFS_I(inode));
-		btrfs_end_transaction(trans);
-		if (ret)
-			return ret;
-
-		/* we don't support swapfiles, so vmtruncate shouldn't fail */
 		truncate_setsize(inode, newsize);
 
 		/* Disable nonlocked read DIO to avoid the end less truncate */
@@ -5125,29 +5202,16 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 		if (ret && inode->i_nlink) {
 			int err;
 
-			/* To get a stable disk_i_size */
-			err = btrfs_wait_ordered_range(inode, 0, (u64)-1);
-			if (err) {
-				btrfs_orphan_del(NULL, BTRFS_I(inode));
-				return err;
-			}
-
 			/*
-			 * failed to truncate, disk_i_size is only adjusted down
-			 * as we remove extents, so it should represent the true
-			 * size of the inode, so reset the in memory size and
-			 * delete our orphan entry.
+			 * Truncate failed, so fix up the in-memory size. We
+			 * adjusted disk_i_size down as we removed extents, so
+			 * wait for disk_i_size to be stable and then update the
+			 * in-memory size to match.
 			 */
-			trans = btrfs_join_transaction(root);
-			if (IS_ERR(trans)) {
-				btrfs_orphan_del(NULL, BTRFS_I(inode));
-				return ret;
-			}
-			i_size_write(inode, BTRFS_I(inode)->disk_i_size);
-			err = btrfs_orphan_del(trans, BTRFS_I(inode));
+			err = btrfs_wait_ordered_range(inode, 0, (u64)-1);
 			if (err)
-				btrfs_abort_transaction(trans, err);
-			btrfs_end_transaction(trans);
+				return err;
+			i_size_write(inode, BTRFS_I(inode)->disk_i_size);
 		}
 	}
 
@@ -5277,13 +5341,52 @@ static void evict_inode_truncate_pages(struct inode *inode)
 	spin_unlock(&io_tree->lock);
 }
 
+static struct btrfs_trans_handle *evict_refill_and_join(struct btrfs_root *root,
+							struct btrfs_block_rsv *rsv,
+							u64 min_size)
+{
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
+	int failures = 0;
+
+	for (;;) {
+		struct btrfs_trans_handle *trans;
+		int ret;
+
+		ret = btrfs_block_rsv_refill(root, rsv, min_size,
+					     BTRFS_RESERVE_FLUSH_LIMIT);
+
+		if (ret && ++failures > 2) {
+			btrfs_warn(fs_info,
+				   "could not allocate space for a delete; will truncate on mount");
+			return ERR_PTR(-ENOSPC);
+		}
+
+		trans = btrfs_join_transaction(root);
+		if (IS_ERR(trans) || !ret)
+			return trans;
+
+		/*
+		 * Try to steal from the global reserve if there is space for
+		 * it.
+		 */
+		if (!btrfs_check_space_for_delayed_refs(trans, fs_info) &&
+		    !btrfs_block_rsv_migrate(global_rsv, rsv, min_size, 0))
+			return trans;
+
+		/* If not, commit and try again. */
+		ret = btrfs_commit_transaction(trans);
+		if (ret)
+			return ERR_PTR(ret);
+	}
+}
+
 void btrfs_evict_inode(struct inode *inode)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_block_rsv *rsv, *global_rsv;
-	int steal_from_global = 0;
+	struct btrfs_block_rsv *rsv;
 	u64 min_size;
 	int ret;
 
@@ -5304,21 +5407,16 @@ void btrfs_evict_inode(struct inode *inode)
 	     btrfs_is_free_space_inode(BTRFS_I(inode))))
 		goto no_delete;
 
-	if (is_bad_inode(inode)) {
-		btrfs_orphan_del(NULL, BTRFS_I(inode));
+	if (is_bad_inode(inode))
 		goto no_delete;
-	}
 	/* do we really want it for ->i_nlink > 0 and zero btrfs_root_refs? */
 	if (!special_file(inode->i_mode))
 		btrfs_wait_ordered_range(inode, 0, (u64)-1);
 
 	btrfs_free_io_failure_record(BTRFS_I(inode), 0, (u64)-1);
 
-	if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) {
-		BUG_ON(test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
-				 &BTRFS_I(inode)->runtime_flags));
+	if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags))
 		goto no_delete;
-	}
 
 	if (inode->i_nlink > 0) {
 		BUG_ON(btrfs_root_refs(&root->root_item) != 0 &&
@@ -5327,130 +5425,63 @@ void btrfs_evict_inode(struct inode *inode)
 	}
 
 	ret = btrfs_commit_inode_delayed_inode(BTRFS_I(inode));
-	if (ret) {
-		btrfs_orphan_del(NULL, BTRFS_I(inode));
+	if (ret)
 		goto no_delete;
-	}
 
 	rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP);
-	if (!rsv) {
-		btrfs_orphan_del(NULL, BTRFS_I(inode));
+	if (!rsv)
 		goto no_delete;
-	}
 	rsv->size = min_size;
 	rsv->failfast = 1;
-	global_rsv = &fs_info->global_block_rsv;
 
 	btrfs_i_size_write(BTRFS_I(inode), 0);
 
-	/*
-	 * This is a bit simpler than btrfs_truncate since we've already
-	 * reserved our space for our orphan item in the unlink, so we just
-	 * need to reserve some slack space in case we add bytes and update
-	 * inode item when doing the truncate.
-	 */
 	while (1) {
-		ret = btrfs_block_rsv_refill(root, rsv, min_size,
-					     BTRFS_RESERVE_FLUSH_LIMIT);
-
-		/*
-		 * Try and steal from the global reserve since we will
-		 * likely not use this space anyway, we want to try as
-		 * hard as possible to get this to work.
-		 */
-		if (ret)
-			steal_from_global++;
-		else
-			steal_from_global = 0;
-		ret = 0;
-
-		/*
-		 * steal_from_global == 0: we reserved stuff, hooray!
-		 * steal_from_global == 1: we didn't reserve stuff, boo!
-		 * steal_from_global == 2: we've committed, still not a lot of
-		 * room but maybe we'll have room in the global reserve this
-		 * time.
-		 * steal_from_global == 3: abandon all hope!
-		 */
-		if (steal_from_global > 2) {
-			btrfs_warn(fs_info,
-				   "Could not get space for a delete, will truncate on mount %d",
-				   ret);
-			btrfs_orphan_del(NULL, BTRFS_I(inode));
-			btrfs_free_block_rsv(fs_info, rsv);
-			goto no_delete;
-		}
-
-		trans = btrfs_join_transaction(root);
-		if (IS_ERR(trans)) {
-			btrfs_orphan_del(NULL, BTRFS_I(inode));
-			btrfs_free_block_rsv(fs_info, rsv);
-			goto no_delete;
-		}
-
-		/*
-		 * We can't just steal from the global reserve, we need to make
-		 * sure there is room to do it, if not we need to commit and try
-		 * again.
-		 */
-		if (steal_from_global) {
-			if (!btrfs_check_space_for_delayed_refs(trans, fs_info))
-				ret = btrfs_block_rsv_migrate(global_rsv, rsv,
-							      min_size, 0);
-			else
-				ret = -ENOSPC;
-		}
-
-		/*
-		 * Couldn't steal from the global reserve, we have too much
-		 * pending stuff built up, commit the transaction and try it
-		 * again.
-		 */
-		if (ret) {
-			ret = btrfs_commit_transaction(trans);
-			if (ret) {
-				btrfs_orphan_del(NULL, BTRFS_I(inode));
-				btrfs_free_block_rsv(fs_info, rsv);
-				goto no_delete;
-			}
-			continue;
-		} else {
-			steal_from_global = 0;
-		}
+		trans = evict_refill_and_join(root, rsv, min_size);
+		if (IS_ERR(trans))
+			goto free_rsv;
 
 		trans->block_rsv = rsv;
 
 		ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0);
-		if (ret != -ENOSPC && ret != -EAGAIN)
-			break;
-
 		trans->block_rsv = &fs_info->trans_block_rsv;
 		btrfs_end_transaction(trans);
-		trans = NULL;
 		btrfs_btree_balance_dirty(fs_info);
+		if (ret && ret != -ENOSPC && ret != -EAGAIN)
+			goto free_rsv;
+		else if (!ret)
+			break;
 	}
 
-	btrfs_free_block_rsv(fs_info, rsv);
-
 	/*
-	 * Errors here aren't a big deal, it just means we leave orphan items
-	 * in the tree.  They will be cleaned up on the next mount.
+	 * Errors here aren't a big deal, it just means we leave orphan items in
+	 * the tree. They will be cleaned up on the next mount. If the inode
+	 * number gets reused, cleanup deletes the orphan item without doing
+	 * anything, and unlink reuses the existing orphan item.
+	 *
+	 * If it turns out that we are dropping too many of these, we might want
+	 * to add a mechanism for retrying these after a commit.
 	 */
-	if (ret == 0) {
-		trans->block_rsv = root->orphan_block_rsv;
+	trans = evict_refill_and_join(root, rsv, min_size);
+	if (!IS_ERR(trans)) {
+		trans->block_rsv = rsv;
 		btrfs_orphan_del(trans, BTRFS_I(inode));
-	} else {
-		btrfs_orphan_del(NULL, BTRFS_I(inode));
+		trans->block_rsv = &fs_info->trans_block_rsv;
+		btrfs_end_transaction(trans);
 	}
 
-	trans->block_rsv = &fs_info->trans_block_rsv;
 	if (!(root == fs_info->tree_root ||
 	      root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID))
 		btrfs_return_ino(root, btrfs_ino(BTRFS_I(inode)));
 
-	btrfs_end_transaction(trans);
-	btrfs_btree_balance_dirty(fs_info);
+free_rsv:
+	btrfs_free_block_rsv(fs_info, rsv);
 no_delete:
+	/*
+	 * If we didn't successfully delete, the orphan item will still be in
+	 * the tree and we'll retry on the next mount. Again, we might also want
+	 * to retry these periodically in the future.
+	 */
 	btrfs_remove_delayed_node(BTRFS_I(inode));
 	clear_inode(inode);
 }
@@ -5612,84 +5643,21 @@ static void inode_tree_del(struct inode *inode)
 	if (!RB_EMPTY_NODE(&BTRFS_I(inode)->rb_node)) {
 		rb_erase(&BTRFS_I(inode)->rb_node, &root->inode_tree);
 		RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node);
-		empty = RB_EMPTY_ROOT(&root->inode_tree);
-	}
-	spin_unlock(&root->inode_lock);
-
-	if (empty && btrfs_root_refs(&root->root_item) == 0) {
-		synchronize_srcu(&fs_info->subvol_srcu);
-		spin_lock(&root->inode_lock);
-		empty = RB_EMPTY_ROOT(&root->inode_tree);
-		spin_unlock(&root->inode_lock);
-		if (empty)
-			btrfs_add_dead_root(root);
-	}
-}
-
-void btrfs_invalidate_inodes(struct btrfs_root *root)
-{
-	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct rb_node *node;
-	struct rb_node *prev;
-	struct btrfs_inode *entry;
-	struct inode *inode;
-	u64 objectid = 0;
-
-	if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
-		WARN_ON(btrfs_root_refs(&root->root_item) != 0);
-
-	spin_lock(&root->inode_lock);
-again:
-	node = root->inode_tree.rb_node;
-	prev = NULL;
-	while (node) {
-		prev = node;
-		entry = rb_entry(node, struct btrfs_inode, rb_node);
-
-		if (objectid < btrfs_ino(BTRFS_I(&entry->vfs_inode)))
-			node = node->rb_left;
-		else if (objectid > btrfs_ino(BTRFS_I(&entry->vfs_inode)))
-			node = node->rb_right;
-		else
-			break;
-	}
-	if (!node) {
-		while (prev) {
-			entry = rb_entry(prev, struct btrfs_inode, rb_node);
-			if (objectid <= btrfs_ino(BTRFS_I(&entry->vfs_inode))) {
-				node = prev;
-				break;
-			}
-			prev = rb_next(prev);
-		}
-	}
-	while (node) {
-		entry = rb_entry(node, struct btrfs_inode, rb_node);
-		objectid = btrfs_ino(BTRFS_I(&entry->vfs_inode)) + 1;
-		inode = igrab(&entry->vfs_inode);
-		if (inode) {
-			spin_unlock(&root->inode_lock);
-			if (atomic_read(&inode->i_count) > 1)
-				d_prune_aliases(inode);
-			/*
-			 * btrfs_drop_inode will have it removed from
-			 * the inode cache when its usage count
-			 * hits zero.
-			 */
-			iput(inode);
-			cond_resched();
-			spin_lock(&root->inode_lock);
-			goto again;
-		}
-
-		if (cond_resched_lock(&root->inode_lock))
-			goto again;
-
-		node = rb_next(node);
+		empty = RB_EMPTY_ROOT(&root->inode_tree);
 	}
 	spin_unlock(&root->inode_lock);
+
+	if (empty && btrfs_root_refs(&root->root_item) == 0) {
+		synchronize_srcu(&fs_info->subvol_srcu);
+		spin_lock(&root->inode_lock);
+		empty = RB_EMPTY_ROOT(&root->inode_tree);
+		spin_unlock(&root->inode_lock);
+		if (empty)
+			btrfs_add_dead_root(root);
+	}
 }
 
+
 static int btrfs_init_locked_inode(struct inode *inode, void *p)
 {
 	struct btrfs_iget_args *args = p;
@@ -5850,11 +5818,6 @@ static int btrfs_dentry_delete(const struct dentry *dentry)
 	return 0;
 }
 
-static void btrfs_dentry_release(struct dentry *dentry)
-{
-	kfree(dentry->d_fsdata);
-}
-
 static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry,
 				   unsigned int flags)
 {
@@ -6270,7 +6233,7 @@ static void btrfs_inherit_iflags(struct inode *inode, struct inode *dir)
 			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM;
 	}
 
-	btrfs_update_iflags(inode);
+	btrfs_sync_inode_flags_to_i_flags(inode);
 }
 
 static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
@@ -6705,8 +6668,9 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	 * 2 items for inode and inode ref
 	 * 2 items for dir items
 	 * 1 item for parent inode
+	 * 1 item for orphan item deletion if O_TMPFILE
 	 */
-	trans = btrfs_start_transaction(root, 5);
+	trans = btrfs_start_transaction(root, inode->i_nlink ? 5 : 6);
 	if (IS_ERR(trans)) {
 		err = PTR_ERR(trans);
 		trans = NULL;
@@ -7083,7 +7047,7 @@ insert:
 
 	err = 0;
 	write_lock(&em_tree->lock);
-	err = btrfs_add_extent_mapping(em_tree, &em, start, len);
+	err = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len);
 	write_unlock(&em_tree->lock);
 out:
 
@@ -7368,6 +7332,14 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
 	    btrfs_file_extent_other_encoding(leaf, fi))
 		goto out;
 
+	/*
+	 * Do the same check as in btrfs_cross_ref_exist but without the
+	 * unnecessary search.
+	 */
+	if (btrfs_file_extent_generation(leaf, fi) <=
+	    btrfs_root_last_snapshot(&root->root_item))
+		goto out;
+
 	backref_offset = btrfs_file_extent_offset(leaf, fi);
 
 	if (orig_start) {
@@ -7568,6 +7540,125 @@ static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len,
 	return em;
 }
 
+
+static int btrfs_get_blocks_direct_read(struct extent_map *em,
+					struct buffer_head *bh_result,
+					struct inode *inode,
+					u64 start, u64 len)
+{
+	if (em->block_start == EXTENT_MAP_HOLE ||
+			test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
+		return -ENOENT;
+
+	len = min(len, em->len - (start - em->start));
+
+	bh_result->b_blocknr = (em->block_start + (start - em->start)) >>
+		inode->i_blkbits;
+	bh_result->b_size = len;
+	bh_result->b_bdev = em->bdev;
+	set_buffer_mapped(bh_result);
+
+	return 0;
+}
+
+static int btrfs_get_blocks_direct_write(struct extent_map **map,
+					 struct buffer_head *bh_result,
+					 struct inode *inode,
+					 struct btrfs_dio_data *dio_data,
+					 u64 start, u64 len)
+{
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct extent_map *em = *map;
+	int ret = 0;
+
+	/*
+	 * We don't allocate a new extent in the following cases
+	 *
+	 * 1) The inode is marked as NODATACOW. In this case we'll just use the
+	 * existing extent.
+	 * 2) The extent is marked as PREALLOC. We're good to go here and can
+	 * just use the extent.
+	 *
+	 */
+	if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) ||
+	    ((BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) &&
+	     em->block_start != EXTENT_MAP_HOLE)) {
+		int type;
+		u64 block_start, orig_start, orig_block_len, ram_bytes;
+
+		if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
+			type = BTRFS_ORDERED_PREALLOC;
+		else
+			type = BTRFS_ORDERED_NOCOW;
+		len = min(len, em->len - (start - em->start));
+		block_start = em->block_start + (start - em->start);
+
+		if (can_nocow_extent(inode, start, &len, &orig_start,
+				     &orig_block_len, &ram_bytes) == 1 &&
+		    btrfs_inc_nocow_writers(fs_info, block_start)) {
+			struct extent_map *em2;
+
+			em2 = btrfs_create_dio_extent(inode, start, len,
+						      orig_start, block_start,
+						      len, orig_block_len,
+						      ram_bytes, type);
+			btrfs_dec_nocow_writers(fs_info, block_start);
+			if (type == BTRFS_ORDERED_PREALLOC) {
+				free_extent_map(em);
+				*map = em = em2;
+			}
+
+			if (em2 && IS_ERR(em2)) {
+				ret = PTR_ERR(em2);
+				goto out;
+			}
+			/*
+			 * For inode marked NODATACOW or extent marked PREALLOC,
+			 * use the existing or preallocated extent, so does not
+			 * need to adjust btrfs_space_info's bytes_may_use.
+			 */
+			btrfs_free_reserved_data_space_noquota(inode, start,
+							       len);
+			goto skip_cow;
+		}
+	}
+
+	/* this will cow the extent */
+	len = bh_result->b_size;
+	free_extent_map(em);
+	*map = em = btrfs_new_extent_direct(inode, start, len);
+	if (IS_ERR(em)) {
+		ret = PTR_ERR(em);
+		goto out;
+	}
+
+	len = min(len, em->len - (start - em->start));
+
+skip_cow:
+	bh_result->b_blocknr = (em->block_start + (start - em->start)) >>
+		inode->i_blkbits;
+	bh_result->b_size = len;
+	bh_result->b_bdev = em->bdev;
+	set_buffer_mapped(bh_result);
+
+	if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
+		set_buffer_new(bh_result);
+
+	/*
+	 * Need to update the i_size under the extent lock so buffered
+	 * readers will get the updated i_size when we unlock.
+	 */
+	if (!dio_data->overwrite && start + len > i_size_read(inode))
+		i_size_write(inode, start + len);
+
+	WARN_ON(dio_data->reserve < len);
+	dio_data->reserve -= len;
+	dio_data->unsubmitted_oe_range_end = start + len;
+	current->journal_info = dio_data;
+out:
+	return ret;
+}
+
 static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
 				   struct buffer_head *bh_result, int create)
 {
@@ -7636,116 +7727,36 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
 		goto unlock_err;
 	}
 
-	/* Just a good old fashioned hole, return */
-	if (!create && (em->block_start == EXTENT_MAP_HOLE ||
-			test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) {
-		free_extent_map(em);
-		goto unlock_err;
-	}
-
-	/*
-	 * We don't allocate a new extent in the following cases
-	 *
-	 * 1) The inode is marked as NODATACOW.  In this case we'll just use the
-	 * existing extent.
-	 * 2) The extent is marked as PREALLOC.  We're good to go here and can
-	 * just use the extent.
-	 *
-	 */
-	if (!create) {
-		len = min(len, em->len - (start - em->start));
-		lockstart = start + len;
-		goto unlock;
-	}
-
-	if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) ||
-	    ((BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) &&
-	     em->block_start != EXTENT_MAP_HOLE)) {
-		int type;
-		u64 block_start, orig_start, orig_block_len, ram_bytes;
-
-		if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
-			type = BTRFS_ORDERED_PREALLOC;
-		else
-			type = BTRFS_ORDERED_NOCOW;
-		len = min(len, em->len - (start - em->start));
-		block_start = em->block_start + (start - em->start);
-
-		if (can_nocow_extent(inode, start, &len, &orig_start,
-				     &orig_block_len, &ram_bytes) == 1 &&
-		    btrfs_inc_nocow_writers(fs_info, block_start)) {
-			struct extent_map *em2;
-
-			em2 = btrfs_create_dio_extent(inode, start, len,
-						      orig_start, block_start,
-						      len, orig_block_len,
-						      ram_bytes, type);
-			btrfs_dec_nocow_writers(fs_info, block_start);
-			if (type == BTRFS_ORDERED_PREALLOC) {
-				free_extent_map(em);
-				em = em2;
-			}
-			if (em2 && IS_ERR(em2)) {
-				ret = PTR_ERR(em2);
-				goto unlock_err;
-			}
-			/*
-			 * For inode marked NODATACOW or extent marked PREALLOC,
-			 * use the existing or preallocated extent, so does not
-			 * need to adjust btrfs_space_info's bytes_may_use.
-			 */
-			btrfs_free_reserved_data_space_noquota(inode,
-					start, len);
-			goto unlock;
-		}
-	}
-
-	/*
-	 * this will cow the extent, reset the len in case we changed
-	 * it above
-	 */
-	len = bh_result->b_size;
-	free_extent_map(em);
-	em = btrfs_new_extent_direct(inode, start, len);
-	if (IS_ERR(em)) {
-		ret = PTR_ERR(em);
-		goto unlock_err;
-	}
-	len = min(len, em->len - (start - em->start));
-unlock:
-	bh_result->b_blocknr = (em->block_start + (start - em->start)) >>
-		inode->i_blkbits;
-	bh_result->b_size = len;
-	bh_result->b_bdev = em->bdev;
-	set_buffer_mapped(bh_result);
 	if (create) {
-		if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
-			set_buffer_new(bh_result);
+		ret = btrfs_get_blocks_direct_write(&em, bh_result, inode,
+						    dio_data, start, len);
+		if (ret < 0)
+			goto unlock_err;
 
+		/* clear and unlock the entire range */
+		clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, lockend,
+				 unlock_bits, 1, 0, &cached_state);
+	} else {
+		ret = btrfs_get_blocks_direct_read(em, bh_result, inode,
+						   start, len);
+		/* Can be negative only if we read from a hole */
+		if (ret < 0) {
+			ret = 0;
+			free_extent_map(em);
+			goto unlock_err;
+		}
 		/*
-		 * Need to update the i_size under the extent lock so buffered
-		 * readers will get the updated i_size when we unlock.
+		 * We need to unlock only the end area that we aren't using.
+		 * The rest is going to be unlocked by the endio routine.
 		 */
-		if (!dio_data->overwrite && start + len > i_size_read(inode))
-			i_size_write(inode, start + len);
-
-		WARN_ON(dio_data->reserve < len);
-		dio_data->reserve -= len;
-		dio_data->unsubmitted_oe_range_end = start + len;
-		current->journal_info = dio_data;
-	}
-
-	/*
-	 * In the case of write we need to clear and unlock the entire range,
-	 * in the case of read we need to unlock only the end area that we
-	 * aren't using if there is any left over space.
-	 */
-	if (lockstart < lockend) {
-		clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
-				 lockend, unlock_bits, 1, 0,
-				 &cached_state);
-	} else {
-		free_extent_state(cached_state);
+		lockstart = start + bh_result->b_size;
+		if (lockstart < lockend) {
+			clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
+					 lockend, unlock_bits, 1, 0,
+					 &cached_state);
+		} else {
+			free_extent_state(cached_state);
+		}
 	}
 
 	free_extent_map(em);
@@ -8131,7 +8142,6 @@ static void __endio_write_update_ordered(struct inode *inode,
 	u64 ordered_offset = offset;
 	u64 ordered_bytes = bytes;
 	u64 last_offset;
-	int ret;
 
 	if (btrfs_is_free_space_inode(BTRFS_I(inode))) {
 		wq = fs_info->endio_freespace_worker;
@@ -8141,32 +8151,31 @@ static void __endio_write_update_ordered(struct inode *inode,
 		func = btrfs_endio_write_helper;
 	}
 
-again:
-	last_offset = ordered_offset;
-	ret = btrfs_dec_test_first_ordered_pending(inode, &ordered,
-						   &ordered_offset,
-						   ordered_bytes,
-						   uptodate);
-	if (!ret)
-		goto out_test;
-
-	btrfs_init_work(&ordered->work, func, finish_ordered_fn, NULL, NULL);
-	btrfs_queue_work(wq, &ordered->work);
-out_test:
-	/*
-	 * If btrfs_dec_test_ordered_pending does not find any ordered extent
-	 * in the range, we can exit.
-	 */
-	if (ordered_offset == last_offset)
-		return;
-	/*
-	 * our bio might span multiple ordered extents.  If we haven't
-	 * completed the accounting for the whole dio, go back and try again
-	 */
-	if (ordered_offset < offset + bytes) {
-		ordered_bytes = offset + bytes - ordered_offset;
-		ordered = NULL;
-		goto again;
+	while (ordered_offset < offset + bytes) {
+		last_offset = ordered_offset;
+		if (btrfs_dec_test_first_ordered_pending(inode, &ordered,
+							   &ordered_offset,
+							   ordered_bytes,
+							   uptodate)) {
+			btrfs_init_work(&ordered->work, func,
+					finish_ordered_fn,
+					NULL, NULL);
+			btrfs_queue_work(wq, &ordered->work);
+		}
+		/*
+		 * If btrfs_dec_test_ordered_pending does not find any ordered
+		 * extent in the range, we can exit.
+		 */
+		if (ordered_offset == last_offset)
+			return;
+		/*
+		 * Our bio might span multiple ordered extents. In this case
+		 * we keep goin until we have accounted the whole dio.
+		 */
+		if (ordered_offset < offset + bytes) {
+			ordered_bytes = offset + bytes - ordered_offset;
+			ordered = NULL;
+		}
 	}
 }
 
@@ -8705,29 +8714,19 @@ static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
 static int btrfs_writepages(struct address_space *mapping,
 			    struct writeback_control *wbc)
 {
-	struct extent_io_tree *tree;
-
-	tree = &BTRFS_I(mapping->host)->io_tree;
-	return extent_writepages(tree, mapping, wbc);
+	return extent_writepages(mapping, wbc);
 }
 
 static int
 btrfs_readpages(struct file *file, struct address_space *mapping,
 		struct list_head *pages, unsigned nr_pages)
 {
-	struct extent_io_tree *tree;
-	tree = &BTRFS_I(mapping->host)->io_tree;
-	return extent_readpages(tree, mapping, pages, nr_pages);
+	return extent_readpages(mapping, pages, nr_pages);
 }
+
 static int __btrfs_releasepage(struct page *page, gfp_t gfp_flags)
 {
-	struct extent_io_tree *tree;
-	struct extent_map_tree *map;
-	int ret;
-
-	tree = &BTRFS_I(page->mapping->host)->io_tree;
-	map = &BTRFS_I(page->mapping->host)->extent_tree;
-	ret = try_release_extent_mapping(map, tree, page, gfp_flags);
+	int ret = try_release_extent_mapping(page, gfp_flags);
 	if (ret == 1) {
 		ClearPagePrivate(page);
 		set_page_private(page, 0);
@@ -8868,8 +8867,8 @@ again:
  *
  * We are not allowed to take the i_mutex here so we have to play games to
  * protect against truncate races as the page could now be beyond EOF.  Because
- * vmtruncate() writes the inode size before removing pages, once we have the
- * page lock we can determine safely if the page is beyond EOF. If it is not
+ * truncate_setsize() writes the inode size before removing pages, once we have
+ * the page lock we can determine safely if the page is beyond EOF. If it is not
  * beyond EOF, then the page is guaranteed safe against truncation until we
  * unlock the page.
  */
@@ -9031,8 +9030,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_block_rsv *rsv;
-	int ret = 0;
-	int err = 0;
+	int ret;
 	struct btrfs_trans_handle *trans;
 	u64 mask = fs_info->sectorsize - 1;
 	u64 min_size = btrfs_calc_trunc_metadata_size(fs_info, 1);
@@ -9045,39 +9043,31 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 	}
 
 	/*
-	 * Yes ladies and gentlemen, this is indeed ugly.  The fact is we have
-	 * 3 things going on here
+	 * Yes ladies and gentlemen, this is indeed ugly.  We have a couple of
+	 * things going on here:
 	 *
-	 * 1) We need to reserve space for our orphan item and the space to
-	 * delete our orphan item.  Lord knows we don't want to have a dangling
-	 * orphan item because we didn't reserve space to remove it.
+	 * 1) We need to reserve space to update our inode.
 	 *
-	 * 2) We need to reserve space to update our inode.
-	 *
-	 * 3) We need to have something to cache all the space that is going to
+	 * 2) We need to have something to cache all the space that is going to
 	 * be free'd up by the truncate operation, but also have some slack
 	 * space reserved in case it uses space during the truncate (thank you
 	 * very much snapshotting).
 	 *
-	 * And we need these to all be separate.  The fact is we can use a lot of
+	 * And we need these to be separate.  The fact is we can use a lot of
 	 * space doing the truncate, and we have no earthly idea how much space
 	 * we will use, so we need the truncate reservation to be separate so it
-	 * doesn't end up using space reserved for updating the inode or
-	 * removing the orphan item.  We also need to be able to stop the
-	 * transaction and start a new one, which means we need to be able to
-	 * update the inode several times, and we have no idea of knowing how
-	 * many times that will be, so we can't just reserve 1 item for the
-	 * entirety of the operation, so that has to be done separately as well.
-	 * Then there is the orphan item, which does indeed need to be held on
-	 * to for the whole operation, and we need nobody to touch this reserved
-	 * space except the orphan code.
+	 * doesn't end up using space reserved for updating the inode.  We also
+	 * need to be able to stop the transaction and start a new one, which
+	 * means we need to be able to update the inode several times, and we
+	 * have no idea of knowing how many times that will be, so we can't just
+	 * reserve 1 item for the entirety of the operation, so that has to be
+	 * done separately as well.
 	 *
 	 * So that leaves us with
 	 *
-	 * 1) root->orphan_block_rsv - for the orphan deletion.
-	 * 2) rsv - for the truncate reservation, which we will steal from the
+	 * 1) rsv - for the truncate reservation, which we will steal from the
 	 * transaction reservation.
-	 * 3) fs_info->trans_block_rsv - this will have 1 items worth left for
+	 * 2) fs_info->trans_block_rsv - this will have 1 items worth left for
 	 * updating the inode.
 	 */
 	rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP);
@@ -9092,7 +9082,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 	 */
 	trans = btrfs_start_transaction(root, 2);
 	if (IS_ERR(trans)) {
-		err = PTR_ERR(trans);
+		ret = PTR_ERR(trans);
 		goto out;
 	}
 
@@ -9116,24 +9106,19 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 						 inode->i_size,
 						 BTRFS_EXTENT_DATA_KEY);
 		trans->block_rsv = &fs_info->trans_block_rsv;
-		if (ret != -ENOSPC && ret != -EAGAIN) {
-			if (ret < 0)
-				err = ret;
+		if (ret != -ENOSPC && ret != -EAGAIN)
 			break;
-		}
 
 		ret = btrfs_update_inode(trans, root, inode);
-		if (ret) {
-			err = ret;
+		if (ret)
 			break;
-		}
 
 		btrfs_end_transaction(trans);
 		btrfs_btree_balance_dirty(fs_info);
 
 		trans = btrfs_start_transaction(root, 2);
 		if (IS_ERR(trans)) {
-			ret = err = PTR_ERR(trans);
+			ret = PTR_ERR(trans);
 			trans = NULL;
 			break;
 		}
@@ -9166,29 +9151,23 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 		btrfs_ordered_update_i_size(inode, inode->i_size, NULL);
 	}
 
-	if (ret == 0 && inode->i_nlink > 0) {
-		trans->block_rsv = root->orphan_block_rsv;
-		ret = btrfs_orphan_del(trans, BTRFS_I(inode));
-		if (ret)
-			err = ret;
-	}
-
 	if (trans) {
+		int ret2;
+
 		trans->block_rsv = &fs_info->trans_block_rsv;
-		ret = btrfs_update_inode(trans, root, inode);
-		if (ret && !err)
-			err = ret;
+		ret2 = btrfs_update_inode(trans, root, inode);
+		if (ret2 && !ret)
+			ret = ret2;
 
-		ret = btrfs_end_transaction(trans);
+		ret2 = btrfs_end_transaction(trans);
+		if (ret2 && !ret)
+			ret = ret2;
 		btrfs_btree_balance_dirty(fs_info);
 	}
 out:
 	btrfs_free_block_rsv(fs_info, rsv);
 
-	if (ret && !err)
-		err = ret;
-
-	return err;
+	return ret;
 }
 
 /*
@@ -9324,13 +9303,6 @@ void btrfs_destroy_inode(struct inode *inode)
 	if (!root)
 		goto free;
 
-	if (test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
-		     &BTRFS_I(inode)->runtime_flags)) {
-		btrfs_info(fs_info, "inode %llu still on the orphan list",
-			   btrfs_ino(BTRFS_I(inode)));
-		atomic_dec(&root->orphan_inodes);
-	}
-
 	while (1) {
 		ordered = btrfs_lookup_first_ordered_extent(inode, (u64)-1);
 		if (!ordered)
@@ -9964,6 +9936,13 @@ static int btrfs_rename2(struct inode *old_dir, struct dentry *old_dentry,
 	return btrfs_rename(old_dir, old_dentry, new_dir, new_dentry, flags);
 }
 
+struct btrfs_delalloc_work {
+	struct inode *inode;
+	struct completion completion;
+	struct list_head list;
+	struct btrfs_work work;
+};
+
 static void btrfs_run_delalloc_work(struct btrfs_work *work)
 {
 	struct btrfs_delalloc_work *delalloc_work;
@@ -9977,15 +9956,11 @@ static void btrfs_run_delalloc_work(struct btrfs_work *work)
 				&BTRFS_I(inode)->runtime_flags))
 		filemap_flush(inode->i_mapping);
 
-	if (delalloc_work->delay_iput)
-		btrfs_add_delayed_iput(inode);
-	else
-		iput(inode);
+	iput(inode);
 	complete(&delalloc_work->completion);
 }
 
-struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
-						    int delay_iput)
+static struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode)
 {
 	struct btrfs_delalloc_work *work;
 
@@ -9996,7 +9971,6 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
 	init_completion(&work->completion);
 	INIT_LIST_HEAD(&work->list);
 	work->inode = inode;
-	work->delay_iput = delay_iput;
 	WARN_ON_ONCE(!inode);
 	btrfs_init_work(&work->work, btrfs_flush_delalloc_helper,
 			btrfs_run_delalloc_work, NULL, NULL);
@@ -10004,18 +9978,11 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
 	return work;
 }
 
-void btrfs_wait_and_free_delalloc_work(struct btrfs_delalloc_work *work)
-{
-	wait_for_completion(&work->completion);
-	kfree(work);
-}
-
 /*
  * some fairly slow code that needs optimization. This walks the list
  * of all the inodes with pending delalloc and forces them to disk.
  */
-static int __start_delalloc_inodes(struct btrfs_root *root, int delay_iput,
-				   int nr)
+static int start_delalloc_inodes(struct btrfs_root *root, int nr)
 {
 	struct btrfs_inode *binode;
 	struct inode *inode;
@@ -10043,12 +10010,9 @@ static int __start_delalloc_inodes(struct btrfs_root *root, int delay_iput,
 		}
 		spin_unlock(&root->delalloc_lock);
 
-		work = btrfs_alloc_delalloc_work(inode, delay_iput);
+		work = btrfs_alloc_delalloc_work(inode);
 		if (!work) {
-			if (delay_iput)
-				btrfs_add_delayed_iput(inode);
-			else
-				iput(inode);
+			iput(inode);
 			ret = -ENOMEM;
 			goto out;
 		}
@@ -10066,10 +10030,11 @@ static int __start_delalloc_inodes(struct btrfs_root *root, int delay_iput,
 out:
 	list_for_each_entry_safe(work, next, &works, list) {
 		list_del_init(&work->list);
-		btrfs_wait_and_free_delalloc_work(work);
+		wait_for_completion(&work->completion);
+		kfree(work);
 	}
 
-	if (!list_empty_careful(&splice)) {
+	if (!list_empty(&splice)) {
 		spin_lock(&root->delalloc_lock);
 		list_splice_tail(&splice, &root->delalloc_inodes);
 		spin_unlock(&root->delalloc_lock);
@@ -10078,7 +10043,7 @@ out:
 	return ret;
 }
 
-int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
+int btrfs_start_delalloc_inodes(struct btrfs_root *root)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret;
@@ -10086,14 +10051,13 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
 	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
 		return -EROFS;
 
-	ret = __start_delalloc_inodes(root, delay_iput, -1);
+	ret = start_delalloc_inodes(root, -1);
 	if (ret > 0)
 		ret = 0;
 	return ret;
 }
 
-int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput,
-			       int nr)
+int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr)
 {
 	struct btrfs_root *root;
 	struct list_head splice;
@@ -10116,7 +10080,7 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput,
 			       &fs_info->delalloc_roots);
 		spin_unlock(&fs_info->delalloc_root_lock);
 
-		ret = __start_delalloc_inodes(root, delay_iput, nr);
+		ret = start_delalloc_inodes(root, nr);
 		btrfs_put_fs_root(root);
 		if (ret < 0)
 			goto out;
@@ -10131,7 +10095,7 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput,
 
 	ret = 0;
 out:
-	if (!list_empty_careful(&splice)) {
+	if (!list_empty(&splice)) {
 		spin_lock(&fs_info->delalloc_root_lock);
 		list_splice_tail(&splice, &fs_info->delalloc_roots);
 		spin_unlock(&fs_info->delalloc_root_lock);
@@ -10669,5 +10633,4 @@ static const struct inode_operations btrfs_symlink_inode_operations = {
 
 const struct dentry_operations btrfs_dentry_operations = {
 	.d_delete	= btrfs_dentry_delete,
-	.d_release	= btrfs_dentry_release,
 };

fs/btrfs/ioctl.c

@@ -93,20 +93,22 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 		       int no_time_update);
 
 /* Mask out flags that are inappropriate for the given type of inode. */
-static unsigned int btrfs_mask_flags(umode_t mode, unsigned int flags)
+static unsigned int btrfs_mask_fsflags_for_type(struct inode *inode,
+		unsigned int flags)
 {
-	if (S_ISDIR(mode))
+	if (S_ISDIR(inode->i_mode))
 		return flags;
-	else if (S_ISREG(mode))
+	else if (S_ISREG(inode->i_mode))
 		return flags & ~FS_DIRSYNC_FL;
 	else
 		return flags & (FS_NODUMP_FL | FS_NOATIME_FL);
 }
 
 /*
- * Export inode flags to the format expected by the FS_IOC_GETFLAGS ioctl.
+ * Export internal inode flags to the format expected by the FS_IOC_GETFLAGS
+ * ioctl.
  */
-static unsigned int btrfs_flags_to_ioctl(unsigned int flags)
+static unsigned int btrfs_inode_flags_to_fsflags(unsigned int flags)
 {
 	unsigned int iflags = 0;
 
@@ -136,20 +138,20 @@ static unsigned int btrfs_flags_to_ioctl(unsigned int flags)
 /*
  * Update inode->i_flags based on the btrfs internal flags.
  */
-void btrfs_update_iflags(struct inode *inode)
+void btrfs_sync_inode_flags_to_i_flags(struct inode *inode)
 {
-	struct btrfs_inode *ip = BTRFS_I(inode);
+	struct btrfs_inode *binode = BTRFS_I(inode);
 	unsigned int new_fl = 0;
 
-	if (ip->flags & BTRFS_INODE_SYNC)
+	if (binode->flags & BTRFS_INODE_SYNC)
 		new_fl |= S_SYNC;
-	if (ip->flags & BTRFS_INODE_IMMUTABLE)
+	if (binode->flags & BTRFS_INODE_IMMUTABLE)
 		new_fl |= S_IMMUTABLE;
-	if (ip->flags & BTRFS_INODE_APPEND)
+	if (binode->flags & BTRFS_INODE_APPEND)
 		new_fl |= S_APPEND;
-	if (ip->flags & BTRFS_INODE_NOATIME)
+	if (binode->flags & BTRFS_INODE_NOATIME)
 		new_fl |= S_NOATIME;
-	if (ip->flags & BTRFS_INODE_DIRSYNC)
+	if (binode->flags & BTRFS_INODE_DIRSYNC)
 		new_fl |= S_DIRSYNC;
 
 	set_mask_bits(&inode->i_flags,
@@ -159,15 +161,16 @@ void btrfs_update_iflags(struct inode *inode)
 
 static int btrfs_ioctl_getflags(struct file *file, void __user *arg)
 {
-	struct btrfs_inode *ip = BTRFS_I(file_inode(file));
-	unsigned int flags = btrfs_flags_to_ioctl(ip->flags);
+	struct btrfs_inode *binode = BTRFS_I(file_inode(file));
+	unsigned int flags = btrfs_inode_flags_to_fsflags(binode->flags);
 
 	if (copy_to_user(arg, &flags, sizeof(flags)))
 		return -EFAULT;
 	return 0;
 }
 
-static int check_flags(unsigned int flags)
+/* Check if @flags are a supported and valid set of FS_*_FL flags */
+static int check_fsflags(unsigned int flags)
 {
 	if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \
 		      FS_NOATIME_FL | FS_NODUMP_FL | \
@@ -186,13 +189,13 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 {
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct btrfs_inode *ip = BTRFS_I(inode);
-	struct btrfs_root *root = ip->root;
+	struct btrfs_inode *binode = BTRFS_I(inode);
+	struct btrfs_root *root = binode->root;
 	struct btrfs_trans_handle *trans;
-	unsigned int flags, oldflags;
+	unsigned int fsflags, old_fsflags;
 	int ret;
-	u64 ip_oldflags;
-	unsigned int i_oldflags;
+	u64 old_flags;
+	unsigned int old_i_flags;
 	umode_t mode;
 
 	if (!inode_owner_or_capable(inode))
@@ -201,10 +204,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 	if (btrfs_root_readonly(root))
 		return -EROFS;
 
-	if (copy_from_user(&flags, arg, sizeof(flags)))
+	if (copy_from_user(&fsflags, arg, sizeof(fsflags)))
 		return -EFAULT;
 
-	ret = check_flags(flags);
+	ret = check_fsflags(fsflags);
 	if (ret)
 		return ret;
 
@@ -214,44 +217,44 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 
 	inode_lock(inode);
 
-	ip_oldflags = ip->flags;
-	i_oldflags = inode->i_flags;
+	old_flags = binode->flags;
+	old_i_flags = inode->i_flags;
 	mode = inode->i_mode;
 
-	flags = btrfs_mask_flags(inode->i_mode, flags);
-	oldflags = btrfs_flags_to_ioctl(ip->flags);
-	if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) {
+	fsflags = btrfs_mask_fsflags_for_type(inode, fsflags);
+	old_fsflags = btrfs_inode_flags_to_fsflags(binode->flags);
+	if ((fsflags ^ old_fsflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) {
 		if (!capable(CAP_LINUX_IMMUTABLE)) {
 			ret = -EPERM;
 			goto out_unlock;
 		}
 	}
 
-	if (flags & FS_SYNC_FL)
-		ip->flags |= BTRFS_INODE_SYNC;
+	if (fsflags & FS_SYNC_FL)
+		binode->flags |= BTRFS_INODE_SYNC;
 	else
-		ip->flags &= ~BTRFS_INODE_SYNC;
-	if (flags & FS_IMMUTABLE_FL)
-		ip->flags |= BTRFS_INODE_IMMUTABLE;
+		binode->flags &= ~BTRFS_INODE_SYNC;
+	if (fsflags & FS_IMMUTABLE_FL)
+		binode->flags |= BTRFS_INODE_IMMUTABLE;
 	else
-		ip->flags &= ~BTRFS_INODE_IMMUTABLE;
-	if (flags & FS_APPEND_FL)
-		ip->flags |= BTRFS_INODE_APPEND;
+		binode->flags &= ~BTRFS_INODE_IMMUTABLE;
+	if (fsflags & FS_APPEND_FL)
+		binode->flags |= BTRFS_INODE_APPEND;
 	else
-		ip->flags &= ~BTRFS_INODE_APPEND;
-	if (flags & FS_NODUMP_FL)
-		ip->flags |= BTRFS_INODE_NODUMP;
+		binode->flags &= ~BTRFS_INODE_APPEND;
+	if (fsflags & FS_NODUMP_FL)
+		binode->flags |= BTRFS_INODE_NODUMP;
 	else
-		ip->flags &= ~BTRFS_INODE_NODUMP;
-	if (flags & FS_NOATIME_FL)
-		ip->flags |= BTRFS_INODE_NOATIME;
+		binode->flags &= ~BTRFS_INODE_NODUMP;
+	if (fsflags & FS_NOATIME_FL)
+		binode->flags |= BTRFS_INODE_NOATIME;
 	else
-		ip->flags &= ~BTRFS_INODE_NOATIME;
-	if (flags & FS_DIRSYNC_FL)
-		ip->flags |= BTRFS_INODE_DIRSYNC;
+		binode->flags &= ~BTRFS_INODE_NOATIME;
+	if (fsflags & FS_DIRSYNC_FL)
+		binode->flags |= BTRFS_INODE_DIRSYNC;
 	else
-		ip->flags &= ~BTRFS_INODE_DIRSYNC;
-	if (flags & FS_NOCOW_FL) {
+		binode->flags &= ~BTRFS_INODE_DIRSYNC;
+	if (fsflags & FS_NOCOW_FL) {
 		if (S_ISREG(mode)) {
 			/*
 			 * It's safe to turn csums off here, no extents exist.
@@ -259,10 +262,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 			 * status of the file and will not set it.
 			 */
 			if (inode->i_size == 0)
-				ip->flags |= BTRFS_INODE_NODATACOW
-					   | BTRFS_INODE_NODATASUM;
+				binode->flags |= BTRFS_INODE_NODATACOW
+					      | BTRFS_INODE_NODATASUM;
 		} else {
-			ip->flags |= BTRFS_INODE_NODATACOW;
+			binode->flags |= BTRFS_INODE_NODATACOW;
 		}
 	} else {
 		/*
@@ -270,10 +273,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 		 */
 		if (S_ISREG(mode)) {
 			if (inode->i_size == 0)
-				ip->flags &= ~(BTRFS_INODE_NODATACOW
+				binode->flags &= ~(BTRFS_INODE_NODATACOW
 				             | BTRFS_INODE_NODATASUM);
 		} else {
-			ip->flags &= ~BTRFS_INODE_NODATACOW;
+			binode->flags &= ~BTRFS_INODE_NODATACOW;
 		}
 	}
 
@@ -282,18 +285,18 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 	 * flag may be changed automatically if compression code won't make
 	 * things smaller.
 	 */
-	if (flags & FS_NOCOMP_FL) {
-		ip->flags &= ~BTRFS_INODE_COMPRESS;
-		ip->flags |= BTRFS_INODE_NOCOMPRESS;
+	if (fsflags & FS_NOCOMP_FL) {
+		binode->flags &= ~BTRFS_INODE_COMPRESS;
+		binode->flags |= BTRFS_INODE_NOCOMPRESS;
 
 		ret = btrfs_set_prop(inode, "btrfs.compression", NULL, 0, 0);
 		if (ret && ret != -ENODATA)
 			goto out_drop;
-	} else if (flags & FS_COMPR_FL) {
+	} else if (fsflags & FS_COMPR_FL) {
 		const char *comp;
 
-		ip->flags |= BTRFS_INODE_COMPRESS;
-		ip->flags &= ~BTRFS_INODE_NOCOMPRESS;
+		binode->flags |= BTRFS_INODE_COMPRESS;
+		binode->flags &= ~BTRFS_INODE_NOCOMPRESS;
 
 		comp = btrfs_compress_type2str(fs_info->compress_type);
 		if (!comp || comp[0] == 0)
@@ -308,7 +311,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 		ret = btrfs_set_prop(inode, "btrfs.compression", NULL, 0, 0);
 		if (ret && ret != -ENODATA)
 			goto out_drop;
-		ip->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS);
+		binode->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS);
 	}
 
 	trans = btrfs_start_transaction(root, 1);
@@ -317,7 +320,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 		goto out_drop;
 	}
 
-	btrfs_update_iflags(inode);
+	btrfs_sync_inode_flags_to_i_flags(inode);
 	inode_inc_iversion(inode);
 	inode->i_ctime = current_time(inode);
 	ret = btrfs_update_inode(trans, root, inode);
@@ -325,8 +328,8 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 	btrfs_end_transaction(trans);
  out_drop:
 	if (ret) {
-		ip->flags = ip_oldflags;
-		inode->i_flags = i_oldflags;
+		binode->flags = old_flags;
+		inode->i_flags = old_i_flags;
 	}
 
  out_unlock:
@@ -335,6 +338,148 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 	return ret;
 }
 
+/*
+ * Translate btrfs internal inode flags to xflags as expected by the
+ * FS_IOC_FSGETXATT ioctl. Filter only the supported ones, unknown flags are
+ * silently dropped.
+ */
+static unsigned int btrfs_inode_flags_to_xflags(unsigned int flags)
+{
+	unsigned int xflags = 0;
+
+	if (flags & BTRFS_INODE_APPEND)
+		xflags |= FS_XFLAG_APPEND;
+	if (flags & BTRFS_INODE_IMMUTABLE)
+		xflags |= FS_XFLAG_IMMUTABLE;
+	if (flags & BTRFS_INODE_NOATIME)
+		xflags |= FS_XFLAG_NOATIME;
+	if (flags & BTRFS_INODE_NODUMP)
+		xflags |= FS_XFLAG_NODUMP;
+	if (flags & BTRFS_INODE_SYNC)
+		xflags |= FS_XFLAG_SYNC;
+
+	return xflags;
+}
+
+/* Check if @flags are a supported and valid set of FS_XFLAGS_* flags */
+static int check_xflags(unsigned int flags)
+{
+	if (flags & ~(FS_XFLAG_APPEND | FS_XFLAG_IMMUTABLE | FS_XFLAG_NOATIME |
+		      FS_XFLAG_NODUMP | FS_XFLAG_SYNC))
+		return -EOPNOTSUPP;
+	return 0;
+}
+
+/*
+ * Set the xflags from the internal inode flags. The remaining items of fsxattr
+ * are zeroed.
+ */
+static int btrfs_ioctl_fsgetxattr(struct file *file, void __user *arg)
+{
+	struct btrfs_inode *binode = BTRFS_I(file_inode(file));
+	struct fsxattr fa;
+
+	memset(&fa, 0, sizeof(fa));
+	fa.fsx_xflags = btrfs_inode_flags_to_xflags(binode->flags);
+
+	if (copy_to_user(arg, &fa, sizeof(fa)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int btrfs_ioctl_fssetxattr(struct file *file, void __user *arg)
+{
+	struct inode *inode = file_inode(file);
+	struct btrfs_inode *binode = BTRFS_I(inode);
+	struct btrfs_root *root = binode->root;
+	struct btrfs_trans_handle *trans;
+	struct fsxattr fa;
+	unsigned old_flags;
+	unsigned old_i_flags;
+	int ret = 0;
+
+	if (!inode_owner_or_capable(inode))
+		return -EPERM;
+
+	if (btrfs_root_readonly(root))
+		return -EROFS;
+
+	memset(&fa, 0, sizeof(fa));
+	if (copy_from_user(&fa, arg, sizeof(fa)))
+		return -EFAULT;
+
+	ret = check_xflags(fa.fsx_xflags);
+	if (ret)
+		return ret;
+
+	if (fa.fsx_extsize != 0 || fa.fsx_projid != 0 || fa.fsx_cowextsize != 0)
+		return -EOPNOTSUPP;
+
+	ret = mnt_want_write_file(file);
+	if (ret)
+		return ret;
+
+	inode_lock(inode);
+
+	old_flags = binode->flags;
+	old_i_flags = inode->i_flags;
+
+	/* We need the capabilities to change append-only or immutable inode */
+	if (((old_flags & (BTRFS_INODE_APPEND | BTRFS_INODE_IMMUTABLE)) ||
+	     (fa.fsx_xflags & (FS_XFLAG_APPEND | FS_XFLAG_IMMUTABLE))) &&
+	    !capable(CAP_LINUX_IMMUTABLE)) {
+		ret = -EPERM;
+		goto out_unlock;
+	}
+
+	if (fa.fsx_xflags & FS_XFLAG_SYNC)
+		binode->flags |= BTRFS_INODE_SYNC;
+	else
+		binode->flags &= ~BTRFS_INODE_SYNC;
+	if (fa.fsx_xflags & FS_XFLAG_IMMUTABLE)
+		binode->flags |= BTRFS_INODE_IMMUTABLE;
+	else
+		binode->flags &= ~BTRFS_INODE_IMMUTABLE;
+	if (fa.fsx_xflags & FS_XFLAG_APPEND)
+		binode->flags |= BTRFS_INODE_APPEND;
+	else
+		binode->flags &= ~BTRFS_INODE_APPEND;
+	if (fa.fsx_xflags & FS_XFLAG_NODUMP)
+		binode->flags |= BTRFS_INODE_NODUMP;
+	else
+		binode->flags &= ~BTRFS_INODE_NODUMP;
+	if (fa.fsx_xflags & FS_XFLAG_NOATIME)
+		binode->flags |= BTRFS_INODE_NOATIME;
+	else
+		binode->flags &= ~BTRFS_INODE_NOATIME;
+
+	/* 1 item for the inode */
+	trans = btrfs_start_transaction(root, 1);
+	if (IS_ERR(trans)) {
+		ret = PTR_ERR(trans);
+		goto out_unlock;
+	}
+
+	btrfs_sync_inode_flags_to_i_flags(inode);
+	inode_inc_iversion(inode);
+	inode->i_ctime = current_time(inode);
+	ret = btrfs_update_inode(trans, root, inode);
+
+	btrfs_end_transaction(trans);
+
+out_unlock:
+	if (ret) {
+		binode->flags = old_flags;
+		inode->i_flags = old_i_flags;
+	}
+
+	inode_unlock(inode);
+	mnt_drop_write_file(file);
+
+	return ret;
+}
+
 static int btrfs_ioctl_getversion(struct file *file, int __user *arg)
 {
 	struct inode *inode = file_inode(file);
@@ -424,7 +569,6 @@ static noinline int create_subvol(struct inode *dir,
 	u64 objectid;
 	u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
 	u64 index = 0;
-	u64 qgroup_reserved;
 	uuid_le new_uuid;
 
 	root_item = kzalloc(sizeof(*root_item), GFP_KERNEL);
@@ -449,8 +593,7 @@ static noinline int create_subvol(struct inode *dir,
 	 * The same as the snapshot creation, please see the comment
 	 * of create_snapshot().
 	 */
-	ret = btrfs_subvolume_reserve_metadata(root, &block_rsv,
-					       8, &qgroup_reserved, false);
+	ret = btrfs_subvolume_reserve_metadata(root, &block_rsv, 8, false);
 	if (ret)
 		goto fail_free;
 
@@ -573,7 +716,7 @@ static noinline int create_subvol(struct inode *dir,
 				 btrfs_ino(BTRFS_I(dir)), index, name, namelen);
 	BUG_ON(ret);
 
-	ret = btrfs_uuid_tree_add(trans, fs_info, root_item->uuid,
+	ret = btrfs_uuid_tree_add(trans, root_item->uuid,
 				  BTRFS_UUID_KEY_SUBVOL, objectid);
 	if (ret)
 		btrfs_abort_transaction(trans, ret);
@@ -640,7 +783,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	wait_event(root->subv_writers->wait,
 		   percpu_counter_sum(&root->subv_writers->counter) == 0);
 
-	ret = btrfs_start_delalloc_inodes(root, 0);
+	ret = btrfs_start_delalloc_inodes(root);
 	if (ret)
 		goto dec_and_free;
 
@@ -658,7 +801,6 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	 */
 	ret = btrfs_subvolume_reserve_metadata(BTRFS_I(dir)->root,
 					&pending_snapshot->block_rsv, 8,
-					&pending_snapshot->qgroup_reserved,
 					false);
 	if (ret)
 		goto dec_and_free;
@@ -1457,7 +1599,6 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 		return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
 	}
 
-	mutex_lock(&fs_info->volume_mutex);
 	vol_args = memdup_user(arg, sizeof(*vol_args));
 	if (IS_ERR(vol_args)) {
 		ret = PTR_ERR(vol_args);
@@ -1565,7 +1706,6 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 out_free:
 	kfree(vol_args);
 out:
-	mutex_unlock(&fs_info->volume_mutex);
 	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
 	mnt_drop_write_file(file);
 	return ret;
@@ -1832,60 +1972,6 @@ out:
 	return ret;
 }
 
-/*
- * helper to check if the subvolume references other subvolumes
- */
-static noinline int may_destroy_subvol(struct btrfs_root *root)
-{
-	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct btrfs_path *path;
-	struct btrfs_dir_item *di;
-	struct btrfs_key key;
-	u64 dir_id;
-	int ret;
-
-	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
-
-	/* Make sure this root isn't set as the default subvol */
-	dir_id = btrfs_super_root_dir(fs_info->super_copy);
-	di = btrfs_lookup_dir_item(NULL, fs_info->tree_root, path,
-				   dir_id, "default", 7, 0);
-	if (di && !IS_ERR(di)) {
-		btrfs_dir_item_key_to_cpu(path->nodes[0], di, &key);
-		if (key.objectid == root->root_key.objectid) {
-			ret = -EPERM;
-			btrfs_err(fs_info,
-				  "deleting default subvolume %llu is not allowed",
-				  key.objectid);
-			goto out;
-		}
-		btrfs_release_path(path);
-	}
-
-	key.objectid = root->root_key.objectid;
-	key.type = BTRFS_ROOT_REF_KEY;
-	key.offset = (u64)-1;
-
-	ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
-	BUG_ON(ret == 0);
-
-	ret = 0;
-	if (path->slots[0] > 0) {
-		path->slots[0]--;
-		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
-		if (key.objectid == root->root_key.objectid &&
-		    key.type == BTRFS_ROOT_REF_KEY)
-			ret = -ENOTEMPTY;
-	}
-out:
-	btrfs_free_path(path);
-	return ret;
-}
-
 static noinline int key_in_sk(struct btrfs_key *key,
 			      struct btrfs_ioctl_search_key *sk)
 {
@@ -2066,7 +2152,7 @@ static noinline int search_ioctl(struct inode *inode,
 		root = btrfs_read_fs_root_no_name(info, &key);
 		if (IS_ERR(root)) {
 			btrfs_free_path(path);
-			return -ENOENT;
+			return PTR_ERR(root);
 		}
 	}
 
@@ -2200,8 +2286,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 	key.offset = (u64)-1;
 	root = btrfs_read_fs_root_no_name(info, &key);
 	if (IS_ERR(root)) {
-		btrfs_err(info, "could not find root %llu", tree_id);
-		ret = -ENOENT;
+		ret = PTR_ERR(root);
 		goto out;
 	}
 
@@ -2237,64 +2322,482 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 			goto out;
 		}
 
-		*(ptr + len) = '/';
-		read_extent_buffer(l, ptr, (unsigned long)(iref + 1), len);
-
-		if (key.offset == BTRFS_FIRST_FREE_OBJECTID)
-			break;
+		*(ptr + len) = '/';
+		read_extent_buffer(l, ptr, (unsigned long)(iref + 1), len);
+
+		if (key.offset == BTRFS_FIRST_FREE_OBJECTID)
+			break;
+
+		btrfs_release_path(path);
+		key.objectid = key.offset;
+		key.offset = (u64)-1;
+		dirid = key.objectid;
+	}
+	memmove(name, ptr, total_len);
+	name[total_len] = '\0';
+	ret = 0;
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
+static int btrfs_search_path_in_tree_user(struct inode *inode,
+				struct btrfs_ioctl_ino_lookup_user_args *args)
+{
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+	struct super_block *sb = inode->i_sb;
+	struct btrfs_key upper_limit = BTRFS_I(inode)->location;
+	u64 treeid = BTRFS_I(inode)->root->root_key.objectid;
+	u64 dirid = args->dirid;
+	unsigned long item_off;
+	unsigned long item_len;
+	struct btrfs_inode_ref *iref;
+	struct btrfs_root_ref *rref;
+	struct btrfs_root *root;
+	struct btrfs_path *path;
+	struct btrfs_key key, key2;
+	struct extent_buffer *leaf;
+	struct inode *temp_inode;
+	char *ptr;
+	int slot;
+	int len;
+	int total_len = 0;
+	int ret;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	/*
+	 * If the bottom subvolume does not exist directly under upper_limit,
+	 * construct the path in from the bottom up.
+	 */
+	if (dirid != upper_limit.objectid) {
+		ptr = &args->path[BTRFS_INO_LOOKUP_USER_PATH_MAX - 1];
+
+		key.objectid = treeid;
+		key.type = BTRFS_ROOT_ITEM_KEY;
+		key.offset = (u64)-1;
+		root = btrfs_read_fs_root_no_name(fs_info, &key);
+		if (IS_ERR(root)) {
+			ret = PTR_ERR(root);
+			goto out;
+		}
+
+		key.objectid = dirid;
+		key.type = BTRFS_INODE_REF_KEY;
+		key.offset = (u64)-1;
+		while (1) {
+			ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+			if (ret < 0) {
+				goto out;
+			} else if (ret > 0) {
+				ret = btrfs_previous_item(root, path, dirid,
+							  BTRFS_INODE_REF_KEY);
+				if (ret < 0) {
+					goto out;
+				} else if (ret > 0) {
+					ret = -ENOENT;
+					goto out;
+				}
+			}
+
+			leaf = path->nodes[0];
+			slot = path->slots[0];
+			btrfs_item_key_to_cpu(leaf, &key, slot);
+
+			iref = btrfs_item_ptr(leaf, slot, struct btrfs_inode_ref);
+			len = btrfs_inode_ref_name_len(leaf, iref);
+			ptr -= len + 1;
+			total_len += len + 1;
+			if (ptr < args->path) {
+				ret = -ENAMETOOLONG;
+				goto out;
+			}
+
+			*(ptr + len) = '/';
+			read_extent_buffer(leaf, ptr,
+					(unsigned long)(iref + 1), len);
+
+			/* Check the read+exec permission of this directory */
+			ret = btrfs_previous_item(root, path, dirid,
+						  BTRFS_INODE_ITEM_KEY);
+			if (ret < 0) {
+				goto out;
+			} else if (ret > 0) {
+				ret = -ENOENT;
+				goto out;
+			}
+
+			leaf = path->nodes[0];
+			slot = path->slots[0];
+			btrfs_item_key_to_cpu(leaf, &key2, slot);
+			if (key2.objectid != dirid) {
+				ret = -ENOENT;
+				goto out;
+			}
+
+			temp_inode = btrfs_iget(sb, &key2, root, NULL);
+			ret = inode_permission(temp_inode, MAY_READ | MAY_EXEC);
+			iput(temp_inode);
+			if (ret) {
+				ret = -EACCES;
+				goto out;
+			}
+
+			if (key.offset == upper_limit.objectid)
+				break;
+			if (key.objectid == BTRFS_FIRST_FREE_OBJECTID) {
+				ret = -EACCES;
+				goto out;
+			}
+
+			btrfs_release_path(path);
+			key.objectid = key.offset;
+			key.offset = (u64)-1;
+			dirid = key.objectid;
+		}
+
+		memmove(args->path, ptr, total_len);
+		args->path[total_len] = '\0';
+		btrfs_release_path(path);
+	}
+
+	/* Get the bottom subvolume's name from ROOT_REF */
+	root = fs_info->tree_root;
+	key.objectid = treeid;
+	key.type = BTRFS_ROOT_REF_KEY;
+	key.offset = args->treeid;
+	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	if (ret < 0) {
+		goto out;
+	} else if (ret > 0) {
+		ret = -ENOENT;
+		goto out;
+	}
+
+	leaf = path->nodes[0];
+	slot = path->slots[0];
+	btrfs_item_key_to_cpu(leaf, &key, slot);
+
+	item_off = btrfs_item_ptr_offset(leaf, slot);
+	item_len = btrfs_item_size_nr(leaf, slot);
+	/* Check if dirid in ROOT_REF corresponds to passed dirid */
+	rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref);
+	if (args->dirid != btrfs_root_ref_dirid(leaf, rref)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/* Copy subvolume's name */
+	item_off += sizeof(struct btrfs_root_ref);
+	item_len -= sizeof(struct btrfs_root_ref);
+	read_extent_buffer(leaf, args->name, item_off, item_len);
+	args->name[item_len] = 0;
+
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
+static noinline int btrfs_ioctl_ino_lookup(struct file *file,
+					   void __user *argp)
+{
+	 struct btrfs_ioctl_ino_lookup_args *args;
+	 struct inode *inode;
+	int ret = 0;
+
+	args = memdup_user(argp, sizeof(*args));
+	if (IS_ERR(args))
+		return PTR_ERR(args);
+
+	inode = file_inode(file);
+
+	/*
+	 * Unprivileged query to obtain the containing subvolume root id. The
+	 * path is reset so it's consistent with btrfs_search_path_in_tree.
+	 */
+	if (args->treeid == 0)
+		args->treeid = BTRFS_I(inode)->root->root_key.objectid;
+
+	if (args->objectid == BTRFS_FIRST_FREE_OBJECTID) {
+		args->name[0] = 0;
+		goto out;
+	}
+
+	if (!capable(CAP_SYS_ADMIN)) {
+		ret = -EPERM;
+		goto out;
+	}
+
+	ret = btrfs_search_path_in_tree(BTRFS_I(inode)->root->fs_info,
+					args->treeid, args->objectid,
+					args->name);
+
+out:
+	if (ret == 0 && copy_to_user(argp, args, sizeof(*args)))
+		ret = -EFAULT;
+
+	kfree(args);
+	return ret;
+}
+
+/*
+ * Version of ino_lookup ioctl (unprivileged)
+ *
+ * The main differences from ino_lookup ioctl are:
+ *
+ *   1. Read + Exec permission will be checked using inode_permission() during
+ *      path construction. -EACCES will be returned in case of failure.
+ *   2. Path construction will be stopped at the inode number which corresponds
+ *      to the fd with which this ioctl is called. If constructed path does not
+ *      exist under fd's inode, -EACCES will be returned.
+ *   3. The name of bottom subvolume is also searched and filled.
+ */
+static int btrfs_ioctl_ino_lookup_user(struct file *file, void __user *argp)
+{
+	struct btrfs_ioctl_ino_lookup_user_args *args;
+	struct inode *inode;
+	int ret;
+
+	args = memdup_user(argp, sizeof(*args));
+	if (IS_ERR(args))
+		return PTR_ERR(args);
+
+	inode = file_inode(file);
+
+	if (args->dirid == BTRFS_FIRST_FREE_OBJECTID &&
+	    BTRFS_I(inode)->location.objectid != BTRFS_FIRST_FREE_OBJECTID) {
+		/*
+		 * The subvolume does not exist under fd with which this is
+		 * called
+		 */
+		kfree(args);
+		return -EACCES;
+	}
+
+	ret = btrfs_search_path_in_tree_user(inode, args);
+
+	if (ret == 0 && copy_to_user(argp, args, sizeof(*args)))
+		ret = -EFAULT;
+
+	kfree(args);
+	return ret;
+}
+
+/* Get the subvolume information in BTRFS_ROOT_ITEM and BTRFS_ROOT_BACKREF */
+static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp)
+{
+	struct btrfs_ioctl_get_subvol_info_args *subvol_info;
+	struct btrfs_fs_info *fs_info;
+	struct btrfs_root *root;
+	struct btrfs_path *path;
+	struct btrfs_key key;
+	struct btrfs_root_item *root_item;
+	struct btrfs_root_ref *rref;
+	struct extent_buffer *leaf;
+	unsigned long item_off;
+	unsigned long item_len;
+	struct inode *inode;
+	int slot;
+	int ret = 0;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	subvol_info = kzalloc(sizeof(*subvol_info), GFP_KERNEL);
+	if (!subvol_info) {
+		btrfs_free_path(path);
+		return -ENOMEM;
+	}
+
+	inode = file_inode(file);
+	fs_info = BTRFS_I(inode)->root->fs_info;
+
+	/* Get root_item of inode's subvolume */
+	key.objectid = BTRFS_I(inode)->root->root_key.objectid;
+	key.type = BTRFS_ROOT_ITEM_KEY;
+	key.offset = (u64)-1;
+	root = btrfs_read_fs_root_no_name(fs_info, &key);
+	if (IS_ERR(root)) {
+		ret = PTR_ERR(root);
+		goto out;
+	}
+	root_item = &root->root_item;
+
+	subvol_info->treeid = key.objectid;
+
+	subvol_info->generation = btrfs_root_generation(root_item);
+	subvol_info->flags = btrfs_root_flags(root_item);
+
+	memcpy(subvol_info->uuid, root_item->uuid, BTRFS_UUID_SIZE);
+	memcpy(subvol_info->parent_uuid, root_item->parent_uuid,
+						    BTRFS_UUID_SIZE);
+	memcpy(subvol_info->received_uuid, root_item->received_uuid,
+						    BTRFS_UUID_SIZE);
+
+	subvol_info->ctransid = btrfs_root_ctransid(root_item);
+	subvol_info->ctime.sec = btrfs_stack_timespec_sec(&root_item->ctime);
+	subvol_info->ctime.nsec = btrfs_stack_timespec_nsec(&root_item->ctime);
+
+	subvol_info->otransid = btrfs_root_otransid(root_item);
+	subvol_info->otime.sec = btrfs_stack_timespec_sec(&root_item->otime);
+	subvol_info->otime.nsec = btrfs_stack_timespec_nsec(&root_item->otime);
+
+	subvol_info->stransid = btrfs_root_stransid(root_item);
+	subvol_info->stime.sec = btrfs_stack_timespec_sec(&root_item->stime);
+	subvol_info->stime.nsec = btrfs_stack_timespec_nsec(&root_item->stime);
+
+	subvol_info->rtransid = btrfs_root_rtransid(root_item);
+	subvol_info->rtime.sec = btrfs_stack_timespec_sec(&root_item->rtime);
+	subvol_info->rtime.nsec = btrfs_stack_timespec_nsec(&root_item->rtime);
+
+	if (key.objectid != BTRFS_FS_TREE_OBJECTID) {
+		/* Search root tree for ROOT_BACKREF of this subvolume */
+		root = fs_info->tree_root;
+
+		key.type = BTRFS_ROOT_BACKREF_KEY;
+		key.offset = 0;
+		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+		if (ret < 0) {
+			goto out;
+		} else if (path->slots[0] >=
+			   btrfs_header_nritems(path->nodes[0])) {
+			ret = btrfs_next_leaf(root, path);
+			if (ret < 0) {
+				goto out;
+			} else if (ret > 0) {
+				ret = -EUCLEAN;
+				goto out;
+			}
+		}
+
+		leaf = path->nodes[0];
+		slot = path->slots[0];
+		btrfs_item_key_to_cpu(leaf, &key, slot);
+		if (key.objectid == subvol_info->treeid &&
+		    key.type == BTRFS_ROOT_BACKREF_KEY) {
+			subvol_info->parent_id = key.offset;
+
+			rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref);
+			subvol_info->dirid = btrfs_root_ref_dirid(leaf, rref);
+
+			item_off = btrfs_item_ptr_offset(leaf, slot)
+					+ sizeof(struct btrfs_root_ref);
+			item_len = btrfs_item_size_nr(leaf, slot)
+					- sizeof(struct btrfs_root_ref);
+			read_extent_buffer(leaf, subvol_info->name,
+					   item_off, item_len);
+		} else {
+			ret = -ENOENT;
+			goto out;
+		}
+	}
+
+	if (copy_to_user(argp, subvol_info, sizeof(*subvol_info)))
+		ret = -EFAULT;
 
-		btrfs_release_path(path);
-		key.objectid = key.offset;
-		key.offset = (u64)-1;
-		dirid = key.objectid;
-	}
-	memmove(name, ptr, total_len);
-	name[total_len] = '\0';
-	ret = 0;
 out:
 	btrfs_free_path(path);
+	kzfree(subvol_info);
 	return ret;
 }
 
-static noinline int btrfs_ioctl_ino_lookup(struct file *file,
-					   void __user *argp)
+/*
+ * Return ROOT_REF information of the subvolume containing this inode
+ * except the subvolume name.
+ */
+static int btrfs_ioctl_get_subvol_rootref(struct file *file, void __user *argp)
 {
-	 struct btrfs_ioctl_ino_lookup_args *args;
-	 struct inode *inode;
-	int ret = 0;
+	struct btrfs_ioctl_get_subvol_rootref_args *rootrefs;
+	struct btrfs_root_ref *rref;
+	struct btrfs_root *root;
+	struct btrfs_path *path;
+	struct btrfs_key key;
+	struct extent_buffer *leaf;
+	struct inode *inode;
+	u64 objectid;
+	int slot;
+	int ret;
+	u8 found;
 
-	args = memdup_user(argp, sizeof(*args));
-	if (IS_ERR(args))
-		return PTR_ERR(args);
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+
+	rootrefs = memdup_user(argp, sizeof(*rootrefs));
+	if (IS_ERR(rootrefs)) {
+		btrfs_free_path(path);
+		return PTR_ERR(rootrefs);
+	}
 
 	inode = file_inode(file);
+	root = BTRFS_I(inode)->root->fs_info->tree_root;
+	objectid = BTRFS_I(inode)->root->root_key.objectid;
 
-	/*
-	 * Unprivileged query to obtain the containing subvolume root id. The
-	 * path is reset so it's consistent with btrfs_search_path_in_tree.
-	 */
-	if (args->treeid == 0)
-		args->treeid = BTRFS_I(inode)->root->root_key.objectid;
+	key.objectid = objectid;
+	key.type = BTRFS_ROOT_REF_KEY;
+	key.offset = rootrefs->min_treeid;
+	found = 0;
 
-	if (args->objectid == BTRFS_FIRST_FREE_OBJECTID) {
-		args->name[0] = 0;
+	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	if (ret < 0) {
 		goto out;
+	} else if (path->slots[0] >=
+		   btrfs_header_nritems(path->nodes[0])) {
+		ret = btrfs_next_leaf(root, path);
+		if (ret < 0) {
+			goto out;
+		} else if (ret > 0) {
+			ret = -EUCLEAN;
+			goto out;
+		}
 	}
+	while (1) {
+		leaf = path->nodes[0];
+		slot = path->slots[0];
 
-	if (!capable(CAP_SYS_ADMIN)) {
-		ret = -EPERM;
-		goto out;
-	}
+		btrfs_item_key_to_cpu(leaf, &key, slot);
+		if (key.objectid != objectid || key.type != BTRFS_ROOT_REF_KEY) {
+			ret = 0;
+			goto out;
+		}
 
-	ret = btrfs_search_path_in_tree(BTRFS_I(inode)->root->fs_info,
-					args->treeid, args->objectid,
-					args->name);
+		if (found == BTRFS_MAX_ROOTREF_BUFFER_NUM) {
+			ret = -EOVERFLOW;
+			goto out;
+		}
+
+		rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref);
+		rootrefs->rootref[found].treeid = key.offset;
+		rootrefs->rootref[found].dirid =
+				  btrfs_root_ref_dirid(leaf, rref);
+		found++;
+
+		ret = btrfs_next_item(root, path);
+		if (ret < 0) {
+			goto out;
+		} else if (ret > 0) {
+			ret = -EUCLEAN;
+			goto out;
+		}
+	}
 
 out:
-	if (ret == 0 && copy_to_user(argp, args, sizeof(*args)))
-		ret = -EFAULT;
+	if (!ret || ret == -EOVERFLOW) {
+		rootrefs->num_items = found;
+		/* update min_treeid for next search */
+		if (found)
+			rootrefs->min_treeid =
+				rootrefs->rootref[found - 1].treeid + 1;
+		if (copy_to_user(argp, rootrefs, sizeof(*rootrefs)))
+			ret = -EFAULT;
+	}
+
+	kfree(rootrefs);
+	btrfs_free_path(path);
 
-	kfree(args);
 	return ret;
 }
 
@@ -2309,12 +2812,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_root *dest = NULL;
 	struct btrfs_ioctl_vol_args *vol_args;
-	struct btrfs_trans_handle *trans;
-	struct btrfs_block_rsv block_rsv;
-	u64 root_flags;
-	u64 qgroup_reserved;
 	int namelen;
-	int ret;
 	int err = 0;
 
 	if (!S_ISDIR(dir->i_mode))
@@ -2398,133 +2896,11 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 	}
 
 	inode_lock(inode);
-
-	/*
-	 * Don't allow to delete a subvolume with send in progress. This is
-	 * inside the i_mutex so the error handling that has to drop the bit
-	 * again is not run concurrently.
-	 */
-	spin_lock(&dest->root_item_lock);
-	root_flags = btrfs_root_flags(&dest->root_item);
-	if (dest->send_in_progress == 0) {
-		btrfs_set_root_flags(&dest->root_item,
-				root_flags | BTRFS_ROOT_SUBVOL_DEAD);
-		spin_unlock(&dest->root_item_lock);
-	} else {
-		spin_unlock(&dest->root_item_lock);
-		btrfs_warn(fs_info,
-			   "Attempt to delete subvolume %llu during send",
-			   dest->root_key.objectid);
-		err = -EPERM;
-		goto out_unlock_inode;
-	}
-
-	down_write(&fs_info->subvol_sem);
-
-	err = may_destroy_subvol(dest);
-	if (err)
-		goto out_up_write;
-
-	btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP);
-	/*
-	 * One for dir inode, two for dir entries, two for root
-	 * ref/backref.
-	 */
-	err = btrfs_subvolume_reserve_metadata(root, &block_rsv,
-					       5, &qgroup_reserved, true);
-	if (err)
-		goto out_up_write;
-
-	trans = btrfs_start_transaction(root, 0);
-	if (IS_ERR(trans)) {
-		err = PTR_ERR(trans);
-		goto out_release;
-	}
-	trans->block_rsv = &block_rsv;
-	trans->bytes_reserved = block_rsv.size;
-
-	btrfs_record_snapshot_destroy(trans, BTRFS_I(dir));
-
-	ret = btrfs_unlink_subvol(trans, root, dir,
-				dest->root_key.objectid,
-				dentry->d_name.name,
-				dentry->d_name.len);
-	if (ret) {
-		err = ret;
-		btrfs_abort_transaction(trans, ret);
-		goto out_end_trans;
-	}
-
-	btrfs_record_root_in_trans(trans, dest);
-
-	memset(&dest->root_item.drop_progress, 0,
-		sizeof(dest->root_item.drop_progress));
-	dest->root_item.drop_level = 0;
-	btrfs_set_root_refs(&dest->root_item, 0);
-
-	if (!test_and_set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &dest->state)) {
-		ret = btrfs_insert_orphan_item(trans,
-					fs_info->tree_root,
-					dest->root_key.objectid);
-		if (ret) {
-			btrfs_abort_transaction(trans, ret);
-			err = ret;
-			goto out_end_trans;
-		}
-	}
-
-	ret = btrfs_uuid_tree_rem(trans, fs_info, dest->root_item.uuid,
-				  BTRFS_UUID_KEY_SUBVOL,
-				  dest->root_key.objectid);
-	if (ret && ret != -ENOENT) {
-		btrfs_abort_transaction(trans, ret);
-		err = ret;
-		goto out_end_trans;
-	}
-	if (!btrfs_is_empty_uuid(dest->root_item.received_uuid)) {
-		ret = btrfs_uuid_tree_rem(trans, fs_info,
-					  dest->root_item.received_uuid,
-					  BTRFS_UUID_KEY_RECEIVED_SUBVOL,
-					  dest->root_key.objectid);
-		if (ret && ret != -ENOENT) {
-			btrfs_abort_transaction(trans, ret);
-			err = ret;
-			goto out_end_trans;
-		}
-	}
-
-out_end_trans:
-	trans->block_rsv = NULL;
-	trans->bytes_reserved = 0;
-	ret = btrfs_end_transaction(trans);
-	if (ret && !err)
-		err = ret;
-	inode->i_flags |= S_DEAD;
-out_release:
-	btrfs_subvolume_release_metadata(fs_info, &block_rsv);
-out_up_write:
-	up_write(&fs_info->subvol_sem);
-	if (err) {
-		spin_lock(&dest->root_item_lock);
-		root_flags = btrfs_root_flags(&dest->root_item);
-		btrfs_set_root_flags(&dest->root_item,
-				root_flags & ~BTRFS_ROOT_SUBVOL_DEAD);
-		spin_unlock(&dest->root_item_lock);
-	}
-out_unlock_inode:
+	err = btrfs_delete_subvolume(dir, dentry);
 	inode_unlock(inode);
-	if (!err) {
-		d_invalidate(dentry);
-		btrfs_invalidate_inodes(dest);
+	if (!err)
 		d_delete(dentry);
-		ASSERT(dest->send_in_progress == 0);
 
-		/* the last ref */
-		if (dest->ino_cache_inode) {
-			iput(dest->ino_cache_inode);
-			dest->ino_cache_inode = NULL;
-		}
-	}
 out_dput:
 	dput(dentry);
 out_unlock_dir:
@@ -2613,7 +2989,6 @@ static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg)
 	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags))
 		return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
 
-	mutex_lock(&fs_info->volume_mutex);
 	vol_args = memdup_user(arg, sizeof(*vol_args));
 	if (IS_ERR(vol_args)) {
 		ret = PTR_ERR(vol_args);
@@ -2628,7 +3003,6 @@ static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg)
 
 	kfree(vol_args);
 out:
-	mutex_unlock(&fs_info->volume_mutex);
 	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
 	return ret;
 }
@@ -2654,8 +3028,10 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 	}
 
 	/* Check for compatibility reject unknown flags */
-	if (vol_args->flags & ~BTRFS_VOL_ARG_V2_FLAGS_SUPPORTED)
-		return -EOPNOTSUPP;
+	if (vol_args->flags & ~BTRFS_VOL_ARG_V2_FLAGS_SUPPORTED) {
+		ret = -EOPNOTSUPP;
+		goto out;
+	}
 
 	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
 		ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
@@ -2954,8 +3330,6 @@ static void btrfs_cmp_data_free(struct cmp_pages *cmp)
 			put_page(pg);
 		}
 	}
-	kfree(cmp->src_pages);
-	kfree(cmp->dst_pages);
 }
 
 static int btrfs_cmp_data_prepare(struct inode *src, u64 loff,
@@ -2964,40 +3338,14 @@ static int btrfs_cmp_data_prepare(struct inode *src, u64 loff,
 {
 	int ret;
 	int num_pages = PAGE_ALIGN(len) >> PAGE_SHIFT;
-	struct page **src_pgarr, **dst_pgarr;
 
-	/*
-	 * We must gather up all the pages before we initiate our
-	 * extent locking. We use an array for the page pointers. Size
-	 * of the array is bounded by len, which is in turn bounded by
-	 * BTRFS_MAX_DEDUPE_LEN.
-	 */
-	src_pgarr = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL);
-	dst_pgarr = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL);
-	if (!src_pgarr || !dst_pgarr) {
-		kfree(src_pgarr);
-		kfree(dst_pgarr);
-		return -ENOMEM;
-	}
 	cmp->num_pages = num_pages;
-	cmp->src_pages = src_pgarr;
-	cmp->dst_pages = dst_pgarr;
-
-	/*
-	 * If deduping ranges in the same inode, locking rules make it mandatory
-	 * to always lock pages in ascending order to avoid deadlocks with
-	 * concurrent tasks (such as starting writeback/delalloc).
-	 */
-	if (src == dst && dst_loff < loff) {
-		swap(src_pgarr, dst_pgarr);
-		swap(loff, dst_loff);
-	}
 
-	ret = gather_extent_pages(src, src_pgarr, cmp->num_pages, loff);
+	ret = gather_extent_pages(src, cmp->src_pages, num_pages, loff);
 	if (ret)
 		goto out;
 
-	ret = gather_extent_pages(dst, dst_pgarr, cmp->num_pages, dst_loff);
+	ret = gather_extent_pages(dst, cmp->dst_pages, num_pages, dst_loff);
 
 out:
 	if (ret)
@@ -3067,31 +3415,23 @@ static int extent_same_check_offsets(struct inode *inode, u64 off, u64 *plen,
 	return 0;
 }
 
-static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
-			     struct inode *dst, u64 dst_loff)
+static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 olen,
+				   struct inode *dst, u64 dst_loff,
+				   struct cmp_pages *cmp)
 {
 	int ret;
 	u64 len = olen;
-	struct cmp_pages cmp;
 	bool same_inode = (src == dst);
 	u64 same_lock_start = 0;
 	u64 same_lock_len = 0;
 
-	if (len == 0)
-		return 0;
-
-	if (same_inode)
-		inode_lock(src);
-	else
-		btrfs_double_inode_lock(src, dst);
-
 	ret = extent_same_check_offsets(src, loff, &len, olen);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	ret = extent_same_check_offsets(dst, dst_loff, &len, olen);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	if (same_inode) {
 		/*
@@ -3108,32 +3448,21 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
 		 * allow an unaligned length so long as it ends at
 		 * i_size.
 		 */
-		if (len != olen) {
-			ret = -EINVAL;
-			goto out_unlock;
-		}
+		if (len != olen)
+			return -EINVAL;
 
 		/* Check for overlapping ranges */
-		if (dst_loff + len > loff && dst_loff < loff + len) {
-			ret = -EINVAL;
-			goto out_unlock;
-		}
+		if (dst_loff + len > loff && dst_loff < loff + len)
+			return -EINVAL;
 
 		same_lock_start = min_t(u64, loff, dst_loff);
 		same_lock_len = max_t(u64, loff, dst_loff) + len - same_lock_start;
 	}
 
-	/* don't make the dst file partly checksummed */
-	if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) !=
-	    (BTRFS_I(dst)->flags & BTRFS_INODE_NODATASUM)) {
-		ret = -EINVAL;
-		goto out_unlock;
-	}
-
 again:
-	ret = btrfs_cmp_data_prepare(src, loff, dst, dst_loff, olen, &cmp);
+	ret = btrfs_cmp_data_prepare(src, loff, dst, dst_loff, olen, cmp);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	if (same_inode)
 		ret = lock_extent_range(src, same_lock_start, same_lock_len,
@@ -3154,7 +3483,7 @@ again:
 		 * Ranges in the io trees already unlocked. Now unlock all
 		 * pages before waiting for all IO to complete.
 		 */
-		btrfs_cmp_data_free(&cmp);
+		btrfs_cmp_data_free(cmp);
 		if (same_inode) {
 			btrfs_wait_ordered_range(src, same_lock_start,
 						 same_lock_len);
@@ -3167,12 +3496,12 @@ again:
 	ASSERT(ret == 0);
 	if (WARN_ON(ret)) {
 		/* ranges in the io trees already unlocked */
-		btrfs_cmp_data_free(&cmp);
+		btrfs_cmp_data_free(cmp);
 		return ret;
 	}
 
 	/* pass original length for comparison so we stay within i_size */
-	ret = btrfs_cmp_data(olen, &cmp);
+	ret = btrfs_cmp_data(olen, cmp);
 	if (ret == 0)
 		ret = btrfs_clone(src, dst, loff, olen, len, dst_loff, 1);
 
@@ -3182,18 +3511,91 @@ again:
 	else
 		btrfs_double_extent_unlock(src, loff, dst, dst_loff, len);
 
-	btrfs_cmp_data_free(&cmp);
+	btrfs_cmp_data_free(cmp);
+
+	return ret;
+}
+
+#define BTRFS_MAX_DEDUPE_LEN	SZ_16M
+
+static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
+			     struct inode *dst, u64 dst_loff)
+{
+	int ret;
+	struct cmp_pages cmp;
+	int num_pages = PAGE_ALIGN(BTRFS_MAX_DEDUPE_LEN) >> PAGE_SHIFT;
+	bool same_inode = (src == dst);
+	u64 i, tail_len, chunk_count;
+
+	if (olen == 0)
+		return 0;
+
+	if (same_inode)
+		inode_lock(src);
+	else
+		btrfs_double_inode_lock(src, dst);
+
+	/* don't make the dst file partly checksummed */
+	if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) !=
+	    (BTRFS_I(dst)->flags & BTRFS_INODE_NODATASUM)) {
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
+	tail_len = olen % BTRFS_MAX_DEDUPE_LEN;
+	chunk_count = div_u64(olen, BTRFS_MAX_DEDUPE_LEN);
+	if (chunk_count == 0)
+		num_pages = PAGE_ALIGN(tail_len) >> PAGE_SHIFT;
+
+	/*
+	 * If deduping ranges in the same inode, locking rules make it
+	 * mandatory to always lock pages in ascending order to avoid deadlocks
+	 * with concurrent tasks (such as starting writeback/delalloc).
+	 */
+	if (same_inode && dst_loff < loff)
+		swap(loff, dst_loff);
+
+	/*
+	 * We must gather up all the pages before we initiate our extent
+	 * locking. We use an array for the page pointers. Size of the array is
+	 * bounded by len, which is in turn bounded by BTRFS_MAX_DEDUPE_LEN.
+	 */
+	cmp.src_pages = kvmalloc_array(num_pages, sizeof(struct page *),
+				       GFP_KERNEL | __GFP_ZERO);
+	cmp.dst_pages = kvmalloc_array(num_pages, sizeof(struct page *),
+				       GFP_KERNEL | __GFP_ZERO);
+	if (!cmp.src_pages || !cmp.dst_pages) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+
+	for (i = 0; i < chunk_count; i++) {
+		ret = btrfs_extent_same_range(src, loff, BTRFS_MAX_DEDUPE_LEN,
+					      dst, dst_loff, &cmp);
+		if (ret)
+			goto out_unlock;
+
+		loff += BTRFS_MAX_DEDUPE_LEN;
+		dst_loff += BTRFS_MAX_DEDUPE_LEN;
+	}
+
+	if (tail_len > 0)
+		ret = btrfs_extent_same_range(src, loff, tail_len, dst,
+					      dst_loff, &cmp);
+
 out_unlock:
 	if (same_inode)
 		inode_unlock(src);
 	else
 		btrfs_double_inode_unlock(src, dst);
 
+out_free:
+	kvfree(cmp.src_pages);
+	kvfree(cmp.dst_pages);
+
 	return ret;
 }
 
-#define BTRFS_MAX_DEDUPE_LEN	SZ_16M
-
 ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen,
 				struct file *dst_file, u64 dst_loff)
 {
@@ -3202,9 +3604,6 @@ ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen,
 	u64 bs = BTRFS_I(src)->root->fs_info->sb->s_blocksize;
 	ssize_t res;
 
-	if (olen > BTRFS_MAX_DEDUPE_LEN)
-		olen = BTRFS_MAX_DEDUPE_LEN;
-
 	if (WARN_ON_ONCE(bs < PAGE_SIZE)) {
 		/*
 		 * Btrfs does not support blocksize < page_size. As a
@@ -3826,11 +4225,6 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src,
 	    src->i_sb != inode->i_sb)
 		return -EXDEV;
 
-	/* don't make the dst file partly checksummed */
-	if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) !=
-	    (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM))
-		return -EINVAL;
-
 	if (S_ISDIR(src->i_mode) || S_ISDIR(inode->i_mode))
 		return -EISDIR;
 
@@ -3840,6 +4234,13 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src,
 		inode_lock(src);
 	}
 
+	/* don't make the dst file partly checksummed */
+	if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) !=
+	    (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
 	/* determine range to clone */
 	ret = -EINVAL;
 	if (off + len > src->i_size || off + len < off)
@@ -4007,8 +4408,8 @@ out:
 	return ret;
 }
 
-void btrfs_get_block_group_info(struct list_head *groups_list,
-				struct btrfs_ioctl_space_info *space)
+static void get_block_group_info(struct list_head *groups_list,
+				 struct btrfs_ioctl_space_info *space)
 {
 	struct btrfs_block_group_cache *block_group;
 
@@ -4124,8 +4525,8 @@ static long btrfs_ioctl_space_info(struct btrfs_fs_info *fs_info,
 		down_read(&info->groups_sem);
 		for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) {
 			if (!list_empty(&info->block_groups[c])) {
-				btrfs_get_block_group_info(
-					&info->block_groups[c], &space);
+				get_block_group_info(&info->block_groups[c],
+						     &space);
 				memcpy(dest, &space, sizeof(space));
 				dest++;
 				space_args.total_spaces++;
@@ -4490,14 +4891,14 @@ out_loi:
 	return ret;
 }
 
-void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock,
+void btrfs_update_ioctl_balance_args(struct btrfs_fs_info *fs_info,
 			       struct btrfs_ioctl_balance_args *bargs)
 {
 	struct btrfs_balance_control *bctl = fs_info->balance_ctl;
 
 	bargs->flags = bctl->flags;
 
-	if (atomic_read(&fs_info->balance_running))
+	if (test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags))
 		bargs->state |= BTRFS_BALANCE_STATE_RUNNING;
 	if (atomic_read(&fs_info->balance_pause_req))
 		bargs->state |= BTRFS_BALANCE_STATE_PAUSE_REQ;
@@ -4508,13 +4909,9 @@ void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock,
 	memcpy(&bargs->meta, &bctl->meta, sizeof(bargs->meta));
 	memcpy(&bargs->sys, &bctl->sys, sizeof(bargs->sys));
 
-	if (lock) {
-		spin_lock(&fs_info->balance_lock);
-		memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat));
-		spin_unlock(&fs_info->balance_lock);
-	} else {
-		memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat));
-	}
+	spin_lock(&fs_info->balance_lock);
+	memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat));
+	spin_unlock(&fs_info->balance_lock);
 }
 
 static long btrfs_ioctl_balance(struct file *file, void __user *arg)
@@ -4535,7 +4932,6 @@ static long btrfs_ioctl_balance(struct file *file, void __user *arg)
 
 again:
 	if (!test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
-		mutex_lock(&fs_info->volume_mutex);
 		mutex_lock(&fs_info->balance_mutex);
 		need_unlock = true;
 		goto locked;
@@ -4550,21 +4946,22 @@ again:
 	mutex_lock(&fs_info->balance_mutex);
 	if (fs_info->balance_ctl) {
 		/* this is either (2) or (3) */
-		if (!atomic_read(&fs_info->balance_running)) {
+		if (!test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) {
 			mutex_unlock(&fs_info->balance_mutex);
-			if (!mutex_trylock(&fs_info->volume_mutex))
-				goto again;
+			/*
+			 * Lock released to allow other waiters to continue,
+			 * we'll reexamine the status again.
+			 */
 			mutex_lock(&fs_info->balance_mutex);
 
 			if (fs_info->balance_ctl &&
-			    !atomic_read(&fs_info->balance_running)) {
+			    !test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) {
 				/* this is (3) */
 				need_unlock = false;
 				goto locked;
 			}
 
 			mutex_unlock(&fs_info->balance_mutex);
-			mutex_unlock(&fs_info->volume_mutex);
 			goto again;
 		} else {
 			/* this is (2) */
@@ -4617,7 +5014,6 @@ locked:
 		goto out_bargs;
 	}
 
-	bctl->fs_info = fs_info;
 	if (arg) {
 		memcpy(&bctl->data, &bargs->data, sizeof(bctl->data));
 		memcpy(&bctl->meta, &bargs->meta, sizeof(bctl->meta));
@@ -4636,14 +5032,14 @@ locked:
 
 do_balance:
 	/*
-	 * Ownership of bctl and filesystem flag BTRFS_FS_EXCL_OP
-	 * goes to to btrfs_balance.  bctl is freed in __cancel_balance,
-	 * or, if restriper was paused all the way until unmount, in
-	 * free_fs_info.  The flag is cleared in __cancel_balance.
+	 * Ownership of bctl and filesystem flag BTRFS_FS_EXCL_OP goes to
+	 * btrfs_balance.  bctl is freed in reset_balance_state, or, if
+	 * restriper was paused all the way until unmount, in free_fs_info.
+	 * The flag should be cleared after reset_balance_state.
 	 */
 	need_unlock = false;
 
-	ret = btrfs_balance(bctl, bargs);
+	ret = btrfs_balance(fs_info, bctl, bargs);
 	bctl = NULL;
 
 	if (arg) {
@@ -4657,7 +5053,6 @@ out_bargs:
 	kfree(bargs);
 out_unlock:
 	mutex_unlock(&fs_info->balance_mutex);
-	mutex_unlock(&fs_info->volume_mutex);
 	if (need_unlock)
 		clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
 out:
@@ -4701,7 +5096,7 @@ static long btrfs_ioctl_balance_progress(struct btrfs_fs_info *fs_info,
 		goto out;
 	}
 
-	update_ioctl_balance_args(fs_info, 1, bargs);
+	btrfs_update_ioctl_balance_args(fs_info, bargs);
 
 	if (copy_to_user(arg, bargs, sizeof(*bargs)))
 		ret = -EFAULT;
@@ -5038,8 +5433,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file,
 				       BTRFS_UUID_SIZE);
 	if (received_uuid_changed &&
 	    !btrfs_is_empty_uuid(root_item->received_uuid)) {
-		ret = btrfs_uuid_tree_rem(trans, fs_info,
-					  root_item->received_uuid,
+		ret = btrfs_uuid_tree_remove(trans, root_item->received_uuid,
 					  BTRFS_UUID_KEY_RECEIVED_SUBVOL,
 					  root->root_key.objectid);
 		if (ret && ret != -ENOENT) {
@@ -5063,7 +5457,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file,
 		goto out;
 	}
 	if (received_uuid_changed && !btrfs_is_empty_uuid(sa->uuid)) {
-		ret = btrfs_uuid_tree_add(trans, fs_info, sa->uuid,
+		ret = btrfs_uuid_tree_add(trans, sa->uuid,
 					  BTRFS_UUID_KEY_RECEIVED_SUBVOL,
 					  root->root_key.objectid);
 		if (ret < 0 && ret != -EEXIST) {
@@ -5497,7 +5891,7 @@ long btrfs_ioctl(struct file *file, unsigned int
 	case BTRFS_IOC_SYNC: {
 		int ret;
 
-		ret = btrfs_start_delalloc_roots(fs_info, 0, -1);
+		ret = btrfs_start_delalloc_roots(fs_info, -1);
 		if (ret)
 			return ret;
 		ret = btrfs_sync_fs(inode->i_sb, 1);
@@ -5565,6 +5959,16 @@ long btrfs_ioctl(struct file *file, unsigned int
 		return btrfs_ioctl_get_features(file, argp);
 	case BTRFS_IOC_SET_FEATURES:
 		return btrfs_ioctl_set_features(file, argp);
+	case FS_IOC_FSGETXATTR:
+		return btrfs_ioctl_fsgetxattr(file, argp);
+	case FS_IOC_FSSETXATTR:
+		return btrfs_ioctl_fssetxattr(file, argp);
+	case BTRFS_IOC_GET_SUBVOL_INFO:
+		return btrfs_ioctl_get_subvol_info(file, argp);
+	case BTRFS_IOC_GET_SUBVOL_ROOTREF:
+		return btrfs_ioctl_get_subvol_rootref(file, argp);
+	case BTRFS_IOC_INO_LOOKUP_USER:
+		return btrfs_ioctl_ino_lookup_user(file, argp);
 	}
 
 	return -ENOTTY;

fs/btrfs/locking.c

@@ -66,22 +66,16 @@ void btrfs_clear_lock_blocking_rw(struct extent_buffer *eb, int rw)
 		write_lock(&eb->lock);
 		WARN_ON(atomic_read(&eb->spinning_writers));
 		atomic_inc(&eb->spinning_writers);
-		/*
-		 * atomic_dec_and_test implies a barrier for waitqueue_active
-		 */
-		if (atomic_dec_and_test(&eb->blocking_writers) &&
-		    waitqueue_active(&eb->write_lock_wq))
-			wake_up(&eb->write_lock_wq);
+		/* atomic_dec_and_test implies a barrier */
+		if (atomic_dec_and_test(&eb->blocking_writers))
+			cond_wake_up_nomb(&eb->write_lock_wq);
 	} else if (rw == BTRFS_READ_LOCK_BLOCKING) {
 		BUG_ON(atomic_read(&eb->blocking_readers) == 0);
 		read_lock(&eb->lock);
 		atomic_inc(&eb->spinning_readers);
-		/*
-		 * atomic_dec_and_test implies a barrier for waitqueue_active
-		 */
-		if (atomic_dec_and_test(&eb->blocking_readers) &&
-		    waitqueue_active(&eb->read_lock_wq))
-			wake_up(&eb->read_lock_wq);
+		/* atomic_dec_and_test implies a barrier */
+		if (atomic_dec_and_test(&eb->blocking_readers))
+			cond_wake_up_nomb(&eb->read_lock_wq);
 	}
 }
 
@@ -221,12 +215,9 @@ void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb)
 	}
 	btrfs_assert_tree_read_locked(eb);
 	WARN_ON(atomic_read(&eb->blocking_readers) == 0);
-	/*
-	 * atomic_dec_and_test implies a barrier for waitqueue_active
-	 */
-	if (atomic_dec_and_test(&eb->blocking_readers) &&
-	    waitqueue_active(&eb->read_lock_wq))
-		wake_up(&eb->read_lock_wq);
+	/* atomic_dec_and_test implies a barrier */
+	if (atomic_dec_and_test(&eb->blocking_readers))
+		cond_wake_up_nomb(&eb->read_lock_wq);
 	atomic_dec(&eb->read_locks);
 }
 
@@ -275,12 +266,9 @@ void btrfs_tree_unlock(struct extent_buffer *eb)
 	if (blockers) {
 		WARN_ON(atomic_read(&eb->spinning_writers));
 		atomic_dec(&eb->blocking_writers);
-		/*
-		 * Make sure counter is updated before we wake up waiters.
-		 */
+		/* Use the lighter barrier after atomic */
 		smp_mb__after_atomic();
-		if (waitqueue_active(&eb->write_lock_wq))
-			wake_up(&eb->write_lock_wq);
+		cond_wake_up_nomb(&eb->write_lock_wq);
 	} else {
 		WARN_ON(atomic_read(&eb->spinning_writers) != 1);
 		atomic_dec(&eb->spinning_writers);

fs/btrfs/lzo.c

@@ -17,6 +17,43 @@
 
 #define LZO_LEN	4
 
+/*
+ * Btrfs LZO compression format
+ *
+ * Regular and inlined LZO compressed data extents consist of:
+ *
+ * 1.  Header
+ *     Fixed size. LZO_LEN (4) bytes long, LE32.
+ *     Records the total size (including the header) of compressed data.
+ *
+ * 2.  Segment(s)
+ *     Variable size. Each segment includes one segment header, followd by data
+ *     payload.
+ *     One regular LZO compressed extent can have one or more segments.
+ *     For inlined LZO compressed extent, only one segment is allowed.
+ *     One segment represents at most one page of uncompressed data.
+ *
+ * 2.1 Segment header
+ *     Fixed size. LZO_LEN (4) bytes long, LE32.
+ *     Records the total size of the segment (not including the header).
+ *     Segment header never crosses page boundary, thus it's possible to
+ *     have at most 3 padding zeros at the end of the page.
+ *
+ * 2.2 Data Payload
+ *     Variable size. Size up limit should be lzo1x_worst_compress(PAGE_SIZE)
+ *     which is 4419 for a 4KiB page.
+ *
+ * Example:
+ * Page 1:
+ *          0     0x2   0x4   0x6   0x8   0xa   0xc   0xe     0x10
+ * 0x0000   |  Header   | SegHdr 01 | Data payload 01 ...     |
+ * ...
+ * 0x0ff0   | SegHdr  N | Data payload  N     ...          |00|
+ *                                                          ^^ padding zeros
+ * Page 2:
+ * 0x1000   | SegHdr N+1| Data payload N+1 ...                |
+ */
+
 struct workspace {
 	void *mem;
 	void *buf;	/* where decompressed data goes */
@@ -258,6 +295,7 @@ static int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 	unsigned long working_bytes;
 	size_t in_len;
 	size_t out_len;
+	const size_t max_segment_len = lzo1x_worst_compress(PAGE_SIZE);
 	unsigned long in_offset;
 	unsigned long in_page_bytes_left;
 	unsigned long tot_in;
@@ -271,10 +309,22 @@ static int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 
 	data_in = kmap(pages_in[0]);
 	tot_len = read_compress_length(data_in);
+	/*
+	 * Compressed data header check.
+	 *
+	 * The real compressed size can't exceed the maximum extent length, and
+	 * all pages should be used (whole unused page with just the segment
+	 * header is not possible).  If this happens it means the compressed
+	 * extent is corrupted.
+	 */
+	if (tot_len > min_t(size_t, BTRFS_MAX_COMPRESSED, srclen) ||
+	    tot_len < srclen - PAGE_SIZE) {
+		ret = -EUCLEAN;
+		goto done;
+	}
 
 	tot_in = LZO_LEN;
 	in_offset = LZO_LEN;
-	tot_len = min_t(size_t, srclen, tot_len);
 	in_page_bytes_left = PAGE_SIZE - LZO_LEN;
 
 	tot_out = 0;
@@ -285,6 +335,17 @@ static int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 		in_offset += LZO_LEN;
 		tot_in += LZO_LEN;
 
+		/*
+		 * Segment header check.
+		 *
+		 * The segment length must not exceed the maximum LZO
+		 * compression size, nor the total compressed size.
+		 */
+		if (in_len > max_segment_len || tot_in + in_len > tot_len) {
+			ret = -EUCLEAN;
+			goto done;
+		}
+
 		tot_in += in_len;
 		working_bytes = in_len;
 		may_late_unmap = need_unmap = false;
@@ -335,7 +396,7 @@ cont:
 			}
 		}
 
-		out_len = lzo1x_worst_compress(PAGE_SIZE);
+		out_len = max_segment_len;
 		ret = lzo1x_decompress_safe(buf, in_len, workspace->buf,
 					    &out_len);
 		if (need_unmap)
@@ -369,15 +430,24 @@ static int lzo_decompress(struct list_head *ws, unsigned char *data_in,
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	size_t in_len;
 	size_t out_len;
+	size_t max_segment_len = lzo1x_worst_compress(PAGE_SIZE);
 	int ret = 0;
 	char *kaddr;
 	unsigned long bytes;
 
-	BUG_ON(srclen < LZO_LEN);
+	if (srclen < LZO_LEN || srclen > max_segment_len + LZO_LEN * 2)
+		return -EUCLEAN;
 
+	in_len = read_compress_length(data_in);
+	if (in_len != srclen)
+		return -EUCLEAN;
 	data_in += LZO_LEN;
 
 	in_len = read_compress_length(data_in);
+	if (in_len != srclen - LZO_LEN * 2) {
+		ret = -EUCLEAN;
+		goto out;
+	}
 	data_in += LZO_LEN;
 
 	out_len = PAGE_SIZE;

fs/btrfs/ordered-data.c

@@ -343,11 +343,8 @@ int btrfs_dec_test_first_ordered_pending(struct inode *inode,
 
 	if (entry->bytes_left == 0) {
 		ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags);
-		/*
-		 * Implicit memory barrier after test_and_set_bit
-		 */
-		if (waitqueue_active(&entry->wait))
-			wake_up(&entry->wait);
+		/* test_and_set_bit implies a barrier */
+		cond_wake_up_nomb(&entry->wait);
 	} else {
 		ret = 1;
 	}
@@ -410,11 +407,8 @@ have_entry:
 
 	if (entry->bytes_left == 0) {
 		ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags);
-		/*
-		 * Implicit memory barrier after test_and_set_bit
-		 */
-		if (waitqueue_active(&entry->wait))
-			wake_up(&entry->wait);
+		/* test_and_set_bit implies a barrier */
+		cond_wake_up_nomb(&entry->wait);
 	} else {
 		ret = 1;
 	}

fs/btrfs/print-tree.c

@@ -166,6 +166,25 @@ static void print_uuid_item(struct extent_buffer *l, unsigned long offset,
 	}
 }
 
+/*
+ * Helper to output refs and locking status of extent buffer.  Useful to debug
+ * race condition related problems.
+ */
+static void print_eb_refs_lock(struct extent_buffer *eb)
+{
+#ifdef CONFIG_BTRFS_DEBUG
+	btrfs_info(eb->fs_info,
+"refs %u lock (w:%d r:%d bw:%d br:%d sw:%d sr:%d) lock_owner %u current %u",
+		   atomic_read(&eb->refs), atomic_read(&eb->write_locks),
+		   atomic_read(&eb->read_locks),
+		   atomic_read(&eb->blocking_writers),
+		   atomic_read(&eb->blocking_readers),
+		   atomic_read(&eb->spinning_writers),
+		   atomic_read(&eb->spinning_readers),
+		   eb->lock_owner, current->pid);
+#endif
+}
+
 void btrfs_print_leaf(struct extent_buffer *l)
 {
 	struct btrfs_fs_info *fs_info;
@@ -193,6 +212,7 @@ void btrfs_print_leaf(struct extent_buffer *l)
 		   "leaf %llu gen %llu total ptrs %d free space %d owner %llu",
 		   btrfs_header_bytenr(l), btrfs_header_generation(l), nr,
 		   btrfs_leaf_free_space(fs_info, l), btrfs_header_owner(l));
+	print_eb_refs_lock(l);
 	for (i = 0 ; i < nr ; i++) {
 		item = btrfs_item_nr(i);
 		btrfs_item_key_to_cpu(l, &key, i);
@@ -347,6 +367,7 @@ void btrfs_print_tree(struct extent_buffer *c, bool follow)
 		   btrfs_header_bytenr(c), level, btrfs_header_generation(c),
 		   nr, (u32)BTRFS_NODEPTRS_PER_BLOCK(fs_info) - nr,
 		   btrfs_header_owner(c));
+	print_eb_refs_lock(c);
 	for (i = 0; i < nr; i++) {
 		btrfs_node_key_to_cpu(c, &key, i);
 		pr_info("\tkey %d (%llu %u %llu) block %llu gen %llu\n",

fs/btrfs/qgroup.c

@@ -1882,8 +1882,8 @@ static int qgroup_update_counters(struct btrfs_fs_info *fs_info,
 		cur_old_count = btrfs_qgroup_get_old_refcnt(qg, seq);
 		cur_new_count = btrfs_qgroup_get_new_refcnt(qg, seq);
 
-		trace_qgroup_update_counters(fs_info, qg->qgroupid,
-					     cur_old_count, cur_new_count);
+		trace_qgroup_update_counters(fs_info, qg, cur_old_count,
+					     cur_new_count);
 
 		/* Rfer update part */
 		if (cur_old_count == 0 && cur_new_count > 0) {
@@ -2014,8 +2014,8 @@ btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans,
 
 	BUG_ON(!fs_info->quota_root);
 
-	trace_btrfs_qgroup_account_extent(fs_info, bytenr, num_bytes,
-					  nr_old_roots, nr_new_roots);
+	trace_btrfs_qgroup_account_extent(fs_info, trans->transid, bytenr,
+					num_bytes, nr_old_roots, nr_new_roots);
 
 	qgroups = ulist_alloc(GFP_NOFS);
 	if (!qgroups) {
@@ -2579,6 +2579,21 @@ out:
 	spin_unlock(&fs_info->qgroup_lock);
 }
 
+/*
+ * Check if the leaf is the last leaf. Which means all node pointers
+ * are at their last position.
+ */
+static bool is_last_leaf(struct btrfs_path *path)
+{
+	int i;
+
+	for (i = 1; i < BTRFS_MAX_LEVEL && path->nodes[i]; i++) {
+		if (path->slots[i] != btrfs_header_nritems(path->nodes[i]) - 1)
+			return false;
+	}
+	return true;
+}
+
 /*
  * returns < 0 on error, 0 when more leafs are to be scanned.
  * returns 1 when done.
@@ -2590,8 +2605,8 @@ qgroup_rescan_leaf(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
 	struct btrfs_key found;
 	struct extent_buffer *scratch_leaf = NULL;
 	struct ulist *roots = NULL;
-	struct seq_list tree_mod_seq_elem = SEQ_LIST_INIT(tree_mod_seq_elem);
 	u64 num_bytes;
+	bool done;
 	int slot;
 	int ret;
 
@@ -2620,12 +2635,12 @@ qgroup_rescan_leaf(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
 		mutex_unlock(&fs_info->qgroup_rescan_lock);
 		return ret;
 	}
+	done = is_last_leaf(path);
 
 	btrfs_item_key_to_cpu(path->nodes[0], &found,
 			      btrfs_header_nritems(path->nodes[0]) - 1);
 	fs_info->qgroup_rescan_progress.objectid = found.objectid + 1;
 
-	btrfs_get_tree_mod_seq(fs_info, &tree_mod_seq_elem);
 	scratch_leaf = btrfs_clone_extent_buffer(path->nodes[0]);
 	if (!scratch_leaf) {
 		ret = -ENOMEM;
@@ -2664,8 +2679,9 @@ out:
 		btrfs_tree_read_unlock_blocking(scratch_leaf);
 		free_extent_buffer(scratch_leaf);
 	}
-	btrfs_put_tree_mod_seq(fs_info, &tree_mod_seq_elem);
 
+	if (done && !ret)
+		ret = 1;
 	return ret;
 }
 
@@ -2681,6 +2697,12 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
 	path = btrfs_alloc_path();
 	if (!path)
 		goto out;
+	/*
+	 * Rescan should only search for commit root, and any later difference
+	 * should be recorded by qgroup
+	 */
+	path->search_commit_root = 1;
+	path->skip_locking = 1;
 
 	err = 0;
 	while (!err && !btrfs_fs_closing(fs_info)) {
@@ -2760,26 +2782,36 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 {
 	int ret = 0;
 
-	if (!init_flags &&
-	    (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) ||
-	     !(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON))) {
-		ret = -EINVAL;
-		goto err;
+	if (!init_flags) {
+		/* we're resuming qgroup rescan at mount time */
+		if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN))
+			btrfs_warn(fs_info,
+			"qgroup rescan init failed, qgroup is not enabled");
+		else if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON))
+			btrfs_warn(fs_info,
+			"qgroup rescan init failed, qgroup rescan is not queued");
+		return -EINVAL;
 	}
 
 	mutex_lock(&fs_info->qgroup_rescan_lock);
 	spin_lock(&fs_info->qgroup_lock);
 
 	if (init_flags) {
-		if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN)
+		if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) {
+			btrfs_warn(fs_info,
+				   "qgroup rescan is already in progress");
 			ret = -EINPROGRESS;
-		else if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON))
+		} else if (!(fs_info->qgroup_flags &
+			     BTRFS_QGROUP_STATUS_FLAG_ON)) {
+			btrfs_warn(fs_info,
+			"qgroup rescan init failed, qgroup is not enabled");
 			ret = -EINVAL;
+		}
 
 		if (ret) {
 			spin_unlock(&fs_info->qgroup_lock);
 			mutex_unlock(&fs_info->qgroup_rescan_lock);
-			goto err;
+			return ret;
 		}
 		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_RESCAN;
 	}
@@ -2798,13 +2830,6 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 	btrfs_init_work(&fs_info->qgroup_rescan_work,
 			btrfs_qgroup_rescan_helper,
 			btrfs_qgroup_rescan_worker, NULL, NULL);
-
-	if (ret) {
-err:
-		btrfs_info(fs_info, "qgroup_rescan_init failed with %d", ret);
-		return ret;
-	}
-
 	return 0;
 }
 

fs/btrfs/raid56.c

@@ -163,6 +163,12 @@ struct btrfs_raid_bio {
 	 * bitmap to record which horizontal stripe has data
 	 */
 	unsigned long *dbitmap;
+
+	/* allocated with real_stripes-many pointers for finish_*() calls */
+	void **finish_pointers;
+
+	/* allocated with stripe_npages-many bits for finish_*() calls */
+	unsigned long *finish_pbitmap;
 };
 
 static int __raid56_parity_recover(struct btrfs_raid_bio *rbio);
@@ -981,9 +987,14 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info,
 	int stripe_npages = DIV_ROUND_UP(stripe_len, PAGE_SIZE);
 	void *p;
 
-	rbio = kzalloc(sizeof(*rbio) + num_pages * sizeof(struct page *) * 2 +
-		       DIV_ROUND_UP(stripe_npages, BITS_PER_LONG) *
-		       sizeof(long), GFP_NOFS);
+	rbio = kzalloc(sizeof(*rbio) +
+		       sizeof(*rbio->stripe_pages) * num_pages +
+		       sizeof(*rbio->bio_pages) * num_pages +
+		       sizeof(*rbio->finish_pointers) * real_stripes +
+		       sizeof(*rbio->dbitmap) * BITS_TO_LONGS(stripe_npages) +
+		       sizeof(*rbio->finish_pbitmap) *
+				BITS_TO_LONGS(stripe_npages),
+		       GFP_NOFS);
 	if (!rbio)
 		return ERR_PTR(-ENOMEM);
 
@@ -1005,13 +1016,20 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info,
 	atomic_set(&rbio->stripes_pending, 0);
 
 	/*
-	 * the stripe_pages and bio_pages array point to the extra
+	 * the stripe_pages, bio_pages, etc arrays point to the extra
 	 * memory we allocated past the end of the rbio
 	 */
 	p = rbio + 1;
-	rbio->stripe_pages = p;
-	rbio->bio_pages = p + sizeof(struct page *) * num_pages;
-	rbio->dbitmap = p + sizeof(struct page *) * num_pages * 2;
+#define CONSUME_ALLOC(ptr, count)	do {				\
+		ptr = p;						\
+		p = (unsigned char *)p + sizeof(*(ptr)) * (count);	\
+	} while (0)
+	CONSUME_ALLOC(rbio->stripe_pages, num_pages);
+	CONSUME_ALLOC(rbio->bio_pages, num_pages);
+	CONSUME_ALLOC(rbio->finish_pointers, real_stripes);
+	CONSUME_ALLOC(rbio->dbitmap, BITS_TO_LONGS(stripe_npages));
+	CONSUME_ALLOC(rbio->finish_pbitmap, BITS_TO_LONGS(stripe_npages));
+#undef  CONSUME_ALLOC
 
 	if (bbio->map_type & BTRFS_BLOCK_GROUP_RAID5)
 		nr_data = real_stripes - 1;
@@ -1180,7 +1198,7 @@ static void index_rbio_pages(struct btrfs_raid_bio *rbio)
 static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
 {
 	struct btrfs_bio *bbio = rbio->bbio;
-	void *pointers[rbio->real_stripes];
+	void **pointers = rbio->finish_pointers;
 	int nr_data = rbio->nr_data;
 	int stripe;
 	int pagenr;
@@ -2350,8 +2368,8 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 					 int need_check)
 {
 	struct btrfs_bio *bbio = rbio->bbio;
-	void *pointers[rbio->real_stripes];
-	DECLARE_BITMAP(pbitmap, rbio->stripe_npages);
+	void **pointers = rbio->finish_pointers;
+	unsigned long *pbitmap = rbio->finish_pbitmap;
 	int nr_data = rbio->nr_data;
 	int stripe;
 	int pagenr;

fs/btrfs/relocation.c

@@ -4299,7 +4299,7 @@ out:
 	return inode;
 }
 
-static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info)
+static struct reloc_control *alloc_reloc_control(void)
 {
 	struct reloc_control *rc;
 
@@ -4344,7 +4344,7 @@ static void describe_relocation(struct btrfs_fs_info *fs_info,
 		DESCRIBE_FLAG(RAID5,    "raid5");
 		DESCRIBE_FLAG(RAID6,    "raid6");
 		if (flags)
-			snprintf(buf, buf - bp + sizeof(buf), "|0x%llx", flags);
+			snprintf(bp, buf - bp + sizeof(buf), "|0x%llx", flags);
 #undef DESCRIBE_FLAG
 	}
 
@@ -4366,7 +4366,7 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 	int rw = 0;
 	int err = 0;
 
-	rc = alloc_reloc_control(fs_info);
+	rc = alloc_reloc_control();
 	if (!rc)
 		return -ENOMEM;
 
@@ -4562,7 +4562,7 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 	if (list_empty(&reloc_roots))
 		goto out;
 
-	rc = alloc_reloc_control(fs_info);
+	rc = alloc_reloc_control();
 	if (!rc) {
 		err = -ENOMEM;
 		goto out;

fs/btrfs/scrub.c

@@ -3984,6 +3984,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			spin_lock(&fs_info->unused_bgs_lock);
 			if (list_empty(&cache->bg_list)) {
 				btrfs_get_block_group(cache);
+				trace_btrfs_add_unused_block_group(cache);
 				list_add_tail(&cache->bg_list,
 					      &fs_info->unused_bgs);
 			}

fs/btrfs/send.c

@@ -235,6 +235,7 @@ struct orphan_dir_info {
 	struct rb_node node;
 	u64 ino;
 	u64 gen;
+	u64 last_dir_index_offset;
 };
 
 struct name_cache_entry {
@@ -2844,12 +2845,6 @@ add_orphan_dir_info(struct send_ctx *sctx, u64 dir_ino)
 	struct rb_node *parent = NULL;
 	struct orphan_dir_info *entry, *odi;
 
-	odi = kmalloc(sizeof(*odi), GFP_KERNEL);
-	if (!odi)
-		return ERR_PTR(-ENOMEM);
-	odi->ino = dir_ino;
-	odi->gen = 0;
-
 	while (*p) {
 		parent = *p;
 		entry = rb_entry(parent, struct orphan_dir_info, node);
@@ -2858,11 +2853,17 @@ add_orphan_dir_info(struct send_ctx *sctx, u64 dir_ino)
 		} else if (dir_ino > entry->ino) {
 			p = &(*p)->rb_right;
 		} else {
-			kfree(odi);
 			return entry;
 		}
 	}
 
+	odi = kmalloc(sizeof(*odi), GFP_KERNEL);
+	if (!odi)
+		return ERR_PTR(-ENOMEM);
+	odi->ino = dir_ino;
+	odi->gen = 0;
+	odi->last_dir_index_offset = 0;
+
 	rb_link_node(&odi->node, parent, p);
 	rb_insert_color(&odi->node, &sctx->orphan_dirs);
 	return odi;
@@ -2917,6 +2918,7 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen,
 	struct btrfs_key found_key;
 	struct btrfs_key loc;
 	struct btrfs_dir_item *di;
+	struct orphan_dir_info *odi = NULL;
 
 	/*
 	 * Don't try to rmdir the top/root subvolume dir.
@@ -2931,6 +2933,11 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen,
 	key.objectid = dir;
 	key.type = BTRFS_DIR_INDEX_KEY;
 	key.offset = 0;
+
+	odi = get_orphan_dir_info(sctx, dir);
+	if (odi)
+		key.offset = odi->last_dir_index_offset;
+
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 	if (ret < 0)
 		goto out;
@@ -2958,30 +2965,33 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen,
 
 		dm = get_waiting_dir_move(sctx, loc.objectid);
 		if (dm) {
-			struct orphan_dir_info *odi;
-
 			odi = add_orphan_dir_info(sctx, dir);
 			if (IS_ERR(odi)) {
 				ret = PTR_ERR(odi);
 				goto out;
 			}
 			odi->gen = dir_gen;
+			odi->last_dir_index_offset = found_key.offset;
 			dm->rmdir_ino = dir;
 			ret = 0;
 			goto out;
 		}
 
 		if (loc.objectid > send_progress) {
-			struct orphan_dir_info *odi;
-
-			odi = get_orphan_dir_info(sctx, dir);
-			free_orphan_dir_info(sctx, odi);
+			odi = add_orphan_dir_info(sctx, dir);
+			if (IS_ERR(odi)) {
+				ret = PTR_ERR(odi);
+				goto out;
+			}
+			odi->gen = dir_gen;
+			odi->last_dir_index_offset = found_key.offset;
 			ret = 0;
 			goto out;
 		}
 
 		path->slots[0]++;
 	}
+	free_orphan_dir_info(sctx, odi);
 
 	ret = 1;
 
@@ -3259,13 +3269,16 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
 
 	if (rmdir_ino) {
 		struct orphan_dir_info *odi;
+		u64 gen;
 
 		odi = get_orphan_dir_info(sctx, rmdir_ino);
 		if (!odi) {
 			/* already deleted */
 			goto finish;
 		}
-		ret = can_rmdir(sctx, rmdir_ino, odi->gen, sctx->cur_ino);
+		gen = odi->gen;
+
+		ret = can_rmdir(sctx, rmdir_ino, gen, sctx->cur_ino);
 		if (ret < 0)
 			goto out;
 		if (!ret)
@@ -3276,13 +3289,12 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
 			ret = -ENOMEM;
 			goto out;
 		}
-		ret = get_cur_path(sctx, rmdir_ino, odi->gen, name);
+		ret = get_cur_path(sctx, rmdir_ino, gen, name);
 		if (ret < 0)
 			goto out;
 		ret = send_rmdir(sctx, name);
 		if (ret < 0)
 			goto out;
-		free_orphan_dir_info(sctx, odi);
 	}
 
 finish:
@@ -6454,7 +6466,7 @@ static void btrfs_root_dec_send_in_progress(struct btrfs_root* root)
 	 */
 	if (root->send_in_progress < 0)
 		btrfs_err(root->fs_info,
-			  "send_in_progres unbalanced %d root %llu",
+			  "send_in_progress unbalanced %d root %llu",
 			  root->send_in_progress, root->root_key.objectid);
 	spin_unlock(&root->root_item_lock);
 }

fs/btrfs/super.c

@@ -323,6 +323,7 @@ enum {
 	Opt_ssd, Opt_nossd,
 	Opt_ssd_spread, Opt_nossd_spread,
 	Opt_subvol,
+	Opt_subvol_empty,
 	Opt_subvolid,
 	Opt_thread_pool,
 	Opt_treelog, Opt_notreelog,
@@ -388,6 +389,7 @@ static const match_table_t tokens = {
 	{Opt_ssd_spread, "ssd_spread"},
 	{Opt_nossd_spread, "nossd_spread"},
 	{Opt_subvol, "subvol=%s"},
+	{Opt_subvol_empty, "subvol="},
 	{Opt_subvolid, "subvolid=%s"},
 	{Opt_thread_pool, "thread_pool=%u"},
 	{Opt_treelog, "treelog"},
@@ -461,6 +463,7 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 			btrfs_set_opt(info->mount_opt, DEGRADED);
 			break;
 		case Opt_subvol:
+		case Opt_subvol_empty:
 		case Opt_subvolid:
 		case Opt_subvolrootid:
 		case Opt_device:
@@ -1782,10 +1785,8 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 	}
 
 	ret = btrfs_parse_options(fs_info, data, *flags);
-	if (ret) {
-		ret = -EINVAL;
+	if (ret)
 		goto restore;
-	}
 
 	btrfs_remount_begin(fs_info, old_opts, *flags);
 	btrfs_resize_thread_pool(fs_info,

fs/btrfs/sysfs.c

@@ -210,12 +210,42 @@ static struct attribute *btrfs_supported_feature_attrs[] = {
 	NULL
 };
 
+/*
+ * Features which depend on feature bits and may differ between each fs.
+ *
+ * /sys/fs/btrfs/features lists all available features of this kernel while
+ * /sys/fs/btrfs/UUID/features shows features of the fs which are enabled or
+ * can be changed online.
+ */
 static const struct attribute_group btrfs_feature_attr_group = {
 	.name = "features",
 	.is_visible = btrfs_feature_visible,
 	.attrs = btrfs_supported_feature_attrs,
 };
 
+static ssize_t rmdir_subvol_show(struct kobject *kobj,
+				 struct kobj_attribute *ka, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "0\n");
+}
+BTRFS_ATTR(static_feature, rmdir_subvol, rmdir_subvol_show);
+
+static struct attribute *btrfs_supported_static_feature_attrs[] = {
+	BTRFS_ATTR_PTR(static_feature, rmdir_subvol),
+	NULL
+};
+
+/*
+ * Features which only depend on kernel version.
+ *
+ * These are listed in /sys/fs/btrfs/features along with
+ * btrfs_feature_attr_group
+ */
+static const struct attribute_group btrfs_static_feature_attr_group = {
+	.name = "features",
+	.attrs = btrfs_supported_static_feature_attrs,
+};
+
 static ssize_t btrfs_show_u64(u64 *value_ptr, spinlock_t *lock, char *buf)
 {
 	u64 val;
@@ -514,10 +544,11 @@ static inline struct btrfs_fs_info *to_fs_info(struct kobject *kobj)
 }
 
 #define NUM_FEATURE_BITS 64
-static char btrfs_unknown_feature_names[3][NUM_FEATURE_BITS][13];
-static struct btrfs_feature_attr btrfs_feature_attrs[3][NUM_FEATURE_BITS];
+#define BTRFS_FEATURE_NAME_MAX 13
+static char btrfs_unknown_feature_names[FEAT_MAX][NUM_FEATURE_BITS][BTRFS_FEATURE_NAME_MAX];
+static struct btrfs_feature_attr btrfs_feature_attrs[FEAT_MAX][NUM_FEATURE_BITS];
 
-static const u64 supported_feature_masks[3] = {
+static const u64 supported_feature_masks[FEAT_MAX] = {
 	[FEAT_COMPAT]    = BTRFS_FEATURE_COMPAT_SUPP,
 	[FEAT_COMPAT_RO] = BTRFS_FEATURE_COMPAT_RO_SUPP,
 	[FEAT_INCOMPAT]  = BTRFS_FEATURE_INCOMPAT_SUPP,
@@ -589,7 +620,7 @@ void btrfs_sysfs_remove_fsid(struct btrfs_fs_devices *fs_devs)
 		return;
 	}
 
-	list_for_each_entry(fs_devs, fs_uuids, list) {
+	list_for_each_entry(fs_devs, fs_uuids, fs_list) {
 		__btrfs_sysfs_remove_fsid(fs_devs);
 	}
 }
@@ -609,7 +640,7 @@ void btrfs_sysfs_remove_mounted(struct btrfs_fs_info *fs_info)
 	btrfs_sysfs_rm_device_link(fs_info->fs_devices, NULL);
 }
 
-const char * const btrfs_feature_set_names[3] = {
+const char * const btrfs_feature_set_names[FEAT_MAX] = {
 	[FEAT_COMPAT]	 = "compat",
 	[FEAT_COMPAT_RO] = "compat_ro",
 	[FEAT_INCOMPAT]	 = "incompat",
@@ -673,7 +704,7 @@ static void init_feature_attrs(void)
 			if (fa->kobj_attr.attr.name)
 				continue;
 
-			snprintf(name, 13, "%s:%u",
+			snprintf(name, BTRFS_FEATURE_NAME_MAX, "%s:%u",
 				 btrfs_feature_set_names[set], i);
 
 			fa->kobj_attr.attr.name = name;
@@ -900,8 +931,15 @@ int __init btrfs_init_sysfs(void)
 	ret = sysfs_create_group(&btrfs_kset->kobj, &btrfs_feature_attr_group);
 	if (ret)
 		goto out2;
+	ret = sysfs_merge_group(&btrfs_kset->kobj,
+				&btrfs_static_feature_attr_group);
+	if (ret)
+		goto out_remove_group;
 
 	return 0;
+
+out_remove_group:
+	sysfs_remove_group(&btrfs_kset->kobj, &btrfs_feature_attr_group);
 out2:
 	debugfs_remove_recursive(btrfs_debugfs_root_dentry);
 out1:
@@ -912,6 +950,8 @@ out1:
 
 void __cold btrfs_exit_sysfs(void)
 {
+	sysfs_unmerge_group(&btrfs_kset->kobj,
+			    &btrfs_static_feature_attr_group);
 	sysfs_remove_group(&btrfs_kset->kobj, &btrfs_feature_attr_group);
 	kset_unregister(btrfs_kset);
 	debugfs_remove_recursive(btrfs_debugfs_root_dentry);

fs/btrfs/sysfs.h

@@ -9,7 +9,7 @@
 extern u64 btrfs_debugfs_test;
 
 enum btrfs_feature_set {
-	FEAT_COMPAT,
+	FEAT_COMPAT = 0,
 	FEAT_COMPAT_RO,
 	FEAT_INCOMPAT,
 	FEAT_MAX
@@ -77,7 +77,7 @@ attr_to_btrfs_feature_attr(struct attribute *attr)
 }
 
 char *btrfs_printable_features(enum btrfs_feature_set set, u64 flags);
-extern const char * const btrfs_feature_set_names[3];
+extern const char * const btrfs_feature_set_names[FEAT_MAX];
 extern struct kobj_type space_info_ktype;
 extern struct kobj_type btrfs_raid_ktype;
 int btrfs_sysfs_add_device_link(struct btrfs_fs_devices *fs_devices,

fs/btrfs/tests/btrfs-tests.c

@@ -219,11 +219,13 @@ void btrfs_free_dummy_block_group(struct btrfs_block_group_cache *cache)
 	kfree(cache);
 }
 
-void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans)
+void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans,
+			    struct btrfs_fs_info *fs_info)
 {
 	memset(trans, 0, sizeof(*trans));
 	trans->transid = 1;
 	trans->type = __TRANS_DUMMY;
+	trans->fs_info = fs_info;
 }
 
 int btrfs_run_sanity_tests(void)

fs/btrfs/tests/btrfs-tests.h

@@ -9,7 +9,8 @@
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 int btrfs_run_sanity_tests(void);
 
-#define test_msg(fmt, ...) pr_info("BTRFS: selftest: " fmt, ##__VA_ARGS__)
+#define test_msg(fmt, ...) pr_info("BTRFS: selftest: " fmt "\n", ##__VA_ARGS__)
+#define test_err(fmt, ...) pr_err("BTRFS: selftest: " fmt "\n", ##__VA_ARGS__)
 
 struct btrfs_root;
 struct btrfs_trans_handle;
@@ -28,7 +29,8 @@ void btrfs_free_dummy_root(struct btrfs_root *root);
 struct btrfs_block_group_cache *
 btrfs_alloc_dummy_block_group(struct btrfs_fs_info *fs_info, unsigned long length);
 void btrfs_free_dummy_block_group(struct btrfs_block_group_cache *cache);
-void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans);
+void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans,
+			    struct btrfs_fs_info *fs_info);
 #else
 static inline int btrfs_run_sanity_tests(void)
 {

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

@@ -26,31 +26,31 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	u32 value_len = strlen(value);
 	int ret = 0;
 
-	test_msg("Running btrfs_split_item tests\n");
+	test_msg("running btrfs_split_item tests");
 
 	fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize);
 	if (!fs_info) {
-		test_msg("Could not allocate fs_info\n");
+		test_err("could not allocate fs_info");
 		return -ENOMEM;
 	}
 
 	root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(root)) {
-		test_msg("Could not allocate root\n");
+		test_err("could not allocate root");
 		ret = PTR_ERR(root);
 		goto out;
 	}
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		test_msg("Could not allocate path\n");
+		test_err("could not allocate path");
 		ret = -ENOMEM;
 		goto out;
 	}
 
 	path->nodes[0] = eb = alloc_dummy_extent_buffer(fs_info, nodesize);
 	if (!eb) {
-		test_msg("Could not allocate dummy buffer\n");
+		test_err("could not allocate dummy buffer");
 		ret = -ENOMEM;
 		goto out;
 	}
@@ -75,7 +75,7 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	 */
 	ret = btrfs_split_item(NULL, root, path, &key, 17);
 	if (ret) {
-		test_msg("Split item failed %d\n", ret);
+		test_err("split item failed %d", ret);
 		goto out;
 	}
 
@@ -86,14 +86,14 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	btrfs_item_key_to_cpu(eb, &key, 0);
 	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
 	    key.offset != 0) {
-		test_msg("Invalid key at slot 0\n");
+		test_err("invalid key at slot 0");
 		ret = -EINVAL;
 		goto out;
 	}
 
 	item = btrfs_item_nr(0);
 	if (btrfs_item_size(eb, item) != strlen(split1)) {
-		test_msg("Invalid len in the first split\n");
+		test_err("invalid len in the first split");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -101,8 +101,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 0),
 			   strlen(split1));
 	if (memcmp(buf, split1, strlen(split1))) {
-		test_msg("Data in the buffer doesn't match what it should "
-			 "in the first split have='%.*s' want '%s'\n",
+		test_err(
+"data in the buffer doesn't match what it should in the first split have='%.*s' want '%s'",
 			 (int)strlen(split1), buf, split1);
 		ret = -EINVAL;
 		goto out;
@@ -111,14 +111,14 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	btrfs_item_key_to_cpu(eb, &key, 1);
 	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
 	    key.offset != 3) {
-		test_msg("Invalid key at slot 1\n");
+		test_err("invalid key at slot 1");
 		ret = -EINVAL;
 		goto out;
 	}
 
 	item = btrfs_item_nr(1);
 	if (btrfs_item_size(eb, item) != strlen(split2)) {
-		test_msg("Invalid len in the second split\n");
+		test_err("invalid len in the second split");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -126,8 +126,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 1),
 			   strlen(split2));
 	if (memcmp(buf, split2, strlen(split2))) {
-		test_msg("Data in the buffer doesn't match what it should "
-			 "in the second split\n");
+		test_err(
+	"data in the buffer doesn't match what it should in the second split");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -136,21 +136,21 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	/* Do it again so we test memmoving the other items in the leaf */
 	ret = btrfs_split_item(NULL, root, path, &key, 4);
 	if (ret) {
-		test_msg("Second split item failed %d\n", ret);
+		test_err("second split item failed %d", ret);
 		goto out;
 	}
 
 	btrfs_item_key_to_cpu(eb, &key, 0);
 	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
 	    key.offset != 0) {
-		test_msg("Invalid key at slot 0\n");
+		test_err("invalid key at slot 0");
 		ret = -EINVAL;
 		goto out;
 	}
 
 	item = btrfs_item_nr(0);
 	if (btrfs_item_size(eb, item) != strlen(split3)) {
-		test_msg("Invalid len in the first split\n");
+		test_err("invalid len in the first split");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -158,8 +158,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 0),
 			   strlen(split3));
 	if (memcmp(buf, split3, strlen(split3))) {
-		test_msg("Data in the buffer doesn't match what it should "
-			 "in the third split");
+		test_err(
+	"data in the buffer doesn't match what it should in the third split");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -167,14 +167,14 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	btrfs_item_key_to_cpu(eb, &key, 1);
 	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
 	    key.offset != 1) {
-		test_msg("Invalid key at slot 1\n");
+		test_err("invalid key at slot 1");
 		ret = -EINVAL;
 		goto out;
 	}
 
 	item = btrfs_item_nr(1);
 	if (btrfs_item_size(eb, item) != strlen(split4)) {
-		test_msg("Invalid len in the second split\n");
+		test_err("invalid len in the second split");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -182,8 +182,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 1),
 			   strlen(split4));
 	if (memcmp(buf, split4, strlen(split4))) {
-		test_msg("Data in the buffer doesn't match what it should "
-			 "in the fourth split\n");
+		test_err(
+	"data in the buffer doesn't match what it should in the fourth split");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -191,14 +191,14 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	btrfs_item_key_to_cpu(eb, &key, 2);
 	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
 	    key.offset != 3) {
-		test_msg("Invalid key at slot 2\n");
+		test_err("invalid key at slot 2");
 		ret = -EINVAL;
 		goto out;
 	}
 
 	item = btrfs_item_nr(2);
 	if (btrfs_item_size(eb, item) != strlen(split2)) {
-		test_msg("Invalid len in the second split\n");
+		test_err("invalid len in the second split");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -206,8 +206,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 2),
 			   strlen(split2));
 	if (memcmp(buf, split2, strlen(split2))) {
-		test_msg("Data in the buffer doesn't match what it should "
-			 "in the last chunk\n");
+		test_err(
+	"data in the buffer doesn't match what it should in the last chunk");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -220,6 +220,6 @@ out:
 
 int btrfs_test_extent_buffer_operations(u32 sectorsize, u32 nodesize)
 {
-	test_msg("Running extent buffer operation tests\n");
+	test_msg("running extent buffer operation tests");
 	return test_btrfs_split_item(sectorsize, nodesize);
 }

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

@@ -46,7 +46,9 @@ static noinline int process_page_range(struct inode *inode, u64 start, u64 end,
 		cond_resched();
 		loops++;
 		if (loops > 100000) {
-			printk(KERN_ERR "stuck in a loop, start %Lu, end %Lu, nr_pages %lu, ret %d\n", start, end, nr_pages, ret);
+			printk(KERN_ERR
+		"stuck in a loop, start %llu, end %llu, nr_pages %lu, ret %d\n",
+				start, end, nr_pages, ret);
 			break;
 		}
 	}
@@ -66,11 +68,11 @@ static int test_find_delalloc(u32 sectorsize)
 	u64 found;
 	int ret = -EINVAL;
 
-	test_msg("Running find delalloc tests\n");
+	test_msg("running find delalloc tests");
 
 	inode = btrfs_new_test_inode();
 	if (!inode) {
-		test_msg("Failed to allocate test inode\n");
+		test_err("failed to allocate test inode");
 		return -ENOMEM;
 	}
 
@@ -84,7 +86,7 @@ static int test_find_delalloc(u32 sectorsize)
 	for (index = 0; index < (total_dirty >> PAGE_SHIFT); index++) {
 		page = find_or_create_page(inode->i_mapping, index, GFP_KERNEL);
 		if (!page) {
-			test_msg("Failed to allocate test page\n");
+			test_err("failed to allocate test page");
 			ret = -ENOMEM;
 			goto out;
 		}
@@ -107,11 +109,11 @@ static int test_find_delalloc(u32 sectorsize)
 	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (!found) {
-		test_msg("Should have found at least one delalloc\n");
+		test_err("should have found at least one delalloc");
 		goto out_bits;
 	}
 	if (start != 0 || end != (sectorsize - 1)) {
-		test_msg("Expected start 0 end %u, got start %llu end %llu\n",
+		test_err("expected start 0 end %u, got start %llu end %llu",
 			sectorsize - 1, start, end);
 		goto out_bits;
 	}
@@ -129,7 +131,7 @@ static int test_find_delalloc(u32 sectorsize)
 	locked_page = find_lock_page(inode->i_mapping,
 				     test_start >> PAGE_SHIFT);
 	if (!locked_page) {
-		test_msg("Couldn't find the locked page\n");
+		test_err("couldn't find the locked page");
 		goto out_bits;
 	}
 	set_extent_delalloc(&tmp, sectorsize, max_bytes - 1, 0, NULL);
@@ -138,17 +140,17 @@ static int test_find_delalloc(u32 sectorsize)
 	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (!found) {
-		test_msg("Couldn't find delalloc in our range\n");
+		test_err("couldn't find delalloc in our range");
 		goto out_bits;
 	}
 	if (start != test_start || end != max_bytes - 1) {
-		test_msg("Expected start %Lu end %Lu, got start %Lu, end "
-			 "%Lu\n", test_start, max_bytes - 1, start, end);
+		test_err("expected start %llu end %llu, got start %llu, end %llu",
+				test_start, max_bytes - 1, start, end);
 		goto out_bits;
 	}
 	if (process_page_range(inode, start, end,
 			       PROCESS_TEST_LOCKED | PROCESS_UNLOCK)) {
-		test_msg("There were unlocked pages in the range\n");
+		test_err("there were unlocked pages in the range");
 		goto out_bits;
 	}
 	unlock_extent(&tmp, start, end);
@@ -164,7 +166,7 @@ static int test_find_delalloc(u32 sectorsize)
 	locked_page = find_lock_page(inode->i_mapping, test_start >>
 				     PAGE_SHIFT);
 	if (!locked_page) {
-		test_msg("Couldn't find the locked page\n");
+		test_err("couldn't find the locked page");
 		goto out_bits;
 	}
 	start = test_start;
@@ -172,11 +174,11 @@ static int test_find_delalloc(u32 sectorsize)
 	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (found) {
-		test_msg("Found range when we shouldn't have\n");
+		test_err("found range when we shouldn't have");
 		goto out_bits;
 	}
 	if (end != (u64)-1) {
-		test_msg("Did not return the proper end offset\n");
+		test_err("did not return the proper end offset");
 		goto out_bits;
 	}
 
@@ -193,17 +195,17 @@ static int test_find_delalloc(u32 sectorsize)
 	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (!found) {
-		test_msg("Didn't find our range\n");
+		test_err("didn't find our range");
 		goto out_bits;
 	}
 	if (start != test_start || end != total_dirty - 1) {
-		test_msg("Expected start %Lu end %Lu, got start %Lu end %Lu\n",
+		test_err("expected start %llu end %llu, got start %llu end %llu",
 			 test_start, total_dirty - 1, start, end);
 		goto out_bits;
 	}
 	if (process_page_range(inode, start, end,
 			       PROCESS_TEST_LOCKED | PROCESS_UNLOCK)) {
-		test_msg("Pages in range were not all locked\n");
+		test_err("pages in range were not all locked");
 		goto out_bits;
 	}
 	unlock_extent(&tmp, start, end);
@@ -215,7 +217,7 @@ static int test_find_delalloc(u32 sectorsize)
 	page = find_get_page(inode->i_mapping,
 			     (max_bytes + SZ_1M) >> PAGE_SHIFT);
 	if (!page) {
-		test_msg("Couldn't find our page\n");
+		test_err("couldn't find our page");
 		goto out_bits;
 	}
 	ClearPageDirty(page);
@@ -234,18 +236,17 @@ static int test_find_delalloc(u32 sectorsize)
 	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (!found) {
-		test_msg("Didn't find our range\n");
+		test_err("didn't find our range");
 		goto out_bits;
 	}
 	if (start != test_start && end != test_start + PAGE_SIZE - 1) {
-		test_msg("Expected start %Lu end %Lu, got start %Lu end %Lu\n",
-			 test_start, test_start + PAGE_SIZE - 1, start,
-			 end);
+		test_err("expected start %llu end %llu, got start %llu end %llu",
+			 test_start, test_start + PAGE_SIZE - 1, start, end);
 		goto out_bits;
 	}
 	if (process_page_range(inode, start, end, PROCESS_TEST_LOCKED |
 			       PROCESS_UNLOCK)) {
-		test_msg("Pages in range were not all locked\n");
+		test_err("pages in range were not all locked");
 		goto out_bits;
 	}
 	ret = 0;
@@ -271,14 +272,14 @@ static int check_eb_bitmap(unsigned long *bitmap, struct extent_buffer *eb,
 		bit = !!test_bit(i, bitmap);
 		bit1 = !!extent_buffer_test_bit(eb, 0, i);
 		if (bit1 != bit) {
-			test_msg("Bits do not match\n");
+			test_err("bits do not match");
 			return -EINVAL;
 		}
 
 		bit1 = !!extent_buffer_test_bit(eb, i / BITS_PER_BYTE,
 						i % BITS_PER_BYTE);
 		if (bit1 != bit) {
-			test_msg("Offset bits do not match\n");
+			test_err("offset bits do not match");
 			return -EINVAL;
 		}
 	}
@@ -295,7 +296,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 	memset(bitmap, 0, len);
 	memzero_extent_buffer(eb, 0, len);
 	if (memcmp_extent_buffer(eb, bitmap, 0, len) != 0) {
-		test_msg("Bitmap was not zeroed\n");
+		test_err("bitmap was not zeroed");
 		return -EINVAL;
 	}
 
@@ -303,7 +304,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 	extent_buffer_bitmap_set(eb, 0, 0, len * BITS_PER_BYTE);
 	ret = check_eb_bitmap(bitmap, eb, len);
 	if (ret) {
-		test_msg("Setting all bits failed\n");
+		test_err("setting all bits failed");
 		return ret;
 	}
 
@@ -311,7 +312,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 	extent_buffer_bitmap_clear(eb, 0, 0, len * BITS_PER_BYTE);
 	ret = check_eb_bitmap(bitmap, eb, len);
 	if (ret) {
-		test_msg("Clearing all bits failed\n");
+		test_err("clearing all bits failed");
 		return ret;
 	}
 
@@ -324,7 +325,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 					sizeof(long) * BITS_PER_BYTE);
 		ret = check_eb_bitmap(bitmap, eb, len);
 		if (ret) {
-			test_msg("Setting straddling pages failed\n");
+			test_err("setting straddling pages failed");
 			return ret;
 		}
 
@@ -337,7 +338,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 					sizeof(long) * BITS_PER_BYTE);
 		ret = check_eb_bitmap(bitmap, eb, len);
 		if (ret) {
-			test_msg("Clearing straddling pages failed\n");
+			test_err("clearing straddling pages failed");
 			return ret;
 		}
 	}
@@ -361,7 +362,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb,
 
 	ret = check_eb_bitmap(bitmap, eb, len);
 	if (ret) {
-		test_msg("Random bit pattern failed\n");
+		test_err("random bit pattern failed");
 		return ret;
 	}
 
@@ -376,7 +377,7 @@ static int test_eb_bitmaps(u32 sectorsize, u32 nodesize)
 	struct extent_buffer *eb;
 	int ret;
 
-	test_msg("Running extent buffer bitmap tests\n");
+	test_msg("running extent buffer bitmap tests");
 
 	/*
 	 * In ppc64, sectorsize can be 64K, thus 4 * 64K will be larger than
@@ -389,13 +390,13 @@ static int test_eb_bitmaps(u32 sectorsize, u32 nodesize)
 
 	bitmap = kmalloc(len, GFP_KERNEL);
 	if (!bitmap) {
-		test_msg("Couldn't allocate test bitmap\n");
+		test_err("couldn't allocate test bitmap");
 		return -ENOMEM;
 	}
 
 	eb = __alloc_dummy_extent_buffer(fs_info, 0, len);
 	if (!eb) {
-		test_msg("Couldn't allocate test extent buffer\n");
+		test_err("couldn't allocate test extent buffer");
 		kfree(bitmap);
 		return -ENOMEM;
 	}
@@ -408,7 +409,7 @@ static int test_eb_bitmaps(u32 sectorsize, u32 nodesize)
 	free_extent_buffer(eb);
 	eb = __alloc_dummy_extent_buffer(NULL, nodesize / 2, len);
 	if (!eb) {
-		test_msg("Couldn't allocate test extent buffer\n");
+		test_err("couldn't allocate test extent buffer");
 		kfree(bitmap);
 		return -ENOMEM;
 	}
@@ -424,7 +425,7 @@ int btrfs_test_extent_io(u32 sectorsize, u32 nodesize)
 {
 	int ret;
 
-	test_msg("Running extent I/O tests\n");
+	test_msg("running extent I/O tests");
 
 	ret = test_find_delalloc(sectorsize);
 	if (ret)
@@ -432,6 +433,6 @@ int btrfs_test_extent_io(u32 sectorsize, u32 nodesize)
 
 	ret = test_eb_bitmaps(sectorsize, nodesize);
 out:
-	test_msg("Extent I/O tests finished\n");
+	test_msg("extent I/O tests finished");
 	return ret;
 }

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

@@ -19,8 +19,8 @@ static void free_extent_map_tree(struct extent_map_tree *em_tree)
 
 #ifdef CONFIG_BTRFS_DEBUG
 		if (refcount_read(&em->refs) != 1) {
-			test_msg(
-"em leak: em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx) refs %d\n",
+			test_err(
+"em leak: em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx) refs %d",
 				 em->start, em->len, em->block_start,
 				 em->block_len, refcount_read(&em->refs));
 
@@ -47,7 +47,8 @@ static void free_extent_map_tree(struct extent_map_tree *em_tree)
  *                                    ->add_extent_mapping(0, 16K)
  *                                    -> #handle -EEXIST
  */
-static void test_case_1(struct extent_map_tree *em_tree)
+static void test_case_1(struct btrfs_fs_info *fs_info,
+		struct extent_map_tree *em_tree)
 {
 	struct extent_map *em;
 	u64 start = 0;
@@ -90,14 +91,14 @@ static void test_case_1(struct extent_map_tree *em_tree)
 	em->len = len;
 	em->block_start = start;
 	em->block_len = len;
-	ret = btrfs_add_extent_mapping(em_tree, &em, em->start, em->len);
+	ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len);
 	if (ret)
-		test_msg("case1 [%llu %llu]: ret %d\n", start, start + len, ret);
+		test_err("case1 [%llu %llu]: ret %d", start, start + len, ret);
 	if (em &&
 	    (em->start != 0 || extent_map_end(em) != SZ_16K ||
 	     em->block_start != 0 || em->block_len != SZ_16K))
-		test_msg(
-"case1 [%llu %llu]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu\n",
+		test_err(
+"case1 [%llu %llu]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu",
 			 start, start + len, ret, em->start, em->len,
 			 em->block_start, em->block_len);
 	free_extent_map(em);
@@ -112,7 +113,8 @@ out:
  * Reading the inline ending up with EEXIST, ie. read an inline
  * extent and discard page cache and read it again.
  */
-static void test_case_2(struct extent_map_tree *em_tree)
+static void test_case_2(struct btrfs_fs_info *fs_info,
+		struct extent_map_tree *em_tree)
 {
 	struct extent_map *em;
 	int ret;
@@ -153,14 +155,14 @@ static void test_case_2(struct extent_map_tree *em_tree)
 	em->len = SZ_1K;
 	em->block_start = EXTENT_MAP_INLINE;
 	em->block_len = (u64)-1;
-	ret = btrfs_add_extent_mapping(em_tree, &em, em->start, em->len);
+	ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len);
 	if (ret)
-		test_msg("case2 [0 1K]: ret %d\n", ret);
+		test_err("case2 [0 1K]: ret %d", ret);
 	if (em &&
 	    (em->start != 0 || extent_map_end(em) != SZ_1K ||
 	     em->block_start != EXTENT_MAP_INLINE || em->block_len != (u64)-1))
-		test_msg(
-"case2 [0 1K]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu\n",
+		test_err(
+"case2 [0 1K]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu",
 			 ret, em->start, em->len, em->block_start,
 			 em->block_len);
 	free_extent_map(em);
@@ -169,7 +171,8 @@ out:
 	free_extent_map_tree(em_tree);
 }
 
-static void __test_case_3(struct extent_map_tree *em_tree, u64 start)
+static void __test_case_3(struct btrfs_fs_info *fs_info,
+		struct extent_map_tree *em_tree, u64 start)
 {
 	struct extent_map *em;
 	u64 len = SZ_4K;
@@ -198,9 +201,9 @@ static void __test_case_3(struct extent_map_tree *em_tree, u64 start)
 	em->len = SZ_16K;
 	em->block_start = 0;
 	em->block_len = SZ_16K;
-	ret = btrfs_add_extent_mapping(em_tree, &em, start, len);
+	ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len);
 	if (ret)
-		test_msg("case3 [0x%llx 0x%llx): ret %d\n",
+		test_err("case3 [0x%llx 0x%llx): ret %d",
 			 start, start + len, ret);
 	/*
 	 * Since bytes within em are contiguous, em->block_start is identical to
@@ -209,8 +212,8 @@ static void __test_case_3(struct extent_map_tree *em_tree, u64 start)
 	if (em &&
 	    (start < em->start || start + len > extent_map_end(em) ||
 	     em->start != em->block_start || em->len != em->block_len))
-		test_msg(
-"case3 [0x%llx 0x%llx): ret %d em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)\n",
+		test_err(
+"case3 [0x%llx 0x%llx): ret %d em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)",
 			 start, start + len, ret, em->start, em->len,
 			 em->block_start, em->block_len);
 	free_extent_map(em);
@@ -235,14 +238,16 @@ out:
  *   -> add_extent_mapping()
  *                            -> add_extent_mapping()
  */
-static void test_case_3(struct extent_map_tree *em_tree)
+static void test_case_3(struct btrfs_fs_info *fs_info,
+		struct extent_map_tree *em_tree)
 {
-	__test_case_3(em_tree, 0);
-	__test_case_3(em_tree, SZ_8K);
-	__test_case_3(em_tree, (12 * 1024ULL));
+	__test_case_3(fs_info, em_tree, 0);
+	__test_case_3(fs_info, em_tree, SZ_8K);
+	__test_case_3(fs_info, em_tree, (12 * 1024ULL));
 }
 
-static void __test_case_4(struct extent_map_tree *em_tree, u64 start)
+static void __test_case_4(struct btrfs_fs_info *fs_info,
+		struct extent_map_tree *em_tree, u64 start)
 {
 	struct extent_map *em;
 	u64 len = SZ_4K;
@@ -283,14 +288,14 @@ static void __test_case_4(struct extent_map_tree *em_tree, u64 start)
 	em->len = SZ_32K;
 	em->block_start = 0;
 	em->block_len = SZ_32K;
-	ret = btrfs_add_extent_mapping(em_tree, &em, start, len);
+	ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len);
 	if (ret)
-		test_msg("case4 [0x%llx 0x%llx): ret %d\n",
+		test_err("case4 [0x%llx 0x%llx): ret %d",
 			 start, len, ret);
 	if (em &&
 	    (start < em->start || start + len > extent_map_end(em)))
-		test_msg(
-"case4 [0x%llx 0x%llx): ret %d, added wrong em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)\n",
+		test_err(
+"case4 [0x%llx 0x%llx): ret %d, added wrong em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)",
 			 start, len, ret, em->start, em->len, em->block_start,
 			 em->block_len);
 	free_extent_map(em);
@@ -324,30 +329,45 @@ out:
  *                                             # handle -EEXIST when adding
  *                                             # [0, 32K)
  */
-static void test_case_4(struct extent_map_tree *em_tree)
+static void test_case_4(struct btrfs_fs_info *fs_info,
+		struct extent_map_tree *em_tree)
 {
-	__test_case_4(em_tree, 0);
-	__test_case_4(em_tree, SZ_4K);
+	__test_case_4(fs_info, em_tree, 0);
+	__test_case_4(fs_info, em_tree, SZ_4K);
 }
 
 int btrfs_test_extent_map(void)
 {
+	struct btrfs_fs_info *fs_info = NULL;
 	struct extent_map_tree *em_tree;
 
-	test_msg("Running extent_map tests\n");
+	test_msg("running extent_map tests");
+
+	/*
+	 * Note: the fs_info is not set up completely, we only need
+	 * fs_info::fsid for the tracepoint.
+	 */
+	fs_info = btrfs_alloc_dummy_fs_info(PAGE_SIZE, PAGE_SIZE);
+	if (!fs_info) {
+		test_msg("Couldn't allocate dummy fs info");
+		return -ENOMEM;
+	}
 
 	em_tree = kzalloc(sizeof(*em_tree), GFP_KERNEL);
 	if (!em_tree)
 		/* Skip the test on error. */
-		return 0;
+		goto out;
 
 	extent_map_tree_init(em_tree);
 
-	test_case_1(em_tree);
-	test_case_2(em_tree);
-	test_case_3(em_tree);
-	test_case_4(em_tree);
+	test_case_1(fs_info, em_tree);
+	test_case_2(fs_info, em_tree);
+	test_case_3(fs_info, em_tree);
+	test_case_4(fs_info, em_tree);
 
 	kfree(em_tree);
+out:
+	btrfs_free_dummy_fs_info(fs_info);
+
 	return 0;
 }

fs/btrfs/tests/free-space-tests.c

@@ -20,63 +20,63 @@ static int test_extents(struct btrfs_block_group_cache *cache)
 {
 	int ret = 0;
 
-	test_msg("Running extent only tests\n");
+	test_msg("running extent only tests");
 
 	/* First just make sure we can remove an entire entry */
 	ret = btrfs_add_free_space(cache, 0, SZ_4M);
 	if (ret) {
-		test_msg("Error adding initial extents %d\n", ret);
+		test_err("error adding initial extents %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, 0, SZ_4M);
 	if (ret) {
-		test_msg("Error removing extent %d\n", ret);
+		test_err("error removing extent %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, 0, SZ_4M)) {
-		test_msg("Full remove left some lingering space\n");
+		test_err("full remove left some lingering space");
 		return -1;
 	}
 
 	/* Ok edge and middle cases now */
 	ret = btrfs_add_free_space(cache, 0, SZ_4M);
 	if (ret) {
-		test_msg("Error adding half extent %d\n", ret);
+		test_err("error adding half extent %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, 3 * SZ_1M, SZ_1M);
 	if (ret) {
-		test_msg("Error removing tail end %d\n", ret);
+		test_err("error removing tail end %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, 0, SZ_1M);
 	if (ret) {
-		test_msg("Error removing front end %d\n", ret);
+		test_err("error removing front end %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, SZ_2M, 4096);
 	if (ret) {
-		test_msg("Error removing middle piece %d\n", ret);
+		test_err("error removing middle piece %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, 0, SZ_1M)) {
-		test_msg("Still have space at the front\n");
+		test_err("still have space at the front");
 		return -1;
 	}
 
 	if (test_check_exists(cache, SZ_2M, 4096)) {
-		test_msg("Still have space in the middle\n");
+		test_err("still have space in the middle");
 		return -1;
 	}
 
 	if (test_check_exists(cache, 3 * SZ_1M, SZ_1M)) {
-		test_msg("Still have space at the end\n");
+		test_err("still have space at the end");
 		return -1;
 	}
 
@@ -92,34 +92,34 @@ static int test_bitmaps(struct btrfs_block_group_cache *cache,
 	u64 next_bitmap_offset;
 	int ret;
 
-	test_msg("Running bitmap only tests\n");
+	test_msg("running bitmap only tests");
 
 	ret = test_add_free_space_entry(cache, 0, SZ_4M, 1);
 	if (ret) {
-		test_msg("Couldn't create a bitmap entry %d\n", ret);
+		test_err("couldn't create a bitmap entry %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, 0, SZ_4M);
 	if (ret) {
-		test_msg("Error removing bitmap full range %d\n", ret);
+		test_err("error removing bitmap full range %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, 0, SZ_4M)) {
-		test_msg("Left some space in bitmap\n");
+		test_err("left some space in bitmap");
 		return -1;
 	}
 
 	ret = test_add_free_space_entry(cache, 0, SZ_4M, 1);
 	if (ret) {
-		test_msg("Couldn't add to our bitmap entry %d\n", ret);
+		test_err("couldn't add to our bitmap entry %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, SZ_1M, SZ_2M);
 	if (ret) {
-		test_msg("Couldn't remove middle chunk %d\n", ret);
+		test_err("couldn't remove middle chunk %d", ret);
 		return ret;
 	}
 
@@ -133,19 +133,19 @@ static int test_bitmaps(struct btrfs_block_group_cache *cache,
 	ret = test_add_free_space_entry(cache, next_bitmap_offset - SZ_2M,
 					SZ_4M, 1);
 	if (ret) {
-		test_msg("Couldn't add space that straddles two bitmaps %d\n",
+		test_err("couldn't add space that straddles two bitmaps %d",
 				ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, next_bitmap_offset - SZ_1M, SZ_2M);
 	if (ret) {
-		test_msg("Couldn't remove overlapping space %d\n", ret);
+		test_err("couldn't remove overlapping space %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, next_bitmap_offset - SZ_1M, SZ_2M)) {
-		test_msg("Left some space when removing overlapping\n");
+		test_err("left some space when removing overlapping");
 		return -1;
 	}
 
@@ -161,7 +161,7 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache,
 	u64 bitmap_offset = (u64)(BITS_PER_BITMAP * sectorsize);
 	int ret;
 
-	test_msg("Running bitmap and extent tests\n");
+	test_msg("running bitmap and extent tests");
 
 	/*
 	 * First let's do something simple, an extent at the same offset as the
@@ -170,42 +170,42 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache,
 	 */
 	ret = test_add_free_space_entry(cache, SZ_4M, SZ_1M, 1);
 	if (ret) {
-		test_msg("Couldn't create bitmap entry %d\n", ret);
+		test_err("couldn't create bitmap entry %d", ret);
 		return ret;
 	}
 
 	ret = test_add_free_space_entry(cache, 0, SZ_1M, 0);
 	if (ret) {
-		test_msg("Couldn't add extent entry %d\n", ret);
+		test_err("couldn't add extent entry %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, 0, SZ_1M);
 	if (ret) {
-		test_msg("Couldn't remove extent entry %d\n", ret);
+		test_err("couldn't remove extent entry %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, 0, SZ_1M)) {
-		test_msg("Left remnants after our remove\n");
+		test_err("left remnants after our remove");
 		return -1;
 	}
 
 	/* Now to add back the extent entry and remove from the bitmap */
 	ret = test_add_free_space_entry(cache, 0, SZ_1M, 0);
 	if (ret) {
-		test_msg("Couldn't re-add extent entry %d\n", ret);
+		test_err("couldn't re-add extent entry %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, SZ_4M, SZ_1M);
 	if (ret) {
-		test_msg("Couldn't remove from bitmap %d\n", ret);
+		test_err("couldn't remove from bitmap %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, SZ_4M, SZ_1M)) {
-		test_msg("Left remnants in the bitmap\n");
+		test_err("left remnants in the bitmap");
 		return -1;
 	}
 
@@ -215,18 +215,18 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache,
 	 */
 	ret = test_add_free_space_entry(cache, SZ_1M, SZ_4M, 1);
 	if (ret) {
-		test_msg("Couldn't add to a bitmap %d\n", ret);
+		test_err("couldn't add to a bitmap %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, SZ_512K, 3 * SZ_1M);
 	if (ret) {
-		test_msg("Couldn't remove overlapping space %d\n", ret);
+		test_err("couldn't remove overlapping space %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, SZ_512K, 3 * SZ_1M)) {
-		test_msg("Left over pieces after removing overlapping\n");
+		test_err("left over pieces after removing overlapping");
 		return -1;
 	}
 
@@ -235,24 +235,24 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache,
 	/* Now with the extent entry offset into the bitmap */
 	ret = test_add_free_space_entry(cache, SZ_4M, SZ_4M, 1);
 	if (ret) {
-		test_msg("Couldn't add space to the bitmap %d\n", ret);
+		test_err("couldn't add space to the bitmap %d", ret);
 		return ret;
 	}
 
 	ret = test_add_free_space_entry(cache, SZ_2M, SZ_2M, 0);
 	if (ret) {
-		test_msg("Couldn't add extent to the cache %d\n", ret);
+		test_err("couldn't add extent to the cache %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, 3 * SZ_1M, SZ_4M);
 	if (ret) {
-		test_msg("Problem removing overlapping space %d\n", ret);
+		test_err("problem removing overlapping space %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, 3 * SZ_1M, SZ_4M)) {
-		test_msg("Left something behind when removing space");
+		test_err("left something behind when removing space");
 		return -1;
 	}
 
@@ -269,25 +269,25 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache,
 	__btrfs_remove_free_space_cache(cache->free_space_ctl);
 	ret = test_add_free_space_entry(cache, bitmap_offset + SZ_4M, SZ_4M, 1);
 	if (ret) {
-		test_msg("Couldn't add bitmap %d\n", ret);
+		test_err("couldn't add bitmap %d", ret);
 		return ret;
 	}
 
 	ret = test_add_free_space_entry(cache, bitmap_offset - SZ_1M,
 					5 * SZ_1M, 0);
 	if (ret) {
-		test_msg("Couldn't add extent entry %d\n", ret);
+		test_err("couldn't add extent entry %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, bitmap_offset + SZ_1M, 5 * SZ_1M);
 	if (ret) {
-		test_msg("Failed to free our space %d\n", ret);
+		test_err("failed to free our space %d", ret);
 		return ret;
 	}
 
 	if (test_check_exists(cache, bitmap_offset + SZ_1M, 5 * SZ_1M)) {
-		test_msg("Left stuff over\n");
+		test_err("left stuff over");
 		return -1;
 	}
 
@@ -301,19 +301,19 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache,
 	 */
 	ret = test_add_free_space_entry(cache, SZ_1M, SZ_2M, 1);
 	if (ret) {
-		test_msg("Couldn't add bitmap entry %d\n", ret);
+		test_err("couldn't add bitmap entry %d", ret);
 		return ret;
 	}
 
 	ret = test_add_free_space_entry(cache, 3 * SZ_1M, SZ_1M, 0);
 	if (ret) {
-		test_msg("Couldn't add extent entry %d\n", ret);
+		test_err("couldn't add extent entry %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_remove_free_space(cache, SZ_1M, 3 * SZ_1M);
 	if (ret) {
-		test_msg("Error removing bitmap and extent overlapping %d\n", ret);
+		test_err("error removing bitmap and extent overlapping %d", ret);
 		return ret;
 	}
 
@@ -335,12 +335,14 @@ check_num_extents_and_bitmaps(const struct btrfs_block_group_cache *cache,
 			      const int num_bitmaps)
 {
 	if (cache->free_space_ctl->free_extents != num_extents) {
-		test_msg("Incorrect # of extent entries in the cache: %d, expected %d\n",
+		test_err(
+		"incorrect # of extent entries in the cache: %d, expected %d",
 			 cache->free_space_ctl->free_extents, num_extents);
 		return -EINVAL;
 	}
 	if (cache->free_space_ctl->total_bitmaps != num_bitmaps) {
-		test_msg("Incorrect # of extent entries in the cache: %d, expected %d\n",
+		test_err(
+		"incorrect # of extent entries in the cache: %d, expected %d",
 			 cache->free_space_ctl->total_bitmaps, num_bitmaps);
 		return -EINVAL;
 	}
@@ -358,7 +360,7 @@ static int check_cache_empty(struct btrfs_block_group_cache *cache)
 	 * allocate.
 	 */
 	if (cache->free_space_ctl->free_space != 0) {
-		test_msg("Cache free space is not 0\n");
+		test_err("cache free space is not 0");
 		return -EINVAL;
 	}
 
@@ -366,7 +368,7 @@ static int check_cache_empty(struct btrfs_block_group_cache *cache)
 	offset = btrfs_find_space_for_alloc(cache, 0, 4096, 0,
 					    &max_extent_size);
 	if (offset != 0) {
-		test_msg("Space allocation did not fail, returned offset: %llu",
+		test_err("space allocation did not fail, returned offset: %llu",
 			 offset);
 		return -EINVAL;
 	}
@@ -402,7 +404,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	};
 	const struct btrfs_free_space_op *orig_free_space_ops;
 
-	test_msg("Running space stealing from bitmap to extent\n");
+	test_msg("running space stealing from bitmap to extent");
 
 	/*
 	 * For this test, we want to ensure we end up with an extent entry
@@ -430,7 +432,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = test_add_free_space_entry(cache, SZ_128M - SZ_256K, SZ_128K, 0);
 	if (ret) {
-		test_msg("Couldn't add extent entry %d\n", ret);
+		test_err("couldn't add extent entry %d", ret);
 		return ret;
 	}
 
@@ -438,7 +440,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	ret = test_add_free_space_entry(cache, SZ_128M + SZ_512K,
 					SZ_128M - SZ_512K, 1);
 	if (ret) {
-		test_msg("Couldn't add bitmap entry %d\n", ret);
+		test_err("couldn't add bitmap entry %d", ret);
 		return ret;
 	}
 
@@ -457,17 +459,17 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 				      SZ_128M + 768 * SZ_1K,
 				      SZ_128M - 768 * SZ_1K);
 	if (ret) {
-		test_msg("Failed to free part of bitmap space %d\n", ret);
+		test_err("failed to free part of bitmap space %d", ret);
 		return ret;
 	}
 
 	/* Confirm that only those 2 ranges are marked as free. */
 	if (!test_check_exists(cache, SZ_128M - SZ_256K, SZ_128K)) {
-		test_msg("Free space range missing\n");
+		test_err("free space range missing");
 		return -ENOENT;
 	}
 	if (!test_check_exists(cache, SZ_128M + SZ_512K, SZ_256K)) {
-		test_msg("Free space range missing\n");
+		test_err("free space range missing");
 		return -ENOENT;
 	}
 
@@ -477,7 +479,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	if (test_check_exists(cache, SZ_128M + 768 * SZ_1K,
 			      SZ_128M - 768 * SZ_1K)) {
-		test_msg("Bitmap region not removed from space cache\n");
+		test_err("bitmap region not removed from space cache");
 		return -EINVAL;
 	}
 
@@ -486,7 +488,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 * covered by the bitmap, isn't marked as free.
 	 */
 	if (test_check_exists(cache, SZ_128M + SZ_256K, SZ_256K)) {
-		test_msg("Invalid bitmap region marked as free\n");
+		test_err("invalid bitmap region marked as free");
 		return -EINVAL;
 	}
 
@@ -495,7 +497,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 * by the bitmap too, isn't marked as free either.
 	 */
 	if (test_check_exists(cache, SZ_128M, SZ_256K)) {
-		test_msg("Invalid bitmap region marked as free\n");
+		test_err("invalid bitmap region marked as free");
 		return -EINVAL;
 	}
 
@@ -506,12 +508,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = btrfs_add_free_space(cache, SZ_128M, SZ_512K);
 	if (ret) {
-		test_msg("Error adding free space: %d\n", ret);
+		test_err("error adding free space: %d", ret);
 		return ret;
 	}
 	/* Confirm the region is marked as free. */
 	if (!test_check_exists(cache, SZ_128M, SZ_512K)) {
-		test_msg("Bitmap region not marked as free\n");
+		test_err("bitmap region not marked as free");
 		return -ENOENT;
 	}
 
@@ -531,7 +533,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = btrfs_add_free_space(cache, SZ_128M + SZ_16M, sectorsize);
 	if (ret) {
-		test_msg("Error adding free space: %d\n", ret);
+		test_err("error adding free space: %d", ret);
 		return ret;
 	}
 
@@ -550,12 +552,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = btrfs_add_free_space(cache, SZ_128M - SZ_128K, SZ_128K);
 	if (ret) {
-		test_msg("Error adding free space: %d\n", ret);
+		test_err("error adding free space: %d", ret);
 		return ret;
 	}
 	/* Confirm the region is marked as free. */
 	if (!test_check_exists(cache, SZ_128M - SZ_128K, SZ_128K)) {
-		test_msg("Extent region not marked as free\n");
+		test_err("extent region not marked as free");
 		return -ENOENT;
 	}
 
@@ -583,12 +585,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 * allocate the whole free space at once.
 	 */
 	if (!test_check_exists(cache, SZ_128M - SZ_256K, SZ_1M)) {
-		test_msg("Expected region not marked as free\n");
+		test_err("expected region not marked as free");
 		return -ENOENT;
 	}
 
 	if (cache->free_space_ctl->free_space != (SZ_1M + sectorsize)) {
-		test_msg("Cache free space is not 1Mb + %u\n", sectorsize);
+		test_err("cache free space is not 1Mb + %u", sectorsize);
 		return -EINVAL;
 	}
 
@@ -596,7 +598,8 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 					    0, SZ_1M, 0,
 					    &max_extent_size);
 	if (offset != (SZ_128M - SZ_256K)) {
-		test_msg("Failed to allocate 1Mb from space cache, returned offset is: %llu\n",
+		test_err(
+	"failed to allocate 1Mb from space cache, returned offset is: %llu",
 			 offset);
 		return -EINVAL;
 	}
@@ -610,7 +613,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 		return ret;
 
 	if (cache->free_space_ctl->free_space != sectorsize) {
-		test_msg("Cache free space is not %u\n", sectorsize);
+		test_err("cache free space is not %u", sectorsize);
 		return -EINVAL;
 	}
 
@@ -618,7 +621,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 					    0, sectorsize, 0,
 					    &max_extent_size);
 	if (offset != (SZ_128M + SZ_16M)) {
-		test_msg("Failed to allocate %u, returned offset : %llu\n",
+		test_err("failed to allocate %u, returned offset : %llu",
 			 sectorsize, offset);
 		return -EINVAL;
 	}
@@ -640,14 +643,14 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = test_add_free_space_entry(cache, SZ_128M + SZ_128K, SZ_128K, 0);
 	if (ret) {
-		test_msg("Couldn't add extent entry %d\n", ret);
+		test_err("couldn't add extent entry %d", ret);
 		return ret;
 	}
 
 	/* Bitmap entry covering free space range [0, 128Mb - 512Kb[ */
 	ret = test_add_free_space_entry(cache, 0, SZ_128M - SZ_512K, 1);
 	if (ret) {
-		test_msg("Couldn't add bitmap entry %d\n", ret);
+		test_err("couldn't add bitmap entry %d", ret);
 		return ret;
 	}
 
@@ -664,17 +667,17 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = btrfs_remove_free_space(cache, 0, SZ_128M - 768 * SZ_1K);
 	if (ret) {
-		test_msg("Failed to free part of bitmap space %d\n", ret);
+		test_err("failed to free part of bitmap space %d", ret);
 		return ret;
 	}
 
 	/* Confirm that only those 2 ranges are marked as free. */
 	if (!test_check_exists(cache, SZ_128M + SZ_128K, SZ_128K)) {
-		test_msg("Free space range missing\n");
+		test_err("free space range missing");
 		return -ENOENT;
 	}
 	if (!test_check_exists(cache, SZ_128M - 768 * SZ_1K, SZ_256K)) {
-		test_msg("Free space range missing\n");
+		test_err("free space range missing");
 		return -ENOENT;
 	}
 
@@ -683,7 +686,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 * as free anymore.
 	 */
 	if (test_check_exists(cache, 0, SZ_128M - 768 * SZ_1K)) {
-		test_msg("Bitmap region not removed from space cache\n");
+		test_err("bitmap region not removed from space cache");
 		return -EINVAL;
 	}
 
@@ -692,7 +695,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 * covered by the bitmap, isn't marked as free.
 	 */
 	if (test_check_exists(cache, SZ_128M - SZ_512K, SZ_512K)) {
-		test_msg("Invalid bitmap region marked as free\n");
+		test_err("invalid bitmap region marked as free");
 		return -EINVAL;
 	}
 
@@ -703,12 +706,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = btrfs_add_free_space(cache, SZ_128M - SZ_512K, SZ_512K);
 	if (ret) {
-		test_msg("Error adding free space: %d\n", ret);
+		test_err("error adding free space: %d", ret);
 		return ret;
 	}
 	/* Confirm the region is marked as free. */
 	if (!test_check_exists(cache, SZ_128M - SZ_512K, SZ_512K)) {
-		test_msg("Bitmap region not marked as free\n");
+		test_err("bitmap region not marked as free");
 		return -ENOENT;
 	}
 
@@ -728,7 +731,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = btrfs_add_free_space(cache, SZ_32M, 2 * sectorsize);
 	if (ret) {
-		test_msg("Error adding free space: %d\n", ret);
+		test_err("error adding free space: %d", ret);
 		return ret;
 	}
 
@@ -739,12 +742,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 */
 	ret = btrfs_add_free_space(cache, SZ_128M, SZ_128K);
 	if (ret) {
-		test_msg("Error adding free space: %d\n", ret);
+		test_err("error adding free space: %d", ret);
 		return ret;
 	}
 	/* Confirm the region is marked as free. */
 	if (!test_check_exists(cache, SZ_128M, SZ_128K)) {
-		test_msg("Extent region not marked as free\n");
+		test_err("extent region not marked as free");
 		return -ENOENT;
 	}
 
@@ -772,19 +775,20 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 	 * allocate the whole free space at once.
 	 */
 	if (!test_check_exists(cache, SZ_128M - 768 * SZ_1K, SZ_1M)) {
-		test_msg("Expected region not marked as free\n");
+		test_err("expected region not marked as free");
 		return -ENOENT;
 	}
 
 	if (cache->free_space_ctl->free_space != (SZ_1M + 2 * sectorsize)) {
-		test_msg("Cache free space is not 1Mb + %u\n", 2 * sectorsize);
+		test_err("cache free space is not 1Mb + %u", 2 * sectorsize);
 		return -EINVAL;
 	}
 
 	offset = btrfs_find_space_for_alloc(cache, 0, SZ_1M, 0,
 					    &max_extent_size);
 	if (offset != (SZ_128M - 768 * SZ_1K)) {
-		test_msg("Failed to allocate 1Mb from space cache, returned offset is: %llu\n",
+		test_err(
+	"failed to allocate 1Mb from space cache, returned offset is: %llu",
 			 offset);
 		return -EINVAL;
 	}
@@ -798,7 +802,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 		return ret;
 
 	if (cache->free_space_ctl->free_space != 2 * sectorsize) {
-		test_msg("Cache free space is not %u\n", 2 * sectorsize);
+		test_err("cache free space is not %u", 2 * sectorsize);
 		return -EINVAL;
 	}
 
@@ -806,9 +810,8 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 					    0, 2 * sectorsize, 0,
 					    &max_extent_size);
 	if (offset != SZ_32M) {
-		test_msg("Failed to allocate %u, offset: %llu\n",
-			 2 * sectorsize,
-			 offset);
+		test_err("failed to allocate %u, offset: %llu",
+			 2 * sectorsize, offset);
 		return -EINVAL;
 	}
 
@@ -829,7 +832,7 @@ int btrfs_test_free_space_cache(u32 sectorsize, u32 nodesize)
 	struct btrfs_root *root = NULL;
 	int ret = -ENOMEM;
 
-	test_msg("Running btrfs free space cache tests\n");
+	test_msg("running btrfs free space cache tests");
 	fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize);
 	if (!fs_info)
 		return -ENOMEM;
@@ -843,7 +846,7 @@ int btrfs_test_free_space_cache(u32 sectorsize, u32 nodesize)
 	cache = btrfs_alloc_dummy_block_group(fs_info,
 				      BITS_PER_BITMAP * sectorsize + PAGE_SIZE);
 	if (!cache) {
-		test_msg("Couldn't run the tests\n");
+		test_err("couldn't run the tests");
 		btrfs_free_dummy_fs_info(fs_info);
 		return 0;
 	}
@@ -871,6 +874,6 @@ out:
 	btrfs_free_dummy_block_group(cache);
 	btrfs_free_dummy_root(root);
 	btrfs_free_dummy_fs_info(fs_info);
-	test_msg("Free space cache tests finished\n");
+	test_msg("free space cache tests finished");
 	return ret;
 }

fs/btrfs/tests/free-space-tree-tests.c

@@ -32,7 +32,7 @@ static int __check_free_space_extents(struct btrfs_trans_handle *trans,
 
 	info = search_free_space_info(trans, fs_info, cache, path, 0);
 	if (IS_ERR(info)) {
-		test_msg("Could not find free space info\n");
+		test_err("could not find free space info");
 		ret = PTR_ERR(info);
 		goto out;
 	}
@@ -40,7 +40,7 @@ static int __check_free_space_extents(struct btrfs_trans_handle *trans,
 	extent_count = btrfs_free_space_extent_count(path->nodes[0], info);
 
 	if (extent_count != num_extents) {
-		test_msg("Extent count is wrong\n");
+		test_err("extent count is wrong");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -99,7 +99,7 @@ out:
 	btrfs_release_path(path);
 	return ret;
 invalid:
-	test_msg("Free space tree is invalid\n");
+	test_err("free space tree is invalid");
 	ret = -EINVAL;
 	goto out;
 }
@@ -117,7 +117,7 @@ static int check_free_space_extents(struct btrfs_trans_handle *trans,
 
 	info = search_free_space_info(trans, fs_info, cache, path, 0);
 	if (IS_ERR(info)) {
-		test_msg("Could not find free space info\n");
+		test_err("could not find free space info");
 		btrfs_release_path(path);
 		return PTR_ERR(info);
 	}
@@ -131,15 +131,15 @@ static int check_free_space_extents(struct btrfs_trans_handle *trans,
 
 	/* Flip it to the other format and check that for good measure. */
 	if (flags & BTRFS_FREE_SPACE_USING_BITMAPS) {
-		ret = convert_free_space_to_extents(trans, fs_info, cache, path);
+		ret = convert_free_space_to_extents(trans, cache, path);
 		if (ret) {
-			test_msg("Could not convert to extents\n");
+			test_err("could not convert to extents");
 			return ret;
 		}
 	} else {
-		ret = convert_free_space_to_bitmaps(trans, fs_info, cache, path);
+		ret = convert_free_space_to_bitmaps(trans, cache, path);
 		if (ret) {
-			test_msg("Could not convert to bitmaps\n");
+			test_err("could not convert to bitmaps");
 			return ret;
 		}
 	}
@@ -170,11 +170,11 @@ static int test_remove_all(struct btrfs_trans_handle *trans,
 	const struct free_space_extent extents[] = {};
 	int ret;
 
-	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
+	ret = __remove_from_free_space_tree(trans, cache, path,
 					    cache->key.objectid,
 					    cache->key.offset);
 	if (ret) {
-		test_msg("Could not remove free space\n");
+		test_err("could not remove free space");
 		return ret;
 	}
 
@@ -194,10 +194,10 @@ static int test_remove_beginning(struct btrfs_trans_handle *trans,
 	};
 	int ret;
 
-	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
+	ret = __remove_from_free_space_tree(trans, cache, path,
 					    cache->key.objectid, alignment);
 	if (ret) {
-		test_msg("Could not remove free space\n");
+		test_err("could not remove free space");
 		return ret;
 	}
 
@@ -217,12 +217,12 @@ static int test_remove_end(struct btrfs_trans_handle *trans,
 	};
 	int ret;
 
-	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
+	ret = __remove_from_free_space_tree(trans, cache, path,
 					    cache->key.objectid +
 					    cache->key.offset - alignment,
 					    alignment);
 	if (ret) {
-		test_msg("Could not remove free space\n");
+		test_err("could not remove free space");
 		return ret;
 	}
 
@@ -243,11 +243,11 @@ static int test_remove_middle(struct btrfs_trans_handle *trans,
 	};
 	int ret;
 
-	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
+	ret = __remove_from_free_space_tree(trans, cache, path,
 					    cache->key.objectid + alignment,
 					    alignment);
 	if (ret) {
-		test_msg("Could not remove free space\n");
+		test_err("could not remove free space");
 		return ret;
 	}
 
@@ -266,26 +266,26 @@ static int test_merge_left(struct btrfs_trans_handle *trans,
 	};
 	int ret;
 
-	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
+	ret = __remove_from_free_space_tree(trans, cache, path,
 					    cache->key.objectid,
 					    cache->key.offset);
 	if (ret) {
-		test_msg("Could not remove free space\n");
+		test_err("could not remove free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid, alignment);
+	ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid,
+				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
+	ret = __add_to_free_space_tree(trans, cache, path,
 				       cache->key.objectid + alignment,
 				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
@@ -304,27 +304,27 @@ static int test_merge_right(struct btrfs_trans_handle *trans,
 	};
 	int ret;
 
-	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
+	ret = __remove_from_free_space_tree(trans, cache, path,
 					    cache->key.objectid,
 					    cache->key.offset);
 	if (ret) {
-		test_msg("Could not remove free space\n");
+		test_err("could not remove free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
+	ret = __add_to_free_space_tree(trans, cache, path,
 				       cache->key.objectid + 2 * alignment,
 				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
+	ret = __add_to_free_space_tree(trans, cache, path,
 				       cache->key.objectid + alignment,
 				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
@@ -343,34 +343,34 @@ static int test_merge_both(struct btrfs_trans_handle *trans,
 	};
 	int ret;
 
-	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
+	ret = __remove_from_free_space_tree(trans, cache, path,
 					    cache->key.objectid,
 					    cache->key.offset);
 	if (ret) {
-		test_msg("Could not remove free space\n");
+		test_err("could not remove free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid, alignment);
+	ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid,
+				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
+	ret = __add_to_free_space_tree(trans, cache, path,
 				       cache->key.objectid + 2 * alignment,
 				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
+	ret = __add_to_free_space_tree(trans, cache, path,
 				       cache->key.objectid + alignment,
 				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
@@ -391,34 +391,34 @@ static int test_merge_none(struct btrfs_trans_handle *trans,
 	};
 	int ret;
 
-	ret = __remove_from_free_space_tree(trans, fs_info, cache, path,
+	ret = __remove_from_free_space_tree(trans, cache, path,
 					    cache->key.objectid,
 					    cache->key.offset);
 	if (ret) {
-		test_msg("Could not remove free space\n");
+		test_err("could not remove free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
-				       cache->key.objectid, alignment);
+	ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid,
+				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
+	ret = __add_to_free_space_tree(trans, cache, path,
 				       cache->key.objectid + 4 * alignment,
 				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, fs_info, cache, path,
+	ret = __add_to_free_space_tree(trans, cache, path,
 				       cache->key.objectid + 2 * alignment,
 				       alignment);
 	if (ret) {
-		test_msg("Could not add free space\n");
+		test_err("could not add free space");
 		return ret;
 	}
 
@@ -444,14 +444,14 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
 
 	fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize);
 	if (!fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
+		test_err("couldn't allocate dummy fs info");
 		ret = -ENOMEM;
 		goto out;
 	}
 
 	root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate dummy root\n");
+		test_err("couldn't allocate dummy root");
 		ret = PTR_ERR(root);
 		goto out;
 	}
@@ -463,7 +463,7 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
 
 	root->node = alloc_test_extent_buffer(root->fs_info, nodesize);
 	if (!root->node) {
-		test_msg("Couldn't allocate dummy buffer\n");
+		test_err("couldn't allocate dummy buffer");
 		ret = -ENOMEM;
 		goto out;
 	}
@@ -473,7 +473,7 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
 
 	cache = btrfs_alloc_dummy_block_group(fs_info, 8 * alignment);
 	if (!cache) {
-		test_msg("Couldn't allocate dummy block group cache\n");
+		test_err("couldn't allocate dummy block group cache");
 		ret = -ENOMEM;
 		goto out;
 	}
@@ -482,26 +482,25 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
 	cache->needs_free_space = 1;
 	cache->fs_info = root->fs_info;
 
-	btrfs_init_dummy_trans(&trans);
+	btrfs_init_dummy_trans(&trans, root->fs_info);
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		test_msg("Couldn't allocate path\n");
+		test_err("couldn't allocate path");
 		ret = -ENOMEM;
 		goto out;
 	}
 
-	ret = add_block_group_free_space(&trans, root->fs_info, cache);
+	ret = add_block_group_free_space(&trans, cache);
 	if (ret) {
-		test_msg("Could not add block group free space\n");
+		test_err("could not add block group free space");
 		goto out;
 	}
 
 	if (bitmaps) {
-		ret = convert_free_space_to_bitmaps(&trans, root->fs_info,
-						    cache, path);
+		ret = convert_free_space_to_bitmaps(&trans, cache, path);
 		if (ret) {
-			test_msg("Could not convert block group to bitmaps\n");
+			test_err("could not convert block group to bitmaps");
 			goto out;
 		}
 	}
@@ -510,14 +509,14 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
 	if (ret)
 		goto out;
 
-	ret = remove_block_group_free_space(&trans, root->fs_info, cache);
+	ret = remove_block_group_free_space(&trans, cache);
 	if (ret) {
-		test_msg("Could not remove block group free space\n");
+		test_err("could not remove block group free space");
 		goto out;
 	}
 
 	if (btrfs_header_nritems(root->node) != 0) {
-		test_msg("Free space tree has leftover items\n");
+		test_err("free space tree has leftover items");
 		ret = -EINVAL;
 		goto out;
 	}
@@ -539,14 +538,16 @@ static int run_test_both_formats(test_func_t test_func, u32 sectorsize,
 
 	ret = run_test(test_func, 0, sectorsize, nodesize, alignment);
 	if (ret) {
-		test_msg("%pf failed with extents, sectorsize=%u, nodesize=%u, alignment=%u\n",
+		test_err(
+	"%pf failed with extents, sectorsize=%u, nodesize=%u, alignment=%u",
 			 test_func, sectorsize, nodesize, alignment);
 		test_ret = ret;
 	}
 
 	ret = run_test(test_func, 1, sectorsize, nodesize, alignment);
 	if (ret) {
-		test_msg("%pf failed with bitmaps, sectorsize=%u, nodesize=%u, alignment=%u\n",
+		test_err(
+	"%pf failed with bitmaps, sectorsize=%u, nodesize=%u, alignment=%u",
 			 test_func, sectorsize, nodesize, alignment);
 		test_ret = ret;
 	}
@@ -577,7 +578,7 @@ int btrfs_test_free_space_tree(u32 sectorsize, u32 nodesize)
 	 */
 	bitmap_alignment = BTRFS_FREE_SPACE_BITMAP_BITS * PAGE_SIZE;
 
-	test_msg("Running free space tree tests\n");
+	test_msg("running free space tree tests");
 	for (i = 0; i < ARRAY_SIZE(tests); i++) {
 		int ret;
 

fs/btrfs/tests/inode-tests.c

@@ -228,7 +228,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	inode = btrfs_new_test_inode();
 	if (!inode) {
-		test_msg("Couldn't allocate inode\n");
+		test_err("couldn't allocate inode");
 		return ret;
 	}
 
@@ -238,19 +238,19 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize);
 	if (!fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
+		test_err("couldn't allocate dummy fs info");
 		goto out;
 	}
 
 	root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate root\n");
+		test_err("couldn't allocate root");
 		goto out;
 	}
 
 	root->node = alloc_dummy_extent_buffer(fs_info, nodesize);
 	if (!root->node) {
-		test_msg("Couldn't allocate dummy buffer\n");
+		test_err("couldn't allocate dummy buffer");
 		goto out;
 	}
 
@@ -268,11 +268,11 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, sectorsize, 0);
 	if (IS_ERR(em)) {
 		em = NULL;
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != EXTENT_MAP_HOLE) {
-		test_msg("Expected a hole, got %llu\n", em->block_start);
+		test_err("expected a hole, got %llu", em->block_start);
 		goto out;
 	}
 	free_extent_map(em);
@@ -287,20 +287,21 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, (u64)-1, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != EXTENT_MAP_HOLE) {
-		test_msg("Expected a hole, got %llu\n", em->block_start);
+		test_err("expected a hole, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != 0 || em->len != 5) {
-		test_msg("Unexpected extent wanted start 0 len 5, got start "
-			 "%llu len %llu\n", em->start, em->len);
+		test_err(
+		"unexpected extent wanted start 0 len 5, got start %llu len %llu",
+			em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	offset = em->start + em->len;
@@ -308,21 +309,22 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != EXTENT_MAP_INLINE) {
-		test_msg("Expected an inline, got %llu\n", em->block_start);
+		test_err("expected an inline, got %llu", em->block_start);
 		goto out;
 	}
 
 	if (em->start != offset || em->len != (sectorsize - 5)) {
-		test_msg("Unexpected extent wanted start %llu len 1, got start "
-			 "%llu len %llu\n", offset, em->start, em->len);
+		test_err(
+	"unexpected extent wanted start %llu len 1, got start %llu len %llu",
+			offset, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	/*
@@ -335,20 +337,21 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != EXTENT_MAP_HOLE) {
-		test_msg("Expected a hole, got %llu\n", em->block_start);
+		test_err("expected a hole, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != 4) {
-		test_msg("Unexpected extent wanted start %llu len 4, got start "
-			 "%llu len %llu\n", offset, em->start, em->len);
+		test_err(
+	"unexpected extent wanted start %llu len 4, got start %llu len %llu",
+			offset, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	offset = em->start + em->len;
@@ -357,24 +360,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	/* Regular extent */
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize - 1) {
-		test_msg("Unexpected extent wanted start %llu len 4095, got "
-			 "start %llu len %llu\n", offset, em->start, em->len);
+		test_err(
+	"unexpected extent wanted start %llu len 4095, got start %llu len %llu",
+			offset, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+		test_err("wrong orig offset, want %llu, have %llu", em->start,
 			 em->orig_start);
 		goto out;
 	}
@@ -384,25 +388,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	/* The next 3 are split extents */
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+		"unexpected extent start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+		test_err("wrong orig offset, want %llu, have %llu", em->start,
 			 em->orig_start);
 		goto out;
 	}
@@ -413,21 +417,21 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != EXTENT_MAP_HOLE) {
-		test_msg("Expected a hole, got %llu\n", em->block_start);
+		test_err("expected a hole, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	offset = em->start + em->len;
@@ -435,31 +439,31 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != 2 * sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, 2 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	if (em->orig_start != orig_start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n",
+		test_err("wrong orig offset, want %llu, have %llu",
 			 orig_start, em->orig_start);
 		goto out;
 	}
 	disk_bytenr += (em->start - orig_start);
 	if (em->block_start != disk_bytenr) {
-		test_msg("Wrong block start, want %llu, have %llu\n",
+		test_err("wrong block start, want %llu, have %llu",
 			 disk_bytenr, em->block_start);
 		goto out;
 	}
@@ -469,26 +473,26 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	/* Prealloc extent */
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != prealloc_only) {
-		test_msg("Unexpected flags set, want %lu have %lu\n",
+		test_err("unexpected flags set, want %lu have %lu",
 			 prealloc_only, em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+		test_err("wrong orig offset, want %llu, have %llu", em->start,
 			 em->orig_start);
 		goto out;
 	}
@@ -498,26 +502,26 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	/* The next 3 are a half written prealloc extent */
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != prealloc_only) {
-		test_msg("Unexpected flags set, want %lu have %lu\n",
+		test_err("unexpected flags set, want %lu have %lu",
 			 prealloc_only, em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+		test_err("wrong orig offset, want %llu, have %llu", em->start,
 			 em->orig_start);
 		goto out;
 	}
@@ -528,30 +532,30 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_HOLE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	if (em->orig_start != orig_start) {
-		test_msg("Unexpected orig offset, wanted %llu, have %llu\n",
+		test_err("unexpected orig offset, wanted %llu, have %llu",
 			 orig_start, em->orig_start);
 		goto out;
 	}
 	if (em->block_start != (disk_bytenr + (em->start - em->orig_start))) {
-		test_msg("Unexpected block start, wanted %llu, have %llu\n",
+		test_err("unexpected block start, wanted %llu, have %llu",
 			 disk_bytenr + (em->start - em->orig_start),
 			 em->block_start);
 		goto out;
@@ -561,31 +565,31 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != 2 * sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, 2 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != prealloc_only) {
-		test_msg("Unexpected flags set, want %lu have %lu\n",
+		test_err("unexpected flags set, want %lu have %lu",
 			 prealloc_only, em->flags);
 		goto out;
 	}
 	if (em->orig_start != orig_start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", orig_start,
+		test_err("wrong orig offset, want %llu, have %llu", orig_start,
 			 em->orig_start);
 		goto out;
 	}
 	if (em->block_start != (disk_bytenr + (em->start - em->orig_start))) {
-		test_msg("Unexpected block start, wanted %llu, have %llu\n",
+		test_err("unexpected block start, wanted %llu, have %llu",
 			 disk_bytenr + (em->start - em->orig_start),
 			 em->block_start);
 		goto out;
@@ -596,31 +600,31 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	/* Now for the compressed extent */
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != 2 * sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u,"
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, 2 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != compressed_only) {
-		test_msg("Unexpected flags set, want %lu have %lu\n",
+		test_err("unexpected flags set, want %lu have %lu",
 			 compressed_only, em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n",
+		test_err("wrong orig offset, want %llu, have %llu",
 			 em->start, em->orig_start);
 		goto out;
 	}
 	if (em->compress_type != BTRFS_COMPRESS_ZLIB) {
-		test_msg("Unexpected compress type, wanted %d, got %d\n",
+		test_err("unexpected compress type, wanted %d, got %d",
 			 BTRFS_COMPRESS_ZLIB, em->compress_type);
 		goto out;
 	}
@@ -630,31 +634,31 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	/* Split compressed extent */
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u,"
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != compressed_only) {
-		test_msg("Unexpected flags set, want %lu have %lu\n",
+		test_err("unexpected flags set, want %lu have %lu",
 			 compressed_only, em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n",
+		test_err("wrong orig offset, want %llu, have %llu",
 			 em->start, em->orig_start);
 		goto out;
 	}
 	if (em->compress_type != BTRFS_COMPRESS_ZLIB) {
-		test_msg("Unexpected compress type, wanted %d, got %d\n",
+		test_err("unexpected compress type, wanted %d, got %d",
 			 BTRFS_COMPRESS_ZLIB, em->compress_type);
 		goto out;
 	}
@@ -665,25 +669,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+		test_err("wrong orig offset, want %llu, have %llu", em->start,
 			 em->orig_start);
 		goto out;
 	}
@@ -692,32 +696,32 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != disk_bytenr) {
-		test_msg("Block start does not match, want %llu got %llu\n",
+		test_err("block start does not match, want %llu got %llu",
 			 disk_bytenr, em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != 2 * sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, 2 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != compressed_only) {
-		test_msg("Unexpected flags set, want %lu have %lu\n",
+		test_err("unexpected flags set, want %lu have %lu",
 			 compressed_only, em->flags);
 		goto out;
 	}
 	if (em->orig_start != orig_start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n",
+		test_err("wrong orig offset, want %llu, have %llu",
 			 em->start, orig_start);
 		goto out;
 	}
 	if (em->compress_type != BTRFS_COMPRESS_ZLIB) {
-		test_msg("Unexpected compress type, wanted %d, got %d\n",
+		test_err("unexpected compress type, wanted %d, got %d",
 			 BTRFS_COMPRESS_ZLIB, em->compress_type);
 		goto out;
 	}
@@ -728,25 +732,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset + 6,
 			sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+		test_err("wrong orig offset, want %llu, have %llu", em->start,
 			 em->orig_start);
 		goto out;
 	}
@@ -755,11 +759,11 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, SZ_4M, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != EXTENT_MAP_HOLE) {
-		test_msg("Expected a hole extent, got %llu\n", em->block_start);
+		test_err("expected a hole extent, got %llu", em->block_start);
 		goto out;
 	}
 	/*
@@ -768,18 +772,18 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	 * test.
 	 */
 	if (em->start != offset || em->len != 3 * sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, 3 * sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != vacancy_only) {
-		test_msg("Unexpected flags set, want %lu have %lu\n",
+		test_err("unexpected flags set, want %lu have %lu",
 			 vacancy_only, em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+		test_err("wrong orig offset, want %llu, have %llu", em->start,
 			 em->orig_start);
 		goto out;
 	}
@@ -788,25 +792,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != offset || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %llu len %u,"
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %llu len %u, got start %llu len %llu",
 			offset, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		test_err("unexpected flags set, want 0 have %lu", em->flags);
 		goto out;
 	}
 	if (em->orig_start != em->start) {
-		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+		test_err("wrong orig offset, want %llu, have %llu", em->start,
 			 em->orig_start);
 		goto out;
 	}
@@ -830,7 +834,7 @@ static int test_hole_first(u32 sectorsize, u32 nodesize)
 
 	inode = btrfs_new_test_inode();
 	if (!inode) {
-		test_msg("Couldn't allocate inode\n");
+		test_err("couldn't allocate inode");
 		return ret;
 	}
 
@@ -840,19 +844,19 @@ static int test_hole_first(u32 sectorsize, u32 nodesize)
 
 	fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize);
 	if (!fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
+		test_err("couldn't allocate dummy fs info");
 		goto out;
 	}
 
 	root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate root\n");
+		test_err("couldn't allocate root");
 		goto out;
 	}
 
 	root->node = alloc_dummy_extent_buffer(fs_info, nodesize);
 	if (!root->node) {
-		test_msg("Couldn't allocate dummy buffer\n");
+		test_err("couldn't allocate dummy buffer");
 		goto out;
 	}
 
@@ -871,21 +875,21 @@ static int test_hole_first(u32 sectorsize, u32 nodesize)
 		      sectorsize, BTRFS_FILE_EXTENT_REG, 0, 1);
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, 2 * sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != EXTENT_MAP_HOLE) {
-		test_msg("Expected a hole, got %llu\n", em->block_start);
+		test_err("expected a hole, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != 0 || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start 0 len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start 0 len %u, got start %llu len %llu",
 			sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != vacancy_only) {
-		test_msg("Wrong flags, wanted %lu, have %lu\n", vacancy_only,
+		test_err("wrong flags, wanted %lu, have %lu", vacancy_only,
 			 em->flags);
 		goto out;
 	}
@@ -894,21 +898,21 @@ static int test_hole_first(u32 sectorsize, u32 nodesize)
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, sectorsize,
 			2 * sectorsize, 0);
 	if (IS_ERR(em)) {
-		test_msg("Got an error when we shouldn't have\n");
+		test_err("got an error when we shouldn't have");
 		goto out;
 	}
 	if (em->block_start != sectorsize) {
-		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		test_err("expected a real extent, got %llu", em->block_start);
 		goto out;
 	}
 	if (em->start != sectorsize || em->len != sectorsize) {
-		test_msg("Unexpected extent wanted start %u len %u, "
-			"got start %llu len %llu\n",
+		test_err(
+	"unexpected extent wanted start %u len %u, got start %llu len %llu",
 			sectorsize, sectorsize, em->start, em->len);
 		goto out;
 	}
 	if (em->flags != 0) {
-		test_msg("Unexpected flags set, wanted 0 got %lu\n",
+		test_err("unexpected flags set, wanted 0 got %lu",
 			 em->flags);
 		goto out;
 	}
@@ -931,19 +935,19 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 
 	inode = btrfs_new_test_inode();
 	if (!inode) {
-		test_msg("Couldn't allocate inode\n");
+		test_err("couldn't allocate inode");
 		return ret;
 	}
 
 	fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize);
 	if (!fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
+		test_err("couldn't allocate dummy fs info");
 		goto out;
 	}
 
 	root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate root\n");
+		test_err("couldn't allocate root");
 		goto out;
 	}
 
@@ -954,12 +958,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 	ret = btrfs_set_extent_delalloc(inode, 0, BTRFS_MAX_EXTENT_SIZE - 1, 0,
 					NULL, 0);
 	if (ret) {
-		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
+		test_err("btrfs_set_extent_delalloc returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents != 1) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 1, got %u\n",
+		test_err("miscount, wanted 1, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -969,12 +973,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 					BTRFS_MAX_EXTENT_SIZE + sectorsize - 1,
 					0, NULL, 0);
 	if (ret) {
-		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
+		test_err("btrfs_set_extent_delalloc returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents != 2) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 2, got %u\n",
+		test_err("miscount, wanted 2, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -986,12 +990,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 			       EXTENT_DELALLOC | EXTENT_DIRTY |
 			       EXTENT_UPTODATE, 0, 0, NULL);
 	if (ret) {
-		test_msg("clear_extent_bit returned %d\n", ret);
+		test_err("clear_extent_bit returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents != 2) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 2, got %u\n",
+		test_err("miscount, wanted 2, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -1002,12 +1006,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 					+ sectorsize - 1,
 					0, NULL, 0);
 	if (ret) {
-		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
+		test_err("btrfs_set_extent_delalloc returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents != 2) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 2, got %u\n",
+		test_err("miscount, wanted 2, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -1020,12 +1024,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 			(BTRFS_MAX_EXTENT_SIZE << 1) + 3 * sectorsize - 1,
 			0, NULL, 0);
 	if (ret) {
-		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
+		test_err("btrfs_set_extent_delalloc returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents != 4) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 4, got %u\n",
+		test_err("miscount, wanted 4, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -1037,12 +1041,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 			BTRFS_MAX_EXTENT_SIZE + sectorsize,
 			BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, 0, NULL, 0);
 	if (ret) {
-		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
+		test_err("btrfs_set_extent_delalloc returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents != 3) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 3, got %u\n",
+		test_err("miscount, wanted 3, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -1054,12 +1058,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 			       EXTENT_DIRTY | EXTENT_DELALLOC |
 			       EXTENT_UPTODATE, 0, 0, NULL);
 	if (ret) {
-		test_msg("clear_extent_bit returned %d\n", ret);
+		test_err("clear_extent_bit returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents != 4) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 4, got %u\n",
+		test_err("miscount, wanted 4, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -1072,12 +1076,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 			BTRFS_MAX_EXTENT_SIZE + sectorsize,
 			BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, 0, NULL, 0);
 	if (ret) {
-		test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
+		test_err("btrfs_set_extent_delalloc returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents != 3) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 3, got %u\n",
+		test_err("miscount, wanted 3, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -1087,12 +1091,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 			       EXTENT_DIRTY | EXTENT_DELALLOC |
 			       EXTENT_UPTODATE, 0, 0, NULL);
 	if (ret) {
-		test_msg("clear_extent_bit returned %d\n", ret);
+		test_err("clear_extent_bit returned %d", ret);
 		goto out;
 	}
 	if (BTRFS_I(inode)->outstanding_extents) {
 		ret = -EINVAL;
-		test_msg("Miscount, wanted 0, got %u\n",
+		test_err("miscount, wanted 0, got %u",
 			 BTRFS_I(inode)->outstanding_extents);
 		goto out;
 	}
@@ -1115,14 +1119,14 @@ int btrfs_test_inodes(u32 sectorsize, u32 nodesize)
 	set_bit(EXTENT_FLAG_COMPRESSED, &compressed_only);
 	set_bit(EXTENT_FLAG_PREALLOC, &prealloc_only);
 
-	test_msg("Running btrfs_get_extent tests\n");
+	test_msg("running btrfs_get_extent tests");
 	ret = test_btrfs_get_extent(sectorsize, nodesize);
 	if (ret)
 		return ret;
-	test_msg("Running hole first btrfs_get_extent test\n");
+	test_msg("running hole first btrfs_get_extent test");
 	ret = test_hole_first(sectorsize, nodesize);
 	if (ret)
 		return ret;
-	test_msg("Running outstanding_extents tests\n");
+	test_msg("running outstanding_extents tests");
 	return test_extent_accounting(sectorsize, nodesize);
 }

fs/btrfs/tests/qgroup-tests.c

@@ -24,7 +24,7 @@ static int insert_normal_tree_ref(struct btrfs_root *root, u64 bytenr,
 	u32 size = sizeof(*item) + sizeof(*iref) + sizeof(*block_info);
 	int ret;
 
-	btrfs_init_dummy_trans(&trans);
+	btrfs_init_dummy_trans(&trans, NULL);
 
 	ins.objectid = bytenr;
 	ins.type = BTRFS_EXTENT_ITEM_KEY;
@@ -32,14 +32,14 @@ static int insert_normal_tree_ref(struct btrfs_root *root, u64 bytenr,
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		test_msg("Couldn't allocate path\n");
+		test_err("couldn't allocate path");
 		return -ENOMEM;
 	}
 
 	path->leave_spinning = 1;
 	ret = btrfs_insert_empty_item(&trans, root, path, &ins, size);
 	if (ret) {
-		test_msg("Couldn't insert ref %d\n", ret);
+		test_err("couldn't insert ref %d", ret);
 		btrfs_free_path(path);
 		return ret;
 	}
@@ -74,7 +74,7 @@ static int add_tree_ref(struct btrfs_root *root, u64 bytenr, u64 num_bytes,
 	u64 refs;
 	int ret;
 
-	btrfs_init_dummy_trans(&trans);
+	btrfs_init_dummy_trans(&trans, NULL);
 
 	key.objectid = bytenr;
 	key.type = BTRFS_EXTENT_ITEM_KEY;
@@ -82,14 +82,14 @@ static int add_tree_ref(struct btrfs_root *root, u64 bytenr, u64 num_bytes,
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		test_msg("Couldn't allocate path\n");
+		test_err("couldn't allocate path");
 		return -ENOMEM;
 	}
 
 	path->leave_spinning = 1;
 	ret = btrfs_search_slot(&trans, root, &key, path, 0, 1);
 	if (ret) {
-		test_msg("Couldn't find extent ref\n");
+		test_err("couldn't find extent ref");
 		btrfs_free_path(path);
 		return ret;
 	}
@@ -111,7 +111,7 @@ static int add_tree_ref(struct btrfs_root *root, u64 bytenr, u64 num_bytes,
 
 	ret = btrfs_insert_empty_item(&trans, root, path, &key, 0);
 	if (ret)
-		test_msg("Failed to insert backref\n");
+		test_err("failed to insert backref");
 	btrfs_free_path(path);
 	return ret;
 }
@@ -124,7 +124,7 @@ static int remove_extent_item(struct btrfs_root *root, u64 bytenr,
 	struct btrfs_path *path;
 	int ret;
 
-	btrfs_init_dummy_trans(&trans);
+	btrfs_init_dummy_trans(&trans, NULL);
 
 	key.objectid = bytenr;
 	key.type = BTRFS_EXTENT_ITEM_KEY;
@@ -132,14 +132,14 @@ static int remove_extent_item(struct btrfs_root *root, u64 bytenr,
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		test_msg("Couldn't allocate path\n");
+		test_err("couldn't allocate path");
 		return -ENOMEM;
 	}
 	path->leave_spinning = 1;
 
 	ret = btrfs_search_slot(&trans, root, &key, path, -1, 1);
 	if (ret) {
-		test_msg("Didn't find our key %d\n", ret);
+		test_err("didn't find our key %d", ret);
 		btrfs_free_path(path);
 		return ret;
 	}
@@ -158,7 +158,7 @@ static int remove_extent_ref(struct btrfs_root *root, u64 bytenr,
 	u64 refs;
 	int ret;
 
-	btrfs_init_dummy_trans(&trans);
+	btrfs_init_dummy_trans(&trans, NULL);
 
 	key.objectid = bytenr;
 	key.type = BTRFS_EXTENT_ITEM_KEY;
@@ -166,14 +166,14 @@ static int remove_extent_ref(struct btrfs_root *root, u64 bytenr,
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		test_msg("Couldn't allocate path\n");
+		test_err("couldn't allocate path");
 		return -ENOMEM;
 	}
 
 	path->leave_spinning = 1;
 	ret = btrfs_search_slot(&trans, root, &key, path, 0, 1);
 	if (ret) {
-		test_msg("Couldn't find extent ref\n");
+		test_err("couldn't find extent ref");
 		btrfs_free_path(path);
 		return ret;
 	}
@@ -195,7 +195,7 @@ static int remove_extent_ref(struct btrfs_root *root, u64 bytenr,
 
 	ret = btrfs_search_slot(&trans, root, &key, path, -1, 1);
 	if (ret) {
-		test_msg("Couldn't find backref %d\n", ret);
+		test_err("couldn't find backref %d", ret);
 		btrfs_free_path(path);
 		return ret;
 	}
@@ -213,12 +213,12 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
 	struct ulist *new_roots = NULL;
 	int ret;
 
-	btrfs_init_dummy_trans(&trans);
+	btrfs_init_dummy_trans(&trans, fs_info);
 
-	test_msg("Qgroup basic add\n");
+	test_msg("qgroup basic add");
 	ret = btrfs_create_qgroup(NULL, fs_info, BTRFS_FS_TREE_OBJECTID);
 	if (ret) {
-		test_msg("Couldn't create a qgroup %d\n", ret);
+		test_err("couldn't create a qgroup %d", ret);
 		return ret;
 	}
 
@@ -231,7 +231,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
 			false);
 	if (ret) {
 		ulist_free(old_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
@@ -245,20 +245,20 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
 					  nodesize, old_roots, new_roots);
 	if (ret) {
-		test_msg("Couldn't account space for a qgroup %d\n", ret);
+		test_err("couldn't account space for a qgroup %d", ret);
 		return ret;
 	}
 
 	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
 				nodesize, nodesize)) {
-		test_msg("Qgroup counts didn't match expected values\n");
+		test_err("qgroup counts didn't match expected values");
 		return -EINVAL;
 	}
 	old_roots = NULL;
@@ -268,7 +268,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
 			false);
 	if (ret) {
 		ulist_free(old_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
@@ -281,19 +281,19 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
 					  nodesize, old_roots, new_roots);
 	if (ret) {
-		test_msg("Couldn't account space for a qgroup %d\n", ret);
+		test_err("couldn't account space for a qgroup %d", ret);
 		return -EINVAL;
 	}
 
 	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID, 0, 0)) {
-		test_msg("Qgroup counts didn't match expected values\n");
+		test_err("qgroup counts didn't match expected values");
 		return -EINVAL;
 	}
 
@@ -314,9 +314,9 @@ static int test_multiple_refs(struct btrfs_root *root,
 	struct ulist *new_roots = NULL;
 	int ret;
 
-	btrfs_init_dummy_trans(&trans);
+	btrfs_init_dummy_trans(&trans, fs_info);
 
-	test_msg("Qgroup multiple refs test\n");
+	test_msg("qgroup multiple refs test");
 
 	/*
 	 * We have BTRFS_FS_TREE_OBJECTID created already from the
@@ -324,7 +324,7 @@ static int test_multiple_refs(struct btrfs_root *root,
 	 */
 	ret = btrfs_create_qgroup(NULL, fs_info, BTRFS_FIRST_FREE_OBJECTID);
 	if (ret) {
-		test_msg("Couldn't create a qgroup %d\n", ret);
+		test_err("couldn't create a qgroup %d", ret);
 		return ret;
 	}
 
@@ -332,7 +332,7 @@ static int test_multiple_refs(struct btrfs_root *root,
 			false);
 	if (ret) {
 		ulist_free(old_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
@@ -346,20 +346,20 @@ static int test_multiple_refs(struct btrfs_root *root,
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
 					  nodesize, old_roots, new_roots);
 	if (ret) {
-		test_msg("Couldn't account space for a qgroup %d\n", ret);
+		test_err("couldn't account space for a qgroup %d", ret);
 		return ret;
 	}
 
 	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
 				       nodesize, nodesize)) {
-		test_msg("Qgroup counts didn't match expected values\n");
+		test_err("qgroup counts didn't match expected values");
 		return -EINVAL;
 	}
 
@@ -367,7 +367,7 @@ static int test_multiple_refs(struct btrfs_root *root,
 			false);
 	if (ret) {
 		ulist_free(old_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
@@ -381,26 +381,26 @@ static int test_multiple_refs(struct btrfs_root *root,
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
 					  nodesize, old_roots, new_roots);
 	if (ret) {
-		test_msg("Couldn't account space for a qgroup %d\n", ret);
+		test_err("couldn't account space for a qgroup %d", ret);
 		return ret;
 	}
 
 	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
 					nodesize, 0)) {
-		test_msg("Qgroup counts didn't match expected values\n");
+		test_err("qgroup counts didn't match expected values");
 		return -EINVAL;
 	}
 
 	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FIRST_FREE_OBJECTID,
 					nodesize, 0)) {
-		test_msg("Qgroup counts didn't match expected values\n");
+		test_err("qgroup counts didn't match expected values");
 		return -EINVAL;
 	}
 
@@ -408,7 +408,7 @@ static int test_multiple_refs(struct btrfs_root *root,
 			false);
 	if (ret) {
 		ulist_free(old_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
@@ -422,26 +422,26 @@ static int test_multiple_refs(struct btrfs_root *root,
 	if (ret) {
 		ulist_free(old_roots);
 		ulist_free(new_roots);
-		test_msg("Couldn't find old roots: %d\n", ret);
+		test_err("couldn't find old roots: %d", ret);
 		return ret;
 	}
 
 	ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize,
 					  nodesize, old_roots, new_roots);
 	if (ret) {
-		test_msg("Couldn't account space for a qgroup %d\n", ret);
+		test_err("couldn't account space for a qgroup %d", ret);
 		return ret;
 	}
 
 	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FIRST_FREE_OBJECTID,
 					0, 0)) {
-		test_msg("Qgroup counts didn't match expected values\n");
+		test_err("qgroup counts didn't match expected values");
 		return -EINVAL;
 	}
 
 	if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
 					nodesize, nodesize)) {
-		test_msg("Qgroup counts didn't match expected values\n");
+		test_err("qgroup counts didn't match expected values");
 		return -EINVAL;
 	}
 
@@ -457,13 +457,13 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 
 	fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize);
 	if (!fs_info) {
-		test_msg("Couldn't allocate dummy fs info\n");
+		test_err("couldn't allocate dummy fs info");
 		return -ENOMEM;
 	}
 
 	root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(root)) {
-		test_msg("Couldn't allocate root\n");
+		test_err("couldn't allocate root");
 		ret = PTR_ERR(root);
 		goto out;
 	}
@@ -485,7 +485,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 	 */
 	root->node = alloc_test_extent_buffer(root->fs_info, nodesize);
 	if (!root->node) {
-		test_msg("Couldn't allocate dummy buffer\n");
+		test_err("couldn't allocate dummy buffer");
 		ret = -ENOMEM;
 		goto out;
 	}
@@ -495,7 +495,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 
 	tmp_root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(tmp_root)) {
-		test_msg("Couldn't allocate a fs root\n");
+		test_err("couldn't allocate a fs root");
 		ret = PTR_ERR(tmp_root);
 		goto out;
 	}
@@ -504,13 +504,13 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 	root->fs_info->fs_root = tmp_root;
 	ret = btrfs_insert_fs_root(root->fs_info, tmp_root);
 	if (ret) {
-		test_msg("Couldn't insert fs root %d\n", ret);
+		test_err("couldn't insert fs root %d", ret);
 		goto out;
 	}
 
 	tmp_root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(tmp_root)) {
-		test_msg("Couldn't allocate a fs root\n");
+		test_err("couldn't allocate a fs root");
 		ret = PTR_ERR(tmp_root);
 		goto out;
 	}
@@ -518,11 +518,11 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 	tmp_root->root_key.objectid = BTRFS_FIRST_FREE_OBJECTID;
 	ret = btrfs_insert_fs_root(root->fs_info, tmp_root);
 	if (ret) {
-		test_msg("Couldn't insert fs root %d\n", ret);
+		test_err("couldn't insert fs root %d", ret);
 		goto out;
 	}
 
-	test_msg("Running qgroup tests\n");
+	test_msg("running qgroup tests");
 	ret = test_no_shared_qgroup(root, sectorsize, nodesize);
 	if (ret)
 		goto out;

fs/btrfs/transaction.c

@@ -877,12 +877,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 	atomic_dec(&cur_trans->num_writers);
 	extwriter_counter_dec(cur_trans, trans->type);
 
-	/*
-	 * Make sure counter is updated before we wake up waiters.
-	 */
-	smp_mb();
-	if (waitqueue_active(&cur_trans->writer_wait))
-		wake_up(&cur_trans->writer_wait);
+	cond_wake_up(&cur_trans->writer_wait);
 	btrfs_put_transaction(cur_trans);
 
 	if (current->journal_info == trans)
@@ -1250,7 +1245,6 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans)
 
 			btrfs_free_log(trans, root);
 			btrfs_update_reloc_root(trans, root);
-			btrfs_orphan_commit_root(trans, root);
 
 			btrfs_save_ino_cache(root, trans);
 
@@ -1640,15 +1634,14 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
-	ret = btrfs_uuid_tree_add(trans, fs_info, new_uuid.b,
-				  BTRFS_UUID_KEY_SUBVOL, objectid);
+	ret = btrfs_uuid_tree_add(trans, new_uuid.b, BTRFS_UUID_KEY_SUBVOL,
+				  objectid);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
 	if (!btrfs_is_empty_uuid(new_root_item->received_uuid)) {
-		ret = btrfs_uuid_tree_add(trans, fs_info,
-					  new_root_item->received_uuid,
+		ret = btrfs_uuid_tree_add(trans, new_root_item->received_uuid,
 					  BTRFS_UUID_KEY_RECEIVED_SUBVOL,
 					  objectid);
 		if (ret && ret != -EEXIST) {

fs/btrfs/transaction.h

@@ -139,7 +139,6 @@ struct btrfs_pending_snapshot {
 	struct btrfs_path *path;
 	/* block reservation for the operation */
 	struct btrfs_block_rsv block_rsv;
-	u64 qgroup_reserved;
 	/* extra metadata reservation for relocation */
 	int error;
 	bool readonly;

fs/btrfs/tree-log.c

@@ -222,11 +222,8 @@ int btrfs_pin_log_trans(struct btrfs_root *root)
 void btrfs_end_log_trans(struct btrfs_root *root)
 {
 	if (atomic_dec_and_test(&root->log_writers)) {
-		/*
-		 * Implicit memory barrier after atomic_dec_and_test
-		 */
-		if (waitqueue_active(&root->log_writer_wait))
-			wake_up(&root->log_writer_wait);
+		/* atomic_dec_and_test implies a barrier */
+		cond_wake_up_nomb(&root->log_writer_wait);
 	}
 }
 
@@ -2988,11 +2985,8 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 
 	mutex_lock(&log_root_tree->log_mutex);
 	if (atomic_dec_and_test(&log_root_tree->log_writers)) {
-		/*
-		 * Implicit memory barrier after atomic_dec_and_test
-		 */
-		if (waitqueue_active(&log_root_tree->log_writer_wait))
-			wake_up(&log_root_tree->log_writer_wait);
+		/* atomic_dec_and_test implies a barrier */
+		cond_wake_up_nomb(&log_root_tree->log_writer_wait);
 	}
 
 	if (ret) {
@@ -3116,10 +3110,11 @@ out_wake_log_root:
 	mutex_unlock(&log_root_tree->log_mutex);
 
 	/*
-	 * The barrier before waitqueue_active is implied by mutex_unlock
+	 * The barrier before waitqueue_active (in cond_wake_up) is needed so
+	 * all the updates above are seen by the woken threads. It might not be
+	 * necessary, but proving that seems to be hard.
 	 */
-	if (waitqueue_active(&log_root_tree->log_commit_wait[index2]))
-		wake_up(&log_root_tree->log_commit_wait[index2]);
+	cond_wake_up(&log_root_tree->log_commit_wait[index2]);
 out:
 	mutex_lock(&root->log_mutex);
 	btrfs_remove_all_log_ctxs(root, index1, ret);
@@ -3128,10 +3123,11 @@ out:
 	mutex_unlock(&root->log_mutex);
 
 	/*
-	 * The barrier before waitqueue_active is implied by mutex_unlock
+	 * The barrier before waitqueue_active (in cond_wake_up) is needed so
+	 * all the updates above are seen by the woken threads. It might not be
+	 * necessary, but proving that seems to be hard.
 	 */
-	if (waitqueue_active(&root->log_commit_wait[index1]))
-		wake_up(&root->log_commit_wait[index1]);
+	cond_wake_up(&root->log_commit_wait[index1]);
 	return ret;
 }
 

fs/btrfs/uuid-tree.c

@@ -79,10 +79,10 @@ out:
 	return ret;
 }
 
-int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, u8 *uuid, u8 type,
+int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 			u64 subid_cpu)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *uuid_root = fs_info->uuid_root;
 	int ret;
 	struct btrfs_path *path = NULL;
@@ -144,10 +144,10 @@ out:
 	return ret;
 }
 
-int btrfs_uuid_tree_rem(struct btrfs_trans_handle *trans,
-			struct btrfs_fs_info *fs_info, u8 *uuid, u8 type,
+int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 			u64 subid)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *uuid_root = fs_info->uuid_root;
 	int ret;
 	struct btrfs_path *path = NULL;
@@ -239,7 +239,7 @@ static int btrfs_uuid_iter_rem(struct btrfs_root *uuid_root, u8 *uuid, u8 type,
 		goto out;
 	}
 
-	ret = btrfs_uuid_tree_rem(trans, uuid_root->fs_info, uuid, type, subid);
+	ret = btrfs_uuid_tree_remove(trans, uuid, type, subid);
 	btrfs_end_transaction(trans);
 
 out:

fs/btrfs/volumes.c

@@ -40,6 +40,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.tolerated_failures = 1,
 		.devs_increment	= 2,
 		.ncopies	= 2,
+		.raid_name	= "raid10",
+		.bg_flag	= BTRFS_BLOCK_GROUP_RAID10,
+		.mindev_error	= BTRFS_ERROR_DEV_RAID10_MIN_NOT_MET,
 	},
 	[BTRFS_RAID_RAID1] = {
 		.sub_stripes	= 1,
@@ -49,6 +52,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.tolerated_failures = 1,
 		.devs_increment	= 2,
 		.ncopies	= 2,
+		.raid_name	= "raid1",
+		.bg_flag	= BTRFS_BLOCK_GROUP_RAID1,
+		.mindev_error	= BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET,
 	},
 	[BTRFS_RAID_DUP] = {
 		.sub_stripes	= 1,
@@ -58,6 +64,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.tolerated_failures = 0,
 		.devs_increment	= 1,
 		.ncopies	= 2,
+		.raid_name	= "dup",
+		.bg_flag	= BTRFS_BLOCK_GROUP_DUP,
+		.mindev_error	= 0,
 	},
 	[BTRFS_RAID_RAID0] = {
 		.sub_stripes	= 1,
@@ -67,6 +76,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.tolerated_failures = 0,
 		.devs_increment	= 1,
 		.ncopies	= 1,
+		.raid_name	= "raid0",
+		.bg_flag	= BTRFS_BLOCK_GROUP_RAID0,
+		.mindev_error	= 0,
 	},
 	[BTRFS_RAID_SINGLE] = {
 		.sub_stripes	= 1,
@@ -76,6 +88,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.tolerated_failures = 0,
 		.devs_increment	= 1,
 		.ncopies	= 1,
+		.raid_name	= "single",
+		.bg_flag	= 0,
+		.mindev_error	= 0,
 	},
 	[BTRFS_RAID_RAID5] = {
 		.sub_stripes	= 1,
@@ -85,6 +100,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.tolerated_failures = 1,
 		.devs_increment	= 1,
 		.ncopies	= 2,
+		.raid_name	= "raid5",
+		.bg_flag	= BTRFS_BLOCK_GROUP_RAID5,
+		.mindev_error	= BTRFS_ERROR_DEV_RAID5_MIN_NOT_MET,
 	},
 	[BTRFS_RAID_RAID6] = {
 		.sub_stripes	= 1,
@@ -94,33 +112,19 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.tolerated_failures = 2,
 		.devs_increment	= 1,
 		.ncopies	= 3,
+		.raid_name	= "raid6",
+		.bg_flag	= BTRFS_BLOCK_GROUP_RAID6,
+		.mindev_error	= BTRFS_ERROR_DEV_RAID6_MIN_NOT_MET,
 	},
 };
 
-const u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES] = {
-	[BTRFS_RAID_RAID10] = BTRFS_BLOCK_GROUP_RAID10,
-	[BTRFS_RAID_RAID1]  = BTRFS_BLOCK_GROUP_RAID1,
-	[BTRFS_RAID_DUP]    = BTRFS_BLOCK_GROUP_DUP,
-	[BTRFS_RAID_RAID0]  = BTRFS_BLOCK_GROUP_RAID0,
-	[BTRFS_RAID_SINGLE] = 0,
-	[BTRFS_RAID_RAID5]  = BTRFS_BLOCK_GROUP_RAID5,
-	[BTRFS_RAID_RAID6]  = BTRFS_BLOCK_GROUP_RAID6,
-};
+const char *get_raid_name(enum btrfs_raid_types type)
+{
+	if (type >= BTRFS_NR_RAID_TYPES)
+		return NULL;
 
-/*
- * Table to convert BTRFS_RAID_* to the error code if minimum number of devices
- * condition is not met. Zero means there's no corresponding
- * BTRFS_ERROR_DEV_*_NOT_MET value.
- */
-const int btrfs_raid_mindev_error[BTRFS_NR_RAID_TYPES] = {
-	[BTRFS_RAID_RAID10] = BTRFS_ERROR_DEV_RAID10_MIN_NOT_MET,
-	[BTRFS_RAID_RAID1]  = BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET,
-	[BTRFS_RAID_DUP]    = 0,
-	[BTRFS_RAID_RAID0]  = 0,
-	[BTRFS_RAID_SINGLE] = 0,
-	[BTRFS_RAID_RAID5]  = BTRFS_ERROR_DEV_RAID5_MIN_NOT_MET,
-	[BTRFS_RAID_RAID6]  = BTRFS_ERROR_DEV_RAID6_MIN_NOT_MET,
-};
+	return btrfs_raid_array[type].raid_name;
+}
 
 static int init_first_rw_device(struct btrfs_trans_handle *trans,
 				struct btrfs_fs_info *fs_info);
@@ -167,12 +171,6 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
  * may be used to exclude some operations from running concurrently without any
  * modifications to the list (see write_all_supers)
  *
- * volume_mutex
- * ------------
- * coarse lock owned by a mounted filesystem; used to exclude some operations
- * that cannot run in parallel and affect the higher-level properties of the
- * filesystem like: device add/deleting/resize/replace, or balance
- *
  * balance_mutex
  * -------------
  * protects balance structures (status, state) and context accessed from
@@ -197,6 +195,41 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
  *     device_list_mutex
  *       chunk_mutex
  *     balance_mutex
+ *
+ *
+ * Exclusive operations, BTRFS_FS_EXCL_OP
+ * ======================================
+ *
+ * Maintains the exclusivity of the following operations that apply to the
+ * whole filesystem and cannot run in parallel.
+ *
+ * - Balance (*)
+ * - Device add
+ * - Device remove
+ * - Device replace (*)
+ * - Resize
+ *
+ * The device operations (as above) can be in one of the following states:
+ *
+ * - Running state
+ * - Paused state
+ * - Completed state
+ *
+ * Only device operations marked with (*) can go into the Paused state for the
+ * following reasons:
+ *
+ * - ioctl (only Balance can be Paused through ioctl)
+ * - filesystem remounted as read-only
+ * - filesystem unmounted and mounted as read-only
+ * - system power-cycle and filesystem mounted as read-only
+ * - filesystem or device errors leading to forced read-only
+ *
+ * BTRFS_FS_EXCL_OP flag is set and cleared using atomic operations.
+ * During the course of Paused state, the BTRFS_FS_EXCL_OP remains set.
+ * A device operation in Paused or Running state can be canceled or resumed
+ * either by ioctl (Balance only) or when remounted as read-write.
+ * BTRFS_FS_EXCL_OP flag is cleared when the device operation is canceled or
+ * completed.
  */
 
 DEFINE_MUTEX(uuid_mutex);
@@ -227,14 +260,14 @@ static struct btrfs_fs_devices *alloc_fs_devices(const u8 *fsid)
 	INIT_LIST_HEAD(&fs_devs->devices);
 	INIT_LIST_HEAD(&fs_devs->resized_devices);
 	INIT_LIST_HEAD(&fs_devs->alloc_list);
-	INIT_LIST_HEAD(&fs_devs->list);
+	INIT_LIST_HEAD(&fs_devs->fs_list);
 	if (fsid)
 		memcpy(fs_devs->fsid, fsid, BTRFS_FSID_SIZE);
 
 	return fs_devs;
 }
 
-static void free_device(struct btrfs_device *device)
+void btrfs_free_device(struct btrfs_device *device)
 {
 	rcu_string_free(device->name);
 	bio_put(device->flush_bio);
@@ -249,7 +282,7 @@ static void free_fs_devices(struct btrfs_fs_devices *fs_devices)
 		device = list_entry(fs_devices->devices.next,
 				    struct btrfs_device, dev_list);
 		list_del(&device->dev_list);
-		free_device(device);
+		btrfs_free_device(device);
 	}
 	kfree(fs_devices);
 }
@@ -273,8 +306,8 @@ void __exit btrfs_cleanup_fs_uuids(void)
 
 	while (!list_empty(&fs_uuids)) {
 		fs_devices = list_entry(fs_uuids.next,
-					struct btrfs_fs_devices, list);
-		list_del(&fs_devices->list);
+					struct btrfs_fs_devices, fs_list);
+		list_del(&fs_devices->fs_list);
 		free_fs_devices(fs_devices);
 	}
 }
@@ -282,7 +315,7 @@ void __exit btrfs_cleanup_fs_uuids(void)
 /*
  * Returns a pointer to a new btrfs_device on success; ERR_PTR() on error.
  * Returned struct is not linked onto any lists and must be destroyed using
- * free_device.
+ * btrfs_free_device.
  */
 static struct btrfs_device *__alloc_device(void)
 {
@@ -327,10 +360,9 @@ static struct btrfs_device *__alloc_device(void)
 static struct btrfs_device *find_device(struct btrfs_fs_devices *fs_devices,
 		u64 devid, const u8 *uuid)
 {
-	struct list_head *head = &fs_devices->devices;
 	struct btrfs_device *dev;
 
-	list_for_each_entry(dev, head, dev_list) {
+	list_for_each_entry(dev, &fs_devices->devices, dev_list) {
 		if (dev->devid == devid &&
 		    (!uuid || !memcmp(dev->uuid, uuid, BTRFS_UUID_SIZE))) {
 			return dev;
@@ -343,7 +375,7 @@ static noinline struct btrfs_fs_devices *find_fsid(u8 *fsid)
 {
 	struct btrfs_fs_devices *fs_devices;
 
-	list_for_each_entry(fs_devices, &fs_uuids, list) {
+	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
 		if (memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE) == 0)
 			return fs_devices;
 	}
@@ -607,7 +639,7 @@ static void btrfs_free_stale_devices(const char *path,
 	struct btrfs_fs_devices *fs_devs, *tmp_fs_devs;
 	struct btrfs_device *dev, *tmp_dev;
 
-	list_for_each_entry_safe(fs_devs, tmp_fs_devs, &fs_uuids, list) {
+	list_for_each_entry_safe(fs_devs, tmp_fs_devs, &fs_uuids, fs_list) {
 
 		if (fs_devs->opened)
 			continue;
@@ -632,13 +664,13 @@ static void btrfs_free_stale_devices(const char *path,
 			/* delete the stale device */
 			if (fs_devs->num_devices == 1) {
 				btrfs_sysfs_remove_fsid(fs_devs);
-				list_del(&fs_devs->list);
+				list_del(&fs_devs->fs_list);
 				free_fs_devices(fs_devs);
 				break;
 			} else {
 				fs_devs->num_devices--;
 				list_del(&dev->dev_list);
-				free_device(dev);
+				btrfs_free_device(dev);
 			}
 		}
 	}
@@ -732,7 +764,7 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 		if (IS_ERR(fs_devices))
 			return ERR_CAST(fs_devices);
 
-		list_add(&fs_devices->list, &fs_uuids);
+		list_add(&fs_devices->fs_list, &fs_uuids);
 
 		device = NULL;
 	} else {
@@ -753,7 +785,7 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 
 		name = rcu_string_strdup(path, GFP_NOFS);
 		if (!name) {
-			free_device(device);
+			btrfs_free_device(device);
 			return ERR_PTR(-ENOMEM);
 		}
 		rcu_assign_pointer(device->name, name);
@@ -866,7 +898,7 @@ static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig)
 			name = rcu_string_strdup(orig_dev->name->str,
 					GFP_KERNEL);
 			if (!name) {
-				free_device(device);
+				btrfs_free_device(device);
 				goto error;
 			}
 			rcu_assign_pointer(device->name, name);
@@ -938,7 +970,7 @@ again:
 		}
 		list_del_init(&device->dev_list);
 		fs_devices->num_devices--;
-		free_device(device);
+		btrfs_free_device(device);
 	}
 
 	if (fs_devices->seed) {
@@ -956,7 +988,7 @@ static void free_device_rcu(struct rcu_head *head)
 	struct btrfs_device *device;
 
 	device = container_of(head, struct btrfs_device, rcu);
-	free_device(device);
+	btrfs_free_device(device);
 }
 
 static void btrfs_close_bdev(struct btrfs_device *device)
@@ -1005,7 +1037,7 @@ static void btrfs_prepare_close_one_device(struct btrfs_device *device)
 	new_device->fs_devices = device->fs_devices;
 }
 
-static int __btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
+static int close_fs_devices(struct btrfs_fs_devices *fs_devices)
 {
 	struct btrfs_device *device, *tmp;
 	struct list_head pending_put;
@@ -1050,7 +1082,7 @@ int btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
 	int ret;
 
 	mutex_lock(&uuid_mutex);
-	ret = __btrfs_close_devices(fs_devices);
+	ret = close_fs_devices(fs_devices);
 	if (!fs_devices->opened) {
 		seed_devices = fs_devices->seed;
 		fs_devices->seed = NULL;
@@ -1060,23 +1092,22 @@ int btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
 	while (seed_devices) {
 		fs_devices = seed_devices;
 		seed_devices = fs_devices->seed;
-		__btrfs_close_devices(fs_devices);
+		close_fs_devices(fs_devices);
 		free_fs_devices(fs_devices);
 	}
 	return ret;
 }
 
-static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
+static int open_fs_devices(struct btrfs_fs_devices *fs_devices,
 				fmode_t flags, void *holder)
 {
-	struct list_head *head = &fs_devices->devices;
 	struct btrfs_device *device;
 	struct btrfs_device *latest_dev = NULL;
 	int ret = 0;
 
 	flags |= FMODE_EXCL;
 
-	list_for_each_entry(device, head, dev_list) {
+	list_for_each_entry(device, &fs_devices->devices, dev_list) {
 		/* Just open everything we can; ignore failures here */
 		if (btrfs_open_one_device(fs_devices, device, flags, holder))
 			continue;
@@ -1115,15 +1146,16 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 {
 	int ret;
 
-	mutex_lock(&uuid_mutex);
+	mutex_lock(&fs_devices->device_list_mutex);
 	if (fs_devices->opened) {
 		fs_devices->opened++;
 		ret = 0;
 	} else {
 		list_sort(NULL, &fs_devices->devices, devid_cmp);
-		ret = __btrfs_open_devices(fs_devices, flags, holder);
+		ret = open_fs_devices(fs_devices, flags, holder);
 	}
-	mutex_unlock(&uuid_mutex);
+	mutex_unlock(&fs_devices->device_list_mutex);
+
 	return ret;
 }
 
@@ -1201,31 +1233,29 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 	 */
 	bytenr = btrfs_sb_offset(0);
 	flags |= FMODE_EXCL;
-	mutex_lock(&uuid_mutex);
 
 	bdev = blkdev_get_by_path(path, flags, holder);
-	if (IS_ERR(bdev)) {
-		ret = PTR_ERR(bdev);
-		goto error;
-	}
+	if (IS_ERR(bdev))
+		return PTR_ERR(bdev);
 
 	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super)) {
 		ret = -EINVAL;
 		goto error_bdev_put;
 	}
 
+	mutex_lock(&uuid_mutex);
 	device = device_list_add(path, disk_super);
 	if (IS_ERR(device))
 		ret = PTR_ERR(device);
 	else
 		*fs_devices_ret = device->fs_devices;
+	mutex_unlock(&uuid_mutex);
 
 	btrfs_release_disk_super(page);
 
 error_bdev_put:
 	blkdev_put(bdev, flags);
-error:
-	mutex_unlock(&uuid_mutex);
+
 	return ret;
 }
 
@@ -1857,11 +1887,11 @@ static int btrfs_check_raid_min_devices(struct btrfs_fs_info *fs_info,
 	} while (read_seqretry(&fs_info->profiles_lock, seq));
 
 	for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
-		if (!(all_avail & btrfs_raid_group[i]))
+		if (!(all_avail & btrfs_raid_array[i].bg_flag))
 			continue;
 
 		if (num_devices < btrfs_raid_array[i].devs_min) {
-			int ret = btrfs_raid_mindev_error[i];
+			int ret = btrfs_raid_array[i].mindev_error;
 
 			if (ret)
 				return ret;
@@ -1917,13 +1947,13 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 {
 	struct btrfs_device *device;
 	struct btrfs_fs_devices *cur_devices;
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
 	u64 num_devices;
 	int ret = 0;
 
-	mutex_lock(&fs_info->volume_mutex);
 	mutex_lock(&uuid_mutex);
 
-	num_devices = fs_info->fs_devices->num_devices;
+	num_devices = fs_devices->num_devices;
 	btrfs_dev_replace_read_lock(&fs_info->dev_replace);
 	if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace)) {
 		WARN_ON(num_devices < 1);
@@ -1986,27 +2016,32 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	 * (super_copy) should hold the device list mutex.
 	 */
 
+	/*
+	 * In normal cases the cur_devices == fs_devices. But in case
+	 * of deleting a seed device, the cur_devices should point to
+	 * its own fs_devices listed under the fs_devices->seed.
+	 */
 	cur_devices = device->fs_devices;
-	mutex_lock(&fs_info->fs_devices->device_list_mutex);
+	mutex_lock(&fs_devices->device_list_mutex);
 	list_del_rcu(&device->dev_list);
 
-	device->fs_devices->num_devices--;
-	device->fs_devices->total_devices--;
+	cur_devices->num_devices--;
+	cur_devices->total_devices--;
 
 	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
-		device->fs_devices->missing_devices--;
+		cur_devices->missing_devices--;
 
 	btrfs_assign_next_active_device(fs_info, device, NULL);
 
 	if (device->bdev) {
-		device->fs_devices->open_devices--;
+		cur_devices->open_devices--;
 		/* remove sysfs entry */
-		btrfs_sysfs_rm_device_link(fs_info->fs_devices, device);
+		btrfs_sysfs_rm_device_link(fs_devices, device);
 	}
 
 	num_devices = btrfs_super_num_devices(fs_info->super_copy) - 1;
 	btrfs_set_super_num_devices(fs_info->super_copy, num_devices);
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+	mutex_unlock(&fs_devices->device_list_mutex);
 
 	/*
 	 * at this point, the device is zero sized and detached from
@@ -2020,8 +2055,6 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	call_rcu(&device->rcu, free_device_rcu);
 
 	if (cur_devices->open_devices == 0) {
-		struct btrfs_fs_devices *fs_devices;
-		fs_devices = fs_info->fs_devices;
 		while (fs_devices) {
 			if (fs_devices->seed == cur_devices) {
 				fs_devices->seed = cur_devices->seed;
@@ -2030,20 +2063,19 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 			fs_devices = fs_devices->seed;
 		}
 		cur_devices->seed = NULL;
-		__btrfs_close_devices(cur_devices);
+		close_fs_devices(cur_devices);
 		free_fs_devices(cur_devices);
 	}
 
 out:
 	mutex_unlock(&uuid_mutex);
-	mutex_unlock(&fs_info->volume_mutex);
 	return ret;
 
 error_undo:
 	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 		mutex_lock(&fs_info->chunk_mutex);
 		list_add(&device->dev_alloc_list,
-			 &fs_info->fs_devices->alloc_list);
+			 &fs_devices->alloc_list);
 		device->fs_devices->rw_devices++;
 		mutex_unlock(&fs_info->chunk_mutex);
 	}
@@ -2112,7 +2144,7 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
 			tmp_fs_devices = tmp_fs_devices->seed;
 		}
 		fs_devices->seed = NULL;
-		__btrfs_close_devices(fs_devices);
+		close_fs_devices(fs_devices);
 		free_fs_devices(fs_devices);
 	}
 }
@@ -2120,23 +2152,23 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
 void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 				      struct btrfs_device *tgtdev)
 {
-	mutex_lock(&uuid_mutex);
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+
 	WARN_ON(!tgtdev);
-	mutex_lock(&fs_info->fs_devices->device_list_mutex);
+	mutex_lock(&fs_devices->device_list_mutex);
 
-	btrfs_sysfs_rm_device_link(fs_info->fs_devices, tgtdev);
+	btrfs_sysfs_rm_device_link(fs_devices, tgtdev);
 
 	if (tgtdev->bdev)
-		fs_info->fs_devices->open_devices--;
+		fs_devices->open_devices--;
 
-	fs_info->fs_devices->num_devices--;
+	fs_devices->num_devices--;
 
 	btrfs_assign_next_active_device(fs_info, tgtdev, NULL);
 
 	list_del_rcu(&tgtdev->dev_list);
 
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-	mutex_unlock(&uuid_mutex);
+	mutex_unlock(&fs_devices->device_list_mutex);
 
 	/*
 	 * The update_dev_time() with in btrfs_scratch_superblocks()
@@ -2188,10 +2220,6 @@ int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
 		struct btrfs_device *tmp;
 
 		devices = &fs_info->fs_devices->devices;
-		/*
-		 * It is safe to read the devices since the volume_mutex
-		 * is held by the caller.
-		 */
 		list_for_each_entry(tmp, devices, dev_list) {
 			if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
 					&tmp->dev_state) && !tmp->bdev) {
@@ -2259,7 +2287,7 @@ static int btrfs_prepare_sprout(struct btrfs_fs_info *fs_info)
 		return PTR_ERR(old_devices);
 	}
 
-	list_add(&old_devices->list, &fs_uuids);
+	list_add(&old_devices->fs_list, &fs_uuids);
 
 	memcpy(seed_devices, fs_devices, sizeof(*seed_devices));
 	seed_devices->opened = 1;
@@ -2570,7 +2598,7 @@ error_trans:
 	if (trans)
 		btrfs_end_transaction(trans);
 error_free_device:
-	free_device(device);
+	btrfs_free_device(device);
 error:
 	blkdev_put(bdev, FMODE_EXCL);
 	if (seeding_dev && !unlocked) {
@@ -2580,99 +2608,6 @@ error:
 	return ret;
 }
 
-int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
-				  const char *device_path,
-				  struct btrfs_device *srcdev,
-				  struct btrfs_device **device_out)
-{
-	struct btrfs_device *device;
-	struct block_device *bdev;
-	struct list_head *devices;
-	struct rcu_string *name;
-	u64 devid = BTRFS_DEV_REPLACE_DEVID;
-	int ret = 0;
-
-	*device_out = NULL;
-	if (fs_info->fs_devices->seeding) {
-		btrfs_err(fs_info, "the filesystem is a seed filesystem!");
-		return -EINVAL;
-	}
-
-	bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL,
-				  fs_info->bdev_holder);
-	if (IS_ERR(bdev)) {
-		btrfs_err(fs_info, "target device %s is invalid!", device_path);
-		return PTR_ERR(bdev);
-	}
-
-	filemap_write_and_wait(bdev->bd_inode->i_mapping);
-
-	devices = &fs_info->fs_devices->devices;
-	list_for_each_entry(device, devices, dev_list) {
-		if (device->bdev == bdev) {
-			btrfs_err(fs_info,
-				  "target device is in the filesystem!");
-			ret = -EEXIST;
-			goto error;
-		}
-	}
-
-
-	if (i_size_read(bdev->bd_inode) <
-	    btrfs_device_get_total_bytes(srcdev)) {
-		btrfs_err(fs_info,
-			  "target device is smaller than source device!");
-		ret = -EINVAL;
-		goto error;
-	}
-
-
-	device = btrfs_alloc_device(NULL, &devid, NULL);
-	if (IS_ERR(device)) {
-		ret = PTR_ERR(device);
-		goto error;
-	}
-
-	name = rcu_string_strdup(device_path, GFP_KERNEL);
-	if (!name) {
-		free_device(device);
-		ret = -ENOMEM;
-		goto error;
-	}
-	rcu_assign_pointer(device->name, name);
-
-	mutex_lock(&fs_info->fs_devices->device_list_mutex);
-	set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
-	device->generation = 0;
-	device->io_width = fs_info->sectorsize;
-	device->io_align = fs_info->sectorsize;
-	device->sector_size = fs_info->sectorsize;
-	device->total_bytes = btrfs_device_get_total_bytes(srcdev);
-	device->disk_total_bytes = btrfs_device_get_disk_total_bytes(srcdev);
-	device->bytes_used = btrfs_device_get_bytes_used(srcdev);
-	device->commit_total_bytes = srcdev->commit_total_bytes;
-	device->commit_bytes_used = device->bytes_used;
-	device->fs_info = fs_info;
-	device->bdev = bdev;
-	set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
-	set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
-	device->mode = FMODE_EXCL;
-	device->dev_stats_valid = 1;
-	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
-	device->fs_devices = fs_info->fs_devices;
-	list_add(&device->dev_list, &fs_info->fs_devices->devices);
-	fs_info->fs_devices->num_devices++;
-	fs_info->fs_devices->open_devices++;
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-
-	*device_out = device;
-	return ret;
-
-error:
-	blkdev_put(bdev, FMODE_EXCL);
-	return ret;
-}
-
 static noinline int btrfs_update_device(struct btrfs_trans_handle *trans,
 					struct btrfs_device *device)
 {
@@ -3273,24 +3208,12 @@ static void update_balance_args(struct btrfs_balance_control *bctl)
 }
 
 /*
- * Should be called with both balance and volume mutexes held to
- * serialize other volume operations (add_dev/rm_dev/resize) with
- * restriper.  Same goes for unset_balance_control.
+ * Clear the balance status in fs_info and delete the balance item from disk.
  */
-static void set_balance_control(struct btrfs_balance_control *bctl)
-{
-	struct btrfs_fs_info *fs_info = bctl->fs_info;
-
-	BUG_ON(fs_info->balance_ctl);
-
-	spin_lock(&fs_info->balance_lock);
-	fs_info->balance_ctl = bctl;
-	spin_unlock(&fs_info->balance_lock);
-}
-
-static void unset_balance_control(struct btrfs_fs_info *fs_info)
+static void reset_balance_state(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_balance_control *bctl = fs_info->balance_ctl;
+	int ret;
 
 	BUG_ON(!fs_info->balance_ctl);
 
@@ -3299,6 +3222,9 @@ static void unset_balance_control(struct btrfs_fs_info *fs_info)
 	spin_unlock(&fs_info->balance_lock);
 
 	kfree(bctl);
+	ret = del_balance_item(fs_info);
+	if (ret)
+		btrfs_handle_fs_error(fs_info, ret, NULL);
 }
 
 /*
@@ -3835,18 +3761,6 @@ static inline int balance_need_close(struct btrfs_fs_info *fs_info)
 		 atomic_read(&fs_info->balance_cancel_req) == 0);
 }
 
-static void __cancel_balance(struct btrfs_fs_info *fs_info)
-{
-	int ret;
-
-	unset_balance_control(fs_info);
-	ret = del_balance_item(fs_info);
-	if (ret)
-		btrfs_handle_fs_error(fs_info, ret, NULL);
-
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
-}
-
 /* Non-zero return value signifies invalidity */
 static inline int validate_convert_profile(struct btrfs_balance_args *bctl_arg,
 		u64 allowed)
@@ -3857,12 +3771,12 @@ static inline int validate_convert_profile(struct btrfs_balance_args *bctl_arg,
 }
 
 /*
- * Should be called with both balance and volume mutexes held
+ * Should be called with balance mutexe held
  */
-int btrfs_balance(struct btrfs_balance_control *bctl,
+int btrfs_balance(struct btrfs_fs_info *fs_info,
+		  struct btrfs_balance_control *bctl,
 		  struct btrfs_ioctl_balance_args *bargs)
 {
-	struct btrfs_fs_info *fs_info = bctl->fs_info;
 	u64 meta_target, data_target;
 	u64 allowed;
 	int mixed = 0;
@@ -3891,7 +3805,7 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 		    !(bctl->flags & BTRFS_BALANCE_METADATA) ||
 		    memcmp(&bctl->data, &bctl->meta, sizeof(bctl->data))) {
 			btrfs_err(fs_info,
-				  "with mixed groups data and metadata balance options must be the same");
+	  "balance: mixed groups data and metadata options must be the same");
 			ret = -EINVAL;
 			goto out;
 		}
@@ -3913,23 +3827,29 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 		allowed |= (BTRFS_BLOCK_GROUP_RAID10 |
 			    BTRFS_BLOCK_GROUP_RAID6);
 	if (validate_convert_profile(&bctl->data, allowed)) {
+		int index = btrfs_bg_flags_to_raid_index(bctl->data.target);
+
 		btrfs_err(fs_info,
-			  "unable to start balance with target data profile %llu",
-			  bctl->data.target);
+			  "balance: invalid convert data profile %s",
+			  get_raid_name(index));
 		ret = -EINVAL;
 		goto out;
 	}
 	if (validate_convert_profile(&bctl->meta, allowed)) {
+		int index = btrfs_bg_flags_to_raid_index(bctl->meta.target);
+
 		btrfs_err(fs_info,
-			  "unable to start balance with target metadata profile %llu",
-			  bctl->meta.target);
+			  "balance: invalid convert metadata profile %s",
+			  get_raid_name(index));
 		ret = -EINVAL;
 		goto out;
 	}
 	if (validate_convert_profile(&bctl->sys, allowed)) {
+		int index = btrfs_bg_flags_to_raid_index(bctl->sys.target);
+
 		btrfs_err(fs_info,
-			  "unable to start balance with target system profile %llu",
-			  bctl->sys.target);
+			  "balance: invalid convert system profile %s",
+			  get_raid_name(index));
 		ret = -EINVAL;
 		goto out;
 	}
@@ -3950,10 +3870,10 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 		     !(bctl->meta.target & allowed))) {
 			if (bctl->flags & BTRFS_BALANCE_FORCE) {
 				btrfs_info(fs_info,
-					   "force reducing metadata integrity");
+				"balance: force reducing metadata integrity");
 			} else {
 				btrfs_err(fs_info,
-					  "balance will reduce metadata integrity, use force if you want this");
+	"balance: reduces metadata integrity, use --force if you want this");
 				ret = -EINVAL;
 				goto out;
 			}
@@ -3967,9 +3887,12 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 		bctl->data.target : fs_info->avail_data_alloc_bits;
 	if (btrfs_get_num_tolerated_disk_barrier_failures(meta_target) <
 		btrfs_get_num_tolerated_disk_barrier_failures(data_target)) {
+		int meta_index = btrfs_bg_flags_to_raid_index(meta_target);
+		int data_index = btrfs_bg_flags_to_raid_index(data_target);
+
 		btrfs_warn(fs_info,
-			   "metadata profile 0x%llx has lower redundancy than data profile 0x%llx",
-			   meta_target, data_target);
+	"balance: metadata profile %s has lower redundancy than data profile %s",
+			   get_raid_name(meta_index), get_raid_name(data_index));
 	}
 
 	ret = insert_balance_item(fs_info, bctl);
@@ -3978,7 +3901,10 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 
 	if (!(bctl->flags & BTRFS_BALANCE_RESUME)) {
 		BUG_ON(ret == -EEXIST);
-		set_balance_control(bctl);
+		BUG_ON(fs_info->balance_ctl);
+		spin_lock(&fs_info->balance_lock);
+		fs_info->balance_ctl = bctl;
+		spin_unlock(&fs_info->balance_lock);
 	} else {
 		BUG_ON(ret != -EEXIST);
 		spin_lock(&fs_info->balance_lock);
@@ -3986,22 +3912,24 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 		spin_unlock(&fs_info->balance_lock);
 	}
 
-	atomic_inc(&fs_info->balance_running);
+	ASSERT(!test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags));
+	set_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags);
 	mutex_unlock(&fs_info->balance_mutex);
 
 	ret = __btrfs_balance(fs_info);
 
 	mutex_lock(&fs_info->balance_mutex);
-	atomic_dec(&fs_info->balance_running);
+	clear_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags);
 
 	if (bargs) {
 		memset(bargs, 0, sizeof(*bargs));
-		update_ioctl_balance_args(fs_info, 0, bargs);
+		btrfs_update_ioctl_balance_args(fs_info, bargs);
 	}
 
 	if ((ret && ret != -ECANCELED && ret != -ENOSPC) ||
 	    balance_need_close(fs_info)) {
-		__cancel_balance(fs_info);
+		reset_balance_state(fs_info);
+		clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
 	}
 
 	wake_up(&fs_info->balance_wait_q);
@@ -4009,11 +3937,11 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
 	return ret;
 out:
 	if (bctl->flags & BTRFS_BALANCE_RESUME)
-		__cancel_balance(fs_info);
-	else {
+		reset_balance_state(fs_info);
+	else
 		kfree(bctl);
-		clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
-	}
+	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+
 	return ret;
 }
 
@@ -4022,16 +3950,12 @@ static int balance_kthread(void *data)
 	struct btrfs_fs_info *fs_info = data;
 	int ret = 0;
 
-	mutex_lock(&fs_info->volume_mutex);
 	mutex_lock(&fs_info->balance_mutex);
-
 	if (fs_info->balance_ctl) {
-		btrfs_info(fs_info, "continuing balance");
-		ret = btrfs_balance(fs_info->balance_ctl, NULL);
+		btrfs_info(fs_info, "balance: resuming");
+		ret = btrfs_balance(fs_info, fs_info->balance_ctl, NULL);
 	}
-
 	mutex_unlock(&fs_info->balance_mutex);
-	mutex_unlock(&fs_info->volume_mutex);
 
 	return ret;
 }
@@ -4040,15 +3964,15 @@ int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info)
 {
 	struct task_struct *tsk;
 
-	spin_lock(&fs_info->balance_lock);
+	mutex_lock(&fs_info->balance_mutex);
 	if (!fs_info->balance_ctl) {
-		spin_unlock(&fs_info->balance_lock);
+		mutex_unlock(&fs_info->balance_mutex);
 		return 0;
 	}
-	spin_unlock(&fs_info->balance_lock);
+	mutex_unlock(&fs_info->balance_mutex);
 
 	if (btrfs_test_opt(fs_info, SKIP_BALANCE)) {
-		btrfs_info(fs_info, "force skipping balance");
+		btrfs_info(fs_info, "balance: resume skipped");
 		return 0;
 	}
 
@@ -4100,7 +4024,6 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info)
 	leaf = path->nodes[0];
 	item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_balance_item);
 
-	bctl->fs_info = fs_info;
 	bctl->flags = btrfs_balance_flags(leaf, item);
 	bctl->flags |= BTRFS_BALANCE_RESUME;
 
@@ -4111,15 +4034,26 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info)
 	btrfs_balance_sys(leaf, item, &disk_bargs);
 	btrfs_disk_balance_args_to_cpu(&bctl->sys, &disk_bargs);
 
-	WARN_ON(test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags));
+	/*
+	 * This should never happen, as the paused balance state is recovered
+	 * during mount without any chance of other exclusive ops to collide.
+	 *
+	 * This gives the exclusive op status to balance and keeps in paused
+	 * state until user intervention (cancel or umount). If the ownership
+	 * cannot be assigned, show a message but do not fail. The balance
+	 * is in a paused state and must have fs_info::balance_ctl properly
+	 * set up.
+	 */
+	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags))
+		btrfs_warn(fs_info,
+	"balance: cannot set exclusive op status, resume manually");
 
-	mutex_lock(&fs_info->volume_mutex);
 	mutex_lock(&fs_info->balance_mutex);
-
-	set_balance_control(bctl);
-
+	BUG_ON(fs_info->balance_ctl);
+	spin_lock(&fs_info->balance_lock);
+	fs_info->balance_ctl = bctl;
+	spin_unlock(&fs_info->balance_lock);
 	mutex_unlock(&fs_info->balance_mutex);
-	mutex_unlock(&fs_info->volume_mutex);
 out:
 	btrfs_free_path(path);
 	return ret;
@@ -4135,16 +4069,16 @@ int btrfs_pause_balance(struct btrfs_fs_info *fs_info)
 		return -ENOTCONN;
 	}
 
-	if (atomic_read(&fs_info->balance_running)) {
+	if (test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) {
 		atomic_inc(&fs_info->balance_pause_req);
 		mutex_unlock(&fs_info->balance_mutex);
 
 		wait_event(fs_info->balance_wait_q,
-			   atomic_read(&fs_info->balance_running) == 0);
+			   !test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags));
 
 		mutex_lock(&fs_info->balance_mutex);
 		/* we are good with balance_ctl ripped off from under us */
-		BUG_ON(atomic_read(&fs_info->balance_running));
+		BUG_ON(test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags));
 		atomic_dec(&fs_info->balance_pause_req);
 	} else {
 		ret = -ENOTCONN;
@@ -4156,38 +4090,49 @@ int btrfs_pause_balance(struct btrfs_fs_info *fs_info)
 
 int btrfs_cancel_balance(struct btrfs_fs_info *fs_info)
 {
-	if (sb_rdonly(fs_info->sb))
-		return -EROFS;
-
 	mutex_lock(&fs_info->balance_mutex);
 	if (!fs_info->balance_ctl) {
 		mutex_unlock(&fs_info->balance_mutex);
 		return -ENOTCONN;
 	}
 
+	/*
+	 * A paused balance with the item stored on disk can be resumed at
+	 * mount time if the mount is read-write. Otherwise it's still paused
+	 * and we must not allow cancelling as it deletes the item.
+	 */
+	if (sb_rdonly(fs_info->sb)) {
+		mutex_unlock(&fs_info->balance_mutex);
+		return -EROFS;
+	}
+
 	atomic_inc(&fs_info->balance_cancel_req);
 	/*
 	 * if we are running just wait and return, balance item is
 	 * deleted in btrfs_balance in this case
 	 */
-	if (atomic_read(&fs_info->balance_running)) {
+	if (test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) {
 		mutex_unlock(&fs_info->balance_mutex);
 		wait_event(fs_info->balance_wait_q,
-			   atomic_read(&fs_info->balance_running) == 0);
+			   !test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags));
 		mutex_lock(&fs_info->balance_mutex);
 	} else {
-		/* __cancel_balance needs volume_mutex */
 		mutex_unlock(&fs_info->balance_mutex);
-		mutex_lock(&fs_info->volume_mutex);
+		/*
+		 * Lock released to allow other waiters to continue, we'll
+		 * reexamine the status again.
+		 */
 		mutex_lock(&fs_info->balance_mutex);
 
-		if (fs_info->balance_ctl)
-			__cancel_balance(fs_info);
-
-		mutex_unlock(&fs_info->volume_mutex);
+		if (fs_info->balance_ctl) {
+			reset_balance_state(fs_info);
+			clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+			btrfs_info(fs_info, "balance: canceled");
+		}
 	}
 
-	BUG_ON(fs_info->balance_ctl || atomic_read(&fs_info->balance_running));
+	BUG_ON(fs_info->balance_ctl ||
+		test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags));
 	atomic_dec(&fs_info->balance_cancel_req);
 	mutex_unlock(&fs_info->balance_mutex);
 	return 0;
@@ -4264,8 +4209,7 @@ static int btrfs_uuid_scan_kthread(void *data)
 		}
 update_tree:
 		if (!btrfs_is_empty_uuid(root_item.uuid)) {
-			ret = btrfs_uuid_tree_add(trans, fs_info,
-						  root_item.uuid,
+			ret = btrfs_uuid_tree_add(trans, root_item.uuid,
 						  BTRFS_UUID_KEY_SUBVOL,
 						  key.objectid);
 			if (ret < 0) {
@@ -4276,7 +4220,7 @@ update_tree:
 		}
 
 		if (!btrfs_is_empty_uuid(root_item.received_uuid)) {
-			ret = btrfs_uuid_tree_add(trans, fs_info,
+			ret = btrfs_uuid_tree_add(trans,
 						  root_item.received_uuid,
 						 BTRFS_UUID_KEY_RECEIVED_SUBVOL,
 						  key.objectid);
@@ -4482,7 +4426,7 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
 	if (!path)
 		return -ENOMEM;
 
-	path->reada = READA_FORWARD;
+	path->reada = READA_BACK;
 
 	mutex_lock(&fs_info->chunk_mutex);
 
@@ -6043,9 +5987,8 @@ int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 	return __btrfs_map_block(fs_info, op, logical, length, bbio_ret, 0, 1);
 }
 
-int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
-		     u64 chunk_start, u64 physical, u64 devid,
-		     u64 **logical, int *naddrs, int *stripe_len)
+int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
+		     u64 physical, u64 **logical, int *naddrs, int *stripe_len)
 {
 	struct extent_map *em;
 	struct map_lookup *map;
@@ -6077,8 +6020,6 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
 	BUG_ON(!buf); /* -ENOMEM */
 
 	for (i = 0; i < map->num_stripes; i++) {
-		if (devid && map->stripes[i].dev->devid != devid)
-			continue;
 		if (map->stripes[i].physical > physical ||
 		    map->stripes[i].physical + length <= physical)
 			continue;
@@ -6410,7 +6351,7 @@ static struct btrfs_device *add_missing_dev(struct btrfs_fs_devices *fs_devices,
  *
  * Return: a pointer to a new &struct btrfs_device on success; ERR_PTR()
  * on error.  Returned struct is not linked onto any lists and must be
- * destroyed with free_device.
+ * destroyed with btrfs_free_device.
  */
 struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 					const u64 *devid,
@@ -6433,7 +6374,7 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 
 		ret = find_next_devid(fs_info, &tmp);
 		if (ret) {
-			free_device(dev);
+			btrfs_free_device(dev);
 			return ERR_PTR(ret);
 		}
 	}
@@ -6684,8 +6625,7 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info,
 	if (IS_ERR(fs_devices))
 		return fs_devices;
 
-	ret = __btrfs_open_devices(fs_devices, FMODE_READ,
-				   fs_info->bdev_holder);
+	ret = open_fs_devices(fs_devices, FMODE_READ, fs_info->bdev_holder);
 	if (ret) {
 		free_fs_devices(fs_devices);
 		fs_devices = ERR_PTR(ret);
@@ -6693,7 +6633,7 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info,
 	}
 
 	if (!fs_devices->seeding) {
-		__btrfs_close_devices(fs_devices);
+		close_fs_devices(fs_devices);
 		free_fs_devices(fs_devices);
 		fs_devices = ERR_PTR(-EINVAL);
 		goto out;
@@ -7002,6 +6942,10 @@ int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info)
 	if (!path)
 		return -ENOMEM;
 
+	/*
+	 * uuid_mutex is needed only if we are mounting a sprout FS
+	 * otherwise we don't need it.
+	 */
 	mutex_lock(&uuid_mutex);
 	mutex_lock(&fs_info->chunk_mutex);
 

fs/btrfs/volumes.h

@@ -208,6 +208,7 @@ BTRFS_DEVICE_GETSET_FUNCS(bytes_used);
 
 struct btrfs_fs_devices {
 	u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */
+	struct list_head fs_list;
 
 	u64 num_devices;
 	u64 open_devices;
@@ -229,7 +230,6 @@ struct btrfs_fs_devices {
 	struct list_head resized_devices;
 	/* devices not currently being allocated */
 	struct list_head alloc_list;
-	struct list_head list;
 
 	struct btrfs_fs_devices *seed;
 	int seeding;
@@ -329,11 +329,12 @@ struct btrfs_raid_attr {
 	int tolerated_failures; /* max tolerated fail devs */
 	int devs_increment;	/* ndevs has to be a multiple of this */
 	int ncopies;		/* how many copies to data has */
+	int mindev_error;	/* error code if min devs requisite is unmet */
+	const char raid_name[8]; /* name of the raid */
+	u64 bg_flag;		/* block group flag of the raid */
 };
 
 extern const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES];
-extern const int btrfs_raid_mindev_error[BTRFS_NR_RAID_TYPES];
-extern const u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES];
 
 struct map_lookup {
 	u64 type;
@@ -351,8 +352,6 @@ struct map_lookup {
 struct btrfs_balance_args;
 struct btrfs_balance_progress;
 struct btrfs_balance_control {
-	struct btrfs_fs_info *fs_info;
-
 	struct btrfs_balance_args data;
 	struct btrfs_balance_args meta;
 	struct btrfs_balance_args sys;
@@ -393,9 +392,8 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 		     u64 logical, u64 *length,
 		     struct btrfs_bio **bbio_ret);
-int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
-		     u64 chunk_start, u64 physical, u64 devid,
-		     u64 **logical, int *naddrs, int *stripe_len);
+int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
+		     u64 physical, u64 **logical, int *naddrs, int *stripe_len);
 int btrfs_read_sys_array(struct btrfs_fs_info *fs_info);
 int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info);
 int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
@@ -421,6 +419,7 @@ int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
 struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 					const u64 *devid,
 					const u8 *uuid);
+void btrfs_free_device(struct btrfs_device *device);
 int btrfs_rm_device(struct btrfs_fs_info *fs_info,
 		    const char *device_path, u64 devid);
 void __exit btrfs_cleanup_fs_uuids(void);
@@ -431,11 +430,8 @@ struct btrfs_device *btrfs_find_device(struct btrfs_fs_info *fs_info, u64 devid,
 				       u8 *uuid, u8 *fsid);
 int btrfs_shrink_device(struct btrfs_device *device, u64 new_size);
 int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *path);
-int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
-				  const char *device_path,
-				  struct btrfs_device *srcdev,
-				  struct btrfs_device **device_out);
-int btrfs_balance(struct btrfs_balance_control *bctl,
+int btrfs_balance(struct btrfs_fs_info *fs_info,
+		  struct btrfs_balance_control *bctl,
 		  struct btrfs_ioctl_balance_args *bargs);
 int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info);
 int btrfs_recover_balance(struct btrfs_fs_info *fs_info);
@@ -553,6 +549,8 @@ static inline enum btrfs_raid_types btrfs_bg_flags_to_raid_index(u64 flags)
 	return BTRFS_RAID_SINGLE; /* BTRFS_BLOCK_GROUP_SINGLE */
 }
 
+const char *get_raid_name(enum btrfs_raid_types type);
+
 void btrfs_update_commit_device_size(struct btrfs_fs_info *fs_info);
 void btrfs_update_commit_device_bytes_used(struct btrfs_transaction *trans);
 

include/trace/events/btrfs.h

@@ -121,9 +121,9 @@ TRACE_EVENT(btrfs_transaction_commit,
 		__entry->root_objectid	= root->root_key.objectid;
 	),
 
-	TP_printk_btrfs("root = %llu(%s), gen = %llu",
+	TP_printk_btrfs("root=%llu(%s) gen=%llu",
 		  show_root_type(__entry->root_objectid),
-		  (unsigned long long)__entry->generation)
+		  __entry->generation)
 );
 
 DECLARE_EVENT_CLASS(btrfs__inode,
@@ -133,7 +133,7 @@ DECLARE_EVENT_CLASS(btrfs__inode,
 	TP_ARGS(inode),
 
 	TP_STRUCT__entry_btrfs(
-		__field(	ino_t,  ino			)
+		__field(	u64,  ino			)
 		__field(	blkcnt_t,  blocks		)
 		__field(	u64,  disk_i_size		)
 		__field(	u64,  generation		)
@@ -143,7 +143,7 @@ DECLARE_EVENT_CLASS(btrfs__inode,
 	),
 
 	TP_fast_assign_btrfs(btrfs_sb(inode->i_sb),
-		__entry->ino	= inode->i_ino;
+		__entry->ino	= btrfs_ino(BTRFS_I(inode));
 		__entry->blocks	= inode->i_blocks;
 		__entry->disk_i_size  = BTRFS_I(inode)->disk_i_size;
 		__entry->generation = BTRFS_I(inode)->generation;
@@ -153,15 +153,15 @@ DECLARE_EVENT_CLASS(btrfs__inode,
 				BTRFS_I(inode)->root->root_key.objectid;
 	),
 
-	TP_printk_btrfs("root=%llu(%s) gen=%llu ino=%lu blocks=%llu "
+	TP_printk_btrfs("root=%llu(%s) gen=%llu ino=%llu blocks=%llu "
 		  "disk_i_size=%llu last_trans=%llu logged_trans=%llu",
 		  show_root_type(__entry->root_objectid),
-		  (unsigned long long)__entry->generation,
-		  (unsigned long)__entry->ino,
+		  __entry->generation,
+		  __entry->ino,
 		  (unsigned long long)__entry->blocks,
-		  (unsigned long long)__entry->disk_i_size,
-		  (unsigned long long)__entry->last_trans,
-		  (unsigned long long)__entry->logged_trans)
+		  __entry->disk_i_size,
+		  __entry->last_trans,
+		  __entry->logged_trans)
 );
 
 DEFINE_EVENT(btrfs__inode, btrfs_inode_new,
@@ -244,23 +244,25 @@ TRACE_EVENT_CONDITION(btrfs_get_extent,
 		  "block_len=%llu flags=%s refs=%u "
 		  "compress_type=%u",
 		  show_root_type(__entry->root_objectid),
-		  (unsigned long long)__entry->ino,
-		  (unsigned long long)__entry->start,
-		  (unsigned long long)__entry->len,
-		  (unsigned long long)__entry->orig_start,
+		  __entry->ino,
+		  __entry->start,
+		  __entry->len,
+		  __entry->orig_start,
 		  show_map_type(__entry->block_start),
-		  (unsigned long long)__entry->block_len,
+		  __entry->block_len,
 		  show_map_flags(__entry->flags),
 		  __entry->refs, __entry->compress_type)
 );
 
 TRACE_EVENT(btrfs_handle_em_exist,
 
-	TP_PROTO(const struct extent_map *existing, const struct extent_map *map, u64 start, u64 len),
+	TP_PROTO(struct btrfs_fs_info *fs_info,
+		const struct extent_map *existing, const struct extent_map *map,
+		u64 start, u64 len),
 
-	TP_ARGS(existing, map, start, len),
+	TP_ARGS(fs_info, existing, map, start, len),
 
-	TP_STRUCT__entry(
+	TP_STRUCT__entry_btrfs(
 		__field(	u64,  e_start		)
 		__field(	u64,  e_len		)
 		__field(	u64,  map_start		)
@@ -269,7 +271,7 @@ TRACE_EVENT(btrfs_handle_em_exist,
 		__field(	u64,  len		)
 	),
 
-	TP_fast_assign(
+	TP_fast_assign_btrfs(fs_info,
 		__entry->e_start	= existing->start;
 		__entry->e_len		= existing->len;
 		__entry->map_start	= map->start;
@@ -278,15 +280,15 @@ TRACE_EVENT(btrfs_handle_em_exist,
 		__entry->len		= len;
 	),
 
-	TP_printk("start=%llu len=%llu "
+	TP_printk_btrfs("start=%llu len=%llu "
 		  "existing(start=%llu len=%llu) "
 		  "em(start=%llu len=%llu)",
-		  (unsigned long long)__entry->start,
-		  (unsigned long long)__entry->len,
-		  (unsigned long long)__entry->e_start,
-		  (unsigned long long)__entry->e_len,
-		  (unsigned long long)__entry->map_start,
-		  (unsigned long long)__entry->map_len)
+		  __entry->start,
+		  __entry->len,
+		  __entry->e_start,
+		  __entry->e_len,
+		  __entry->map_start,
+		  __entry->map_len)
 );
 
 /* file extent item */
@@ -443,7 +445,7 @@ DECLARE_EVENT_CLASS(btrfs__ordered_extent,
 	TP_ARGS(inode, ordered),
 
 	TP_STRUCT__entry_btrfs(
-		__field(	ino_t,  ino		)
+		__field(	u64,  ino		)
 		__field(	u64,  file_offset	)
 		__field(	u64,  start		)
 		__field(	u64,  len		)
@@ -457,7 +459,7 @@ DECLARE_EVENT_CLASS(btrfs__ordered_extent,
 	),
 
 	TP_fast_assign_btrfs(btrfs_sb(inode->i_sb),
-		__entry->ino 		= inode->i_ino;
+		__entry->ino 		= btrfs_ino(BTRFS_I(inode));
 		__entry->file_offset	= ordered->file_offset;
 		__entry->start		= ordered->start;
 		__entry->len		= ordered->len;
@@ -477,13 +479,13 @@ DECLARE_EVENT_CLASS(btrfs__ordered_extent,
 		  "bytes_left=%llu flags=%s compress_type=%d "
 		  "refs=%d",
 		  show_root_type(__entry->root_objectid),
-		  (unsigned long long)__entry->ino,
-		  (unsigned long long)__entry->file_offset,
-		  (unsigned long long)__entry->start,
-		  (unsigned long long)__entry->len,
-		  (unsigned long long)__entry->disk_len,
-		  (unsigned long long)__entry->truncated_len,
-		  (unsigned long long)__entry->bytes_left,
+		  __entry->ino,
+		  __entry->file_offset,
+		  __entry->start,
+		  __entry->len,
+		  __entry->disk_len,
+		  __entry->truncated_len,
+		  __entry->bytes_left,
 		  show_ordered_flags(__entry->flags),
 		  __entry->compress_type, __entry->refs)
 );
@@ -528,7 +530,7 @@ DECLARE_EVENT_CLASS(btrfs__writepage,
 	TP_ARGS(page, inode, wbc),
 
 	TP_STRUCT__entry_btrfs(
-		__field(	ino_t,  ino			)
+		__field(	u64,	ino			)
 		__field(	pgoff_t,  index			)
 		__field(	long,   nr_to_write		)
 		__field(	long,   pages_skipped		)
@@ -542,7 +544,7 @@ DECLARE_EVENT_CLASS(btrfs__writepage,
 	),
 
 	TP_fast_assign_btrfs(btrfs_sb(inode->i_sb),
-		__entry->ino		= inode->i_ino;
+		__entry->ino		= btrfs_ino(BTRFS_I(inode));
 		__entry->index		= page->index;
 		__entry->nr_to_write	= wbc->nr_to_write;
 		__entry->pages_skipped	= wbc->pages_skipped;
@@ -556,12 +558,12 @@ DECLARE_EVENT_CLASS(btrfs__writepage,
 				 BTRFS_I(inode)->root->root_key.objectid;
 	),
 
-	TP_printk_btrfs("root=%llu(%s) ino=%lu page_index=%lu "
+	TP_printk_btrfs("root=%llu(%s) ino=%llu page_index=%lu "
 		  "nr_to_write=%ld pages_skipped=%ld range_start=%llu "
 		  "range_end=%llu for_kupdate=%d "
 		  "for_reclaim=%d range_cyclic=%d writeback_index=%lu",
 		  show_root_type(__entry->root_objectid),
-		  (unsigned long)__entry->ino, __entry->index,
+		  __entry->ino, __entry->index,
 		  __entry->nr_to_write, __entry->pages_skipped,
 		  __entry->range_start, __entry->range_end,
 		  __entry->for_kupdate,
@@ -584,7 +586,7 @@ TRACE_EVENT(btrfs_writepage_end_io_hook,
 	TP_ARGS(page, start, end, uptodate),
 
 	TP_STRUCT__entry_btrfs(
-		__field(	ino_t,	 ino		)
+		__field(	u64,	 ino		)
 		__field(	pgoff_t, index		)
 		__field(	u64,	 start		)
 		__field(	u64,	 end		)
@@ -593,7 +595,7 @@ TRACE_EVENT(btrfs_writepage_end_io_hook,
 	),
 
 	TP_fast_assign_btrfs(btrfs_sb(page->mapping->host->i_sb),
-		__entry->ino	= page->mapping->host->i_ino;
+		__entry->ino	= btrfs_ino(BTRFS_I(page->mapping->host));
 		__entry->index	= page->index;
 		__entry->start	= start;
 		__entry->end	= end;
@@ -602,12 +604,12 @@ TRACE_EVENT(btrfs_writepage_end_io_hook,
 			 BTRFS_I(page->mapping->host)->root->root_key.objectid;
 	),
 
-	TP_printk_btrfs("root=%llu(%s) ino=%lu page_index=%lu start=%llu "
+	TP_printk_btrfs("root=%llu(%s) ino=%llu page_index=%lu start=%llu "
 		  "end=%llu uptodate=%d",
 		  show_root_type(__entry->root_objectid),
-		  (unsigned long)__entry->ino, (unsigned long)__entry->index,
-		  (unsigned long long)__entry->start,
-		  (unsigned long long)__entry->end, __entry->uptodate)
+		  __entry->ino, (unsigned long)__entry->index,
+		  __entry->start,
+		  __entry->end, __entry->uptodate)
 );
 
 TRACE_EVENT(btrfs_sync_file,
@@ -617,8 +619,8 @@ TRACE_EVENT(btrfs_sync_file,
 	TP_ARGS(file, datasync),
 
 	TP_STRUCT__entry_btrfs(
-		__field(	ino_t,  ino		)
-		__field(	ino_t,  parent		)
+		__field(	u64,	ino		)
+		__field(	u64,	parent		)
 		__field(	int,    datasync	)
 		__field(	u64,    root_objectid	)
 	),
@@ -628,16 +630,17 @@ TRACE_EVENT(btrfs_sync_file,
 		const struct inode *inode = d_inode(dentry);
 
 		TP_fast_assign_fsid(btrfs_sb(file->f_path.dentry->d_sb));
-		__entry->ino		= inode->i_ino;
-		__entry->parent		= d_inode(dentry->d_parent)->i_ino;
+		__entry->ino		= btrfs_ino(BTRFS_I(inode));
+		__entry->parent		= btrfs_ino(BTRFS_I(d_inode(dentry->d_parent)));
 		__entry->datasync	= datasync;
 		__entry->root_objectid	=
 				 BTRFS_I(inode)->root->root_key.objectid;
 	),
 
-	TP_printk_btrfs("root=%llu(%s) ino=%ld parent=%ld datasync=%d",
+	TP_printk_btrfs("root=%llu(%s) ino=%llu parent=%llu datasync=%d",
 		  show_root_type(__entry->root_objectid),
-		  (unsigned long)__entry->ino, (unsigned long)__entry->parent,
+		  __entry->ino,
+		  __entry->parent,
 		  __entry->datasync)
 );
 
@@ -655,7 +658,7 @@ TRACE_EVENT(btrfs_sync_fs,
 		__entry->wait	= wait;
 	),
 
-	TP_printk_btrfs("wait = %d", __entry->wait)
+	TP_printk_btrfs("wait=%d", __entry->wait)
 );
 
 TRACE_EVENT(btrfs_add_block_group,
@@ -665,8 +668,7 @@ TRACE_EVENT(btrfs_add_block_group,
 
 	TP_ARGS(fs_info, block_group, create),
 
-	TP_STRUCT__entry(
-		__array(	u8,	fsid,	BTRFS_FSID_SIZE	)
+	TP_STRUCT__entry_btrfs(
 		__field(	u64,	offset			)
 		__field(	u64,	size			)
 		__field(	u64,	flags			)
@@ -675,8 +677,7 @@ TRACE_EVENT(btrfs_add_block_group,
 		__field(	int,	create			)
 	),
 
-	TP_fast_assign(
-		memcpy(__entry->fsid, fs_info->fsid, BTRFS_FSID_SIZE);
+	TP_fast_assign_btrfs(fs_info,
 		__entry->offset		= block_group->key.objectid;
 		__entry->size		= block_group->key.offset;
 		__entry->flags		= block_group->flags;
@@ -686,16 +687,16 @@ TRACE_EVENT(btrfs_add_block_group,
 		__entry->create		= create;
 	),
 
-	TP_printk("%pU: block_group offset=%llu size=%llu "
+	TP_printk_btrfs("block_group offset=%llu size=%llu "
 		  "flags=%llu(%s) bytes_used=%llu bytes_super=%llu "
-		  "create=%d", __entry->fsid,
-		  (unsigned long long)__entry->offset,
-		  (unsigned long long)__entry->size,
-		  (unsigned long long)__entry->flags,
+		  "create=%d",
+		  __entry->offset,
+		  __entry->size,
+		  __entry->flags,
 		  __print_flags((unsigned long)__entry->flags, "|",
 				BTRFS_GROUP_FLAGS),
-		  (unsigned long long)__entry->bytes_used,
-		  (unsigned long long)__entry->bytes_super, __entry->create)
+		  __entry->bytes_used,
+		  __entry->bytes_super, __entry->create)
 );
 
 #define show_ref_action(action)						\
@@ -740,13 +741,13 @@ DECLARE_EVENT_CLASS(btrfs_delayed_tree_ref,
 	TP_printk_btrfs("bytenr=%llu num_bytes=%llu action=%s "
 		  "parent=%llu(%s) ref_root=%llu(%s) level=%d "
 		  "type=%s seq=%llu",
-		  (unsigned long long)__entry->bytenr,
-		  (unsigned long long)__entry->num_bytes,
+		  __entry->bytenr,
+		  __entry->num_bytes,
 		  show_ref_action(__entry->action),
 		  show_root_type(__entry->parent),
 		  show_root_type(__entry->ref_root),
 		  __entry->level, show_ref_type(__entry->type),
-		  (unsigned long long)__entry->seq)
+		  __entry->seq)
 );
 
 DEFINE_EVENT(btrfs_delayed_tree_ref,  add_delayed_tree_ref,
@@ -805,15 +806,15 @@ DECLARE_EVENT_CLASS(btrfs_delayed_data_ref,
 	TP_printk_btrfs("bytenr=%llu num_bytes=%llu action=%s "
 		  "parent=%llu(%s) ref_root=%llu(%s) owner=%llu "
 		  "offset=%llu type=%s seq=%llu",
-		  (unsigned long long)__entry->bytenr,
-		  (unsigned long long)__entry->num_bytes,
+		  __entry->bytenr,
+		  __entry->num_bytes,
 		  show_ref_action(__entry->action),
 		  show_root_type(__entry->parent),
 		  show_root_type(__entry->ref_root),
-		  (unsigned long long)__entry->owner,
-		  (unsigned long long)__entry->offset,
+		  __entry->owner,
+		  __entry->offset,
 		  show_ref_type(__entry->type),
-		  (unsigned long long)__entry->seq)
+		  __entry->seq)
 );
 
 DEFINE_EVENT(btrfs_delayed_data_ref,  add_delayed_data_ref,
@@ -859,8 +860,8 @@ DECLARE_EVENT_CLASS(btrfs_delayed_ref_head,
 	),
 
 	TP_printk_btrfs("bytenr=%llu num_bytes=%llu action=%s is_data=%d",
-		  (unsigned long long)__entry->bytenr,
-		  (unsigned long long)__entry->num_bytes,
+		  __entry->bytenr,
+		  __entry->num_bytes,
 		  show_ref_action(__entry->action),
 		  __entry->is_data)
 );
@@ -923,8 +924,8 @@ DECLARE_EVENT_CLASS(btrfs__chunk,
 	TP_printk_btrfs("root=%llu(%s) offset=%llu size=%llu "
 		  "num_stripes=%d sub_stripes=%d type=%s",
 		  show_root_type(__entry->root_objectid),
-		  (unsigned long long)__entry->offset,
-		  (unsigned long long)__entry->size,
+		  __entry->offset,
+		  __entry->size,
 		  __entry->num_stripes, __entry->sub_stripes,
 		  show_chunk_type(__entry->type))
 );
@@ -974,9 +975,9 @@ TRACE_EVENT(btrfs_cow_block,
 		  "(orig_level=%d) cow_buf=%llu (cow_level=%d)",
 		  show_root_type(__entry->root_objectid),
 		  __entry->refs,
-		  (unsigned long long)__entry->buf_start,
+		  __entry->buf_start,
 		  __entry->buf_level,
-		  (unsigned long long)__entry->cow_start,
+		  __entry->cow_start,
 		  __entry->cow_level)
 );
 
@@ -1001,7 +1002,7 @@ TRACE_EVENT(btrfs_space_reservation,
 		__entry->reserve	= reserve;
 	),
 
-	TP_printk_btrfs("%s: %Lu %s %Lu", __get_str(type), __entry->val,
+	TP_printk_btrfs("%s: %llu %s %llu", __get_str(type), __entry->val,
 			__entry->reserve ? "reserve" : "release",
 			__entry->bytes)
 );
@@ -1019,29 +1020,27 @@ TRACE_EVENT(btrfs_trigger_flush,
 
 	TP_ARGS(fs_info, flags, bytes, flush, reason),
 
-	TP_STRUCT__entry(
-		__array(	u8,	fsid,	BTRFS_FSID_SIZE	)
+	TP_STRUCT__entry_btrfs(
 		__field(	u64,	flags			)
 		__field(	u64,	bytes			)
 		__field(	int,	flush			)
 		__string(	reason,	reason			)
 	),
 
-	TP_fast_assign(
-		memcpy(__entry->fsid, fs_info->fsid, BTRFS_FSID_SIZE);
+	TP_fast_assign_btrfs(fs_info,
 		__entry->flags	= flags;
 		__entry->bytes	= bytes;
 		__entry->flush	= flush;
 		__assign_str(reason, reason)
 	),
 
-	TP_printk("%pU: %s: flush=%d(%s) flags=%llu(%s) bytes=%llu",
-		  __entry->fsid, __get_str(reason), __entry->flush,
+	TP_printk_btrfs("%s: flush=%d(%s) flags=%llu(%s) bytes=%llu",
+		  __get_str(reason), __entry->flush,
 		  show_flush_action(__entry->flush),
-		  (unsigned long long)__entry->flags,
+		  __entry->flags,
 		  __print_flags((unsigned long)__entry->flags, "|",
 				BTRFS_GROUP_FLAGS),
-		  (unsigned long long)__entry->bytes)
+		  __entry->bytes)
 );
 
 #define show_flush_state(state)							\
@@ -1060,29 +1059,27 @@ TRACE_EVENT(btrfs_flush_space,
 
 	TP_ARGS(fs_info, flags, num_bytes, state, ret),
 
-	TP_STRUCT__entry(
-		__array(	u8,	fsid,	BTRFS_FSID_SIZE	)
+	TP_STRUCT__entry_btrfs(
 		__field(	u64,	flags			)
 		__field(	u64,	num_bytes		)
 		__field(	int,	state			)
 		__field(	int,	ret			)
 	),
 
-	TP_fast_assign(
-		memcpy(__entry->fsid, fs_info->fsid, BTRFS_FSID_SIZE);
+	TP_fast_assign_btrfs(fs_info,
 		__entry->flags		=	flags;
 		__entry->num_bytes	=	num_bytes;
 		__entry->state		=	state;
 		__entry->ret		=	ret;
 	),
 
-	TP_printk("%pU: state=%d(%s) flags=%llu(%s) num_bytes=%llu ret=%d",
-		  __entry->fsid, __entry->state,
+	TP_printk_btrfs("state=%d(%s) flags=%llu(%s) num_bytes=%llu ret=%d",
+		  __entry->state,
 		  show_flush_state(__entry->state),
-		  (unsigned long long)__entry->flags,
+		  __entry->flags,
 		  __print_flags((unsigned long)__entry->flags, "|",
 				BTRFS_GROUP_FLAGS),
-		  (unsigned long long)__entry->num_bytes, __entry->ret)
+		  __entry->num_bytes, __entry->ret)
 );
 
 DECLARE_EVENT_CLASS(btrfs__reserved_extent,
@@ -1103,8 +1100,8 @@ DECLARE_EVENT_CLASS(btrfs__reserved_extent,
 
 	TP_printk_btrfs("root=%llu(%s) start=%llu len=%llu",
 		  show_root_type(BTRFS_EXTENT_TREE_OBJECTID),
-		  (unsigned long long)__entry->start,
-		  (unsigned long long)__entry->len)
+		  __entry->start,
+		  __entry->len)
 );
 
 DEFINE_EVENT(btrfs__reserved_extent,  btrfs_reserved_extent_alloc,
@@ -1140,7 +1137,7 @@ TRACE_EVENT(find_free_extent,
 		__entry->data		= data;
 	),
 
-	TP_printk_btrfs("root=%Lu(%s) len=%Lu empty_size=%Lu flags=%Lu(%s)",
+	TP_printk_btrfs("root=%llu(%s) len=%llu empty_size=%llu flags=%llu(%s)",
 		  show_root_type(BTRFS_EXTENT_TREE_OBJECTID),
 		  __entry->num_bytes, __entry->empty_size, __entry->data,
 		  __print_flags((unsigned long)__entry->data, "|",
@@ -1149,11 +1146,10 @@ TRACE_EVENT(find_free_extent,
 
 DECLARE_EVENT_CLASS(btrfs__reserve_extent,
 
-	TP_PROTO(const struct btrfs_fs_info *fs_info,
-		 const struct btrfs_block_group_cache *block_group, u64 start,
+	TP_PROTO(const struct btrfs_block_group_cache *block_group, u64 start,
 		 u64 len),
 
-	TP_ARGS(fs_info, block_group, start, len),
+	TP_ARGS(block_group, start, len),
 
 	TP_STRUCT__entry_btrfs(
 		__field(	u64,	bg_objectid		)
@@ -1162,15 +1158,15 @@ DECLARE_EVENT_CLASS(btrfs__reserve_extent,
 		__field(	u64,	len			)
 	),
 
-	TP_fast_assign_btrfs(fs_info,
+	TP_fast_assign_btrfs(block_group->fs_info,
 		__entry->bg_objectid	= block_group->key.objectid;
 		__entry->flags		= block_group->flags;
 		__entry->start		= start;
 		__entry->len		= len;
 	),
 
-	TP_printk_btrfs("root=%Lu(%s) block_group=%Lu flags=%Lu(%s) "
-		  "start=%Lu len=%Lu",
+	TP_printk_btrfs("root=%llu(%s) block_group=%llu flags=%llu(%s) "
+		  "start=%llu len=%llu",
 		  show_root_type(BTRFS_EXTENT_TREE_OBJECTID),
 		  __entry->bg_objectid,
 		  __entry->flags, __print_flags((unsigned long)__entry->flags,
@@ -1180,20 +1176,18 @@ DECLARE_EVENT_CLASS(btrfs__reserve_extent,
 
 DEFINE_EVENT(btrfs__reserve_extent, btrfs_reserve_extent,
 
-	TP_PROTO(const struct btrfs_fs_info *fs_info,
-		 const struct btrfs_block_group_cache *block_group, u64 start,
+	TP_PROTO(const struct btrfs_block_group_cache *block_group, u64 start,
 		 u64 len),
 
-	TP_ARGS(fs_info, block_group, start, len)
+	TP_ARGS(block_group, start, len)
 );
 
 DEFINE_EVENT(btrfs__reserve_extent, btrfs_reserve_extent_cluster,
 
-	TP_PROTO(const struct btrfs_fs_info *fs_info,
-		 const struct btrfs_block_group_cache *block_group, u64 start,
+	TP_PROTO(const struct btrfs_block_group_cache *block_group, u64 start,
 		 u64 len),
 
-	TP_ARGS(fs_info, block_group, start, len)
+	TP_ARGS(block_group, start, len)
 );
 
 TRACE_EVENT(btrfs_find_cluster,
@@ -1221,8 +1215,8 @@ TRACE_EVENT(btrfs_find_cluster,
 		__entry->min_bytes	= min_bytes;
 	),
 
-	TP_printk_btrfs("block_group=%Lu flags=%Lu(%s) start=%Lu len=%Lu "
-		  "empty_size=%Lu min_bytes=%Lu", __entry->bg_objectid,
+	TP_printk_btrfs("block_group=%llu flags=%llu(%s) start=%llu len=%llu "
+		  "empty_size=%llu min_bytes=%llu", __entry->bg_objectid,
 		  __entry->flags,
 		  __print_flags((unsigned long)__entry->flags, "|",
 				BTRFS_GROUP_FLAGS), __entry->start,
@@ -1243,7 +1237,7 @@ TRACE_EVENT(btrfs_failed_cluster_setup,
 		__entry->bg_objectid	= block_group->key.objectid;
 	),
 
-	TP_printk_btrfs("block_group=%Lu", __entry->bg_objectid)
+	TP_printk_btrfs("block_group=%llu", __entry->bg_objectid)
 );
 
 TRACE_EVENT(btrfs_setup_cluster,
@@ -1272,8 +1266,8 @@ TRACE_EVENT(btrfs_setup_cluster,
 		__entry->bitmap		= bitmap;
 	),
 
-	TP_printk_btrfs("block_group=%Lu flags=%Lu(%s) window_start=%Lu "
-		  "size=%Lu max_size=%Lu bitmap=%d",
+	TP_printk_btrfs("block_group=%llu flags=%llu(%s) window_start=%llu "
+		  "size=%llu max_size=%llu bitmap=%d",
 		  __entry->bg_objectid,
 		  __entry->flags,
 		  __print_flags((unsigned long)__entry->flags, "|",
@@ -1476,7 +1470,7 @@ DECLARE_EVENT_CLASS(btrfs__qgroup_rsv_data,
 
 	TP_STRUCT__entry_btrfs(
 		__field(	u64,		rootid		)
-		__field(	unsigned long,	ino		)
+		__field(	u64,		ino		)
 		__field(	u64,		start		)
 		__field(	u64,		len		)
 		__field(	u64,		reserved	)
@@ -1485,14 +1479,14 @@ DECLARE_EVENT_CLASS(btrfs__qgroup_rsv_data,
 
 	TP_fast_assign_btrfs(btrfs_sb(inode->i_sb),
 		__entry->rootid		= BTRFS_I(inode)->root->objectid;
-		__entry->ino		= inode->i_ino;
+		__entry->ino		= btrfs_ino(BTRFS_I(inode));
 		__entry->start		= start;
 		__entry->len		= len;
 		__entry->reserved	= reserved;
 		__entry->op		= op;
 	),
 
-	TP_printk_btrfs("root=%llu ino=%lu start=%llu len=%llu reserved=%llu op=%s",
+	TP_printk_btrfs("root=%llu ino=%llu start=%llu len=%llu reserved=%llu op=%s",
 		  __entry->rootid, __entry->ino, __entry->start, __entry->len,
 		  __entry->reserved,
 		  __print_flags((unsigned long)__entry->op, "",
@@ -1584,12 +1578,14 @@ DEFINE_EVENT(btrfs_qgroup_extent, btrfs_qgroup_trace_extent,
 
 TRACE_EVENT(btrfs_qgroup_account_extent,
 
-	TP_PROTO(const struct btrfs_fs_info *fs_info, u64 bytenr,
+	TP_PROTO(const struct btrfs_fs_info *fs_info, u64 transid, u64 bytenr,
 		 u64 num_bytes, u64 nr_old_roots, u64 nr_new_roots),
 
-	TP_ARGS(fs_info, bytenr, num_bytes, nr_old_roots, nr_new_roots),
+	TP_ARGS(fs_info, transid, bytenr, num_bytes, nr_old_roots,
+		nr_new_roots),
 
 	TP_STRUCT__entry_btrfs(
+		__field(	u64,  transid			)
 		__field(	u64,  bytenr			)
 		__field(	u64,  num_bytes			)
 		__field(	u64,  nr_old_roots		)
@@ -1597,43 +1593,49 @@ TRACE_EVENT(btrfs_qgroup_account_extent,
 	),
 
 	TP_fast_assign_btrfs(fs_info,
+		__entry->transid	= transid;
 		__entry->bytenr		= bytenr;
 		__entry->num_bytes	= num_bytes;
 		__entry->nr_old_roots	= nr_old_roots;
 		__entry->nr_new_roots	= nr_new_roots;
 	),
 
-	TP_printk_btrfs("bytenr=%llu num_bytes=%llu nr_old_roots=%llu "
-		  "nr_new_roots=%llu",
-		  __entry->bytenr,
-		  __entry->num_bytes,
-		  __entry->nr_old_roots,
-		  __entry->nr_new_roots)
+	TP_printk_btrfs(
+"transid=%llu bytenr=%llu num_bytes=%llu nr_old_roots=%llu nr_new_roots=%llu",
+		__entry->transid,
+		__entry->bytenr,
+		__entry->num_bytes,
+		__entry->nr_old_roots,
+		__entry->nr_new_roots)
 );
 
 TRACE_EVENT(qgroup_update_counters,
 
-	TP_PROTO(const struct btrfs_fs_info *fs_info, u64 qgid,
+	TP_PROTO(const struct btrfs_fs_info *fs_info,
+		 struct btrfs_qgroup *qgroup,
 		 u64 cur_old_count, u64 cur_new_count),
 
-	TP_ARGS(fs_info, qgid, cur_old_count, cur_new_count),
+	TP_ARGS(fs_info, qgroup, cur_old_count, cur_new_count),
 
 	TP_STRUCT__entry_btrfs(
 		__field(	u64,  qgid			)
+		__field(	u64,  old_rfer			)
+		__field(	u64,  old_excl			)
 		__field(	u64,  cur_old_count		)
 		__field(	u64,  cur_new_count		)
 	),
 
 	TP_fast_assign_btrfs(fs_info,
-		__entry->qgid		= qgid;
+		__entry->qgid		= qgroup->qgroupid;
+		__entry->old_rfer	= qgroup->rfer;
+		__entry->old_excl	= qgroup->excl;
 		__entry->cur_old_count	= cur_old_count;
 		__entry->cur_new_count	= cur_new_count;
 	),
 
-	TP_printk_btrfs("qgid=%llu cur_old_count=%llu cur_new_count=%llu",
-		  __entry->qgid,
-		  __entry->cur_old_count,
-		  __entry->cur_new_count)
+	TP_printk_btrfs("qgid=%llu old_rfer=%llu old_excl=%llu cur_old_count=%llu cur_new_count=%llu",
+		  __entry->qgid, __entry->old_rfer, __entry->old_excl,
+		  __entry->cur_old_count, __entry->cur_new_count)
 );
 
 TRACE_EVENT(qgroup_update_reserve,
@@ -1765,14 +1767,14 @@ DECLARE_EVENT_CLASS(btrfs__prelim_ref,
 	),
 
 	TP_printk_btrfs("root_id=%llu key=[%llu,%u,%llu] level=%d count=[%d+%d=%d] parent=%llu wanted_disk_byte=%llu nodes=%llu",
-			(unsigned long long)__entry->root_id,
-			(unsigned long long)__entry->objectid, __entry->type,
-			(unsigned long long)__entry->offset, __entry->level,
+			__entry->root_id,
+			__entry->objectid, __entry->type,
+			__entry->offset, __entry->level,
 			__entry->old_count, __entry->mod_count,
 			__entry->old_count + __entry->mod_count,
-			(unsigned long long)__entry->parent,
-			(unsigned long long)__entry->bytenr,
-			(unsigned long long)__entry->tree_size)
+			__entry->parent,
+			__entry->bytenr,
+			__entry->tree_size)
 );
 
 DEFINE_EVENT(btrfs__prelim_ref, btrfs_prelim_ref_merge,
@@ -1808,8 +1810,51 @@ TRACE_EVENT(btrfs_inode_mod_outstanding_extents,
 
 	TP_printk_btrfs("root=%llu(%s) ino=%llu mod=%d",
 			show_root_type(__entry->root_objectid),
-			(unsigned long long)__entry->ino, __entry->mod)
+			__entry->ino, __entry->mod)
 );
+
+DECLARE_EVENT_CLASS(btrfs__block_group,
+	TP_PROTO(const struct btrfs_block_group_cache *bg_cache),
+
+	TP_ARGS(bg_cache),
+
+	TP_STRUCT__entry_btrfs(
+		__field(	u64,	bytenr		)
+		__field(	u64,	len		)
+		__field(	u64,	used		)
+		__field(	u64,	flags		)
+	),
+
+	TP_fast_assign_btrfs(bg_cache->fs_info,
+		__entry->bytenr = bg_cache->key.objectid,
+		__entry->len	= bg_cache->key.offset,
+		__entry->used	= btrfs_block_group_used(&bg_cache->item);
+		__entry->flags	= bg_cache->flags;
+	),
+
+	TP_printk_btrfs("bg bytenr=%llu len=%llu used=%llu flags=%llu(%s)",
+		__entry->bytenr, __entry->len, __entry->used, __entry->flags,
+		__print_flags(__entry->flags, "|", BTRFS_GROUP_FLAGS))
+);
+
+DEFINE_EVENT(btrfs__block_group, btrfs_remove_block_group,
+	TP_PROTO(const struct btrfs_block_group_cache *bg_cache),
+
+	TP_ARGS(bg_cache)
+);
+
+DEFINE_EVENT(btrfs__block_group, btrfs_add_unused_block_group,
+	TP_PROTO(const struct btrfs_block_group_cache *bg_cache),
+
+	TP_ARGS(bg_cache)
+);
+
+DEFINE_EVENT(btrfs__block_group, btrfs_skip_unused_block_group,
+	TP_PROTO(const struct btrfs_block_group_cache *bg_cache),
+
+	TP_ARGS(bg_cache)
+);
+
 #endif /* _TRACE_BTRFS_H */
 
 /* This part must be outside protection */

include/uapi/linux/btrfs.h

@@ -422,6 +422,21 @@ struct btrfs_ioctl_ino_lookup_args {
 	char name[BTRFS_INO_LOOKUP_PATH_MAX];
 };
 
+#define BTRFS_INO_LOOKUP_USER_PATH_MAX (4080 - BTRFS_VOL_NAME_MAX - 1)
+struct btrfs_ioctl_ino_lookup_user_args {
+	/* in, inode number containing the subvolume of 'subvolid' */
+	__u64 dirid;
+	/* in */
+	__u64 treeid;
+	/* out, name of the subvolume of 'treeid' */
+	char name[BTRFS_VOL_NAME_MAX + 1];
+	/*
+	 * out, constructed path from the directory with which the ioctl is
+	 * called to dirid
+	 */
+	char path[BTRFS_INO_LOOKUP_USER_PATH_MAX];
+};
+
 /* Search criteria for the btrfs SEARCH ioctl family. */
 struct btrfs_ioctl_search_key {
 	/*
@@ -725,6 +740,82 @@ struct btrfs_ioctl_send_args {
 	__u64 reserved[4];		/* in */
 };
 
+/*
+ * Information about a fs tree root.
+ *
+ * All items are filled by the ioctl
+ */
+struct btrfs_ioctl_get_subvol_info_args {
+	/* Id of this subvolume */
+	__u64 treeid;
+
+	/* Name of this subvolume, used to get the real name at mount point */
+	char name[BTRFS_VOL_NAME_MAX + 1];
+
+	/*
+	 * Id of the subvolume which contains this subvolume.
+	 * Zero for top-level subvolume or a deleted subvolume.
+	 */
+	__u64 parent_id;
+
+	/*
+	 * Inode number of the directory which contains this subvolume.
+	 * Zero for top-level subvolume or a deleted subvolume
+	 */
+	__u64 dirid;
+
+	/* Latest transaction id of this subvolume */
+	__u64 generation;
+
+	/* Flags of this subvolume */
+	__u64 flags;
+
+	/* UUID of this subvolume */
+	__u8 uuid[BTRFS_UUID_SIZE];
+
+	/*
+	 * UUID of the subvolume of which this subvolume is a snapshot.
+	 * All zero for a non-snapshot subvolume.
+	 */
+	__u8 parent_uuid[BTRFS_UUID_SIZE];
+
+	/*
+	 * UUID of the subvolume from which this subvolume was received.
+	 * All zero for non-received subvolume.
+	 */
+	__u8 received_uuid[BTRFS_UUID_SIZE];
+
+	/* Transaction id indicating when change/create/send/receive happened */
+	__u64 ctransid;
+	__u64 otransid;
+	__u64 stransid;
+	__u64 rtransid;
+	/* Time corresponding to c/o/s/rtransid */
+	struct btrfs_ioctl_timespec ctime;
+	struct btrfs_ioctl_timespec otime;
+	struct btrfs_ioctl_timespec stime;
+	struct btrfs_ioctl_timespec rtime;
+
+	/* Must be zero */
+	__u64 reserved[8];
+};
+
+#define BTRFS_MAX_ROOTREF_BUFFER_NUM 255
+struct btrfs_ioctl_get_subvol_rootref_args {
+		/* in/out, minimum id of rootref's treeid to be searched */
+		__u64 min_treeid;
+
+		/* out */
+		struct {
+			__u64 treeid;
+			__u64 dirid;
+		} rootref[BTRFS_MAX_ROOTREF_BUFFER_NUM];
+
+		/* out, number of found items */
+		__u8 num_items;
+		__u8 align[7];
+};
+
 /* Error codes as returned by the kernel */
 enum btrfs_err_code {
 	BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET = 1,
@@ -843,5 +934,11 @@ enum btrfs_err_code {
 				   struct btrfs_ioctl_vol_args_v2)
 #define BTRFS_IOC_LOGICAL_INO_V2 _IOWR(BTRFS_IOCTL_MAGIC, 59, \
 					struct btrfs_ioctl_logical_ino_args)
+#define BTRFS_IOC_GET_SUBVOL_INFO _IOR(BTRFS_IOCTL_MAGIC, 60, \
+				struct btrfs_ioctl_get_subvol_info_args)
+#define BTRFS_IOC_GET_SUBVOL_ROOTREF _IOWR(BTRFS_IOCTL_MAGIC, 61, \
+				struct btrfs_ioctl_get_subvol_rootref_args)
+#define BTRFS_IOC_INO_LOOKUP_USER _IOWR(BTRFS_IOCTL_MAGIC, 62, \
+				struct btrfs_ioctl_ino_lookup_user_args)
 
 #endif /* _UAPI_LINUX_BTRFS_H */