Merge tag 'for-5.10-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "Mostly core updates with a few user visible bits and fixes.

  Hilights:

   - fsync performance improvements
      - less contention of log mutex (throughput +4%, latency -14%,
        dbench with 32 clients)
      - skip unnecessary commits for link and rename (throughput +6%,
        latency -30%, rename latency -75%, dbench with 16 clients)
      - make fast fsync wait only for writeback (throughput +10..40%,
        runtime -1..-20%, dbench with 1 to 64 clients on various
        file/block sizes)

   - direct io is now implemented using the iomap infrastructure, that's
     the main part, we still have a workaround that requires an iomap
     API update, coming in 5.10

   - new sysfs exports:
      - information about the exclusive filesystem operation status
        (balance, device add/remove/replace, ...)
      - supported send stream version

  Core:

   - use ticket space reservations for data, fair policy using the same
     infrastructure as metadata

   - preparatory work to switch locking from our custom tree locks to
     standard rwsem, now the locking context is propagated to all
     callers, actual switch is expected to happen in the next dev cycle

   - seed device structures are now using list API

   - extent tracepoints print proper tree id

   - unified range checks for extent buffer helpers

   - send: avoid using temporary buffer for copying data

   - remove unnecessary RCU protection from space infos

   - remove unused readpage callback for metadata, enabling several
     cleanups

   - replace indirect function calls for end io hooks and remove
     extent_io_ops completely

  Fixes:

   - more lockdep warning fixes

   - fix qgroup reservation for delayed inode and an occasional
     reservation leak for preallocated files

   - fix device replace of a seed device

   - fix metadata reservation for fallocate that leads to transaction
     aborts

   - reschedule if necessary when logging directory items or when
     cloning lots of extents

   - tree-checker: fix false alert caused by legacy btrfs root item

   - send: fix rename/link conflicts for orphanized inodes

   - properly initialize device stats for seed devices

   - skip devices without magic signature when mounting

  Other:

   - error handling improvements, BUG_ONs replaced by proper handling,
     fuzz fixes

   - various function parameter cleanups

   - various W=1 cleanups

   - error/info messages improved

  Mishaps:

   - commit 62cf5391209a ("btrfs: move btrfs_rm_dev_replace_free_srcdev
     outside of all locks") is a rebase leftover after the patch got
     merged to 5.9-rc8 as a466c85edc6f ("btrfs: move
     btrfs_rm_dev_replace_free_srcdev outside of all locks"), the
     remaining part is trivial and the patch is in the middle of the
     series so I'm keeping it there instead of rebasing"

* tag 'for-5.10-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (161 commits)
  btrfs: rename BTRFS_INODE_ORDERED_DATA_CLOSE flag
  btrfs: annotate device name rcu_string with __rcu
  btrfs: skip devices without magic signature when mounting
  btrfs: cleanup cow block on error
  btrfs: remove BTRFS_INODE_READDIO_NEED_LOCK
  fs: remove no longer used dio_end_io()
  btrfs: return error if we're unable to read device stats
  btrfs: init device stats for seed devices
  btrfs: remove struct extent_io_ops
  btrfs: call submit_bio_hook directly for metadata pages
  btrfs: stop calling submit_bio_hook for data inodes
  btrfs: don't opencode is_data_inode in end_bio_extent_readpage
  btrfs: call submit_bio_hook directly in submit_one_bio
  btrfs: remove extent_io_ops::readpage_end_io_hook
  btrfs: replace readpage_end_io_hook with direct calls
  btrfs: send, recompute reference path after orphanization of a directory
  btrfs: send, orphanize first all conflicting inodes when processing references
  btrfs: tree-checker: fix false alert caused by legacy btrfs root item
  btrfs: use unaligned helpers for stack and header set/get helpers
  btrfs: free-space-cache: use unaligned helpers to access data
  ...

55 files changed, 2719 insertions, 1993 deletions

diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig
index 575636f6491ef6..68b95ad82126ed 100644
--- a/fs/btrfs/Kconfig
+++ b/fs/btrfs/Kconfig
@@ -14,6 +14,7 @@ config BTRFS_FS
 	select LZO_DECOMPRESS
 	select ZSTD_COMPRESS
 	select ZSTD_DECOMPRESS
+	select FS_IOMAP
 	select RAID6_PQ
 	select XOR_BLOCKS
 	select SRCU
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index ea1c28ccb44ff9..b3268f4ea5f34c 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -2997,7 +2997,6 @@ int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache,
 	while (!list_empty(&pending_edge)) {
 		struct btrfs_backref_node *upper;
 		struct btrfs_backref_node *lower;
-		struct rb_node *rb_node;
 
 		edge = list_first_entry(&pending_edge,
 				struct btrfs_backref_edge, list[UPPER]);
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index ea8aaf36647eea..c0f1d6818df769 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -1766,16 +1766,10 @@ static void link_block_group(struct btrfs_block_group *cache)
 {
 	struct btrfs_space_info *space_info = cache->space_info;
 	int index = btrfs_bg_flags_to_raid_index(cache->flags);
-	bool first = false;
 
 	down_write(&space_info->groups_sem);
-	if (list_empty(&space_info->block_groups[index]))
-		first = true;
 	list_add_tail(&cache->list, &space_info->block_groups[index]);
 	up_write(&space_info->groups_sem);
-
-	if (first)
-		btrfs_sysfs_add_block_group_type(cache);
 }
 
 static struct btrfs_block_group *btrfs_create_block_group_cache(
@@ -1873,7 +1867,7 @@ static int check_chunk_block_group_mappings(struct btrfs_fs_info *fs_info)
 	return ret;
 }
 
-static int read_block_group_item(struct btrfs_block_group *cache,
+static void read_block_group_item(struct btrfs_block_group *cache,
 				 struct btrfs_path *path,
 				 const struct btrfs_key *key)
 {
@@ -1887,8 +1881,6 @@ static int read_block_group_item(struct btrfs_block_group *cache,
 			   sizeof(bgi));
 	cache->used = btrfs_stack_block_group_used(&bgi);
 	cache->flags = btrfs_stack_block_group_flags(&bgi);
-
-	return 0;
 }
 
 static int read_one_block_group(struct btrfs_fs_info *info,
@@ -1907,9 +1899,7 @@ static int read_one_block_group(struct btrfs_fs_info *info,
 	if (!cache)
 		return -ENOMEM;
 
-	ret = read_block_group_item(cache, path, key);
-	if (ret < 0)
-		goto error;
+	read_block_group_item(cache, path, key);
 
 	set_free_space_tree_thresholds(cache);
 
@@ -2035,8 +2025,18 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 		btrfs_release_path(path);
 	}
 
-	rcu_read_lock();
-	list_for_each_entry_rcu(space_info, &info->space_info, list) {
+	list_for_each_entry(space_info, &info->space_info, list) {
+		int i;
+
+		for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
+			if (list_empty(&space_info->block_groups[i]))
+				continue;
+			cache = list_first_entry(&space_info->block_groups[i],
+						 struct btrfs_block_group,
+						 list);
+			btrfs_sysfs_add_block_group_type(cache);
+		}
+
 		if (!(btrfs_get_alloc_profile(info, space_info->flags) &
 		      (BTRFS_BLOCK_GROUP_RAID10 |
 		       BTRFS_BLOCK_GROUP_RAID1_MASK |
@@ -2056,7 +2056,6 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 				list)
 			inc_block_group_ro(cache, 1);
 	}
-	rcu_read_unlock();
 
 	btrfs_init_global_block_rsv(info);
 	ret = check_chunk_block_group_mappings(info);
@@ -2097,12 +2096,16 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 		return;
 
 	while (!list_empty(&trans->new_bgs)) {
+		int index;
+
 		block_group = list_first_entry(&trans->new_bgs,
 					       struct btrfs_block_group,
 					       bg_list);
 		if (ret)
 			goto next;
 
+		index = btrfs_bg_flags_to_raid_index(block_group->flags);
+
 		ret = insert_block_group_item(trans, block_group);
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
@@ -2111,6 +2114,16 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
 		add_block_group_free_space(trans, block_group);
+
+		/*
+		 * If we restriped during balance, we may have added a new raid
+		 * type, so now add the sysfs entries when it is safe to do so.
+		 * We don't have to worry about locking here as it's handled in
+		 * btrfs_sysfs_add_block_group_type.
+		 */
+		if (block_group->space_info->block_group_kobjs[index] == NULL)
+			btrfs_sysfs_add_block_group_type(block_group);
+
 		/* Already aborted the transaction if it failed. */
 next:
 		btrfs_delayed_refs_rsv_release(fs_info, 1);
@@ -2785,7 +2798,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 			 * finished yet (no block group item in the extent tree
 			 * yet, etc). If this is the case, wait for all free
 			 * space endio workers to finish and retry. This is a
-			 * a very rare case so no need for a more efficient and
+			 * very rare case so no need for a more efficient and
 			 * complex approach.
 			 */
 			if (ret == -ENOENT) {
@@ -2961,6 +2974,13 @@ int btrfs_add_reserved_bytes(struct btrfs_block_group *cache,
 						      space_info, -ram_bytes);
 		if (delalloc)
 			cache->delalloc_bytes += num_bytes;
+
+		/*
+		 * Compression can use less space than we reserved, so wake
+		 * tickets if that happens
+		 */
+		if (num_bytes < ram_bytes)
+			btrfs_try_granting_tickets(cache->fs_info, space_info);
 	}
 	spin_unlock(&cache->lock);
 	spin_unlock(&space_info->lock);
@@ -2994,6 +3014,8 @@ void btrfs_free_reserved_bytes(struct btrfs_block_group *cache,
 	if (delalloc)
 		cache->delalloc_bytes -= num_bytes;
 	spin_unlock(&cache->lock);
+
+	btrfs_try_granting_tickets(cache->fs_info, space_info);
 	spin_unlock(&space_info->lock);
 }
 
@@ -3002,12 +3024,10 @@ static void force_metadata_allocation(struct btrfs_fs_info *info)
 	struct list_head *head = &info->space_info;
 	struct btrfs_space_info *found;
 
-	rcu_read_lock();
-	list_for_each_entry_rcu(found, head, list) {
+	list_for_each_entry(found, head, list) {
 		if (found->flags & BTRFS_BLOCK_GROUP_METADATA)
 			found->force_alloc = CHUNK_ALLOC_FORCE;
 	}
-	rcu_read_unlock();
 }
 
 static int should_alloc_chunk(struct btrfs_fs_info *fs_info,
@@ -3338,14 +3358,6 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 	}
 	spin_unlock(&info->block_group_cache_lock);
 
-	/*
-	 * Now that all the block groups are freed, go through and free all the
-	 * space_info structs.  This is only called during the final stages of
-	 * unmount, and so we know nobody is using them.  We call
-	 * synchronize_rcu() once before we start, just to be on the safe side.
-	 */
-	synchronize_rcu();
-
 	btrfs_release_global_block_rsv(info);
 
 	while (!list_empty(&info->space_info)) {
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index c47b6c6fea9fa9..92dd86bceae310 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -21,14 +21,18 @@
  * new data the application may have written before commit.
  */
 enum {
-	BTRFS_INODE_ORDERED_DATA_CLOSE,
+	BTRFS_INODE_FLUSH_ON_CLOSE,
 	BTRFS_INODE_DUMMY,
 	BTRFS_INODE_IN_DEFRAG,
 	BTRFS_INODE_HAS_ASYNC_EXTENT,
+	 /*
+	  * Always set under the VFS' inode lock, otherwise it can cause races
+	  * during fsync (we start as a fast fsync and then end up in a full
+	  * fsync racing with ordered extent completion).
+	  */
 	BTRFS_INODE_NEEDS_FULL_SYNC,
 	BTRFS_INODE_COPY_EVERYTHING,
 	BTRFS_INODE_IN_DELALLOC_LIST,
-	BTRFS_INODE_READDIO_NEED_LOCK,
 	BTRFS_INODE_HAS_PROPS,
 	BTRFS_INODE_SNAPSHOT_FLUSH,
 };
@@ -212,6 +216,11 @@ struct btrfs_inode {
 	struct inode vfs_inode;
 };
 
+static inline u32 btrfs_inode_sectorsize(const struct btrfs_inode *inode)
+{
+	return inode->root->fs_info->sectorsize;
+}
+
 static inline struct btrfs_inode *BTRFS_I(const struct inode *inode)
 {
 	return container_of(inode, struct btrfs_inode, vfs_inode);
@@ -324,23 +333,6 @@ struct btrfs_dio_private {
 	u8 csums[];
 };
 
-/*
- * Disable DIO read nolock optimization, so new dio readers will be forced
- * to grab i_mutex. It is used to avoid the endless truncate due to
- * nonlocked dio read.
- */
-static inline void btrfs_inode_block_unlocked_dio(struct btrfs_inode *inode)
-{
-	set_bit(BTRFS_INODE_READDIO_NEED_LOCK, &inode->runtime_flags);
-	smp_mb();
-}
-
-static inline void btrfs_inode_resume_unlocked_dio(struct btrfs_inode *inode)
-{
-	smp_mb__before_atomic();
-	clear_bit(BTRFS_INODE_READDIO_NEED_LOCK, &inode->runtime_flags);
-}
-
 /* Array of bytes with variable length, hexadecimal format 0x1234 */
 #define CSUM_FMT				"0x%*phN"
 #define CSUM_FMT_VALUE(size, bytes)		size, bytes
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 1ab56a734e7063..eeface30facd3e 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -29,41 +29,6 @@
 #include "extent_io.h"
 #include "extent_map.h"
 
-int zlib_compress_pages(struct list_head *ws, struct address_space *mapping,
-		u64 start, struct page **pages, unsigned long *out_pages,
-		unsigned long *total_in, unsigned long *total_out);
-int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
-int zlib_decompress(struct list_head *ws, unsigned char *data_in,
-		struct page *dest_page, unsigned long start_byte, size_t srclen,
-		size_t destlen);
-struct list_head *zlib_alloc_workspace(unsigned int level);
-void zlib_free_workspace(struct list_head *ws);
-struct list_head *zlib_get_workspace(unsigned int level);
-
-int lzo_compress_pages(struct list_head *ws, struct address_space *mapping,
-		u64 start, struct page **pages, unsigned long *out_pages,
-		unsigned long *total_in, unsigned long *total_out);
-int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
-int lzo_decompress(struct list_head *ws, unsigned char *data_in,
-		struct page *dest_page, unsigned long start_byte, size_t srclen,
-		size_t destlen);
-struct list_head *lzo_alloc_workspace(unsigned int level);
-void lzo_free_workspace(struct list_head *ws);
-
-int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
-		u64 start, struct page **pages, unsigned long *out_pages,
-		unsigned long *total_in, unsigned long *total_out);
-int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
-int zstd_decompress(struct list_head *ws, unsigned char *data_in,
-		struct page *dest_page, unsigned long start_byte, size_t srclen,
-		size_t destlen);
-void zstd_init_workspace_manager(void);
-void zstd_cleanup_workspace_manager(void);
-struct list_head *zstd_alloc_workspace(unsigned int level);
-void zstd_free_workspace(struct list_head *ws);
-struct list_head *zstd_get_workspace(unsigned int level);
-void zstd_put_workspace(struct list_head *ws);
-
 static const char* const btrfs_compress_types[] = { "", "zlib", "lzo", "zstd" };
 
 const char* btrfs_compress_type2str(enum btrfs_compression_type type)
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index 9f3dbe372631be..8001b700ea3ae3 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -144,4 +144,39 @@ bool btrfs_compress_is_valid_type(const char *str, size_t len);
 
 int btrfs_compress_heuristic(struct inode *inode, u64 start, u64 end);
 
+int zlib_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out);
+int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
+int zlib_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen);
+struct list_head *zlib_alloc_workspace(unsigned int level);
+void zlib_free_workspace(struct list_head *ws);
+struct list_head *zlib_get_workspace(unsigned int level);
+
+int lzo_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out);
+int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
+int lzo_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen);
+struct list_head *lzo_alloc_workspace(unsigned int level);
+void lzo_free_workspace(struct list_head *ws);
+
+int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out);
+int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
+int zstd_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen);
+void zstd_init_workspace_manager(void);
+void zstd_cleanup_workspace_manager(void);
+struct list_head *zstd_alloc_workspace(unsigned int level);
+void zstd_free_workspace(struct list_head *ws);
+struct list_head *zstd_get_workspace(unsigned int level);
+void zstd_put_workspace(struct list_head *ws);
+
 #endif
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index cd392da69b8190..113da62dc17f65 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -198,7 +198,8 @@ int btrfs_copy_root(struct btrfs_trans_handle *trans,
 		btrfs_node_key(buf, &disk_key, 0);
 
 	cow = btrfs_alloc_tree_block(trans, root, 0, new_root_objectid,
-			&disk_key, level, buf->start, 0);
+				     &disk_key, level, buf->start, 0,
+				     BTRFS_NESTING_NEW_ROOT);
 	if (IS_ERR(cow))
 		return PTR_ERR(cow);
 
@@ -957,7 +958,8 @@ static struct extent_buffer *alloc_tree_block_no_bg_flush(
 					  const struct btrfs_disk_key *disk_key,
 					  int level,
 					  u64 hint,
-					  u64 empty_size)
+					  u64 empty_size,
+					  enum btrfs_lock_nesting nest)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_buffer *ret;
@@ -986,7 +988,7 @@ static struct extent_buffer *alloc_tree_block_no_bg_flush(
 
 	ret = btrfs_alloc_tree_block(trans, root, parent_start,
 				     root->root_key.objectid, disk_key, level,
-				     hint, empty_size);
+				     hint, empty_size, nest);
 	trans->can_flush_pending_bgs = true;
 
 	return ret;
@@ -1009,7 +1011,8 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 			     struct extent_buffer *buf,
 			     struct extent_buffer *parent, int parent_slot,
 			     struct extent_buffer **cow_ret,
-			     u64 search_start, u64 empty_size)
+			     u64 search_start, u64 empty_size,
+			     enum btrfs_lock_nesting nest)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_disk_key disk_key;
@@ -1040,7 +1043,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		parent_start = parent->start;
 
 	cow = alloc_tree_block_no_bg_flush(trans, root, parent_start, &disk_key,
-					   level, search_start, empty_size);
+					   level, search_start, empty_size, nest);
 	if (IS_ERR(cow))
 		return PTR_ERR(cow);
 
@@ -1061,6 +1064,8 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 
 	ret = update_ref_for_cow(trans, root, buf, cow, &last_ref);
 	if (ret) {
+		btrfs_tree_unlock(cow);
+		free_extent_buffer(cow);
 		btrfs_abort_transaction(trans, ret);
 		return ret;
 	}
@@ -1068,6 +1073,8 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 	if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) {
 		ret = btrfs_reloc_cow_block(trans, root, buf, cow);
 		if (ret) {
+			btrfs_tree_unlock(cow);
+			free_extent_buffer(cow);
 			btrfs_abort_transaction(trans, ret);
 			return ret;
 		}
@@ -1100,6 +1107,8 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		if (last_ref) {
 			ret = tree_mod_log_free_eb(buf);
 			if (ret) {
+				btrfs_tree_unlock(cow);
+				free_extent_buffer(cow);
 				btrfs_abort_transaction(trans, ret);
 				return ret;
 			}
@@ -1446,7 +1455,8 @@ static inline int should_cow_block(struct btrfs_trans_handle *trans,
 noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,
 		    struct btrfs_root *root, struct extent_buffer *buf,
 		    struct extent_buffer *parent, int parent_slot,
-		    struct extent_buffer **cow_ret)
+		    struct extent_buffer **cow_ret,
+		    enum btrfs_lock_nesting nest)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	u64 search_start;
@@ -1485,7 +1495,7 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,
 	 */
 	btrfs_qgroup_trace_subtree_after_cow(trans, root, buf);
 	ret = __btrfs_cow_block(trans, root, buf, parent,
-				 parent_slot, cow_ret, search_start, 0);
+				 parent_slot, cow_ret, search_start, 0, nest);
 
 	trace_btrfs_cow_block(root, buf, *cow_ret);
 
@@ -1657,7 +1667,8 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
 		err = __btrfs_cow_block(trans, root, cur, parent, i,
 					&cur, search_start,
 					min(16 * blocksize,
-					    (end_slot - i) * blocksize));
+					    (end_slot - i) * blocksize),
+					BTRFS_NESTING_COW);
 		if (err) {
 			btrfs_tree_unlock(cur);
 			free_extent_buffer(cur);
@@ -1855,7 +1866,8 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 
 		btrfs_tree_lock(child);
 		btrfs_set_lock_blocking_write(child);
-		ret = btrfs_cow_block(trans, root, child, mid, 0, &child);
+		ret = btrfs_cow_block(trans, root, child, mid, 0, &child,
+				      BTRFS_NESTING_COW);
 		if (ret) {
 			btrfs_tree_unlock(child);
 			free_extent_buffer(child);
@@ -1891,10 +1903,11 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		left = NULL;
 
 	if (left) {
-		btrfs_tree_lock(left);
+		__btrfs_tree_lock(left, BTRFS_NESTING_LEFT);
 		btrfs_set_lock_blocking_write(left);
 		wret = btrfs_cow_block(trans, root, left,
-				       parent, pslot - 1, &left);
+				       parent, pslot - 1, &left,
+				       BTRFS_NESTING_LEFT_COW);
 		if (wret) {
 			ret = wret;
 			goto enospc;
@@ -1906,10 +1919,11 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		right = NULL;
 
 	if (right) {
-		btrfs_tree_lock(right);
+		__btrfs_tree_lock(right, BTRFS_NESTING_RIGHT);
 		btrfs_set_lock_blocking_write(right);
 		wret = btrfs_cow_block(trans, root, right,
-				       parent, pslot + 1, &right);
+				       parent, pslot + 1, &right,
+				       BTRFS_NESTING_RIGHT_COW);
 		if (wret) {
 			ret = wret;
 			goto enospc;
@@ -2069,7 +2083,7 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
 	if (left) {
 		u32 left_nr;
 
-		btrfs_tree_lock(left);
+		__btrfs_tree_lock(left, BTRFS_NESTING_LEFT);
 		btrfs_set_lock_blocking_write(left);
 
 		left_nr = btrfs_header_nritems(left);
@@ -2077,7 +2091,8 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
 			wret = 1;
 		} else {
 			ret = btrfs_cow_block(trans, root, left, parent,
-					      pslot - 1, &left);
+					      pslot - 1, &left,
+					      BTRFS_NESTING_LEFT_COW);
 			if (ret)
 				wret = 1;
 			else {
@@ -2123,7 +2138,7 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
 	if (right) {
 		u32 right_nr;
 
-		btrfs_tree_lock(right);
+		__btrfs_tree_lock(right, BTRFS_NESTING_RIGHT);
 		btrfs_set_lock_blocking_write(right);
 
 		right_nr = btrfs_header_nritems(right);
@@ -2132,7 +2147,7 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
 		} else {
 			ret = btrfs_cow_block(trans, root, right,
 					      parent, pslot + 1,
-					      &right);
+					      &right, BTRFS_NESTING_RIGHT_COW);
 			if (ret)
 				wret = 1;
 			else {
@@ -2601,7 +2616,7 @@ static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
 		 * We don't know the level of the root node until we actually
 		 * have it read locked
 		 */
-		b = btrfs_read_lock_root_node(root);
+		b = __btrfs_read_lock_root_node(root, p->recurse);
 		level = btrfs_header_level(b);
 		if (level > write_lock_level)
 			goto out;
@@ -2740,11 +2755,13 @@ again:
 			btrfs_set_path_blocking(p);
 			if (last_level)
 				err = btrfs_cow_block(trans, root, b, NULL, 0,
-						      &b);
+						      &b,
+						      BTRFS_NESTING_COW);
 			else
 				err = btrfs_cow_block(trans, root, b,
 						      p->nodes[level + 1],
-						      p->slots[level + 1], &b);
+						      p->slots[level + 1], &b,
+						      BTRFS_NESTING_COW);
 			if (err) {
 				ret = err;
 				goto done;
@@ -2875,7 +2892,8 @@ cow_done:
 			} else {
 				if (!btrfs_tree_read_lock_atomic(b)) {
 					btrfs_set_path_blocking(p);
-					btrfs_tree_read_lock(b);
+					__btrfs_tree_read_lock(b, BTRFS_NESTING_NORMAL,
+							       p->recurse);
 				}
 				p->locks[level] = BTRFS_READ_LOCK;
 			}
@@ -3164,6 +3182,58 @@ void btrfs_set_item_key_safe(struct btrfs_fs_info *fs_info,
 }
 
 /*
+ * Check key order of two sibling extent buffers.
+ *
+ * Return true if something is wrong.
+ * Return false if everything is fine.
+ *
+ * Tree-checker only works inside one tree block, thus the following
+ * corruption can not be detected by tree-checker:
+ *
+ * Leaf @left			| Leaf @right
+ * --------------------------------------------------------------
+ * | 1 | 2 | 3 | 4 | 5 | f6 |   | 7 | 8 |
+ *
+ * Key f6 in leaf @left itself is valid, but not valid when the next
+ * key in leaf @right is 7.
+ * This can only be checked at tree block merge time.
+ * And since tree checker has ensured all key order in each tree block
+ * is correct, we only need to bother the last key of @left and the first
+ * key of @right.
+ */
+static bool check_sibling_keys(struct extent_buffer *left,
+			       struct extent_buffer *right)
+{
+	struct btrfs_key left_last;
+	struct btrfs_key right_first;
+	int level = btrfs_header_level(left);
+	int nr_left = btrfs_header_nritems(left);
+	int nr_right = btrfs_header_nritems(right);
+
+	/* No key to check in one of the tree blocks */
+	if (!nr_left || !nr_right)
+		return false;
+
+	if (level) {
+		btrfs_node_key_to_cpu(left, &left_last, nr_left - 1);
+		btrfs_node_key_to_cpu(right, &right_first, 0);
+	} else {
+		btrfs_item_key_to_cpu(left, &left_last, nr_left - 1);
+		btrfs_item_key_to_cpu(right, &right_first, 0);
+	}
+
+	if (btrfs_comp_cpu_keys(&left_last, &right_first) >= 0) {
+		btrfs_crit(left->fs_info,
+"bad key order, sibling blocks, left last (%llu %u %llu) right first (%llu %u %llu)",
+			   left_last.objectid, left_last.type,
+			   left_last.offset, right_first.objectid,
+			   right_first.type, right_first.offset);
+		return true;
+	}
+	return false;
+}
+
+/*
  * try to push data from one node into the next node left in the
  * tree.
  *
@@ -3207,6 +3277,12 @@ static int push_node_left(struct btrfs_trans_handle *trans,
 	} else
 		push_items = min(src_nritems - 8, push_items);
 
+	/* dst is the left eb, src is the middle eb */
+	if (check_sibling_keys(dst, src)) {
+		ret = -EUCLEAN;
+		btrfs_abort_transaction(trans, ret);
+		return ret;
+	}
 	ret = tree_mod_log_eb_copy(dst, src, dst_nritems, 0, push_items);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
@@ -3275,6 +3351,12 @@ static int balance_node_right(struct btrfs_trans_handle *trans,
 	if (max_push < push_items)
 		push_items = max_push;
 
+	/* dst is the right eb, src is the middle eb */
+	if (check_sibling_keys(src, dst)) {
+		ret = -EUCLEAN;
+		btrfs_abort_transaction(trans, ret);
+		return ret;
+	}
 	ret = tree_mod_log_insert_move(dst, push_items, 0, dst_nritems);
 	BUG_ON(ret < 0);
 	memmove_extent_buffer(dst, btrfs_node_key_ptr_offset(push_items),
@@ -3331,7 +3413,8 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans,
 		btrfs_node_key(lower, &lower_key, 0);
 
 	c = alloc_tree_block_no_bg_flush(trans, root, 0, &lower_key, level,
-					 root->node->start, 0);
+					 root->node->start, 0,
+					 BTRFS_NESTING_NEW_ROOT);
 	if (IS_ERR(c))
 		return PTR_ERR(c);
 
@@ -3461,7 +3544,7 @@ static noinline int split_node(struct btrfs_trans_handle *trans,
 	btrfs_node_key(c, &disk_key, mid);
 
 	split = alloc_tree_block_no_bg_flush(trans, root, 0, &disk_key, level,
-					     c->start, 0);
+					     c->start, 0, BTRFS_NESTING_SPLIT);
 	if (IS_ERR(split))
 		return PTR_ERR(split);
 
@@ -3730,7 +3813,7 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
 	if (IS_ERR(right))
 		return 1;
 
-	btrfs_tree_lock(right);
+	__btrfs_tree_lock(right, BTRFS_NESTING_RIGHT);
 	btrfs_set_lock_blocking_write(right);
 
 	free_space = btrfs_leaf_free_space(right);
@@ -3739,7 +3822,7 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
 
 	/* cow and double check */
 	ret = btrfs_cow_block(trans, root, right, upper,
-			      slot + 1, &right);
+			      slot + 1, &right, BTRFS_NESTING_RIGHT_COW);
 	if (ret)
 		goto out_unlock;
 
@@ -3751,6 +3834,12 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
 	if (left_nritems == 0)
 		goto out_unlock;
 
+	if (check_sibling_keys(left, right)) {
+		ret = -EUCLEAN;
+		btrfs_tree_unlock(right);
+		free_extent_buffer(right);
+		return ret;
+	}
 	if (path->slots[0] == left_nritems && !empty) {
 		/* Key greater than all keys in the leaf, right neighbor has
 		 * enough room for it and we're not emptying our leaf to delete
@@ -3963,7 +4052,7 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 	if (IS_ERR(left))
 		return 1;
 
-	btrfs_tree_lock(left);
+	__btrfs_tree_lock(left, BTRFS_NESTING_LEFT);
 	btrfs_set_lock_blocking_write(left);
 
 	free_space = btrfs_leaf_free_space(left);
@@ -3974,7 +4063,8 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 
 	/* cow and double check */
 	ret = btrfs_cow_block(trans, root, left,
-			      path->nodes[1], slot - 1, &left);
+			      path->nodes[1], slot - 1, &left,
+			      BTRFS_NESTING_LEFT_COW);
 	if (ret) {
 		/* we hit -ENOSPC, but it isn't fatal here */
 		if (ret == -ENOSPC)
@@ -3988,6 +4078,10 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 		goto out;
 	}
 
+	if (check_sibling_keys(left, right)) {
+		ret = -EUCLEAN;
+		goto out;
+	}
 	return __push_leaf_left(path, min_data_size,
 			       empty, left, free_space, right_nritems,
 			       max_slot);
@@ -4236,8 +4330,18 @@ again:
 	else
 		btrfs_item_key(l, &disk_key, mid);
 
+	/*
+	 * We have to about BTRFS_NESTING_NEW_ROOT here if we've done a double
+	 * split, because we're only allowed to have MAX_LOCKDEP_SUBCLASSES
+	 * subclasses, which is 8 at the time of this patch, and we've maxed it
+	 * out.  In the future we could add a
+	 * BTRFS_NESTING_SPLIT_THE_SPLITTENING if we need to, but for now just
+	 * use BTRFS_NESTING_NEW_ROOT.
+	 */
 	right = alloc_tree_block_no_bg_flush(trans, root, 0, &disk_key, 0,
-					     l->start, 0);
+					     l->start, 0, num_doubles ?
+					     BTRFS_NESTING_NEW_ROOT :
+					     BTRFS_NESTING_SPLIT);
 	if (IS_ERR(right))
 		return PTR_ERR(right);
 
@@ -4482,9 +4586,7 @@ int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
 		return ret;
 
 	path->slots[0]++;
-	setup_items_for_insert(root, path, new_key, &item_size,
-			       item_size, item_size +
-			       sizeof(struct btrfs_item), 1);
+	setup_items_for_insert(root, path, new_key, &item_size, 1);
 	leaf = path->nodes[0];
 	memcpy_extent_buffer(leaf,
 			     btrfs_item_ptr_offset(leaf, path->slots[0]),
@@ -4657,14 +4759,20 @@ void btrfs_extend_item(struct btrfs_path *path, u32 data_size)
 	}
 }
 
-/*
- * this is a helper for btrfs_insert_empty_items, the main goal here is
- * to save stack depth by doing the bulk of the work in a function
- * that doesn't call btrfs_search_slot
+/**
+ * setup_items_for_insert - Helper called before inserting one or more items
+ * to a leaf. Main purpose is to save stack depth by doing the bulk of the work
+ * in a function that doesn't call btrfs_search_slot
+ *
+ * @root:	root we are inserting items to
+ * @path:	points to the leaf/slot where we are going to insert new items
+ * @cpu_key:	array of keys for items to be inserted
+ * @data_size:	size of the body of each item we are going to insert
+ * @nr:		size of @cpu_key/@data_size arrays
  */
 void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 			    const struct btrfs_key *cpu_key, u32 *data_size,
-			    u32 total_data, u32 total_size, int nr)
+			    int nr)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_item *item;
@@ -4675,6 +4783,12 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 	struct extent_buffer *leaf;
 	int slot;
 	struct btrfs_map_token token;
+	u32 total_size;
+	u32 total_data = 0;
+
+	for (i = 0; i < nr; i++)
+		total_data += data_size[i];
+	total_size = total_data + (nr * sizeof(struct btrfs_item));
 
 	if (path->slots[0] == 0) {
 		btrfs_cpu_key_to_disk(&disk_key, cpu_key);
@@ -4701,7 +4815,8 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 
 		if (old_data < data_end) {
 			btrfs_print_leaf(leaf);
-			btrfs_crit(fs_info, "slot %d old_data %d data_end %d",
+			btrfs_crit(fs_info,
+		"item at slot %d with data offset %u beyond data end of leaf %u",
 				   slot, old_data, data_end);
 			BUG();
 		}
@@ -4734,8 +4849,8 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 		btrfs_cpu_key_to_disk(&disk_key, cpu_key + i);
 		btrfs_set_item_key(leaf, &disk_key, slot + i);
 		item = btrfs_item_nr(slot + i);
-		btrfs_set_token_item_offset(&token, item, data_end - data_size[i]);
 		data_end -= data_size[i];
+		btrfs_set_token_item_offset(&token, item, data_end);
 		btrfs_set_token_item_size(&token, item, data_size[i]);
 	}
 
@@ -4777,8 +4892,7 @@ int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
 	slot = path->slots[0];
 	BUG_ON(slot < 0);
 
-	setup_items_for_insert(root, path, cpu_key, data_size,
-			       total_data, total_size, nr);
+	setup_items_for_insert(root, path, cpu_key, data_size, nr);
 	return 0;
 }
 
@@ -5115,7 +5229,7 @@ again:
 			slot--;
 		/*
 		 * check this node pointer against the min_trans parameters.
-		 * If it is too old, old, skip to the next one.
+		 * If it is too old, skip to the next one.
 		 */
 		while (slot < nritems) {
 			u64 gen;
@@ -5379,7 +5493,9 @@ again:
 			}
 			if (!ret) {
 				btrfs_set_path_blocking(path);
-				btrfs_tree_read_lock(next);
+				__btrfs_tree_read_lock(next,
+						       BTRFS_NESTING_RIGHT,
+						       path->recurse);
 			}
 			next_rw_lock = BTRFS_READ_LOCK;
 		}
@@ -5414,7 +5530,9 @@ again:
 			ret = btrfs_try_tree_read_lock(next);
 			if (!ret) {
 				btrfs_set_path_blocking(path);
-				btrfs_tree_read_lock(next);
+				__btrfs_tree_read_lock(next,
+						       BTRFS_NESTING_RIGHT,
+						       path->recurse);
 			}
 			next_rw_lock = BTRFS_READ_LOCK;
 		}
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 9a72896bed2ee4..aac3d6f4e35b17 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -374,6 +374,7 @@ struct btrfs_path {
 	unsigned int search_commit_root:1;
 	unsigned int need_commit_sem:1;
 	unsigned int skip_release_on_error:1;
+	unsigned int recurse:1;
 };
 #define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r->fs_info) >> 4) - \
 					sizeof(struct btrfs_item))
@@ -494,7 +495,7 @@ enum btrfs_orphan_cleanup_state {
 	ORPHAN_CLEANUP_DONE	= 2,
 };
 
-void btrfs_init_async_reclaim_work(struct work_struct *work);
+void btrfs_init_async_reclaim_work(struct btrfs_fs_info *fs_info);
 
 /* fs_info */
 struct reloc_control;
@@ -541,11 +542,6 @@ enum {
 	/* Used to record internally whether fs has been frozen */
 	BTRFS_FS_FROZEN,
 	/*
-	 * Indicate that a whole-filesystem exclusive operation is running
-	 * (device replace, resize, device add/delete, balance)
-	 */
-	BTRFS_FS_EXCL_OP,
-	/*
 	 * Indicate that balance has been set up from the ioctl and is in the
 	 * main phase. The fs_info::balance_ctl is initialized.
 	 * Set and cleared while holding fs_info::balance_mutex.
@@ -565,6 +561,19 @@ enum {
 	BTRFS_FS_DISCARD_RUNNING,
 };
 
+/*
+ * Exclusive operations (device replace, resize, device add/remove, balance)
+ */
+enum btrfs_exclusive_operation {
+	BTRFS_EXCLOP_NONE,
+	BTRFS_EXCLOP_BALANCE,
+	BTRFS_EXCLOP_DEV_ADD,
+	BTRFS_EXCLOP_DEV_REMOVE,
+	BTRFS_EXCLOP_DEV_REPLACE,
+	BTRFS_EXCLOP_RESIZE,
+	BTRFS_EXCLOP_SWAP_ACTIVATE,
+};
+
 struct btrfs_fs_info {
 	u8 chunk_tree_uuid[BTRFS_UUID_SIZE];
 	unsigned long flags;
@@ -912,6 +921,7 @@ struct btrfs_fs_info {
 
 	/* Used to reclaim the metadata space in the background. */
 	struct work_struct async_reclaim_work;
+	struct work_struct async_data_reclaim_work;
 
 	spinlock_t unused_bgs_lock;
 	struct list_head unused_bgs;
@@ -935,6 +945,9 @@ struct btrfs_fs_info {
 	 */
 	int send_in_progress;
 
+	/* Type of exclusive operation running */
+	unsigned long exclusive_operation;
+
 #ifdef CONFIG_BTRFS_FS_REF_VERIFY
 	spinlock_t ref_verify_lock;
 	struct rb_root block_tree;
@@ -1181,24 +1194,40 @@ struct btrfs_root {
 #endif
 };
 
-struct btrfs_clone_extent_info {
+/*
+ * Structure that conveys information about an extent that is going to replace
+ * all the extents in a file range.
+ */
+struct btrfs_replace_extent_info {
 	u64 disk_offset;
 	u64 disk_len;
 	u64 data_offset;
 	u64 data_len;
 	u64 file_offset;
+	/* Pointer to a file extent item of type regular or prealloc. */
 	char *extent_buf;
-	u32 item_size;
+	/*
+	 * Set to true when attempting to replace a file range with a new extent
+	 * described by this structure, set to false when attempting to clone an
+	 * existing extent into a file range.
+	 */
+	bool is_new_extent;
+	/* Meaningful only if is_new_extent is true. */
+	int qgroup_reserved;
+	/*
+	 * Meaningful only if is_new_extent is true.
+	 * Used to track how many extent items we have already inserted in a
+	 * subvolume tree that refer to the extent described by this structure,
+	 * so that we know when to create a new delayed ref or update an existing
+	 * one.
+	 */
+	int insertions;
 };
 
 struct btrfs_file_private {
 	void *filldir_buf;
 };
 
-static inline u32 btrfs_inode_sectorsize(const struct inode *inode)
-{
-	return btrfs_sb(inode->i_sb)->sectorsize;
-}
 
 static inline u32 BTRFS_LEAF_DATA_SIZE(const struct btrfs_fs_info *info)
 {
@@ -1391,6 +1420,16 @@ static inline void btrfs_init_map_token(struct btrfs_map_token *token,
 #define cpu_to_le8(v) (v)
 #define __le8 u8
 
+static inline u8 get_unaligned_le8(const void *p)
+{
+       return *(u8 *)p;
+}
+
+static inline void put_unaligned_le8(u8 val, void *p)
+{
+       *(u8 *)p = val;
+}
+
 #define read_eb_member(eb, ptr, type, member, result) (\
 	read_extent_buffer(eb, (char *)(result),			\
 			   ((unsigned long)(ptr)) +			\
@@ -1449,27 +1488,25 @@ static inline void btrfs_set_token_##name(struct btrfs_map_token *token,\
 static inline u##bits btrfs_##name(const struct extent_buffer *eb)	\
 {									\
 	const type *p = page_address(eb->pages[0]);			\
-	u##bits res = le##bits##_to_cpu(p->member);			\
-	return res;							\
+	return get_unaligned_le##bits(&p->member);			\
 }									\
 static inline void btrfs_set_##name(const struct extent_buffer *eb,	\
 				    u##bits val)			\
 {									\
 	type *p = page_address(eb->pages[0]);				\
-	p->member = cpu_to_le##bits(val);				\
+	put_unaligned_le##bits(val, &p->member);			\
 }
 
 #define BTRFS_SETGET_STACK_FUNCS(name, type, member, bits)		\
 static inline u##bits btrfs_##name(const type *s)			\
 {									\
-	return le##bits##_to_cpu(s->member);				\
+	return get_unaligned_le##bits(&s->member);			\
 }									\
 static inline void btrfs_set_##name(type *s, u##bits val)		\
 {									\
-	s->member = cpu_to_le##bits(val);				\
+	put_unaligned_le##bits(val, &s->member);			\
 }
 
-
 static inline u64 btrfs_device_total_bytes(const struct extent_buffer *eb,
 					   struct btrfs_dev_item *s)
 {
@@ -2524,7 +2561,8 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 					     u64 parent, u64 root_objectid,
 					     const struct btrfs_disk_key *key,
 					     int level, u64 hint,
-					     u64 empty_size);
+					     u64 empty_size,
+					     enum btrfs_lock_nesting nest);
 void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 			   struct btrfs_root *root,
 			   struct extent_buffer *buf,
@@ -2592,6 +2630,8 @@ enum btrfs_reserve_flush_enum {
 	 *
 	 * Can be interruped by fatal signal.
 	 */
+	BTRFS_RESERVE_FLUSH_DATA,
+	BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE,
 	BTRFS_RESERVE_FLUSH_ALL,
 
 	/*
@@ -2619,7 +2659,7 @@ enum btrfs_flush_state {
 int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 				     struct btrfs_block_rsv *rsv,
 				     int nitems, bool use_global_rsv);
-void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info,
+void btrfs_subvolume_release_metadata(struct btrfs_root *root,
 				      struct btrfs_block_rsv *rsv);
 void btrfs_delalloc_release_extents(struct btrfs_inode *inode, u64 num_bytes);
 
@@ -2651,8 +2691,6 @@ void btrfs_set_item_key_safe(struct btrfs_fs_info *fs_info,
 			     struct btrfs_path *path,
 			     const struct btrfs_key *new_key);
 struct extent_buffer *btrfs_root_node(struct btrfs_root *root);
-struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root);
-struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root);
 int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path,
 			struct btrfs_key *key, int lowest_level,
 			u64 min_trans);
@@ -2665,7 +2703,8 @@ struct extent_buffer *btrfs_read_node_slot(struct extent_buffer *parent,
 int btrfs_cow_block(struct btrfs_trans_handle *trans,
 		    struct btrfs_root *root, struct extent_buffer *buf,
 		    struct extent_buffer *parent, int parent_slot,
-		    struct extent_buffer **cow_ret);
+		    struct extent_buffer **cow_ret,
+		    enum btrfs_lock_nesting nest);
 int btrfs_copy_root(struct btrfs_trans_handle *trans,
 		      struct btrfs_root *root,
 		      struct extent_buffer *buf,
@@ -2713,7 +2752,7 @@ static inline int btrfs_del_item(struct btrfs_trans_handle *trans,
 
 void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 			    const struct btrfs_key *cpu_key, u32 *data_size,
-			    u32 total_data, u32 total_size, int nr);
+			    int nr);
 int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		      const struct btrfs_key *key, void *data, u32 data_size);
 int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
@@ -2930,6 +2969,10 @@ void btrfs_inode_safe_disk_i_size_write(struct inode *inode, u64 new_i_size);
 u64 btrfs_file_extent_end(const struct btrfs_path *path);
 
 /* inode.c */
+blk_status_t btrfs_submit_data_bio(struct inode *inode, struct bio *bio,
+				   int mirror_num, unsigned long bio_flags);
+int btrfs_verify_data_csum(struct btrfs_io_bio *io_bio, u64 phy_offset,
+			   struct page *page, u64 start, u64 end, int mirror);
 struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
 					   u64 start, u64 len);
 noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
@@ -2956,7 +2999,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 			       u32 min_type);
 
 int btrfs_start_delalloc_snapshot(struct btrfs_root *root);
-int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr);
+int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr);
 int btrfs_set_extent_delalloc(struct btrfs_inode *inode, u64 start, u64 end,
 			      unsigned int extra_bits,
 			      struct extent_state **cached_state);
@@ -3017,6 +3060,7 @@ int btrfs_writepage_cow_fixup(struct page *page, u64 start, u64 end);
 void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start,
 					  u64 end, int uptodate);
 extern const struct dentry_operations btrfs_dentry_operations;
+ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter);
 
 /* ioctl.c */
 long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
@@ -3031,6 +3075,9 @@ void btrfs_get_block_group_info(struct list_head *groups_list,
 				struct btrfs_ioctl_space_info *space);
 void btrfs_update_ioctl_balance_args(struct btrfs_fs_info *fs_info,
 			       struct btrfs_ioctl_balance_args *bargs);
+bool btrfs_exclop_start(struct btrfs_fs_info *fs_info,
+			enum btrfs_exclusive_operation type);
+void btrfs_exclop_finish(struct btrfs_fs_info *fs_info);
 
 /* file.c */
 int __init btrfs_auto_defrag_init(void);
@@ -3053,9 +3100,9 @@ int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
 int btrfs_drop_extents(struct btrfs_trans_handle *trans,
 		       struct btrfs_root *root, struct inode *inode, u64 start,
 		       u64 end, int drop_cache);
-int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
+int btrfs_replace_file_extents(struct inode *inode, struct btrfs_path *path,
 			   const u64 start, const u64 end,
-			   struct btrfs_clone_extent_info *clone_info,
+			   struct btrfs_replace_extent_info *extent_info,
 			   struct btrfs_trans_handle **trans_out);
 int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
 			      struct btrfs_inode *inode, u64 start, u64 end);
@@ -3536,9 +3583,7 @@ static inline int btrfs_defrag_cancelled(struct btrfs_fs_info *fs_info)
 
 /* Sanity test specific functions */
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-void btrfs_test_inode_set_ops(struct inode *inode);
 void btrfs_test_destroy_inode(struct inode *inode);
-
 static inline int btrfs_is_testing(struct btrfs_fs_info *fs_info)
 {
 	return test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state);
diff --git a/fs/btrfs/delalloc-space.c b/fs/btrfs/delalloc-space.c
index 0e354e9e57d003..bacee09b7bfdd7 100644
--- a/fs/btrfs/delalloc-space.c
+++ b/fs/btrfs/delalloc-space.c
@@ -115,126 +115,15 @@ int btrfs_alloc_data_chunk_ondemand(struct btrfs_inode *inode, u64 bytes)
 {
 	struct btrfs_root *root = inode->root;
 	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct btrfs_space_info *data_sinfo = fs_info->data_sinfo;
-	u64 used;
-	int ret = 0;
-	int need_commit = 2;
-	int have_pinned_space;
+	enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_FLUSH_DATA;
 
 	/* Make sure bytes are sectorsize aligned */
 	bytes = ALIGN(bytes, fs_info->sectorsize);
 
-	if (btrfs_is_free_space_inode(inode)) {
-		need_commit = 0;
-		ASSERT(current->journal_info);
-	}
-
-again:
-	/* Make sure we have enough space to handle the data first */
-	spin_lock(&data_sinfo->lock);
-	used = btrfs_space_info_used(data_sinfo, true);
-
-	if (used + bytes > data_sinfo->total_bytes) {
-		struct btrfs_trans_handle *trans;
-
-		/*
-		 * If we don't have enough free bytes in this space then we need
-		 * to alloc a new chunk.
-		 */
-		if (!data_sinfo->full) {
-			u64 alloc_target;
-
-			data_sinfo->force_alloc = CHUNK_ALLOC_FORCE;
-			spin_unlock(&data_sinfo->lock);
-
-			alloc_target = btrfs_data_alloc_profile(fs_info);
-			/*
-			 * It is ugly that we don't call nolock join
-			 * transaction for the free space inode case here.
-			 * But it is safe because we only do the data space
-			 * reservation for the free space cache in the
-			 * transaction context, the common join transaction
-			 * just increase the counter of the current transaction
-			 * handler, doesn't try to acquire the trans_lock of
-			 * the fs.
-			 */
-			trans = btrfs_join_transaction(root);
-			if (IS_ERR(trans))
-				return PTR_ERR(trans);
-
-			ret = btrfs_chunk_alloc(trans, alloc_target,
-						CHUNK_ALLOC_NO_FORCE);
-			btrfs_end_transaction(trans);
-			if (ret < 0) {
-				if (ret != -ENOSPC)
-					return ret;
-				else {
-					have_pinned_space = 1;
-					goto commit_trans;
-				}
-			}
-
-			goto again;
-		}
-
-		/*
-		 * If we don't have enough pinned space to deal with this
-		 * allocation, and no removed chunk in current transaction,
-		 * don't bother committing the transaction.
-		 */
-		have_pinned_space = __percpu_counter_compare(
-			&data_sinfo->total_bytes_pinned,
-			used + bytes - data_sinfo->total_bytes,
-			BTRFS_TOTAL_BYTES_PINNED_BATCH);
-		spin_unlock(&data_sinfo->lock);
-
-		/* Commit the current transaction and try again */
-commit_trans:
-		if (need_commit) {
-			need_commit--;
-
-			if (need_commit > 0) {
-				btrfs_start_delalloc_roots(fs_info, -1);
-				btrfs_wait_ordered_roots(fs_info, U64_MAX, 0,
-							 (u64)-1);
-			}
-
-			trans = btrfs_join_transaction(root);
-			if (IS_ERR(trans))
-				return PTR_ERR(trans);
-			if (have_pinned_space >= 0 ||
-			    test_bit(BTRFS_TRANS_HAVE_FREE_BGS,
-				     &trans->transaction->flags) ||
-			    need_commit > 0) {
-				ret = btrfs_commit_transaction(trans);
-				if (ret)
-					return ret;
-				/*
-				 * The cleaner kthread might still be doing iput
-				 * operations. Wait for it to finish so that
-				 * more space is released.  We don't need to
-				 * explicitly run the delayed iputs here because
-				 * the commit_transaction would have woken up
-				 * the cleaner.
-				 */
-				ret = btrfs_wait_on_delayed_iputs(fs_info);
-				if (ret)
-					return ret;
-				goto again;
-			} else {
-				btrfs_end_transaction(trans);
-			}
-		}
-
-		trace_btrfs_space_reservation(fs_info,
-					      "space_info:enospc",
-					      data_sinfo->flags, bytes, 1);
-		return -ENOSPC;
-	}
-	btrfs_space_info_update_bytes_may_use(fs_info, data_sinfo, bytes);
-	spin_unlock(&data_sinfo->lock);
+	if (btrfs_is_free_space_inode(inode))
+		flush = BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE;
 
-	return 0;
+	return btrfs_reserve_data_bytes(fs_info, bytes, flush);
 }
 
 int btrfs_check_data_free_space(struct btrfs_inode *inode,
@@ -277,9 +166,7 @@ void btrfs_free_reserved_data_space_noquota(struct btrfs_fs_info *fs_info,
 	ASSERT(IS_ALIGNED(len, fs_info->sectorsize));
 
 	data_sinfo = fs_info->data_sinfo;
-	spin_lock(&data_sinfo->lock);
-	btrfs_space_info_update_bytes_may_use(fs_info, data_sinfo, -len);
-	spin_unlock(&data_sinfo->lock);
+	btrfs_space_info_free_bytes_may_use(fs_info, data_sinfo, len);
 }
 
 /*
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index bf1595a42a988c..5aba81e161132b 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -627,8 +627,7 @@ static int btrfs_delayed_inode_reserve_metadata(
 	 */
 	if (!src_rsv || (!trans->bytes_reserved &&
 			 src_rsv->type != BTRFS_BLOCK_RSV_DELALLOC)) {
-		ret = btrfs_qgroup_reserve_meta_prealloc(root,
-				fs_info->nodesize, true);
+		ret = btrfs_qgroup_reserve_meta_prealloc(root, num_bytes, true);
 		if (ret < 0)
 			return ret;
 		ret = btrfs_block_rsv_add(root, dst_rsv, num_bytes,
@@ -769,8 +768,7 @@ static int btrfs_batch_insert_items(struct btrfs_root *root,
 	}
 
 	/* insert the keys of the items */
-	setup_items_for_insert(root, path, keys, data_size,
-			       total_data_size, total_size, nitems);
+	setup_items_for_insert(root, path, keys, data_size, nitems);
 
 	/* insert the dir index items */
 	slot = path->slots[0];
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index e4a1c6afe35dcd..4a0243cb9d9730 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -64,10 +64,6 @@
 
 static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 				       int scrub_ret);
-static void btrfs_dev_replace_update_device_in_mapping_tree(
-						struct btrfs_fs_info *fs_info,
-						struct btrfs_device *srcdev,
-						struct btrfs_device *tgtdev);
 static int btrfs_dev_replace_kthread(void *data);
 
 int btrfs_init_dev_replace(struct btrfs_fs_info *fs_info)
@@ -224,13 +220,12 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 {
 	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) {
+	if (srcdev->fs_devices->seeding) {
 		btrfs_err(fs_info, "the filesystem is a seed filesystem!");
 		return -EINVAL;
 	}
@@ -244,8 +239,7 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 
 	sync_blockdev(bdev);
 
-	devices = &fs_info->fs_devices->devices;
-	list_for_each_entry(device, devices, dev_list) {
+	list_for_each_entry(device, &fs_info->fs_devices->devices, dev_list) {
 		if (device->bdev == bdev) {
 			btrfs_err(fs_info,
 				  "target device is in the filesystem!");
@@ -512,7 +506,7 @@ static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
 	up_write(&dev_replace->rwsem);
 
-	ret = btrfs_sysfs_add_devices_dir(tgt_device->fs_devices, tgt_device);
+	ret = btrfs_sysfs_add_device(tgt_device);
 	if (ret)
 		btrfs_err(fs_info, "kobj add dev failed %d", ret);
 
@@ -630,6 +624,32 @@ static int btrfs_set_target_alloc_state(struct btrfs_device *srcdev,
 	return ret;
 }
 
+static void btrfs_dev_replace_update_device_in_mapping_tree(
+						struct btrfs_fs_info *fs_info,
+						struct btrfs_device *srcdev,
+						struct btrfs_device *tgtdev)
+{
+	struct extent_map_tree *em_tree = &fs_info->mapping_tree;
+	struct extent_map *em;
+	struct map_lookup *map;
+	u64 start = 0;
+	int i;
+
+	write_lock(&em_tree->lock);
+	do {
+		em = lookup_extent_mapping(em_tree, start, (u64)-1);
+		if (!em)
+			break;
+		map = em->map_lookup;
+		for (i = 0; i < map->num_stripes; i++)
+			if (srcdev == map->stripes[i].dev)
+				map->stripes[i].dev = tgtdev;
+		start = em->start + em->len;
+		free_extent_map(em);
+	} while (start);
+	write_unlock(&em_tree->lock);
+}
+
 static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 				       int scrub_ret)
 {
@@ -661,7 +681,7 @@ 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, -1);
+	ret = btrfs_start_delalloc_roots(fs_info, U64_MAX);
 	if (ret) {
 		mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
 		return ret;
@@ -781,7 +801,7 @@ error:
 	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
 	/* replace the sysfs entry */
-	btrfs_sysfs_remove_devices_dir(fs_info->fs_devices, src_device);
+	btrfs_sysfs_remove_device(src_device);
 	btrfs_sysfs_update_devid(tgt_device);
 	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &src_device->dev_state))
 		btrfs_scratch_superblocks(fs_info, src_device->bdev,
@@ -799,32 +819,6 @@ error:
 	return 0;
 }
 
-static void btrfs_dev_replace_update_device_in_mapping_tree(
-						struct btrfs_fs_info *fs_info,
-						struct btrfs_device *srcdev,
-						struct btrfs_device *tgtdev)
-{
-	struct extent_map_tree *em_tree = &fs_info->mapping_tree;
-	struct extent_map *em;
-	struct map_lookup *map;
-	u64 start = 0;
-	int i;
-
-	write_lock(&em_tree->lock);
-	do {
-		em = lookup_extent_mapping(em_tree, start, (u64)-1);
-		if (!em)
-			break;
-		map = em->map_lookup;
-		for (i = 0; i < map->num_stripes; i++)
-			if (srcdev == map->stripes[i].dev)
-				map->stripes[i].dev = tgtdev;
-		start = em->start + em->len;
-		free_extent_map(em);
-	} while (start);
-	write_unlock(&em_tree->lock);
-}
-
 /*
  * Read progress of device replace status according to the state and last
  * stored position. The value format is the same as for
@@ -1025,7 +1019,7 @@ int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info)
 	 * 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)) {
+	if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_REPLACE)) {
 		down_write(&dev_replace->rwsem);
 		dev_replace->replace_state =
 					BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED;
@@ -1062,7 +1056,7 @@ static int btrfs_dev_replace_kthread(void *data)
 	ret = btrfs_dev_replace_finishing(fs_info, ret);
 	WARN_ON(ret && ret != -ECANCELED);
 
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+	btrfs_exclop_finish(fs_info);
 	return 0;
 }
 
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 9f72b092bc2281..764001609a154f 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -50,7 +50,6 @@
 				 BTRFS_SUPER_FLAG_METADUMP |\
 				 BTRFS_SUPER_FLAG_METADUMP_V2)
 
-static const struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
 static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
 static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
@@ -205,53 +204,6 @@ void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb,
 #endif
 
 /*
- * extents on the btree inode are pretty simple, there's one extent
- * that covers the entire device
- */
-struct extent_map *btree_get_extent(struct btrfs_inode *inode,
-				    struct page *page, size_t pg_offset,
-				    u64 start, u64 len)
-{
-	struct extent_map_tree *em_tree = &inode->extent_tree;
-	struct extent_map *em;
-	int ret;
-
-	read_lock(&em_tree->lock);
-	em = lookup_extent_mapping(em_tree, start, len);
-	if (em) {
-		read_unlock(&em_tree->lock);
-		goto out;
-	}
-	read_unlock(&em_tree->lock);
-
-	em = alloc_extent_map();
-	if (!em) {
-		em = ERR_PTR(-ENOMEM);
-		goto out;
-	}
-	em->start = 0;
-	em->len = (u64)-1;
-	em->block_len = (u64)-1;
-	em->block_start = 0;
-
-	write_lock(&em_tree->lock);
-	ret = add_extent_mapping(em_tree, em, 0);
-	if (ret == -EEXIST) {
-		free_extent_map(em);
-		em = lookup_extent_mapping(em_tree, start, len);
-		if (!em)
-			em = ERR_PTR(-EIO);
-	} else if (ret) {
-		free_extent_map(em);
-		em = ERR_PTR(ret);
-	}
-	write_unlock(&em_tree->lock);
-
-out:
-	return em;
-}
-
-/*
  * Compute the csum of a btree block and store the result to provided buffer.
  */
 static void csum_tree_block(struct extent_buffer *buf, u8 *result)
@@ -545,38 +497,35 @@ static int csum_dirty_buffer(struct btrfs_fs_info *fs_info, struct page *page)
 static int check_tree_block_fsid(struct extent_buffer *eb)
 {
 	struct btrfs_fs_info *fs_info = eb->fs_info;
-	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices, *seed_devs;
 	u8 fsid[BTRFS_FSID_SIZE];
-	int ret = 1;
+	u8 *metadata_uuid;
 
 	read_extent_buffer(eb, fsid, offsetof(struct btrfs_header, fsid),
 			   BTRFS_FSID_SIZE);
-	while (fs_devices) {
-		u8 *metadata_uuid;
+	/*
+	 * Checking the incompat flag is only valid for the current fs. For
+	 * seed devices it's forbidden to have their uuid changed so reading
+	 * ->fsid in this case is fine
+	 */
+	if (btrfs_fs_incompat(fs_info, METADATA_UUID))
+		metadata_uuid = fs_devices->metadata_uuid;
+	else
+		metadata_uuid = fs_devices->fsid;
 
-		/*
-		 * Checking the incompat flag is only valid for the current
-		 * fs. For seed devices it's forbidden to have their uuid
-		 * changed so reading ->fsid in this case is fine
-		 */
-		if (fs_devices == fs_info->fs_devices &&
-		    btrfs_fs_incompat(fs_info, METADATA_UUID))
-			metadata_uuid = fs_devices->metadata_uuid;
-		else
-			metadata_uuid = fs_devices->fsid;
+	if (!memcmp(fsid, metadata_uuid, BTRFS_FSID_SIZE))
+		return 0;
 
-		if (!memcmp(fsid, metadata_uuid, BTRFS_FSID_SIZE)) {
-			ret = 0;
-			break;
-		}
-		fs_devices = fs_devices->seed;
-	}
-	return ret;
+	list_for_each_entry(seed_devs, &fs_devices->seed_list, seed_list)
+		if (!memcmp(fsid, seed_devs->fsid, BTRFS_FSID_SIZE))
+			return 0;
+
+	return 1;
 }
 
-static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
-				      u64 phy_offset, struct page *page,
-				      u64 start, u64 end, int mirror)
+int btrfs_validate_metadata_buffer(struct btrfs_io_bio *io_bio, u64 phy_offset,
+				   struct page *page, u64 start, u64 end,
+				   int mirror)
 {
 	u64 found_start;
 	int found_level;
@@ -864,9 +813,8 @@ static int check_async_write(struct btrfs_fs_info *fs_info,
 	return 1;
 }
 
-static blk_status_t btree_submit_bio_hook(struct inode *inode, struct bio *bio,
-					  int mirror_num,
-					  unsigned long bio_flags)
+blk_status_t btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio,
+				       int mirror_num, unsigned long bio_flags)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	int async = check_async_write(fs_info, BTRFS_I(inode));
@@ -951,11 +899,6 @@ static int btree_writepages(struct address_space *mapping,
 	return btree_write_cache_pages(mapping, wbc);
 }
 
-static int btree_readpage(struct file *file, struct page *page)
-{
-	return extent_read_full_page(page, btree_get_extent, 0);
-}
-
 static int btree_releasepage(struct page *page, gfp_t gfp_flags)
 {
 	if (PageWriteback(page) || PageDirty(page))
@@ -995,7 +938,6 @@ static int btree_set_page_dirty(struct page *page)
 }
 
 static const struct address_space_operations btree_aops = {
-	.readpage	= btree_readpage,
 	.writepages	= btree_writepages,
 	.releasepage	= btree_releasepage,
 	.invalidatepage = btree_invalidatepage,
@@ -1208,7 +1150,8 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 	root->root_key.type = BTRFS_ROOT_ITEM_KEY;
 	root->root_key.offset = 0;
 
-	leaf = btrfs_alloc_tree_block(trans, root, 0, objectid, NULL, 0, 0, 0);
+	leaf = btrfs_alloc_tree_block(trans, root, 0, objectid, NULL, 0, 0, 0,
+				      BTRFS_NESTING_NORMAL);
 	if (IS_ERR(leaf)) {
 		ret = PTR_ERR(leaf);
 		leaf = NULL;
@@ -1280,7 +1223,7 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
 	 */
 
 	leaf = btrfs_alloc_tree_block(trans, root, 0, BTRFS_TREE_LOG_OBJECTID,
-			NULL, 0, 0, 0);
+			NULL, 0, 0, 0, BTRFS_NESTING_NORMAL);
 	if (IS_ERR(leaf)) {
 		btrfs_put_root(root);
 		return ERR_CAST(leaf);
@@ -1505,10 +1448,12 @@ void btrfs_check_leaked_roots(struct btrfs_fs_info *fs_info)
 	struct btrfs_root *root;
 
 	while (!list_empty(&fs_info->allocated_roots)) {
+		char buf[BTRFS_ROOT_NAME_BUF_LEN];
+
 		root = list_first_entry(&fs_info->allocated_roots,
 					struct btrfs_root, leak_list);
-		btrfs_err(fs_info, "leaked root %llu-%llu refcount %d",
-			  root->root_key.objectid, root->root_key.offset,
+		btrfs_err(fs_info, "leaked root %s refcount %d",
+			  btrfs_root_name(root->root_key.objectid, buf),
 			  refcount_read(&root->refs));
 		while (refcount_read(&root->refs) > 1)
 			btrfs_put_root(root);
@@ -2115,12 +2060,10 @@ static void btrfs_init_btree_inode(struct btrfs_fs_info *fs_info)
 
 	RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node);
 	extent_io_tree_init(fs_info, &BTRFS_I(inode)->io_tree,
-			    IO_TREE_INODE_IO, inode);
+			    IO_TREE_BTREE_INODE_IO, inode);
 	BTRFS_I(inode)->io_tree.track_uptodate = false;
 	extent_map_tree_init(&BTRFS_I(inode)->extent_tree);
 
-	BTRFS_I(inode)->io_tree.ops = &btree_extent_io_ops;
-
 	BTRFS_I(inode)->root = btrfs_grab_root(fs_info->tree_root);
 	memset(&BTRFS_I(inode)->location, 0, sizeof(struct btrfs_key));
 	set_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags);
@@ -2626,18 +2569,17 @@ static int __cold init_tree_roots(struct btrfs_fs_info *fs_info)
 		level = btrfs_super_root_level(sb);
 		tree_root->node = read_tree_block(fs_info, btrfs_super_root(sb),
 						  generation, level, NULL);
-		if (IS_ERR(tree_root->node) ||
-		    !extent_buffer_uptodate(tree_root->node)) {
+		if (IS_ERR(tree_root->node)) {
 			handle_error = true;
+			ret = PTR_ERR(tree_root->node);
+			tree_root->node = NULL;
+			btrfs_warn(fs_info, "couldn't read tree root");
+			continue;
 
-			if (IS_ERR(tree_root->node)) {
-				ret = PTR_ERR(tree_root->node);
-				tree_root->node = NULL;
-			} else if (!extent_buffer_uptodate(tree_root->node)) {
-				ret = -EUCLEAN;
-			}
-
-			btrfs_warn(fs_info, "failed to read tree root");
+		} else if (!extent_buffer_uptodate(tree_root->node)) {
+			handle_error = true;
+			ret = -EIO;
+			btrfs_warn(fs_info, "error while reading tree root");
 			continue;
 		}
 
@@ -2753,7 +2695,7 @@ void btrfs_init_fs_info(struct btrfs_fs_info *fs_info)
 	fs_info->check_integrity_print_mask = 0;
 #endif
 	btrfs_init_balance(fs_info);
-	btrfs_init_async_reclaim_work(&fs_info->async_reclaim_work);
+	btrfs_init_async_reclaim_work(fs_info);
 
 	spin_lock_init(&fs_info->block_group_cache_lock);
 	fs_info->block_group_cache_tree = RB_ROOT;
@@ -2928,7 +2870,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
 	}
 
 	/*
-	 * Verify the type first, if that or the the checksum value are
+	 * Verify the type first, if that or the checksum value are
 	 * corrupted, we'll find out
 	 */
 	csum_type = btrfs_super_csum_type(disk_super);
@@ -3482,8 +3424,12 @@ struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev,
 		return ERR_CAST(page);
 
 	super = page_address(page);
-	if (btrfs_super_bytenr(super) != bytenr ||
-		    btrfs_super_magic(super) != BTRFS_MAGIC) {
+	if (btrfs_super_magic(super) != BTRFS_MAGIC) {
+		btrfs_release_disk_super(super);
+		return ERR_PTR(-ENODATA);
+	}
+
+	if (btrfs_super_bytenr(super) != bytenr) {
 		btrfs_release_disk_super(super);
 		return ERR_PTR(-EINVAL);
 	}
@@ -4056,6 +4002,7 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info)
 	btrfs_cleanup_defrag_inodes(fs_info);
 
 	cancel_work_sync(&fs_info->async_reclaim_work);
+	cancel_work_sync(&fs_info->async_data_reclaim_work);
 
 	/* Cancel or finish ongoing discard work */
 	btrfs_discard_cleanup(fs_info);
@@ -4687,9 +4634,3 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info)
 
 	return 0;
 }
-
-static const struct extent_io_ops btree_extent_io_ops = {
-	/* mandatory callbacks */
-	.submit_bio_hook = btree_submit_bio_hook,
-	.readpage_end_io_hook = btree_readpage_end_io_hook,
-};
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 00dc39d47ed346..fee69ced58b4b9 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -76,7 +76,11 @@ void btrfs_btree_balance_dirty(struct btrfs_fs_info *fs_info);
 void btrfs_btree_balance_dirty_nodelay(struct btrfs_fs_info *fs_info);
 void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 				 struct btrfs_root *root);
-
+int btrfs_validate_metadata_buffer(struct btrfs_io_bio *io_bio, u64 phy_offset,
+				   struct page *page, u64 start, u64 end,
+				   int mirror);
+blk_status_t btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio,
+				       int mirror_num, unsigned long bio_flags);
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info);
 #endif
@@ -123,9 +127,6 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 				     u64 objectid);
 int btree_lock_page_hook(struct page *page, void *data,
 				void (*flush_fn)(void *));
-struct extent_map *btree_get_extent(struct btrfs_inode *inode,
-				    struct page *page, size_t pg_offset,
-				    u64 start, u64 len);
 int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags);
 int __init btrfs_end_io_wq_init(void);
 void __cold btrfs_end_io_wq_exit(void);
diff --git a/fs/btrfs/extent-io-tree.h b/fs/btrfs/extent-io-tree.h
index 219a09a2b7340a..9800a830636871 100644
--- a/fs/btrfs/extent-io-tree.h
+++ b/fs/btrfs/extent-io-tree.h
@@ -40,6 +40,7 @@ struct io_failure_record;
 enum {
 	IO_TREE_FS_PINNED_EXTENTS,
 	IO_TREE_FS_EXCLUDED_EXTENTS,
+	IO_TREE_BTREE_INODE_IO,
 	IO_TREE_INODE_IO,
 	IO_TREE_INODE_IO_FAILURE,
 	IO_TREE_RELOC_BLOCKS,
@@ -48,6 +49,7 @@ enum {
 	IO_TREE_INODE_FILE_EXTENT,
 	IO_TREE_LOG_CSUM_RANGE,
 	IO_TREE_SELFTEST,
+	IO_TREE_DEVICE_ALLOC_STATE,
 };
 
 struct extent_io_tree {
@@ -61,7 +63,6 @@ struct extent_io_tree {
 	u8 owner;
 
 	spinlock_t lock;
-	const struct extent_io_ops *ops;
 };
 
 struct extent_state {
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 780b9c9a98fe37..3b21fee13e7781 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -1177,7 +1177,22 @@ int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
 					   num_bytes, parent, root_objectid,
 					   owner, offset, 1);
 	if (ret == 0) {
-		BUG_ON(owner < BTRFS_FIRST_FREE_OBJECTID);
+		/*
+		 * We're adding refs to a tree block we already own, this
+		 * should not happen at all.
+		 */
+		if (owner < BTRFS_FIRST_FREE_OBJECTID) {
+			btrfs_crit(trans->fs_info,
+"adding refs to an existing tree ref, bytenr %llu num_bytes %llu root_objectid %llu",
+				   bytenr, num_bytes, root_objectid);
+			if (IS_ENABLED(CONFIG_BTRFS_DEBUG)) {
+				WARN_ON(1);
+				btrfs_crit(trans->fs_info,
+			"path->slots[0]=%d path->nodes[0]:", path->slots[0]);
+				btrfs_print_leaf(path->nodes[0]);
+			}
+			return -EUCLEAN;
+		}
 		update_inline_extent_backref(path, iref, refs_to_add,
 					     extent_op, NULL);
 	} else if (ret == -ENOENT) {
@@ -1397,6 +1412,9 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 /*
  * __btrfs_inc_extent_ref - insert backreference for a given extent
  *
+ * The counterpart is in __btrfs_free_extent(), with examples and more details
+ * how it works.
+ *
  * @trans:	    Handle of transaction
  *
  * @node:	    The delayed ref node used to get the bytenr/length for
@@ -2849,11 +2867,10 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info,
 				len -= to_add;
 			}
 			spin_unlock(&global_rsv->lock);
-			/* Add to any tickets we may have */
-			if (len)
-				btrfs_try_granting_tickets(fs_info,
-							   space_info);
 		}
+		/* Add to any tickets we may have */
+		if (!readonly && return_free_space && len)
+			btrfs_try_granting_tickets(fs_info, space_info);
 		spin_unlock(&space_info->lock);
 	}
 
@@ -2935,6 +2952,65 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 	return 0;
 }
 
+/*
+ * Drop one or more refs of @node.
+ *
+ * 1. Locate the extent refs.
+ *    It's either inline in EXTENT/METADATA_ITEM or in keyed SHARED_* item.
+ *    Locate it, then reduce the refs number or remove the ref line completely.
+ *
+ * 2. Update the refs count in EXTENT/METADATA_ITEM
+ *
+ * Inline backref case:
+ *
+ * in extent tree we have:
+ *
+ * 	item 0 key (13631488 EXTENT_ITEM 1048576) itemoff 16201 itemsize 82
+ *		refs 2 gen 6 flags DATA
+ *		extent data backref root FS_TREE objectid 258 offset 0 count 1
+ *		extent data backref root FS_TREE objectid 257 offset 0 count 1
+ *
+ * This function gets called with:
+ *
+ *    node->bytenr = 13631488
+ *    node->num_bytes = 1048576
+ *    root_objectid = FS_TREE
+ *    owner_objectid = 257
+ *    owner_offset = 0
+ *    refs_to_drop = 1
+ *
+ * Then we should get some like:
+ *
+ * 	item 0 key (13631488 EXTENT_ITEM 1048576) itemoff 16201 itemsize 82
+ *		refs 1 gen 6 flags DATA
+ *		extent data backref root FS_TREE objectid 258 offset 0 count 1
+ *
+ * Keyed backref case:
+ *
+ * in extent tree we have:
+ *
+ *	item 0 key (13631488 EXTENT_ITEM 1048576) itemoff 3971 itemsize 24
+ *		refs 754 gen 6 flags DATA
+ *	[...]
+ *	item 2 key (13631488 EXTENT_DATA_REF <HASH>) itemoff 3915 itemsize 28
+ *		extent data backref root FS_TREE objectid 866 offset 0 count 1
+ *
+ * This function get called with:
+ *
+ *    node->bytenr = 13631488
+ *    node->num_bytes = 1048576
+ *    root_objectid = FS_TREE
+ *    owner_objectid = 866
+ *    owner_offset = 0
+ *    refs_to_drop = 1
+ *
+ * Then we should get some like:
+ *
+ *	item 0 key (13631488 EXTENT_ITEM 1048576) itemoff 3971 itemsize 24
+ *		refs 753 gen 6 flags DATA
+ *
+ * And that (13631488 EXTENT_DATA_REF <HASH>) gets removed.
+ */
 static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 			       struct btrfs_delayed_ref_node *node, u64 parent,
 			       u64 root_objectid, u64 owner_objectid,
@@ -2967,7 +3043,15 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 	path->leave_spinning = 1;
 
 	is_data = owner_objectid >= BTRFS_FIRST_FREE_OBJECTID;
-	BUG_ON(!is_data && refs_to_drop != 1);
+
+	if (!is_data && refs_to_drop != 1) {
+		btrfs_crit(info,
+"invalid refs_to_drop, dropping more than 1 refs for tree block %llu refs_to_drop %u",
+			   node->bytenr, refs_to_drop);
+		ret = -EINVAL;
+		btrfs_abort_transaction(trans, ret);
+		goto out;
+	}
 
 	if (is_data)
 		skinny_metadata = false;
@@ -2976,6 +3060,13 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 				    parent, root_objectid, owner_objectid,
 				    owner_offset);
 	if (ret == 0) {
+		/*
+		 * Either the inline backref or the SHARED_DATA_REF/
+		 * SHARED_BLOCK_REF is found
+		 *
+		 * Here is a quick path to locate EXTENT/METADATA_ITEM.
+		 * It's possible the EXTENT/METADATA_ITEM is near current slot.
+		 */
 		extent_slot = path->slots[0];
 		while (extent_slot >= 0) {
 			btrfs_item_key_to_cpu(path->nodes[0], &key,
@@ -2992,13 +3083,21 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 				found_extent = 1;
 				break;
 			}
+
+			/* Quick path didn't find the EXTEMT/METADATA_ITEM */
 			if (path->slots[0] - extent_slot > 5)
 				break;
 			extent_slot--;
 		}
 
 		if (!found_extent) {
-			BUG_ON(iref);
+			if (iref) {
+				btrfs_crit(info,
+"invalid iref, no EXTENT/METADATA_ITEM found but has inline extent ref");
+				btrfs_abort_transaction(trans, -EUCLEAN);
+				goto err_dump;
+			}
+			/* Must be SHARED_* item, remove the backref first */
 			ret = remove_extent_backref(trans, path, NULL,
 						    refs_to_drop,
 						    is_data, &last_ref);
@@ -3009,6 +3108,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 			btrfs_release_path(path);
 			path->leave_spinning = 1;
 
+			/* Slow path to locate EXTENT/METADATA_ITEM */
 			key.objectid = bytenr;
 			key.type = BTRFS_EXTENT_ITEM_KEY;
 			key.offset = num_bytes;
@@ -3083,19 +3183,26 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 	if (owner_objectid < BTRFS_FIRST_FREE_OBJECTID &&
 	    key.type == BTRFS_EXTENT_ITEM_KEY) {
 		struct btrfs_tree_block_info *bi;
-		BUG_ON(item_size < sizeof(*ei) + sizeof(*bi));
+		if (item_size < sizeof(*ei) + sizeof(*bi)) {
+			btrfs_crit(info,
+"invalid extent item size for key (%llu, %u, %llu) owner %llu, has %u expect >= %lu",
+				   key.objectid, key.type, key.offset,
+				   owner_objectid, item_size,
+				   sizeof(*ei) + sizeof(*bi));
+			btrfs_abort_transaction(trans, -EUCLEAN);
+			goto err_dump;
+		}
 		bi = (struct btrfs_tree_block_info *)(ei + 1);
 		WARN_ON(owner_objectid != btrfs_tree_block_level(leaf, bi));
 	}
 
 	refs = btrfs_extent_refs(leaf, ei);
 	if (refs < refs_to_drop) {
-		btrfs_err(info,
-			  "trying to drop %d refs but we only have %Lu for bytenr %Lu",
+		btrfs_crit(info,
+		"trying to drop %d refs but we only have %llu for bytenr %llu",
 			  refs_to_drop, refs, bytenr);
-		ret = -EINVAL;
-		btrfs_abort_transaction(trans, ret);
-		goto out;
+		btrfs_abort_transaction(trans, -EUCLEAN);
+		goto err_dump;
 	}
 	refs -= refs_to_drop;
 
@@ -3107,7 +3214,12 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		 * be updated by remove_extent_backref
 		 */
 		if (iref) {
-			BUG_ON(!found_extent);
+			if (!found_extent) {
+				btrfs_crit(info,
+"invalid iref, got inlined extent ref but no EXTENT/METADATA_ITEM found");
+				btrfs_abort_transaction(trans, -EUCLEAN);
+				goto err_dump;
+			}
 		} else {
 			btrfs_set_extent_refs(leaf, ei, refs);
 			btrfs_mark_buffer_dirty(leaf);
@@ -3122,13 +3234,39 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 			}
 		}
 	} else {
+		/* In this branch refs == 1 */
 		if (found_extent) {
-			BUG_ON(is_data && refs_to_drop !=
-			       extent_data_ref_count(path, iref));
+			if (is_data && refs_to_drop !=
+			    extent_data_ref_count(path, iref)) {
+				btrfs_crit(info,
+		"invalid refs_to_drop, current refs %u refs_to_drop %u",
+					   extent_data_ref_count(path, iref),
+					   refs_to_drop);
+				btrfs_abort_transaction(trans, -EUCLEAN);
+				goto err_dump;
+			}
 			if (iref) {
-				BUG_ON(path->slots[0] != extent_slot);
+				if (path->slots[0] != extent_slot) {
+					btrfs_crit(info,
+"invalid iref, extent item key (%llu %u %llu) doesn't have wanted iref",
+						   key.objectid, key.type,
+						   key.offset);
+					btrfs_abort_transaction(trans, -EUCLEAN);
+					goto err_dump;
+				}
 			} else {
-				BUG_ON(path->slots[0] != extent_slot + 1);
+				/*
+				 * No inline ref, we must be at SHARED_* item,
+				 * And it's single ref, it must be:
+				 * |	extent_slot	  ||extent_slot + 1|
+				 * [ EXTENT/METADATA_ITEM ][ SHARED_* ITEM ]
+				 */
+				if (path->slots[0] != extent_slot + 1) {
+					btrfs_crit(info,
+	"invalid SHARED_* item, previous item is not EXTENT/METADATA_ITEM");
+					btrfs_abort_transaction(trans, -EUCLEAN);
+					goto err_dump;
+				}
 				path->slots[0] = extent_slot;
 				num_to_del = 2;
 			}
@@ -3169,6 +3307,19 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 out:
 	btrfs_free_path(path);
 	return ret;
+err_dump:
+	/*
+	 * Leaf dump can take up a lot of log buffer, so we only do full leaf
+	 * dump for debug build.
+	 */
+	if (IS_ENABLED(CONFIG_BTRFS_DEBUG)) {
+		btrfs_crit(info, "path->slots[0]=%d extent_slot=%d",
+			   path->slots[0], extent_slot);
+		btrfs_print_leaf(path->nodes[0]);
+	}
+
+	btrfs_free_path(path);
+	return -EUCLEAN;
 }
 
 /*
@@ -3918,11 +4069,12 @@ static int prepare_allocation(struct btrfs_fs_info *fs_info,
  * |- Push harder to find free extents
  *    |- If not found, re-iterate all block groups
  */
-static noinline int find_free_extent(struct btrfs_fs_info *fs_info,
+static noinline int find_free_extent(struct btrfs_root *root,
 				u64 ram_bytes, u64 num_bytes, u64 empty_size,
 				u64 hint_byte_orig, struct btrfs_key *ins,
 				u64 flags, int delalloc)
 {
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret = 0;
 	int cache_block_group_error = 0;
 	struct btrfs_block_group *block_group = NULL;
@@ -3954,7 +4106,7 @@ static noinline int find_free_extent(struct btrfs_fs_info *fs_info,
 	ins->objectid = 0;
 	ins->offset = 0;
 
-	trace_find_free_extent(fs_info, num_bytes, empty_size, flags);
+	trace_find_free_extent(root, num_bytes, empty_size, flags);
 
 	space_info = btrfs_find_space_info(fs_info, flags);
 	if (!space_info) {
@@ -4203,7 +4355,7 @@ int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
 	flags = get_alloc_profile_by_root(root, is_data);
 again:
 	WARN_ON(num_bytes < fs_info->sectorsize);
-	ret = find_free_extent(fs_info, ram_bytes, num_bytes, empty_size,
+	ret = find_free_extent(root, ram_bytes, num_bytes, empty_size,
 			       hint_byte, ins, flags, delalloc);
 	if (!ret && !is_data) {
 		btrfs_dec_block_group_reservations(fs_info, ins->objectid);
@@ -4504,7 +4656,8 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
 
 static struct extent_buffer *
 btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		      u64 bytenr, int level, u64 owner)
+		      u64 bytenr, int level, u64 owner,
+		      enum btrfs_lock_nesting nest)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_buffer *buf;
@@ -4527,7 +4680,7 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	}
 
 	btrfs_set_buffer_lockdep_class(owner, buf, level);
-	btrfs_tree_lock(buf);
+	__btrfs_tree_lock(buf, nest);
 	btrfs_clean_tree_block(buf);
 	clear_bit(EXTENT_BUFFER_STALE, &buf->bflags);
 
@@ -4573,7 +4726,8 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 					     u64 parent, u64 root_objectid,
 					     const struct btrfs_disk_key *key,
 					     int level, u64 hint,
-					     u64 empty_size)
+					     u64 empty_size,
+					     enum btrfs_lock_nesting nest)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_key ins;
@@ -4589,7 +4743,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 	if (btrfs_is_testing(fs_info)) {
 		buf = btrfs_init_new_buffer(trans, root, root->alloc_bytenr,
-					    level, root_objectid);
+					    level, root_objectid, nest);
 		if (!IS_ERR(buf))
 			root->alloc_bytenr += blocksize;
 		return buf;
@@ -4606,7 +4760,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 		goto out_unuse;
 
 	buf = btrfs_init_new_buffer(trans, root, ins.objectid, level,
-				    root_objectid);
+				    root_objectid, nest);
 	if (IS_ERR(buf)) {
 		ret = PTR_ERR(buf);
 		goto out_free_reserved;
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index a940edb1e64f2e..60f5f68d892dff 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -160,19 +160,20 @@ static int add_extent_changeset(struct extent_state *state, unsigned bits,
 	return ret;
 }
 
-static int __must_check submit_one_bio(struct bio *bio, int mirror_num,
-				       unsigned long bio_flags)
+int __must_check submit_one_bio(struct bio *bio, int mirror_num,
+				unsigned long bio_flags)
 {
 	blk_status_t ret = 0;
 	struct extent_io_tree *tree = bio->bi_private;
 
 	bio->bi_private = NULL;
 
-	if (tree->ops)
-		ret = tree->ops->submit_bio_hook(tree->private_data, bio,
-						 mirror_num, bio_flags);
+	if (is_data_inode(tree->private_data))
+		ret = btrfs_submit_data_bio(tree->private_data, bio, mirror_num,
+					    bio_flags);
 	else
-		btrfsic_submit_bio(bio);
+		ret = btrfs_submit_metadata_bio(tree->private_data, bio,
+						mirror_num, bio_flags);
 
 	return blk_status_to_errno(ret);
 }
@@ -280,7 +281,6 @@ void extent_io_tree_init(struct btrfs_fs_info *fs_info,
 {
 	tree->fs_info = fs_info;
 	tree->state = RB_ROOT;
-	tree->ops = NULL;
 	tree->dirty_bytes = 0;
 	spin_lock_init(&tree->lock);
 	tree->private_data = private_data;
@@ -2819,8 +2819,6 @@ static void end_bio_extent_readpage(struct bio *bio)
 		struct page *page = bvec->bv_page;
 		struct inode *inode = page->mapping->host;
 		struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-		bool data_inode = btrfs_ino(BTRFS_I(inode))
-			!= BTRFS_BTREE_INODE_OBJECTID;
 
 		btrfs_debug(fs_info,
 			"end_bio_extent_readpage: bi_sector=%llu, err=%d, mirror=%u",
@@ -2851,9 +2849,12 @@ static void end_bio_extent_readpage(struct bio *bio)
 
 		mirror = io_bio->mirror_num;
 		if (likely(uptodate)) {
-			ret = tree->ops->readpage_end_io_hook(io_bio, offset,
-							      page, start, end,
-							      mirror);
+			if (is_data_inode(inode))
+				ret = btrfs_verify_data_csum(io_bio, offset, page,
+							     start, end, mirror);
+			else
+				ret = btrfs_validate_metadata_buffer(io_bio,
+					offset, page, start, end, mirror);
 			if (ret)
 				uptodate = 0;
 			else
@@ -2866,7 +2867,7 @@ static void end_bio_extent_readpage(struct bio *bio)
 		if (likely(uptodate))
 			goto readpage_ok;
 
-		if (data_inode) {
+		if (is_data_inode(inode)) {
 
 			/*
 			 * The generic bio_readpage_error handles errors the
@@ -2881,7 +2882,7 @@ static void end_bio_extent_readpage(struct bio *bio)
 			if (!btrfs_submit_read_repair(inode, bio, offset, page,
 						start - page_offset(page),
 						start, end, mirror,
-						tree->ops->submit_bio_hook)) {
+						btrfs_submit_data_bio)) {
 				uptodate = !bio->bi_status;
 				offset += len;
 				continue;
@@ -3053,7 +3054,6 @@ static int submit_extent_page(unsigned int opf,
 		else
 			contig = bio_end_sector(bio) == sector;
 
-		ASSERT(tree->ops);
 		if (btrfs_bio_fits_in_stripe(page, page_size, bio, bio_flags))
 			can_merge = false;
 
@@ -3110,8 +3110,7 @@ void set_page_extent_mapped(struct page *page)
 
 static struct extent_map *
 __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
-		 u64 start, u64 len, get_extent_t *get_extent,
-		 struct extent_map **em_cached)
+		 u64 start, u64 len, struct extent_map **em_cached)
 {
 	struct extent_map *em;
 
@@ -3127,7 +3126,7 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
 		*em_cached = NULL;
 	}
 
-	em = get_extent(BTRFS_I(inode), page, pg_offset, start, len);
+	em = btrfs_get_extent(BTRFS_I(inode), page, pg_offset, start, len);
 	if (em_cached && !IS_ERR_OR_NULL(em)) {
 		BUG_ON(*em_cached);
 		refcount_inc(&em->refs);
@@ -3142,12 +3141,9 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
  * XXX JDM: This needs looking at to ensure proper page locking
  * return 0 on success, otherwise return error
  */
-static int __do_readpage(struct page *page,
-			 get_extent_t *get_extent,
-			 struct extent_map **em_cached,
-			 struct bio **bio, int mirror_num,
-			 unsigned long *bio_flags, unsigned int read_flags,
-			 u64 *prev_em_start)
+int btrfs_do_readpage(struct page *page, struct extent_map **em_cached,
+		      struct bio **bio, unsigned long *bio_flags,
+		      unsigned int read_flags, u64 *prev_em_start)
 {
 	struct inode *inode = page->mapping->host;
 	u64 start = page_offset(page);
@@ -3209,7 +3205,7 @@ static int __do_readpage(struct page *page,
 			break;
 		}
 		em = __get_extent_map(inode, page, pg_offset, cur,
-				      end - cur + 1, get_extent, em_cached);
+				      end - cur + 1, em_cached);
 		if (IS_ERR_OR_NULL(em)) {
 			SetPageError(page);
 			unlock_extent(tree, cur, end);
@@ -3241,7 +3237,7 @@ static int __do_readpage(struct page *page,
 
 		/*
 		 * If we have a file range that points to a compressed extent
-		 * and it's followed by a consecutive file range that points to
+		 * and it's followed by a consecutive file range that points
 		 * to the same compressed extent (possibly with a different
 		 * offset and/or length, so it either points to the whole extent
 		 * or only part of it), we must make sure we do not submit a
@@ -3325,7 +3321,7 @@ static int __do_readpage(struct page *page,
 		ret = submit_extent_page(REQ_OP_READ | read_flags, NULL,
 					 page, offset, disk_io_size,
 					 pg_offset, bio,
-					 end_bio_extent_readpage, mirror_num,
+					 end_bio_extent_readpage, 0,
 					 *bio_flags,
 					 this_bio_flag,
 					 force_bio_submit);
@@ -3362,44 +3358,12 @@ static inline void contiguous_readpages(struct page *pages[], int nr_pages,
 	btrfs_lock_and_flush_ordered_range(inode, start, end, NULL);
 
 	for (index = 0; index < nr_pages; index++) {
-		__do_readpage(pages[index], btrfs_get_extent, em_cached,
-				bio, 0, bio_flags, REQ_RAHEAD, prev_em_start);
+		btrfs_do_readpage(pages[index], em_cached, bio, bio_flags,
+				  REQ_RAHEAD, prev_em_start);
 		put_page(pages[index]);
 	}
 }
 
-static int __extent_read_full_page(struct page *page,
-				   get_extent_t *get_extent,
-				   struct bio **bio, int mirror_num,
-				   unsigned long *bio_flags,
-				   unsigned int read_flags)
-{
-	struct btrfs_inode *inode = BTRFS_I(page->mapping->host);
-	u64 start = page_offset(page);
-	u64 end = start + PAGE_SIZE - 1;
-	int ret;
-
-	btrfs_lock_and_flush_ordered_range(inode, start, end, NULL);
-
-	ret = __do_readpage(page, get_extent, NULL, bio, mirror_num,
-			    bio_flags, read_flags, NULL);
-	return ret;
-}
-
-int extent_read_full_page(struct page *page, get_extent_t *get_extent,
-			  int mirror_num)
-{
-	struct bio *bio = NULL;
-	unsigned long bio_flags = 0;
-	int ret;
-
-	ret = __extent_read_full_page(page, get_extent, &bio, mirror_num,
-				      &bio_flags, 0);
-	if (bio)
-		ret = submit_one_bio(bio, mirror_num, bio_flags);
-	return ret;
-}
-
 static void update_nr_written(struct writeback_control *wbc,
 			      unsigned long nr_written)
 {
@@ -4552,7 +4516,7 @@ next:
  * helper function for fiemap, which doesn't want to see any holes.
  * This maps until we find something past 'last'
  */
-static struct extent_map *get_extent_skip_holes(struct inode *inode,
+static struct extent_map *get_extent_skip_holes(struct btrfs_inode *inode,
 						u64 offset, u64 last)
 {
 	u64 sectorsize = btrfs_inode_sectorsize(inode);
@@ -4567,7 +4531,7 @@ static struct extent_map *get_extent_skip_holes(struct inode *inode,
 		if (len == 0)
 			break;
 		len = ALIGN(len, sectorsize);
-		em = btrfs_get_extent_fiemap(BTRFS_I(inode), offset, len);
+		em = btrfs_get_extent_fiemap(inode, offset, len);
 		if (IS_ERR_OR_NULL(em))
 			return em;
 
@@ -4696,7 +4660,7 @@ static int emit_last_fiemap_cache(struct fiemap_extent_info *fieinfo,
 	return ret;
 }
 
-int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+int extent_fiemap(struct btrfs_inode *inode, struct fiemap_extent_info *fieinfo,
 		  u64 start, u64 len)
 {
 	int ret = 0;
@@ -4707,12 +4671,12 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	u64 last;
 	u64 last_for_get_extent = 0;
 	u64 disko = 0;
-	u64 isize = i_size_read(inode);
+	u64 isize = i_size_read(&inode->vfs_inode);
 	struct btrfs_key found_key;
 	struct extent_map *em = NULL;
 	struct extent_state *cached_state = NULL;
 	struct btrfs_path *path;
-	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_root *root = inode->root;
 	struct fiemap_cache cache = { 0 };
 	struct ulist *roots;
 	struct ulist *tmp_ulist;
@@ -4743,8 +4707,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	 * lookup the last file extent.  We're not using i_size here
 	 * because there might be preallocation past i_size
 	 */
-	ret = btrfs_lookup_file_extent(NULL, root, path,
-			btrfs_ino(BTRFS_I(inode)), -1, 0);
+	ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(inode), -1,
+				       0);
 	if (ret < 0) {
 		goto out_free_ulist;
 	} else {
@@ -4758,7 +4722,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	found_type = found_key.type;
 
 	/* No extents, but there might be delalloc bits */
-	if (found_key.objectid != btrfs_ino(BTRFS_I(inode)) ||
+	if (found_key.objectid != btrfs_ino(inode) ||
 	    found_type != BTRFS_EXTENT_DATA_KEY) {
 		/* have to trust i_size as the end */
 		last = (u64)-1;
@@ -4784,7 +4748,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		last_for_get_extent = isize;
 	}
 
-	lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len - 1,
+	lock_extent_bits(&inode->io_tree, start, start + len - 1,
 			 &cached_state);
 
 	em = get_extent_skip_holes(inode, start, last_for_get_extent);
@@ -4853,8 +4817,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			 * then we're just getting a count and we can skip the
 			 * lookup stuff.
 			 */
-			ret = btrfs_check_shared(root,
-						 btrfs_ino(BTRFS_I(inode)),
+			ret = btrfs_check_shared(root, btrfs_ino(inode),
 						 bytenr, roots, tmp_ulist);
 			if (ret < 0)
 				goto out_free;
@@ -4898,7 +4861,7 @@ out_free:
 		ret = emit_last_fiemap_cache(fieinfo, &cache);
 	free_extent_map(em);
 out:
-	unlock_extent_cached(&BTRFS_I(inode)->io_tree, start, start + len - 1,
+	unlock_extent_cached(&inode->io_tree, start, start + len - 1,
 			     &cached_state);
 
 out_free_ulist:
@@ -4990,7 +4953,7 @@ __alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start,
 	rwlock_init(&eb->lock);
 	atomic_set(&eb->blocking_readers, 0);
 	eb->blocking_writers = 0;
-	eb->lock_nested = false;
+	eb->lock_recursed = false;
 	init_waitqueue_head(&eb->write_lock_wq);
 	init_waitqueue_head(&eb->read_lock_wq);
 
@@ -5574,20 +5537,19 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num)
 			}
 
 			ClearPageError(page);
-			err = __extent_read_full_page(page,
-						      btree_get_extent, &bio,
-						      mirror_num, &bio_flags,
-						      REQ_META);
+			err = submit_extent_page(REQ_OP_READ | REQ_META, NULL,
+					 page, page_offset(page), PAGE_SIZE, 0,
+					 &bio, end_bio_extent_readpage,
+					 mirror_num, 0, 0, false);
 			if (err) {
-				ret = err;
 				/*
-				 * We use &bio in above __extent_read_full_page,
-				 * so we ensure that if it returns error, the
-				 * current page fails to add itself to bio and
-				 * it's been unlocked.
-				 *
-				 * We must dec io_pages by ourselves.
+				 * We failed to submit the bio so it's the
+				 * caller's responsibility to perform cleanup
+				 * i.e unlock page/set error bit.
 				 */
+				ret = err;
+				SetPageError(page);
+				unlock_page(page);
 				atomic_dec(&eb->io_pages);
 			}
 		} else {
@@ -5622,6 +5584,36 @@ unlock_exit:
 	return ret;
 }
 
+static bool report_eb_range(const struct extent_buffer *eb, unsigned long start,
+			    unsigned long len)
+{
+	btrfs_warn(eb->fs_info,
+		"access to eb bytenr %llu len %lu out of range start %lu len %lu",
+		eb->start, eb->len, start, len);
+	WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG));
+
+	return true;
+}
+
+/*
+ * Check if the [start, start + len) range is valid before reading/writing
+ * the eb.
+ * NOTE: @start and @len are offset inside the eb, not logical address.
+ *
+ * Caller should not touch the dst/src memory if this function returns error.
+ */
+static inline int check_eb_range(const struct extent_buffer *eb,
+				 unsigned long start, unsigned long len)
+{
+	unsigned long offset;
+
+	/* start, start + len should not go beyond eb->len nor overflow */
+	if (unlikely(check_add_overflow(start, len, &offset) || offset > eb->len))
+		return report_eb_range(eb, start, len);
+
+	return false;
+}
+
 void read_extent_buffer(const struct extent_buffer *eb, void *dstv,
 			unsigned long start, unsigned long len)
 {
@@ -5632,12 +5624,8 @@ void read_extent_buffer(const struct extent_buffer *eb, void *dstv,
 	char *dst = (char *)dstv;
 	unsigned long i = start >> PAGE_SHIFT;
 
-	if (start + len > eb->len) {
-		WARN(1, KERN_ERR "btrfs bad mapping eb start %llu len %lu, wanted %lu %lu\n",
-		     eb->start, eb->len, start, len);
-		memset(dst, 0, len);
+	if (check_eb_range(eb, start, len))
 		return;
-	}
 
 	offset = offset_in_page(start);
 
@@ -5702,8 +5690,8 @@ int memcmp_extent_buffer(const struct extent_buffer *eb, const void *ptrv,
 	unsigned long i = start >> PAGE_SHIFT;
 	int ret = 0;
 
-	WARN_ON(start > eb->len);
-	WARN_ON(start + len > eb->start + eb->len);
+	if (check_eb_range(eb, start, len))
+		return -EINVAL;
 
 	offset = offset_in_page(start);
 
@@ -5756,8 +5744,8 @@ void write_extent_buffer(const struct extent_buffer *eb, const void *srcv,
 	char *src = (char *)srcv;
 	unsigned long i = start >> PAGE_SHIFT;
 
-	WARN_ON(start > eb->len);
-	WARN_ON(start + len > eb->start + eb->len);
+	if (check_eb_range(eb, start, len))
+		return;
 
 	offset = offset_in_page(start);
 
@@ -5785,8 +5773,8 @@ void memzero_extent_buffer(const struct extent_buffer *eb, unsigned long start,
 	char *kaddr;
 	unsigned long i = start >> PAGE_SHIFT;
 
-	WARN_ON(start > eb->len);
-	WARN_ON(start + len > eb->start + eb->len);
+	if (check_eb_range(eb, start, len))
+		return;
 
 	offset = offset_in_page(start);
 
@@ -5830,6 +5818,10 @@ void copy_extent_buffer(const struct extent_buffer *dst,
 	char *kaddr;
 	unsigned long i = dst_offset >> PAGE_SHIFT;
 
+	if (check_eb_range(dst, dst_offset, len) ||
+	    check_eb_range(src, src_offset, len))
+		return;
+
 	WARN_ON(src->len != dst_len);
 
 	offset = offset_in_page(dst_offset);
@@ -6019,25 +6011,15 @@ void memcpy_extent_buffer(const struct extent_buffer *dst,
 			  unsigned long dst_offset, unsigned long src_offset,
 			  unsigned long len)
 {
-	struct btrfs_fs_info *fs_info = dst->fs_info;
 	size_t cur;
 	size_t dst_off_in_page;
 	size_t src_off_in_page;
 	unsigned long dst_i;
 	unsigned long src_i;
 
-	if (src_offset + len > dst->len) {
-		btrfs_err(fs_info,
-			"memmove bogus src_offset %lu move len %lu dst len %lu",
-			 src_offset, len, dst->len);
-		BUG();
-	}
-	if (dst_offset + len > dst->len) {
-		btrfs_err(fs_info,
-			"memmove bogus dst_offset %lu move len %lu dst len %lu",
-			 dst_offset, len, dst->len);
-		BUG();
-	}
+	if (check_eb_range(dst, dst_offset, len) ||
+	    check_eb_range(dst, src_offset, len))
+		return;
 
 	while (len > 0) {
 		dst_off_in_page = offset_in_page(dst_offset);
@@ -6064,7 +6046,6 @@ void memmove_extent_buffer(const struct extent_buffer *dst,
 			   unsigned long dst_offset, unsigned long src_offset,
 			   unsigned long len)
 {
-	struct btrfs_fs_info *fs_info = dst->fs_info;
 	size_t cur;
 	size_t dst_off_in_page;
 	size_t src_off_in_page;
@@ -6073,18 +6054,9 @@ void memmove_extent_buffer(const struct extent_buffer *dst,
 	unsigned long dst_i;
 	unsigned long src_i;
 
-	if (src_offset + len > dst->len) {
-		btrfs_err(fs_info,
-			  "memmove bogus src_offset %lu move len %lu len %lu",
-			  src_offset, len, dst->len);
-		BUG();
-	}
-	if (dst_offset + len > dst->len) {
-		btrfs_err(fs_info,
-			  "memmove bogus dst_offset %lu move len %lu len %lu",
-			  dst_offset, len, dst->len);
-		BUG();
-	}
+	if (check_eb_range(dst, dst_offset, len) ||
+	    check_eb_range(dst, src_offset, len))
+		return;
 	if (dst_offset < src_offset) {
 		memcpy_extent_buffer(dst, dst_offset, src_offset, len);
 		return;
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 30794ae58498bb..f39d02e7f7efe1 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -74,18 +74,6 @@ typedef blk_status_t (submit_bio_hook_t)(struct inode *inode, struct bio *bio,
 typedef blk_status_t (extent_submit_bio_start_t)(void *private_data,
 		struct bio *bio, u64 bio_offset);
 
-struct extent_io_ops {
-	/*
-	 * The following callbacks must be always defined, the function
-	 * pointer will be called unconditionally.
-	 */
-	submit_bio_hook_t *submit_bio_hook;
-	int (*readpage_end_io_hook)(struct btrfs_io_bio *io_bio, u64 phy_offset,
-				    struct page *page, u64 start, u64 end,
-				    int mirror);
-};
-
-
 #define INLINE_EXTENT_BUFFER_PAGES 16
 #define MAX_INLINE_EXTENT_BUFFER_SIZE (INLINE_EXTENT_BUFFER_PAGES * PAGE_SIZE)
 struct extent_buffer {
@@ -102,7 +90,7 @@ struct extent_buffer {
 
 	int blocking_writers;
 	atomic_t blocking_readers;
-	bool lock_nested;
+	bool lock_recursed;
 	/* >= 0 if eb belongs to a log tree, -1 otherwise */
 	short log_index;
 
@@ -193,8 +181,11 @@ typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode,
 int try_release_extent_mapping(struct page *page, gfp_t mask);
 int try_release_extent_buffer(struct page *page);
 
-int extent_read_full_page(struct page *page, get_extent_t *get_extent,
-			  int mirror_num);
+int __must_check submit_one_bio(struct bio *bio, int mirror_num,
+				unsigned long bio_flags);
+int btrfs_do_readpage(struct page *page, struct extent_map **em_cached,
+		      struct bio **bio, unsigned long *bio_flags,
+		      unsigned int read_flags, u64 *prev_em_start);
 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);
@@ -203,7 +194,7 @@ int extent_writepages(struct address_space *mapping,
 int btree_write_cache_pages(struct address_space *mapping,
 			    struct writeback_control *wbc);
 void extent_readahead(struct readahead_control *rac);
-int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+int extent_fiemap(struct btrfs_inode *inode, struct fiemap_extent_info *fieinfo,
 		  u64 start, u64 len);
 void set_page_extent_mapped(struct page *page);
 
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 7d5ec71615b8f0..8f4f2bd6d9b95c 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -318,8 +318,8 @@ blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
 
 		if (page_offsets)
 			offset = page_offset(bvec.bv_page) + bvec.bv_offset;
-		count = btrfs_find_ordered_sum(inode, offset, disk_bytenr,
-					       csum, nblocks);
+		count = btrfs_find_ordered_sum(BTRFS_I(inode), offset,
+					       disk_bytenr, csum, nblocks);
 		if (count)
 			goto found;
 
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 4507c3d093994f..0ff659455b1ebc 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1057,11 +1057,7 @@ delete_extent_item:
 			if (btrfs_comp_cpu_keys(&key, &slot_key) > 0)
 				path->slots[0]++;
 		}
-		setup_items_for_insert(root, path, &key,
-				       &extent_item_size,
-				       extent_item_size,
-				       sizeof(struct btrfs_item) +
-				       extent_item_size, 1);
+		setup_items_for_insert(root, path, &key, &extent_item_size, 1);
 		*key_inserted = 1;
 	}
 
@@ -1477,9 +1473,7 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 	int ret = 0;
 
 	start_pos = round_down(pos, fs_info->sectorsize);
-	last_pos = start_pos
-		+ round_up(pos + write_bytes - start_pos,
-			   fs_info->sectorsize) - 1;
+	last_pos = round_up(pos + write_bytes, fs_info->sectorsize) - 1;
 
 	if (start_pos < inode->vfs_inode.i_size) {
 		struct btrfs_ordered_extent *ordered;
@@ -1497,8 +1491,7 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 				unlock_page(pages[i]);
 				put_page(pages[i]);
 			}
-			btrfs_start_ordered_extent(&inode->vfs_inode,
-					ordered, 1);
+			btrfs_start_ordered_extent(ordered, 1);
 			btrfs_put_ordered_extent(ordered);
 			return -EAGAIN;
 		}
@@ -1872,7 +1865,7 @@ static ssize_t __btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	loff_t endbyte;
 	int err;
 
-	written = generic_file_direct_write(iocb, from);
+	written = btrfs_direct_IO(iocb, from);
 
 	if (written < 0 || !iov_iter_count(from))
 		return written;
@@ -2025,7 +2018,40 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 		atomic_inc(&BTRFS_I(inode)->sync_writers);
 
 	if (iocb->ki_flags & IOCB_DIRECT) {
+		/*
+		 * 1. We must always clear IOCB_DSYNC in order to not deadlock
+		 *    in iomap, as it calls generic_write_sync() in this case.
+		 * 2. If we are async, we can call iomap_dio_complete() either
+		 *    in
+		 *
+		 *    2.1. A worker thread from the last bio completed.  In this
+		 *	   case we need to mark the btrfs_dio_data that it is
+		 *	   async in order to call generic_write_sync() properly.
+		 *	   This is handled by setting BTRFS_DIO_SYNC_STUB in the
+		 *	   current->journal_info.
+		 *    2.2  The submitter context, because all IO completed
+		 *         before we exited iomap_dio_rw().  In this case we can
+		 *         just re-set the IOCB_DSYNC on the iocb and we'll do
+		 *         the sync below.  If our ->end_io() gets called and
+		 *         current->journal_info is set, then we know we're in
+		 *         our current context and we will clear
+		 *         current->journal_info to indicate that we need to
+		 *         sync below.
+		 */
+		if (sync) {
+			ASSERT(current->journal_info == NULL);
+			iocb->ki_flags &= ~IOCB_DSYNC;
+			current->journal_info = BTRFS_DIO_SYNC_STUB;
+		}
 		num_written = __btrfs_direct_write(iocb, from);
+
+		/*
+		 * As stated above, we cleared journal_info, so we need to do
+		 * the sync ourselves.
+		 */
+		if (sync && current->journal_info == NULL)
+			iocb->ki_flags |= IOCB_DSYNC;
+		current->journal_info = NULL;
 	} else {
 		num_written = btrfs_buffered_write(iocb, from);
 		if (num_written > 0)
@@ -2065,12 +2091,12 @@ int btrfs_release_file(struct inode *inode, struct file *filp)
 	filp->private_data = NULL;
 
 	/*
-	 * ordered_data_close is set by setattr when we are about to truncate
-	 * a file from a non-zero size to a zero size.  This tries to
-	 * flush down new bytes that may have been written if the
-	 * application were using truncate to replace a file in place.
+	 * Set by setattr when we are about to truncate a file from a non-zero
+	 * size to a zero size.  This tries to flush down new bytes that may
+	 * have been written if the application were using truncate to replace
+	 * a file in place.
 	 */
-	if (test_and_clear_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
+	if (test_and_clear_bit(BTRFS_INODE_FLUSH_ON_CLOSE,
 			       &BTRFS_I(inode)->runtime_flags))
 			filemap_flush(inode->i_mapping);
 	return 0;
@@ -2116,20 +2142,24 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_log_ctx ctx;
 	int ret = 0, err;
+	u64 len;
+	bool full_sync;
 
 	trace_btrfs_sync_file(file, datasync);
 
 	btrfs_init_log_ctx(&ctx, inode);
 
 	/*
-	 * Set the range to full if the NO_HOLES feature is not enabled.
-	 * This is to avoid missing file extent items representing holes after
-	 * replaying the log.
+	 * Always set the range to a full range, otherwise we can get into
+	 * several problems, from missing file extent items to represent holes
+	 * when not using the NO_HOLES feature, to log tree corruption due to
+	 * races between hole detection during logging and completion of ordered
+	 * extents outside the range, to missing checksums due to ordered extents
+	 * for which we flushed only a subset of their pages.
 	 */
-	if (!btrfs_fs_incompat(fs_info, NO_HOLES)) {
-		start = 0;
-		end = LLONG_MAX;
-	}
+	start = 0;
+	end = LLONG_MAX;
+	len = (u64)LLONG_MAX + 1;
 
 	/*
 	 * We write the dirty pages in the range and wait until they complete
@@ -2153,19 +2183,12 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	atomic_inc(&root->log_batch);
 
 	/*
-	 * If the inode needs a full sync, make sure we use a full range to
-	 * avoid log tree corruption, due to hole detection racing with ordered
-	 * extent completion for adjacent ranges and races between logging and
-	 * completion of ordered extents for adjancent ranges - both races
-	 * could lead to file extent items in the log with overlapping ranges.
-	 * Do this while holding the inode lock, to avoid races with other
-	 * tasks.
+	 * Always check for the full sync flag while holding the inode's lock,
+	 * to avoid races with other tasks. The flag must be either set all the
+	 * time during logging or always off all the time while logging.
 	 */
-	if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
-		     &BTRFS_I(inode)->runtime_flags)) {
-		start = 0;
-		end = LLONG_MAX;
-	}
+	full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+			     &BTRFS_I(inode)->runtime_flags);
 
 	/*
 	 * Before we acquired the inode's lock, someone may have dirtied more
@@ -2196,20 +2219,42 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 * We have to do this here to avoid the priority inversion of waiting on
 	 * IO of a lower priority task while holding a transaction open.
 	 *
-	 * Also, the range length can be represented by u64, we have to do the
-	 * typecasts to avoid signed overflow if it's [0, LLONG_MAX].
+	 * For a full fsync we wait for the ordered extents to complete while
+	 * for a fast fsync we wait just for writeback to complete, and then
+	 * attach the ordered extents to the transaction so that a transaction
+	 * commit waits for their completion, to avoid data loss if we fsync,
+	 * the current transaction commits before the ordered extents complete
+	 * and a power failure happens right after that.
 	 */
-	ret = btrfs_wait_ordered_range(inode, start, (u64)end - (u64)start + 1);
-	if (ret) {
-		up_write(&BTRFS_I(inode)->dio_sem);
-		inode_unlock(inode);
-		goto out;
+	if (full_sync) {
+		ret = btrfs_wait_ordered_range(inode, start, len);
+	} else {
+		/*
+		 * Get our ordered extents as soon as possible to avoid doing
+		 * checksum lookups in the csum tree, and use instead the
+		 * checksums attached to the ordered extents.
+		 */
+		btrfs_get_ordered_extents_for_logging(BTRFS_I(inode),
+						      &ctx.ordered_extents);
+		ret = filemap_fdatawait_range(inode->i_mapping, start, end);
 	}
+
+	if (ret)
+		goto out_release_extents;
+
 	atomic_inc(&root->log_batch);
 
+	/*
+	 * If we are doing a fast fsync we can not bail out if the inode's
+	 * last_trans is <= then the last committed transaction, because we only
+	 * update the last_trans of the inode during ordered extent completion,
+	 * and for a fast fsync we don't wait for that, we only wait for the
+	 * writeback to complete.
+	 */
 	smp_mb();
 	if (btrfs_inode_in_log(BTRFS_I(inode), fs_info->generation) ||
-	    BTRFS_I(inode)->last_trans <= fs_info->last_trans_committed) {
+	    (BTRFS_I(inode)->last_trans <= fs_info->last_trans_committed &&
+	     (full_sync || list_empty(&ctx.ordered_extents)))) {
 		/*
 		 * We've had everything committed since the last time we were
 		 * modified so clear this flag in case it was set for whatever
@@ -2225,9 +2270,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 		 * checked called fsync.
 		 */
 		ret = filemap_check_wb_err(inode->i_mapping, file->f_wb_err);
-		up_write(&BTRFS_I(inode)->dio_sem);
-		inode_unlock(inode);
-		goto out;
+		goto out_release_extents;
 	}
 
 	/*
@@ -2244,12 +2287,11 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
-		up_write(&BTRFS_I(inode)->dio_sem);
-		inode_unlock(inode);
-		goto out;
+		goto out_release_extents;
 	}
 
-	ret = btrfs_log_dentry_safe(trans, dentry, start, end, &ctx);
+	ret = btrfs_log_dentry_safe(trans, dentry, &ctx);
+	btrfs_release_log_ctx_extents(&ctx);
 	if (ret < 0) {
 		/* Fallthrough and commit/free transaction. */
 		ret = 1;
@@ -2276,6 +2318,13 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 				goto out;
 			}
 		}
+		if (!full_sync) {
+			ret = btrfs_wait_ordered_range(inode, start, len);
+			if (ret) {
+				btrfs_end_transaction(trans);
+				goto out;
+			}
+		}
 		ret = btrfs_commit_transaction(trans);
 	} else {
 		ret = btrfs_end_transaction(trans);
@@ -2286,6 +2335,12 @@ out:
 	if (!ret)
 		ret = err;
 	return ret > 0 ? -EIO : ret;
+
+out_release_extents:
+	btrfs_release_log_ctx_extents(&ctx);
+	up_write(&BTRFS_I(inode)->dio_sem);
+	inode_unlock(inode);
+	goto out;
 }
 
 static const struct vm_operations_struct btrfs_file_vm_ops = {
@@ -2481,7 +2536,8 @@ static int btrfs_punch_hole_lock_range(struct inode *inode,
 
 		lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 				 cached_state);
-		ordered = btrfs_lookup_first_ordered_extent(inode, lockend);
+		ordered = btrfs_lookup_first_ordered_extent(BTRFS_I(inode),
+							    lockend);
 
 		/*
 		 * We need to make sure we have no ordered extents in this range
@@ -2509,11 +2565,11 @@ static int btrfs_punch_hole_lock_range(struct inode *inode,
 	return 0;
 }
 
-static int btrfs_insert_clone_extent(struct btrfs_trans_handle *trans,
+static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans,
 				     struct inode *inode,
 				     struct btrfs_path *path,
-				     struct btrfs_clone_extent_info *clone_info,
-				     const u64 clone_len)
+				     struct btrfs_replace_extent_info *extent_info,
+				     const u64 replace_len)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -2522,51 +2578,69 @@ static int btrfs_insert_clone_extent(struct btrfs_trans_handle *trans,
 	struct btrfs_key key;
 	int slot;
 	struct btrfs_ref ref = { 0 };
-	u64 ref_offset;
 	int ret;
 
-	if (clone_len == 0)
+	if (replace_len == 0)
 		return 0;
 
-	if (clone_info->disk_offset == 0 &&
+	if (extent_info->disk_offset == 0 &&
 	    btrfs_fs_incompat(fs_info, NO_HOLES))
 		return 0;
 
 	key.objectid = btrfs_ino(BTRFS_I(inode));
 	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = clone_info->file_offset;
+	key.offset = extent_info->file_offset;
 	ret = btrfs_insert_empty_item(trans, root, path, &key,
-				      clone_info->item_size);
+				      sizeof(struct btrfs_file_extent_item));
 	if (ret)
 		return ret;
 	leaf = path->nodes[0];
 	slot = path->slots[0];
-	write_extent_buffer(leaf, clone_info->extent_buf,
+	write_extent_buffer(leaf, extent_info->extent_buf,
 			    btrfs_item_ptr_offset(leaf, slot),
-			    clone_info->item_size);
+			    sizeof(struct btrfs_file_extent_item));
 	extent = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
-	btrfs_set_file_extent_offset(leaf, extent, clone_info->data_offset);
-	btrfs_set_file_extent_num_bytes(leaf, extent, clone_len);
+	ASSERT(btrfs_file_extent_type(leaf, extent) != BTRFS_FILE_EXTENT_INLINE);
+	btrfs_set_file_extent_offset(leaf, extent, extent_info->data_offset);
+	btrfs_set_file_extent_num_bytes(leaf, extent, replace_len);
+	if (extent_info->is_new_extent)
+		btrfs_set_file_extent_generation(leaf, extent, trans->transid);
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_release_path(path);
 
 	ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode),
-			clone_info->file_offset, clone_len);
+			extent_info->file_offset, replace_len);
 	if (ret)
 		return ret;
 
 	/* If it's a hole, nothing more needs to be done. */
-	if (clone_info->disk_offset == 0)
+	if (extent_info->disk_offset == 0)
 		return 0;
 
-	inode_add_bytes(inode, clone_len);
-	btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF,
-			       clone_info->disk_offset,
-			       clone_info->disk_len, 0);
-	ref_offset = clone_info->file_offset - clone_info->data_offset;
-	btrfs_init_data_ref(&ref, root->root_key.objectid,
-			    btrfs_ino(BTRFS_I(inode)), ref_offset);
-	ret = btrfs_inc_extent_ref(trans, &ref);
+	inode_add_bytes(inode, replace_len);
+
+	if (extent_info->is_new_extent && extent_info->insertions == 0) {
+		key.objectid = extent_info->disk_offset;
+		key.type = BTRFS_EXTENT_ITEM_KEY;
+		key.offset = extent_info->disk_len;
+		ret = btrfs_alloc_reserved_file_extent(trans, root,
+						       btrfs_ino(BTRFS_I(inode)),
+						       extent_info->file_offset,
+						       extent_info->qgroup_reserved,
+						       &key);
+	} else {
+		u64 ref_offset;
+
+		btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF,
+				       extent_info->disk_offset,
+				       extent_info->disk_len, 0);
+		ref_offset = extent_info->file_offset - extent_info->data_offset;
+		btrfs_init_data_ref(&ref, root->root_key.objectid,
+				    btrfs_ino(BTRFS_I(inode)), ref_offset);
+		ret = btrfs_inc_extent_ref(trans, &ref);
+	}
+
+	extent_info->insertions++;
 
 	return ret;
 }
@@ -2574,15 +2648,15 @@ static int btrfs_insert_clone_extent(struct btrfs_trans_handle *trans,
 /*
  * The respective range must have been previously locked, as well as the inode.
  * The end offset is inclusive (last byte of the range).
- * @clone_info is NULL for fallocate's hole punching and non-NULL for extent
- * cloning.
- * When cloning, we don't want to end up in a state where we dropped extents
- * without inserting a new one, so we must abort the transaction to avoid a
- * corruption.
+ * @extent_info is NULL for fallocate's hole punching and non-NULL when replacing
+ * the file range with an extent.
+ * When not punching a hole, we don't want to end up in a state where we dropped
+ * extents without inserting a new one, so we must abort the transaction to avoid
+ * a corruption.
  */
-int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
+int btrfs_replace_file_extents(struct inode *inode, struct btrfs_path *path,
 			   const u64 start, const u64 end,
-			   struct btrfs_clone_extent_info *clone_info,
+			   struct btrfs_replace_extent_info *extent_info,
 			   struct btrfs_trans_handle **trans_out)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
@@ -2611,10 +2685,10 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 	/*
 	 * 1 - update the inode
 	 * 1 - removing the extents in the range
-	 * 1 - adding the hole extent if no_holes isn't set or if we are cloning
-	 *     an extent
+	 * 1 - adding the hole extent if no_holes isn't set or if we are
+	 *     replacing the range with a new extent
 	 */
-	if (!btrfs_fs_incompat(fs_info, NO_HOLES) || clone_info)
+	if (!btrfs_fs_incompat(fs_info, NO_HOLES) || extent_info)
 		rsv_count = 3;
 	else
 		rsv_count = 2;
@@ -2644,14 +2718,15 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 			 * returned by __btrfs_drop_extents() without having
 			 * changed anything in the file.
 			 */
-			if (clone_info && ret && ret != -EOPNOTSUPP)
+			if (extent_info && !extent_info->is_new_extent &&
+			    ret && ret != -EOPNOTSUPP)
 				btrfs_abort_transaction(trans, ret);
 			break;
 		}
 
 		trans->block_rsv = &fs_info->trans_block_rsv;
 
-		if (!clone_info && cur_offset < drop_end &&
+		if (!extent_info && cur_offset < drop_end &&
 		    cur_offset < ino_size) {
 			ret = fill_holes(trans, BTRFS_I(inode), path,
 					cur_offset, drop_end);
@@ -2665,7 +2740,7 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 				btrfs_abort_transaction(trans, ret);
 				break;
 			}
-		} else if (!clone_info && cur_offset < drop_end) {
+		} else if (!extent_info && cur_offset < drop_end) {
 			/*
 			 * We are past the i_size here, but since we didn't
 			 * insert holes we need to clear the mapped area so we
@@ -2685,18 +2760,18 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 			}
 		}
 
-		if (clone_info && drop_end > clone_info->file_offset) {
-			u64 clone_len = drop_end - clone_info->file_offset;
+		if (extent_info && drop_end > extent_info->file_offset) {
+			u64 replace_len = drop_end - extent_info->file_offset;
 
-			ret = btrfs_insert_clone_extent(trans, inode, path,
-							clone_info, clone_len);
+			ret = btrfs_insert_replace_extent(trans, inode, path,
+							extent_info, replace_len);
 			if (ret) {
 				btrfs_abort_transaction(trans, ret);
 				break;
 			}
-			clone_info->data_len -= clone_len;
-			clone_info->data_offset += clone_len;
-			clone_info->file_offset += clone_len;
+			extent_info->data_len -= replace_len;
+			extent_info->data_offset += replace_len;
+			extent_info->file_offset += replace_len;
 		}
 
 		cur_offset = drop_end;
@@ -2720,7 +2795,7 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 		BUG_ON(ret);	/* shouldn't happen */
 		trans->block_rsv = rsv;
 
-		if (!clone_info) {
+		if (!extent_info) {
 			ret = find_first_non_hole(inode, &cur_offset, &len);
 			if (unlikely(ret < 0))
 				break;
@@ -2739,7 +2814,7 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 	 * than 16Mb would force the full fsync any way (when
 	 * try_release_extent_mapping() is invoked during page cache truncation.
 	 */
-	if (clone_info)
+	if (extent_info && !extent_info->is_new_extent)
 		set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
 			&BTRFS_I(inode)->runtime_flags);
 
@@ -2765,7 +2840,7 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 	 * (because it's useless) or if it represents a 0 bytes range (when
 	 * cur_offset == drop_end).
 	 */
-	if (!clone_info && cur_offset < ino_size && cur_offset < drop_end) {
+	if (!extent_info && cur_offset < ino_size && cur_offset < drop_end) {
 		ret = fill_holes(trans, BTRFS_I(inode), path,
 				cur_offset, drop_end);
 		if (ret) {
@@ -2773,7 +2848,7 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 			btrfs_abort_transaction(trans, ret);
 			goto out_trans;
 		}
-	} else if (!clone_info && cur_offset < drop_end) {
+	} else if (!extent_info && cur_offset < drop_end) {
 		/* See the comment in the loop above for the reasoning here. */
 		ret = btrfs_inode_clear_file_extent_range(BTRFS_I(inode),
 					cur_offset, drop_end - cur_offset);
@@ -2783,9 +2858,9 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 		}
 
 	}
-	if (clone_info) {
-		ret = btrfs_insert_clone_extent(trans, inode, path, clone_info,
-						clone_info->data_len);
+	if (extent_info) {
+		ret = btrfs_insert_replace_extent(trans, inode, path, extent_info,
+						extent_info->data_len);
 		if (ret) {
 			btrfs_abort_transaction(trans, ret);
 			goto out_trans;
@@ -2840,9 +2915,9 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 		goto out_only_mutex;
 	}
 
-	lockstart = round_up(offset, btrfs_inode_sectorsize(inode));
+	lockstart = round_up(offset, btrfs_inode_sectorsize(BTRFS_I(inode)));
 	lockend = round_down(offset + len,
-			     btrfs_inode_sectorsize(inode)) - 1;
+			     btrfs_inode_sectorsize(BTRFS_I(inode))) - 1;
 	same_block = (BTRFS_BYTES_TO_BLKS(fs_info, offset))
 		== (BTRFS_BYTES_TO_BLKS(fs_info, offset + len - 1));
 	/*
@@ -2927,7 +3002,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 		goto out;
 	}
 
-	ret = btrfs_punch_hole_range(inode, path, lockstart, lockend, NULL,
+	ret = btrfs_replace_file_extents(inode, path, lockstart, lockend, NULL,
 				     &trans);
 	btrfs_free_path(path);
 	if (ret)
@@ -3044,7 +3119,7 @@ enum {
 	RANGE_BOUNDARY_HOLE,
 };
 
-static int btrfs_zero_range_check_range_boundary(struct inode *inode,
+static int btrfs_zero_range_check_range_boundary(struct btrfs_inode *inode,
 						 u64 offset)
 {
 	const u64 sectorsize = btrfs_inode_sectorsize(inode);
@@ -3052,7 +3127,7 @@ static int btrfs_zero_range_check_range_boundary(struct inode *inode,
 	int ret;
 
 	offset = round_down(offset, sectorsize);
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
+	em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize);
 	if (IS_ERR(em))
 		return PTR_ERR(em);
 
@@ -3077,7 +3152,7 @@ static int btrfs_zero_range(struct inode *inode,
 	struct extent_changeset *data_reserved = NULL;
 	int ret;
 	u64 alloc_hint = 0;
-	const u64 sectorsize = btrfs_inode_sectorsize(inode);
+	const u64 sectorsize = btrfs_inode_sectorsize(BTRFS_I(inode));
 	u64 alloc_start = round_down(offset, sectorsize);
 	u64 alloc_end = round_up(offset + len, sectorsize);
 	u64 bytes_to_reserve = 0;
@@ -3167,7 +3242,8 @@ static int btrfs_zero_range(struct inode *inode,
 	 * to cover them.
 	 */
 	if (!IS_ALIGNED(offset, sectorsize)) {
-		ret = btrfs_zero_range_check_range_boundary(inode, offset);
+		ret = btrfs_zero_range_check_range_boundary(BTRFS_I(inode),
+							    offset);
 		if (ret < 0)
 			goto out;
 		if (ret == RANGE_BOUNDARY_HOLE) {
@@ -3183,7 +3259,7 @@ static int btrfs_zero_range(struct inode *inode,
 	}
 
 	if (!IS_ALIGNED(offset + len, sectorsize)) {
-		ret = btrfs_zero_range_check_range_boundary(inode,
+		ret = btrfs_zero_range_check_range_boundary(BTRFS_I(inode),
 							    offset + len);
 		if (ret < 0)
 			goto out;
@@ -3258,7 +3334,7 @@ static long btrfs_fallocate(struct file *file, int mode,
 	u64 locked_end;
 	u64 actual_end = 0;
 	struct extent_map *em;
-	int blocksize = btrfs_inode_sectorsize(inode);
+	int blocksize = btrfs_inode_sectorsize(BTRFS_I(inode));
 	int ret;
 
 	alloc_start = round_down(offset, blocksize);
@@ -3340,7 +3416,8 @@ static long btrfs_fallocate(struct file *file, int mode,
 		 */
 		lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start,
 				 locked_end, &cached_state);
-		ordered = btrfs_lookup_first_ordered_extent(inode, locked_end);
+		ordered = btrfs_lookup_first_ordered_extent(BTRFS_I(inode),
+							    locked_end);
 
 		if (ordered &&
 		    ordered->file_offset + ordered->num_bytes > alloc_start &&
@@ -3541,9 +3618,26 @@ static int btrfs_file_open(struct inode *inode, struct file *filp)
 	return generic_file_open(inode, filp);
 }
 
+static ssize_t btrfs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	ssize_t ret = 0;
+
+	if (iocb->ki_flags & IOCB_DIRECT) {
+		struct inode *inode = file_inode(iocb->ki_filp);
+
+		inode_lock_shared(inode);
+		ret = btrfs_direct_IO(iocb, to);
+		inode_unlock_shared(inode);
+		if (ret < 0)
+			return ret;
+	}
+
+	return generic_file_buffered_read(iocb, to, ret);
+}
+
 const struct file_operations btrfs_file_operations = {
 	.llseek		= btrfs_file_llseek,
-	.read_iter      = generic_file_read_iter,
+	.read_iter      = btrfs_file_read_iter,
 	.splice_read	= generic_file_splice_read,
 	.write_iter	= btrfs_file_write_iter,
 	.splice_write	= iter_file_splice_write,
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index dc82fd0c80cbba..af0013d3df63fd 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -413,8 +413,6 @@ static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, bool uptodate)
 
 static void io_ctl_set_generation(struct btrfs_io_ctl *io_ctl, u64 generation)
 {
-	__le64 *val;
-
 	io_ctl_map_page(io_ctl, 1);
 
 	/*
@@ -429,14 +427,13 @@ static void io_ctl_set_generation(struct btrfs_io_ctl *io_ctl, u64 generation)
 		io_ctl->size -= sizeof(u64) * 2;
 	}
 
-	val = io_ctl->cur;
-	*val = cpu_to_le64(generation);
+	put_unaligned_le64(generation, io_ctl->cur);
 	io_ctl->cur += sizeof(u64);
 }
 
 static int io_ctl_check_generation(struct btrfs_io_ctl *io_ctl, u64 generation)
 {
-	__le64 *gen;
+	u64 cache_gen;
 
 	/*
 	 * Skip the crc area.  If we don't check crcs then we just have a 64bit
@@ -451,11 +448,11 @@ static int io_ctl_check_generation(struct btrfs_io_ctl *io_ctl, u64 generation)
 		io_ctl->size -= sizeof(u64) * 2;
 	}
 
-	gen = io_ctl->cur;
-	if (le64_to_cpu(*gen) != generation) {
+	cache_gen = get_unaligned_le64(io_ctl->cur);
+	if (cache_gen != generation) {
 		btrfs_err_rl(io_ctl->fs_info,
 			"space cache generation (%llu) does not match inode (%llu)",
-				*gen, generation);
+				cache_gen, generation);
 		io_ctl_unmap_page(io_ctl);
 		return -EIO;
 	}
@@ -525,8 +522,8 @@ static int io_ctl_add_entry(struct btrfs_io_ctl *io_ctl, u64 offset, u64 bytes,
 		return -ENOSPC;
 
 	entry = io_ctl->cur;
-	entry->offset = cpu_to_le64(offset);
-	entry->bytes = cpu_to_le64(bytes);
+	put_unaligned_le64(offset, &entry->offset);
+	put_unaligned_le64(bytes, &entry->bytes);
 	entry->type = (bitmap) ? BTRFS_FREE_SPACE_BITMAP :
 		BTRFS_FREE_SPACE_EXTENT;
 	io_ctl->cur += sizeof(struct btrfs_free_space_entry);
@@ -599,8 +596,8 @@ static int io_ctl_read_entry(struct btrfs_io_ctl *io_ctl,
 	}
 
 	e = io_ctl->cur;
-	entry->offset = le64_to_cpu(e->offset);
-	entry->bytes = le64_to_cpu(e->bytes);
+	entry->offset = get_unaligned_le64(&e->offset);
+	entry->bytes = get_unaligned_le64(&e->bytes);
 	*type = e->type;
 	io_ctl->cur += sizeof(struct btrfs_free_space_entry);
 	io_ctl->size -= sizeof(struct btrfs_free_space_entry);
@@ -1353,7 +1350,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 
 	/*
 	 * at this point the pages are under IO and we're happy,
-	 * The caller is responsible for waiting on them and updating the
+	 * The caller is responsible for waiting on them and updating
 	 * the cache and the inode
 	 */
 	io_ctl->entries = entries;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 9570458aa8471d..936c3137c6467d 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6,7 +6,6 @@
 #include <crypto/hash.h>
 #include <linux/kernel.h>
 #include <linux/bio.h>
-#include <linux/buffer_head.h>
 #include <linux/file.h>
 #include <linux/fs.h>
 #include <linux/pagemap.h>
@@ -31,6 +30,7 @@
 #include <linux/swap.h>
 #include <linux/migrate.h>
 #include <linux/sched/mm.h>
+#include <linux/iomap.h>
 #include <asm/unaligned.h>
 #include "misc.h"
 #include "ctree.h"
@@ -59,9 +59,10 @@ struct btrfs_iget_args {
 
 struct btrfs_dio_data {
 	u64 reserve;
-	u64 unsubmitted_oe_range_start;
-	u64 unsubmitted_oe_range_end;
-	int overwrite;
+	loff_t length;
+	ssize_t submitted;
+	struct extent_changeset *data_reserved;
+	bool sync;
 };
 
 static const struct inode_operations btrfs_dir_inode_operations;
@@ -70,7 +71,6 @@ static const struct inode_operations btrfs_special_inode_operations;
 static const struct inode_operations btrfs_file_inode_operations;
 static const struct address_space_operations btrfs_aops;
 static const struct file_operations btrfs_dir_file_operations;
-static const struct extent_io_ops btrfs_extent_io_ops;
 
 static struct kmem_cache *btrfs_inode_cachep;
 struct kmem_cache *btrfs_trans_handle_cachep;
@@ -140,13 +140,6 @@ static inline void btrfs_cleanup_ordered_extents(struct btrfs_inode *inode,
 
 static int btrfs_dirty_inode(struct inode *inode);
 
-#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-void btrfs_test_inode_set_ops(struct inode *inode)
-{
-	BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
-}
-#endif
-
 static int btrfs_init_inode_security(struct btrfs_trans_handle *trans,
 				     struct inode *inode,  struct inode *dir,
 				     const struct qstr *qstr)
@@ -2183,9 +2176,8 @@ static blk_status_t btrfs_submit_bio_start(void *private_data, struct bio *bio,
  *
  *    c-3) otherwise:			async submit
  */
-static blk_status_t btrfs_submit_bio_hook(struct inode *inode, struct bio *bio,
-					  int mirror_num,
-					  unsigned long bio_flags)
+blk_status_t btrfs_submit_data_bio(struct inode *inode, struct bio *bio,
+				   int mirror_num, unsigned long bio_flags)
 
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
@@ -2245,16 +2237,15 @@ out:
  * given a list of ordered sums record them in the inode.  This happens
  * at IO completion time based on sums calculated at bio submission time.
  */
-static noinline int add_pending_csums(struct btrfs_trans_handle *trans,
-			     struct inode *inode, struct list_head *list)
+static int add_pending_csums(struct btrfs_trans_handle *trans,
+			     struct list_head *list)
 {
 	struct btrfs_ordered_sum *sum;
 	int ret;
 
 	list_for_each_entry(sum, list, list) {
 		trans->adding_csums = true;
-		ret = btrfs_csum_file_blocks(trans,
-		       BTRFS_I(inode)->root->fs_info->csum_root, sum);
+		ret = btrfs_csum_file_blocks(trans, trans->fs_info->csum_root, sum);
 		trans->adding_csums = false;
 		if (ret)
 			return ret;
@@ -2357,7 +2348,7 @@ again:
 		unlock_extent_cached(&inode->io_tree, page_start, page_end,
 				     &cached_state);
 		unlock_page(page);
-		btrfs_start_ordered_extent(&inode->vfs_inode, ordered, 1);
+		btrfs_start_ordered_extent(ordered, 1);
 		btrfs_put_ordered_extent(ordered);
 		goto again;
 	}
@@ -2548,7 +2539,6 @@ static void btrfs_release_delalloc_bytes(struct btrfs_fs_info *fs_info,
 }
 
 static int insert_ordered_extent_file_extent(struct btrfs_trans_handle *trans,
-					     struct inode *inode,
 					     struct btrfs_ordered_extent *oe)
 {
 	struct btrfs_file_extent_item stack_fi;
@@ -2568,8 +2558,9 @@ static int insert_ordered_extent_file_extent(struct btrfs_trans_handle *trans,
 	btrfs_set_stack_file_extent_compression(&stack_fi, oe->compress_type);
 	/* Encryption and other encoding is reserved and all 0 */
 
-	return insert_reserved_file_extent(trans, BTRFS_I(inode), oe->file_offset,
-					   &stack_fi, oe->qgroup_rsv);
+	return insert_reserved_file_extent(trans, BTRFS_I(oe->inode),
+					   oe->file_offset, &stack_fi,
+					   oe->qgroup_rsv);
 }
 
 /*
@@ -2666,8 +2657,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 						logical_len);
 	} else {
 		BUG_ON(root == fs_info->tree_root);
-		ret = insert_ordered_extent_file_extent(trans, inode,
-							ordered_extent);
+		ret = insert_ordered_extent_file_extent(trans, ordered_extent);
 		if (!ret) {
 			clear_reserved_extent = false;
 			btrfs_release_delalloc_bytes(fs_info,
@@ -2683,7 +2673,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		goto out;
 	}
 
-	ret = add_pending_csums(trans, inode, &ordered_extent->list);
+	ret = add_pending_csums(trans, &ordered_extent->list);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
 		goto out;
@@ -2752,7 +2742,7 @@ out:
 	 * This needs to be done to make sure anybody waiting knows we are done
 	 * updating everything for this ordered extent.
 	 */
-	btrfs_remove_ordered_extent(inode, ordered_extent);
+	btrfs_remove_ordered_extent(BTRFS_I(inode), ordered_extent);
 
 	/* once for us */
 	btrfs_put_ordered_extent(ordered_extent);
@@ -2772,8 +2762,8 @@ static void finish_ordered_fn(struct btrfs_work *work)
 void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start,
 					  u64 end, int uptodate)
 {
-	struct inode *inode = page->mapping->host;
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct btrfs_inode *inode = BTRFS_I(page->mapping->host);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_ordered_extent *ordered_extent = NULL;
 	struct btrfs_workqueue *wq;
 
@@ -2784,7 +2774,7 @@ void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start,
 					    end - start + 1, uptodate))
 		return;
 
-	if (btrfs_is_free_space_inode(BTRFS_I(inode)))
+	if (btrfs_is_free_space_inode(inode))
 		wq = fs_info->endio_freespace_worker;
 	else
 		wq = fs_info->endio_write_workers;
@@ -2833,9 +2823,8 @@ zeroit:
  * if there's a match, we allow the bio to finish.  If not, the code in
  * extent_io.c will try to find good copies for us.
  */
-static int btrfs_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
-				      u64 phy_offset, struct page *page,
-				      u64 start, u64 end, int mirror)
+int btrfs_verify_data_csum(struct btrfs_io_bio *io_bio, u64 phy_offset,
+			   struct page *page, u64 start, u64 end, int mirror)
 {
 	size_t offset = start - page_offset(page);
 	struct inode *inode = page->mapping->host;
@@ -3055,7 +3044,6 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 
 		if (ret == -ENOENT && root == fs_info->tree_root) {
 			struct btrfs_root *dead_root;
-			struct btrfs_fs_info *fs_info = root->fs_info;
 			int is_dead_root = 0;
 
 			/*
@@ -3395,7 +3383,6 @@ cache_acl:
 	switch (inode->i_mode & S_IFMT) {
 	case S_IFREG:
 		inode->i_mapping->a_ops = &btrfs_aops;
-		BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
 		inode->i_fop = &btrfs_file_operations;
 		inode->i_op = &btrfs_file_inode_operations;
 		break;
@@ -4051,7 +4038,7 @@ out_end_trans:
 		err = ret;
 	inode->i_flags |= S_DEAD;
 out_release:
-	btrfs_subvolume_release_metadata(fs_info, &block_rsv);
+	btrfs_subvolume_release_metadata(root, &block_rsv);
 out_up_write:
 	up_write(&fs_info->subvol_sem);
 	if (err) {
@@ -4583,7 +4570,7 @@ again:
 				     &cached_state);
 		unlock_page(page);
 		put_page(page);
-		btrfs_start_ordered_extent(inode, ordered, 1);
+		btrfs_start_ordered_extent(ordered, 1);
 		btrfs_put_ordered_extent(ordered);
 		goto again;
 	}
@@ -4848,19 +4835,16 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 
 		/*
 		 * We're truncating a file that used to have good data down to
-		 * zero. Make sure it gets into the ordered flush list so that
-		 * any new writes get down to disk quickly.
+		 * zero. Make sure any new writes to the file get on disk
+		 * on close.
 		 */
 		if (newsize == 0)
-			set_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
+			set_bit(BTRFS_INODE_FLUSH_ON_CLOSE,
 				&BTRFS_I(inode)->runtime_flags);
 
 		truncate_setsize(inode, newsize);
 
-		/* Disable nonlocked read DIO to avoid the endless truncate */
-		btrfs_inode_block_unlocked_dio(BTRFS_I(inode));
 		inode_dio_wait(inode);
-		btrfs_inode_resume_unlocked_dio(BTRFS_I(inode));
 
 		ret = btrfs_truncate(inode, newsize == oldsize);
 		if (ret && inode->i_nlink) {
@@ -5305,15 +5289,15 @@ static void inode_tree_add(struct inode *inode)
 	spin_unlock(&root->inode_lock);
 }
 
-static void inode_tree_del(struct inode *inode)
+static void inode_tree_del(struct btrfs_inode *inode)
 {
-	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_root *root = inode->root;
 	int empty = 0;
 
 	spin_lock(&root->inode_lock);
-	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);
+	if (!RB_EMPTY_NODE(&inode->rb_node)) {
+		rb_erase(&inode->rb_node, &root->inode_tree);
+		RB_CLEAR_NODE(&inode->rb_node);
 		empty = RB_EMPTY_ROOT(&root->inode_tree);
 	}
 	spin_unlock(&root->inode_lock);
@@ -6311,7 +6295,6 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
 	if (err)
 		goto out_unlock;
 
-	BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
 	d_instantiate_new(dentry, inode);
 
 out_unlock:
@@ -6374,7 +6357,6 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 		drop_inode = 1;
 	} else {
 		struct dentry *parent = dentry->d_parent;
-		int ret;
 
 		err = btrfs_update_inode(trans, root, inode);
 		if (err)
@@ -6389,12 +6371,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 				goto fail;
 		}
 		d_instantiate(dentry, inode);
-		ret = btrfs_log_new_name(trans, BTRFS_I(inode), NULL, parent,
-					 true, NULL);
-		if (ret == BTRFS_NEED_TRANS_COMMIT) {
-			err = btrfs_commit_transaction(trans);
-			trans = NULL;
-		}
+		btrfs_log_new_name(trans, BTRFS_I(inode), NULL, parent);
 	}
 
 fail:
@@ -6540,8 +6517,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 				    u64 start, u64 len)
 {
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
-	int ret;
-	int err = 0;
+	int ret = 0;
 	u64 extent_start = 0;
 	u64 extent_end = 0;
 	u64 objectid = btrfs_ino(inode);
@@ -6569,7 +6545,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	}
 	em = alloc_extent_map();
 	if (!em) {
-		err = -ENOMEM;
+		ret = -ENOMEM;
 		goto out;
 	}
 	em->start = EXTENT_MAP_HOLE;
@@ -6579,7 +6555,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		err = -ENOMEM;
+		ret = -ENOMEM;
 		goto out;
 	}
 
@@ -6592,14 +6568,16 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	 */
 	path->leave_spinning = 1;
 
+	path->recurse = btrfs_is_free_space_inode(inode);
+
 	ret = btrfs_lookup_file_extent(NULL, root, path, objectid, start, 0);
 	if (ret < 0) {
-		err = ret;
 		goto out;
 	} else if (ret > 0) {
 		if (path->slots[0] == 0)
 			goto not_found;
 		path->slots[0]--;
+		ret = 0;
 	}
 
 	leaf = path->nodes[0];
@@ -6625,7 +6603,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	    extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
 		/* Only regular file could have regular/prealloc extent */
 		if (!S_ISREG(inode->vfs_inode.i_mode)) {
-			err = -EUCLEAN;
+			ret = -EUCLEAN;
 			btrfs_crit(fs_info,
 		"regular/prealloc extent found for non-regular inode %llu",
 				   btrfs_ino(inode));
@@ -6643,12 +6621,11 @@ next:
 		path->slots[0]++;
 		if (path->slots[0] >= btrfs_header_nritems(leaf)) {
 			ret = btrfs_next_leaf(root, path);
-			if (ret < 0) {
-				err = ret;
+			if (ret < 0)
 				goto out;
-			} else if (ret > 0) {
+			else if (ret > 0)
 				goto not_found;
-			}
+
 			leaf = path->nodes[0];
 		}
 		btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
@@ -6699,10 +6676,8 @@ next:
 			    BTRFS_COMPRESS_NONE) {
 				ret = uncompress_inline(path, page, pg_offset,
 							extent_offset, item);
-				if (ret) {
-					err = ret;
+				if (ret)
 					goto out;
-				}
 			} else {
 				map = kmap(page);
 				read_extent_buffer(leaf, map + pg_offset, ptr,
@@ -6726,29 +6701,28 @@ not_found:
 	em->len = len;
 	em->block_start = EXTENT_MAP_HOLE;
 insert:
+	ret = 0;
 	btrfs_release_path(path);
 	if (em->start > start || extent_map_end(em) <= start) {
 		btrfs_err(fs_info,
 			  "bad extent! em: [%llu %llu] passed [%llu %llu]",
 			  em->start, em->len, start, len);
-		err = -EIO;
+		ret = -EIO;
 		goto out;
 	}
 
-	err = 0;
 	write_lock(&em_tree->lock);
-	err = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len);
+	ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len);
 	write_unlock(&em_tree->lock);
 out:
 	btrfs_free_path(path);
 
 	trace_btrfs_get_extent(root, inode, em);
 
-	if (err) {
+	if (ret) {
 		free_extent_map(em);
-		return ERR_PTR(err);
+		return ERR_PTR(ret);
 	}
-	BUG_ON(!em); /* Error is always set */
 	return em;
 }
 
@@ -7111,7 +7085,7 @@ out:
 }
 
 static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
-			      struct extent_state **cached_state, int writing)
+			      struct extent_state **cached_state, bool writing)
 {
 	struct btrfs_ordered_extent *ordered;
 	int ret = 0;
@@ -7160,7 +7134,7 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
 			 */
 			if (writing ||
 			    test_bit(BTRFS_ORDERED_DIRECT, &ordered->flags))
-				btrfs_start_ordered_extent(inode, ordered, 1);
+				btrfs_start_ordered_extent(ordered, 1);
 			else
 				ret = -ENOTBLK;
 			btrfs_put_ordered_extent(ordered);
@@ -7249,30 +7223,7 @@ static struct extent_map *create_io_em(struct btrfs_inode *inode, u64 start,
 }
 
 
-static int btrfs_get_blocks_direct_read(struct extent_map *em,
-					struct buffer_head *bh_result,
-					struct inode *inode,
-					u64 start, u64 len)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-
-	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 = fs_info->fs_devices->latest_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)
@@ -7333,7 +7284,6 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 	}
 
 	/* this will cow the extent */
-	len = bh_result->b_size;
 	free_extent_map(em);
 	*map = em = btrfs_new_extent_direct(BTRFS_I(inode), start, len);
 	if (IS_ERR(em)) {
@@ -7344,64 +7294,88 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 	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 = fs_info->fs_devices->latest_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))
+	if (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)
+static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start,
+		loff_t length, unsigned int flags, struct iomap *iomap,
+		struct iomap *srcmap)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_map *em;
 	struct extent_state *cached_state = NULL;
 	struct btrfs_dio_data *dio_data = NULL;
-	u64 start = iblock << inode->i_blkbits;
 	u64 lockstart, lockend;
-	u64 len = bh_result->b_size;
+	const bool write = !!(flags & IOMAP_WRITE);
 	int ret = 0;
+	u64 len = length;
+	bool unlock_extents = false;
+	bool sync = (current->journal_info == BTRFS_DIO_SYNC_STUB);
+
+	/*
+	 * We used current->journal_info here to see if we were sync, but
+	 * there's a lot of tests in the enospc machinery to not do flushing if
+	 * we have a journal_info set, so we need to clear this out and re-set
+	 * it in iomap_end.
+	 */
+	ASSERT(current->journal_info == NULL ||
+	       current->journal_info == BTRFS_DIO_SYNC_STUB);
+	current->journal_info = NULL;
 
-	if (!create)
+	if (!write)
 		len = min_t(u64, len, fs_info->sectorsize);
 
 	lockstart = start;
 	lockend = start + len - 1;
 
-	if (current->journal_info) {
-		/*
-		 * Need to pull our outstanding extents and set journal_info to NULL so
-		 * that anything that needs to check if there's a transaction doesn't get
-		 * confused.
-		 */
-		dio_data = current->journal_info;
-		current->journal_info = NULL;
+	/*
+	 * The generic stuff only does filemap_write_and_wait_range, which
+	 * isn't enough if we've written compressed pages to this area, so we
+	 * need to flush the dirty pages again to make absolutely sure that any
+	 * outstanding dirty pages are on disk.
+	 */
+	if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
+		     &BTRFS_I(inode)->runtime_flags)) {
+		ret = filemap_fdatawrite_range(inode->i_mapping, start,
+					       start + length - 1);
+		if (ret)
+			return ret;
+	}
+
+	dio_data = kzalloc(sizeof(*dio_data), GFP_NOFS);
+	if (!dio_data)
+		return -ENOMEM;
+
+	dio_data->sync = sync;
+	dio_data->length = length;
+	if (write) {
+		dio_data->reserve = round_up(length, fs_info->sectorsize);
+		ret = btrfs_delalloc_reserve_space(BTRFS_I(inode),
+				&dio_data->data_reserved,
+				start, dio_data->reserve);
+		if (ret) {
+			extent_changeset_free(dio_data->data_reserved);
+			kfree(dio_data);
+			return ret;
+		}
 	}
+	iomap->private = dio_data;
+
 
 	/*
 	 * If this errors out it's because we couldn't invalidate pagecache for
 	 * this range and we need to fallback to buffered.
 	 */
-	if (lock_extent_direct(inode, lockstart, lockend, &cached_state,
-			       create)) {
+	if (lock_extent_direct(inode, lockstart, lockend, &cached_state, write)) {
 		ret = -ENOTBLK;
 		goto err;
 	}
@@ -7433,35 +7407,47 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
 		goto unlock_err;
 	}
 
-	if (create) {
-		ret = btrfs_get_blocks_direct_write(&em, bh_result, inode,
-						    dio_data, start, len);
+	len = min(len, em->len - (start - em->start));
+	if (write) {
+		ret = btrfs_get_blocks_direct_write(&em, inode, dio_data,
+						    start, len);
 		if (ret < 0)
 			goto unlock_err;
-
-		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart,
-				     lockend, &cached_state);
+		unlock_extents = true;
+		/* Recalc len in case the new em is smaller than requested */
+		len = min(len, em->len - (start - em->start));
 	} 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;
-		}
 		/*
 		 * We need to unlock only the end area that we aren't using.
 		 * The rest is going to be unlocked by the endio routine.
 		 */
-		lockstart = start + bh_result->b_size;
-		if (lockstart < lockend) {
-			unlock_extent_cached(&BTRFS_I(inode)->io_tree,
-					     lockstart, lockend, &cached_state);
-		} else {
-			free_extent_state(cached_state);
-		}
+		lockstart = start + len;
+		if (lockstart < lockend)
+			unlock_extents = true;
+	}
+
+	if (unlock_extents)
+		unlock_extent_cached(&BTRFS_I(inode)->io_tree,
+				     lockstart, lockend, &cached_state);
+	else
+		free_extent_state(cached_state);
+
+	/*
+	 * Translate extent map information to iomap.
+	 * We trim the extents (and move the addr) even though iomap code does
+	 * that, since we have locked only the parts we are performing I/O in.
+	 */
+	if ((em->block_start == EXTENT_MAP_HOLE) ||
+	    (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) && !write)) {
+		iomap->addr = IOMAP_NULL_ADDR;
+		iomap->type = IOMAP_HOLE;
+	} else {
+		iomap->addr = em->block_start + (start - em->start);
+		iomap->type = IOMAP_MAPPED;
 	}
+	iomap->offset = start;
+	iomap->bdev = fs_info->fs_devices->latest_bdev;
+	iomap->length = len;
 
 	free_extent_map(em);
 
@@ -7471,8 +7457,63 @@ unlock_err:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 			     &cached_state);
 err:
-	if (dio_data)
-		current->journal_info = dio_data;
+	if (dio_data) {
+		btrfs_delalloc_release_space(BTRFS_I(inode),
+				dio_data->data_reserved, start,
+				dio_data->reserve, true);
+		btrfs_delalloc_release_extents(BTRFS_I(inode), dio_data->reserve);
+		extent_changeset_free(dio_data->data_reserved);
+		kfree(dio_data);
+	}
+	return ret;
+}
+
+static int btrfs_dio_iomap_end(struct inode *inode, loff_t pos, loff_t length,
+		ssize_t written, unsigned int flags, struct iomap *iomap)
+{
+	int ret = 0;
+	struct btrfs_dio_data *dio_data = iomap->private;
+	size_t submitted = dio_data->submitted;
+	const bool write = !!(flags & IOMAP_WRITE);
+
+	if (!write && (iomap->type == IOMAP_HOLE)) {
+		/* If reading from a hole, unlock and return */
+		unlock_extent(&BTRFS_I(inode)->io_tree, pos, pos + length - 1);
+		goto out;
+	}
+
+	if (submitted < length) {
+		pos += submitted;
+		length -= submitted;
+		if (write)
+			__endio_write_update_ordered(BTRFS_I(inode), pos,
+					length, false);
+		else
+			unlock_extent(&BTRFS_I(inode)->io_tree, pos,
+				      pos + length - 1);
+		ret = -ENOTBLK;
+	}
+
+	if (write) {
+		if (dio_data->reserve)
+			btrfs_delalloc_release_space(BTRFS_I(inode),
+					dio_data->data_reserved, pos,
+					dio_data->reserve, true);
+		btrfs_delalloc_release_extents(BTRFS_I(inode), dio_data->length);
+		extent_changeset_free(dio_data->data_reserved);
+	}
+out:
+	/*
+	 * We're all done, we can re-set the current->journal_info now safely
+	 * for our endio.
+	 */
+	if (dio_data->sync) {
+		ASSERT(current->journal_info == NULL);
+		current->journal_info = BTRFS_DIO_SYNC_STUB;
+	}
+	kfree(dio_data);
+	iomap->private = NULL;
+
 	return ret;
 }
 
@@ -7496,7 +7537,7 @@ static void btrfs_dio_private_put(struct btrfs_dio_private *dip)
 			      dip->logical_offset + dip->bytes - 1);
 	}
 
-	dio_end_io(dip->dio_bio);
+	bio_endio(dip->dio_bio);
 	kfree(dip);
 }
 
@@ -7730,24 +7771,11 @@ static struct btrfs_dio_private *btrfs_create_dio_private(struct bio *dio_bio,
 	dip->disk_bytenr = (u64)dio_bio->bi_iter.bi_sector << 9;
 	dip->dio_bio = dio_bio;
 	refcount_set(&dip->refs, 1);
-
-	if (write) {
-		struct btrfs_dio_data *dio_data = current->journal_info;
-
-		/*
-		 * Setting range start and end to the same value means that
-		 * no cleanup will happen in btrfs_direct_IO
-		 */
-		dio_data->unsubmitted_oe_range_end = dip->logical_offset +
-			dip->bytes;
-		dio_data->unsubmitted_oe_range_start =
-			dio_data->unsubmitted_oe_range_end;
-	}
 	return dip;
 }
 
-static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
-				loff_t file_offset)
+static blk_qc_t btrfs_submit_direct(struct inode *inode, struct iomap *iomap,
+		struct bio *dio_bio, loff_t file_offset)
 {
 	const bool write = (bio_op(dio_bio) == REQ_OP_WRITE);
 	const bool csum = !(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM);
@@ -7764,6 +7792,7 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
 	int ret;
 	blk_status_t status;
 	struct btrfs_io_geometry geom;
+	struct btrfs_dio_data *dio_data = iomap->private;
 
 	dip = btrfs_create_dio_private(dio_bio, inode, file_offset);
 	if (!dip) {
@@ -7772,8 +7801,8 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
 				file_offset + dio_bio->bi_iter.bi_size - 1);
 		}
 		dio_bio->bi_status = BLK_STS_RESOURCE;
-		dio_end_io(dio_bio);
-		return;
+		bio_endio(dio_bio);
+		return BLK_QC_T_NONE;
 	}
 
 	if (!write && csum) {
@@ -7844,15 +7873,17 @@ static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
 			goto out_err;
 		}
 
+		dio_data->submitted += clone_len;
 		clone_offset += clone_len;
 		start_sector += clone_len >> 9;
 		file_offset += clone_len;
 	} while (submit_len > 0);
-	return;
+	return BLK_QC_T_NONE;
 
 out_err:
 	dip->dio_bio->bi_status = status;
 	btrfs_dio_private_put(dip);
+	return BLK_QC_T_NONE;
 }
 
 static ssize_t check_direct_IO(struct btrfs_fs_info *fs_info,
@@ -7888,37 +7919,59 @@ out:
 	return retval;
 }
 
-static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
+static inline int btrfs_maybe_fsync_end_io(struct kiocb *iocb, ssize_t size,
+					   int error, unsigned flags)
+{
+	/*
+	 * Now if we're still in the context of our submitter we know we can't
+	 * safely run generic_write_sync(), so clear our flag here so that the
+	 * caller knows to follow up with a sync.
+	 */
+	if (current->journal_info == BTRFS_DIO_SYNC_STUB) {
+		current->journal_info = NULL;
+		return error;
+	}
+
+	if (error)
+		return error;
+
+	if (size) {
+		iocb->ki_flags |= IOCB_DSYNC;
+		return generic_write_sync(iocb, size);
+	}
+
+	return 0;
+}
+
+static const struct iomap_ops btrfs_dio_iomap_ops = {
+	.iomap_begin            = btrfs_dio_iomap_begin,
+	.iomap_end              = btrfs_dio_iomap_end,
+};
+
+static const struct iomap_dio_ops btrfs_dio_ops = {
+	.submit_io		= btrfs_submit_direct,
+};
+
+static const struct iomap_dio_ops btrfs_sync_dops = {
+	.submit_io		= btrfs_submit_direct,
+	.end_io			= btrfs_maybe_fsync_end_io,
+};
+
+ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file->f_mapping->host;
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct btrfs_dio_data dio_data = { 0 };
 	struct extent_changeset *data_reserved = NULL;
 	loff_t offset = iocb->ki_pos;
 	size_t count = 0;
-	int flags = 0;
-	bool wakeup = true;
 	bool relock = false;
 	ssize_t ret;
 
 	if (check_direct_IO(fs_info, iter, offset))
 		return 0;
 
-	inode_dio_begin(inode);
-
-	/*
-	 * The generic stuff only does filemap_write_and_wait_range, which
-	 * isn't enough if we've written compressed pages to this area, so
-	 * we need to flush the dirty pages again to make absolutely sure
-	 * that any outstanding dirty pages are on disk.
-	 */
 	count = iov_iter_count(iter);
-	if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
-		     &BTRFS_I(inode)->runtime_flags))
-		filemap_fdatawrite_range(inode->i_mapping, offset,
-					 offset + count - 1);
-
 	if (iov_iter_rw(iter) == WRITE) {
 		/*
 		 * If the write DIO is beyond the EOF, we need update
@@ -7926,66 +7979,29 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 		 * not unlock the i_mutex at this case.
 		 */
 		if (offset + count <= inode->i_size) {
-			dio_data.overwrite = 1;
 			inode_unlock(inode);
 			relock = true;
 		}
-		ret = btrfs_delalloc_reserve_space(BTRFS_I(inode), &data_reserved,
-						   offset, count);
-		if (ret)
-			goto out;
-
-		/*
-		 * We need to know how many extents we reserved so that we can
-		 * do the accounting properly if we go over the number we
-		 * originally calculated.  Abuse current->journal_info for this.
-		 */
-		dio_data.reserve = round_up(count,
-					    fs_info->sectorsize);
-		dio_data.unsubmitted_oe_range_start = (u64)offset;
-		dio_data.unsubmitted_oe_range_end = (u64)offset;
-		current->journal_info = &dio_data;
 		down_read(&BTRFS_I(inode)->dio_sem);
-	} else if (test_bit(BTRFS_INODE_READDIO_NEED_LOCK,
-				     &BTRFS_I(inode)->runtime_flags)) {
-		inode_dio_end(inode);
-		flags = DIO_LOCKING | DIO_SKIP_HOLES;
-		wakeup = false;
 	}
 
-	ret = __blockdev_direct_IO(iocb, inode,
-				   fs_info->fs_devices->latest_bdev,
-				   iter, btrfs_get_blocks_direct, NULL,
-				   btrfs_submit_direct, flags);
-	if (iov_iter_rw(iter) == WRITE) {
+	/*
+	 * We have are actually a sync iocb, so we need our fancy endio to know
+	 * if we need to sync.
+	 */
+	if (current->journal_info)
+		ret = iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops,
+				   &btrfs_sync_dops, is_sync_kiocb(iocb));
+	else
+		ret = iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops,
+				   &btrfs_dio_ops, is_sync_kiocb(iocb));
+
+	if (ret == -ENOTBLK)
+		ret = 0;
+
+	if (iov_iter_rw(iter) == WRITE)
 		up_read(&BTRFS_I(inode)->dio_sem);
-		current->journal_info = NULL;
-		if (ret < 0 && ret != -EIOCBQUEUED) {
-			if (dio_data.reserve)
-				btrfs_delalloc_release_space(BTRFS_I(inode),
-					data_reserved, offset, dio_data.reserve,
-					true);
-			/*
-			 * On error we might have left some ordered extents
-			 * without submitting corresponding bios for them, so
-			 * cleanup them up to avoid other tasks getting them
-			 * and waiting for them to complete forever.
-			 */
-			if (dio_data.unsubmitted_oe_range_start <
-			    dio_data.unsubmitted_oe_range_end)
-				__endio_write_update_ordered(BTRFS_I(inode),
-					dio_data.unsubmitted_oe_range_start,
-					dio_data.unsubmitted_oe_range_end -
-					dio_data.unsubmitted_oe_range_start,
-					false);
-		} else if (ret >= 0 && (size_t)ret < count)
-			btrfs_delalloc_release_space(BTRFS_I(inode), data_reserved,
-					offset, count - (size_t)ret, true);
-		btrfs_delalloc_release_extents(BTRFS_I(inode), count);
-	}
-out:
-	if (wakeup)
-		inode_dio_end(inode);
+
 	if (relock)
 		inode_lock(inode);
 
@@ -8002,12 +8018,24 @@ static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	if (ret)
 		return ret;
 
-	return extent_fiemap(inode, fieinfo, start, len);
+	return extent_fiemap(BTRFS_I(inode), fieinfo, start, len);
 }
 
 int btrfs_readpage(struct file *file, struct page *page)
 {
-	return extent_read_full_page(page, btrfs_get_extent, 0);
+	struct btrfs_inode *inode = BTRFS_I(page->mapping->host);
+	u64 start = page_offset(page);
+	u64 end = start + PAGE_SIZE - 1;
+	unsigned long bio_flags = 0;
+	struct bio *bio = NULL;
+	int ret;
+
+	btrfs_lock_and_flush_ordered_range(inode, start, end, NULL);
+
+	ret = btrfs_do_readpage(page, NULL, &bio, &bio_flags, 0, NULL);
+	if (bio)
+		ret = submit_one_bio(bio, 0, bio_flags);
+	return ret;
 }
 
 static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
@@ -8091,15 +8119,15 @@ static int btrfs_migratepage(struct address_space *mapping,
 static void btrfs_invalidatepage(struct page *page, unsigned int offset,
 				 unsigned int length)
 {
-	struct inode *inode = page->mapping->host;
-	struct extent_io_tree *tree;
+	struct btrfs_inode *inode = BTRFS_I(page->mapping->host);
+	struct extent_io_tree *tree = &inode->io_tree;
 	struct btrfs_ordered_extent *ordered;
 	struct extent_state *cached_state = NULL;
 	u64 page_start = page_offset(page);
 	u64 page_end = page_start + PAGE_SIZE - 1;
 	u64 start;
 	u64 end;
-	int inode_evicting = inode->i_state & I_FREEING;
+	int inode_evicting = inode->vfs_inode.i_state & I_FREEING;
 
 	/*
 	 * we have the page locked, so new writeback can't start,
@@ -8110,7 +8138,6 @@ static void btrfs_invalidatepage(struct page *page, unsigned int offset,
 	 */
 	wait_on_page_writeback(page);
 
-	tree = &BTRFS_I(inode)->io_tree;
 	if (offset) {
 		btrfs_releasepage(page, GFP_NOFS);
 		return;
@@ -8120,8 +8147,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned int offset,
 		lock_extent_bits(tree, page_start, page_end, &cached_state);
 again:
 	start = page_start;
-	ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start,
-					page_end - start + 1);
+	ordered = btrfs_lookup_ordered_range(inode, start, page_end - start + 1);
 	if (ordered) {
 		end = min(page_end,
 			  ordered->file_offset + ordered->num_bytes - 1);
@@ -8142,7 +8168,7 @@ again:
 			struct btrfs_ordered_inode_tree *tree;
 			u64 new_len;
 
-			tree = &BTRFS_I(inode)->ordered_tree;
+			tree = &inode->ordered_tree;
 
 			spin_lock_irq(&tree->lock);
 			set_bit(BTRFS_ORDERED_TRUNCATED, &ordered->flags);
@@ -8181,7 +8207,7 @@ again:
 	 *    bit of its io_tree, and free the qgroup reserved data space.
 	 *    Since the IO will never happen for this page.
 	 */
-	btrfs_qgroup_free_data(BTRFS_I(inode), NULL, page_start, PAGE_SIZE);
+	btrfs_qgroup_free_data(inode, NULL, page_start, PAGE_SIZE);
 	if (!inode_evicting) {
 		clear_extent_bit(tree, page_start, page_end, EXTENT_LOCKED |
 				 EXTENT_DELALLOC | EXTENT_DELALLOC_NEW |
@@ -8283,7 +8309,7 @@ again:
 		unlock_extent_cached(io_tree, page_start, page_end,
 				     &cached_state);
 		unlock_page(page);
-		btrfs_start_ordered_extent(inode, ordered, 1);
+		btrfs_start_ordered_extent(ordered, 1);
 		btrfs_put_ordered_extent(ordered);
 		goto again;
 	}
@@ -8614,21 +8640,21 @@ void btrfs_free_inode(struct inode *inode)
 	kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode));
 }
 
-void btrfs_destroy_inode(struct inode *inode)
+void btrfs_destroy_inode(struct inode *vfs_inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ordered_extent *ordered;
-	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_inode *inode = BTRFS_I(vfs_inode);
+	struct btrfs_root *root = inode->root;
 
-	WARN_ON(!hlist_empty(&inode->i_dentry));
-	WARN_ON(inode->i_data.nrpages);
-	WARN_ON(BTRFS_I(inode)->block_rsv.reserved);
-	WARN_ON(BTRFS_I(inode)->block_rsv.size);
-	WARN_ON(BTRFS_I(inode)->outstanding_extents);
-	WARN_ON(BTRFS_I(inode)->delalloc_bytes);
-	WARN_ON(BTRFS_I(inode)->new_delalloc_bytes);
-	WARN_ON(BTRFS_I(inode)->csum_bytes);
-	WARN_ON(BTRFS_I(inode)->defrag_bytes);
+	WARN_ON(!hlist_empty(&vfs_inode->i_dentry));
+	WARN_ON(vfs_inode->i_data.nrpages);
+	WARN_ON(inode->block_rsv.reserved);
+	WARN_ON(inode->block_rsv.size);
+	WARN_ON(inode->outstanding_extents);
+	WARN_ON(inode->delalloc_bytes);
+	WARN_ON(inode->new_delalloc_bytes);
+	WARN_ON(inode->csum_bytes);
+	WARN_ON(inode->defrag_bytes);
 
 	/*
 	 * This can happen where we create an inode, but somebody else also
@@ -8643,7 +8669,7 @@ void btrfs_destroy_inode(struct inode *inode)
 		if (!ordered)
 			break;
 		else {
-			btrfs_err(fs_info,
+			btrfs_err(root->fs_info,
 				  "found ordered extent %llu %llu on inode cleanup",
 				  ordered->file_offset, ordered->num_bytes);
 			btrfs_remove_ordered_extent(inode, ordered);
@@ -8651,11 +8677,11 @@ void btrfs_destroy_inode(struct inode *inode)
 			btrfs_put_ordered_extent(ordered);
 		}
 	}
-	btrfs_qgroup_check_reserved_leak(BTRFS_I(inode));
+	btrfs_qgroup_check_reserved_leak(inode);
 	inode_tree_del(inode);
-	btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0);
-	btrfs_inode_clear_file_extent_range(BTRFS_I(inode), 0, (u64)-1);
-	btrfs_put_root(BTRFS_I(inode)->root);
+	btrfs_drop_extent_cache(inode, 0, (u64)-1, 0);
+	btrfs_inode_clear_file_extent_range(inode, 0, (u64)-1);
+	btrfs_put_root(inode->root);
 }
 
 int btrfs_drop_inode(struct inode *inode)
@@ -8780,27 +8806,19 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 	struct inode *new_inode = new_dentry->d_inode;
 	struct inode *old_inode = old_dentry->d_inode;
 	struct timespec64 ctime = current_time(old_inode);
-	struct dentry *parent;
 	u64 old_ino = btrfs_ino(BTRFS_I(old_inode));
 	u64 new_ino = btrfs_ino(BTRFS_I(new_inode));
 	u64 old_idx = 0;
 	u64 new_idx = 0;
 	int ret;
+	int ret2;
 	bool root_log_pinned = false;
 	bool dest_log_pinned = false;
-	struct btrfs_log_ctx ctx_root;
-	struct btrfs_log_ctx ctx_dest;
-	bool sync_log_root = false;
-	bool sync_log_dest = false;
-	bool commit_transaction = false;
 
 	/* we only allow rename subvolume link between subvolumes */
 	if (old_ino != BTRFS_FIRST_FREE_OBJECTID && root != dest)
 		return -EXDEV;
 
-	btrfs_init_log_ctx(&ctx_root, old_inode);
-	btrfs_init_log_ctx(&ctx_dest, new_inode);
-
 	/* close the race window with snapshot create/destroy ioctl */
 	if (old_ino == BTRFS_FIRST_FREE_OBJECTID ||
 	    new_ino == BTRFS_FIRST_FREE_OBJECTID)
@@ -8942,30 +8960,14 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 		BTRFS_I(new_inode)->dir_index = new_idx;
 
 	if (root_log_pinned) {
-		parent = new_dentry->d_parent;
-		ret = btrfs_log_new_name(trans, BTRFS_I(old_inode),
-					 BTRFS_I(old_dir), parent,
-					 false, &ctx_root);
-		if (ret == BTRFS_NEED_LOG_SYNC)
-			sync_log_root = true;
-		else if (ret == BTRFS_NEED_TRANS_COMMIT)
-			commit_transaction = true;
-		ret = 0;
+		btrfs_log_new_name(trans, BTRFS_I(old_inode), BTRFS_I(old_dir),
+				   new_dentry->d_parent);
 		btrfs_end_log_trans(root);
 		root_log_pinned = false;
 	}
 	if (dest_log_pinned) {
-		if (!commit_transaction) {
-			parent = old_dentry->d_parent;
-			ret = btrfs_log_new_name(trans, BTRFS_I(new_inode),
-						 BTRFS_I(new_dir), parent,
-						 false, &ctx_dest);
-			if (ret == BTRFS_NEED_LOG_SYNC)
-				sync_log_dest = true;
-			else if (ret == BTRFS_NEED_TRANS_COMMIT)
-				commit_transaction = true;
-			ret = 0;
-		}
+		btrfs_log_new_name(trans, BTRFS_I(new_inode), BTRFS_I(new_dir),
+				   old_dentry->d_parent);
 		btrfs_end_log_trans(dest);
 		dest_log_pinned = false;
 	}
@@ -8998,46 +9000,13 @@ out_fail:
 			dest_log_pinned = false;
 		}
 	}
-	if (!ret && sync_log_root && !commit_transaction) {
-		ret = btrfs_sync_log(trans, BTRFS_I(old_inode)->root,
-				     &ctx_root);
-		if (ret)
-			commit_transaction = true;
-	}
-	if (!ret && sync_log_dest && !commit_transaction) {
-		ret = btrfs_sync_log(trans, BTRFS_I(new_inode)->root,
-				     &ctx_dest);
-		if (ret)
-			commit_transaction = true;
-	}
-	if (commit_transaction) {
-		/*
-		 * We may have set commit_transaction when logging the new name
-		 * in the destination root, in which case we left the source
-		 * root context in the list of log contextes. So make sure we
-		 * remove it to avoid invalid memory accesses, since the context
-		 * was allocated in our stack frame.
-		 */
-		if (sync_log_root) {
-			mutex_lock(&root->log_mutex);
-			list_del_init(&ctx_root.list);
-			mutex_unlock(&root->log_mutex);
-		}
-		ret = btrfs_commit_transaction(trans);
-	} else {
-		int ret2;
-
-		ret2 = btrfs_end_transaction(trans);
-		ret = ret ? ret : ret2;
-	}
+	ret2 = btrfs_end_transaction(trans);
+	ret = ret ? ret : ret2;
 out_notrans:
 	if (new_ino == BTRFS_FIRST_FREE_OBJECTID ||
 	    old_ino == BTRFS_FIRST_FREE_OBJECTID)
 		up_read(&fs_info->subvol_sem);
 
-	ASSERT(list_empty(&ctx_root.list));
-	ASSERT(list_empty(&ctx_dest.list));
-
 	return ret;
 }
 
@@ -9105,11 +9074,9 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct inode *old_inode = d_inode(old_dentry);
 	u64 index = 0;
 	int ret;
+	int ret2;
 	u64 old_ino = btrfs_ino(BTRFS_I(old_inode));
 	bool log_pinned = false;
-	struct btrfs_log_ctx ctx;
-	bool sync_log = false;
-	bool commit_transaction = false;
 
 	if (btrfs_ino(BTRFS_I(new_dir)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)
 		return -EPERM;
@@ -9259,17 +9226,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		BTRFS_I(old_inode)->dir_index = index;
 
 	if (log_pinned) {
-		struct dentry *parent = new_dentry->d_parent;
-
-		btrfs_init_log_ctx(&ctx, old_inode);
-		ret = btrfs_log_new_name(trans, BTRFS_I(old_inode),
-					 BTRFS_I(old_dir), parent,
-					 false, &ctx);
-		if (ret == BTRFS_NEED_LOG_SYNC)
-			sync_log = true;
-		else if (ret == BTRFS_NEED_TRANS_COMMIT)
-			commit_transaction = true;
-		ret = 0;
+		btrfs_log_new_name(trans, BTRFS_I(old_inode), BTRFS_I(old_dir),
+				   new_dentry->d_parent);
 		btrfs_end_log_trans(root);
 		log_pinned = false;
 	}
@@ -9306,23 +9264,8 @@ out_fail:
 		btrfs_end_log_trans(root);
 		log_pinned = false;
 	}
-	if (!ret && sync_log) {
-		ret = btrfs_sync_log(trans, BTRFS_I(old_inode)->root, &ctx);
-		if (ret)
-			commit_transaction = true;
-	} else if (sync_log) {
-		mutex_lock(&root->log_mutex);
-		list_del(&ctx.list);
-		mutex_unlock(&root->log_mutex);
-	}
-	if (commit_transaction) {
-		ret = btrfs_commit_transaction(trans);
-	} else {
-		int ret2;
-
-		ret2 = btrfs_end_transaction(trans);
-		ret = ret ? ret : ret2;
-	}
+	ret2 = btrfs_end_transaction(trans);
+	ret = ret ? ret : ret2;
 out_notrans:
 	if (old_ino == BTRFS_FIRST_FREE_OBJECTID)
 		up_read(&fs_info->subvol_sem);
@@ -9388,7 +9331,7 @@ static struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode
  * 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 nr, bool snapshot)
+static int start_delalloc_inodes(struct btrfs_root *root, u64 *nr, bool snapshot)
 {
 	struct btrfs_inode *binode;
 	struct inode *inode;
@@ -9428,9 +9371,11 @@ static int start_delalloc_inodes(struct btrfs_root *root, int nr, bool snapshot)
 		list_add_tail(&work->list, &works);
 		btrfs_queue_work(root->fs_info->flush_workers,
 				 &work->work);
-		ret++;
-		if (nr != -1 && ret >= nr)
-			goto out;
+		if (*nr != U64_MAX) {
+			(*nr)--;
+			if (*nr == 0)
+				goto out;
+		}
 		cond_resched();
 		spin_lock(&root->delalloc_lock);
 	}
@@ -9455,18 +9400,15 @@ out:
 int btrfs_start_delalloc_snapshot(struct btrfs_root *root)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
-	int ret;
+	u64 nr = U64_MAX;
 
 	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
 		return -EROFS;
 
-	ret = start_delalloc_inodes(root, -1, true);
-	if (ret > 0)
-		ret = 0;
-	return ret;
+	return start_delalloc_inodes(root, &nr, true);
 }
 
-int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr)
+int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr)
 {
 	struct btrfs_root *root;
 	struct list_head splice;
@@ -9489,15 +9431,10 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr)
 			       &fs_info->delalloc_roots);
 		spin_unlock(&fs_info->delalloc_root_lock);
 
-		ret = start_delalloc_inodes(root, nr, false);
+		ret = start_delalloc_inodes(root, &nr, false);
 		btrfs_put_root(root);
 		if (ret < 0)
 			goto out;
-
-		if (nr != -1) {
-			nr -= ret;
-			WARN_ON(nr < 0);
-		}
 		spin_lock(&fs_info->delalloc_root_lock);
 	}
 	spin_unlock(&fs_info->delalloc_root_lock);
@@ -9568,7 +9505,6 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	inode->i_fop = &btrfs_file_operations;
 	inode->i_op = &btrfs_file_inode_operations;
 	inode->i_mapping->a_ops = &btrfs_aops;
-	BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
 
 	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
 	if (err)
@@ -9633,11 +9569,15 @@ out_unlock:
 	return err;
 }
 
-static int insert_prealloc_file_extent(struct btrfs_trans_handle *trans,
+static struct btrfs_trans_handle *insert_prealloc_file_extent(
+				       struct btrfs_trans_handle *trans_in,
 				       struct inode *inode, struct btrfs_key *ins,
 				       u64 file_offset)
 {
 	struct btrfs_file_extent_item stack_fi;
+	struct btrfs_replace_extent_info extent_info;
+	struct btrfs_trans_handle *trans = trans_in;
+	struct btrfs_path *path;
 	u64 start = ins->objectid;
 	u64 len = ins->offset;
 	int ret;
@@ -9654,10 +9594,40 @@ static int insert_prealloc_file_extent(struct btrfs_trans_handle *trans,
 
 	ret = btrfs_qgroup_release_data(BTRFS_I(inode), file_offset, len);
 	if (ret < 0)
-		return ret;
-	return insert_reserved_file_extent(trans, BTRFS_I(inode), file_offset,
-					   &stack_fi, ret);
+		return ERR_PTR(ret);
+
+	if (trans) {
+		ret = insert_reserved_file_extent(trans, BTRFS_I(inode),
+						  file_offset, &stack_fi, ret);
+		if (ret)
+			return ERR_PTR(ret);
+		return trans;
+	}
+
+	extent_info.disk_offset = start;
+	extent_info.disk_len = len;
+	extent_info.data_offset = 0;
+	extent_info.data_len = len;
+	extent_info.file_offset = file_offset;
+	extent_info.extent_buf = (char *)&stack_fi;
+	extent_info.is_new_extent = true;
+	extent_info.qgroup_reserved = ret;
+	extent_info.insertions = 0;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return ERR_PTR(-ENOMEM);
+
+	ret = btrfs_replace_file_extents(inode, path, file_offset,
+				     file_offset + len - 1, &extent_info,
+				     &trans);
+	btrfs_free_path(path);
+	if (ret)
+		return ERR_PTR(ret);
+
+	return trans;
 }
+
 static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 				       u64 start, u64 num_bytes, u64 min_size,
 				       loff_t actual_len, u64 *alloc_hint,
@@ -9680,14 +9650,6 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 	if (trans)
 		own_trans = false;
 	while (num_bytes > 0) {
-		if (own_trans) {
-			trans = btrfs_start_transaction(root, 3);
-			if (IS_ERR(trans)) {
-				ret = PTR_ERR(trans);
-				break;
-			}
-		}
-
 		cur_bytes = min_t(u64, num_bytes, SZ_256M);
 		cur_bytes = max(cur_bytes, min_size);
 		/*
@@ -9699,11 +9661,8 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 		cur_bytes = min(cur_bytes, last_alloc);
 		ret = btrfs_reserve_extent(root, cur_bytes, cur_bytes,
 				min_size, 0, *alloc_hint, &ins, 1, 0);
-		if (ret) {
-			if (own_trans)
-				btrfs_end_transaction(trans);
+		if (ret)
 			break;
-		}
 
 		/*
 		 * We've reserved this space, and thus converted it from
@@ -9716,13 +9675,11 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 		btrfs_dec_block_group_reservations(fs_info, ins.objectid);
 
 		last_alloc = ins.offset;
-		ret = insert_prealloc_file_extent(trans, inode, &ins, cur_offset);
-		if (ret) {
+		trans = insert_prealloc_file_extent(trans, inode, &ins, cur_offset);
+		if (IS_ERR(trans)) {
+			ret = PTR_ERR(trans);
 			btrfs_free_reserved_extent(fs_info, ins.objectid,
 						   ins.offset, 0);
-			btrfs_abort_transaction(trans, ret);
-			if (own_trans)
-				btrfs_end_transaction(trans);
 			break;
 		}
 
@@ -9785,8 +9742,10 @@ next:
 			break;
 		}
 
-		if (own_trans)
+		if (own_trans) {
 			btrfs_end_transaction(trans);
+			trans = NULL;
+		}
 	}
 	if (clear_offset < end)
 		btrfs_free_reserved_data_space(BTRFS_I(inode), NULL, clear_offset,
@@ -9865,7 +9824,6 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
 	inode->i_op = &btrfs_file_inode_operations;
 
 	inode->i_mapping->a_ops = &btrfs_aops;
-	BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
 
 	ret = btrfs_init_inode_security(trans, inode, dir, NULL);
 	if (ret)
@@ -10072,14 +10030,14 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 
 	/*
 	 * Balance or device remove/replace/resize can move stuff around from
-	 * under us. The EXCL_OP flag makes sure they aren't running/won't run
-	 * concurrently while we are mapping the swap extents, and
-	 * fs_info->swapfile_pins prevents them from running while the swap file
-	 * is active and moving the extents. Note that this also prevents a
-	 * concurrent device add which isn't actually necessary, but it's not
+	 * under us. The exclop protection makes sure they aren't running/won't
+	 * run concurrently while we are mapping the swap extents, and
+	 * fs_info->swapfile_pins prevents them from running while the swap
+	 * file is active and moving the extents. Note that this also prevents
+	 * a concurrent device add which isn't actually necessary, but it's not
 	 * really worth the trouble to allow it.
 	 */
-	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
+	if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_SWAP_ACTIVATE)) {
 		btrfs_warn(fs_info,
 	   "cannot activate swapfile while exclusive operation is running");
 		return -EBUSY;
@@ -10225,7 +10183,7 @@ out:
 	if (ret)
 		btrfs_swap_deactivate(file);
 
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+	btrfs_exclop_finish(fs_info);
 
 	if (ret)
 		return ret;
@@ -10283,12 +10241,6 @@ static const struct file_operations btrfs_dir_file_operations = {
 	.fsync		= btrfs_sync_file,
 };
 
-static const struct extent_io_ops btrfs_extent_io_ops = {
-	/* mandatory callbacks */
-	.submit_bio_hook = btrfs_submit_bio_hook,
-	.readpage_end_io_hook = btrfs_readpage_end_io_hook,
-};
-
 /*
  * btrfs doesn't support the bmap operation because swapfiles
  * use bmap to make a mapping of extents in the file.  They assume
@@ -10306,7 +10258,7 @@ static const struct address_space_operations btrfs_aops = {
 	.writepage	= btrfs_writepage,
 	.writepages	= btrfs_writepages,
 	.readahead	= btrfs_readahead,
-	.direct_IO	= btrfs_direct_IO,
+	.direct_IO	= noop_direct_IO,
 	.invalidatepage = btrfs_invalidatepage,
 	.releasepage	= btrfs_releasepage,
 #ifdef CONFIG_MIGRATION
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 2d9109d9e98f9d..ab408a23ba32dc 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -378,6 +378,18 @@ static int check_xflags(unsigned int flags)
 	return 0;
 }
 
+bool btrfs_exclop_start(struct btrfs_fs_info *fs_info,
+			enum btrfs_exclusive_operation type)
+{
+	return !cmpxchg(&fs_info->exclusive_operation, BTRFS_EXCLOP_NONE, type);
+}
+
+void btrfs_exclop_finish(struct btrfs_fs_info *fs_info)
+{
+	WRITE_ONCE(fs_info->exclusive_operation, BTRFS_EXCLOP_NONE);
+	sysfs_notify(&fs_info->fs_devices->fsid_kobj, NULL, "exclusive_operation");
+}
+
 /*
  * Set the xflags from the internal inode flags. The remaining items of fsxattr
  * are zeroed.
@@ -618,7 +630,7 @@ static noinline int create_subvol(struct inode *dir,
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
-		btrfs_subvolume_release_metadata(fs_info, &block_rsv);
+		btrfs_subvolume_release_metadata(root, &block_rsv);
 		goto fail_free;
 	}
 	trans->block_rsv = &block_rsv;
@@ -628,7 +640,8 @@ static noinline int create_subvol(struct inode *dir,
 	if (ret)
 		goto fail;
 
-	leaf = btrfs_alloc_tree_block(trans, root, 0, objectid, NULL, 0, 0, 0);
+	leaf = btrfs_alloc_tree_block(trans, root, 0, objectid, NULL, 0, 0, 0,
+				      BTRFS_NESTING_NORMAL);
 	if (IS_ERR(leaf)) {
 		ret = PTR_ERR(leaf);
 		goto fail;
@@ -742,7 +755,7 @@ fail:
 	kfree(root_item);
 	trans->block_rsv = NULL;
 	trans->bytes_reserved = 0;
-	btrfs_subvolume_release_metadata(fs_info, &block_rsv);
+	btrfs_subvolume_release_metadata(root, &block_rsv);
 
 	err = btrfs_commit_transaction(trans);
 	if (err && !ret)
@@ -856,7 +869,7 @@ fail:
 	if (ret && pending_snapshot->snap)
 		pending_snapshot->snap->anon_dev = 0;
 	btrfs_put_root(pending_snapshot->snap);
-	btrfs_subvolume_release_metadata(fs_info, &pending_snapshot->block_rsv);
+	btrfs_subvolume_release_metadata(root, &pending_snapshot->block_rsv);
 free_pending:
 	if (pending_snapshot->anon_dev)
 		free_anon_bdev(pending_snapshot->anon_dev);
@@ -1306,7 +1319,7 @@ again:
 				break;
 
 			unlock_page(page);
-			btrfs_start_ordered_extent(inode, ordered, 1);
+			btrfs_start_ordered_extent(ordered, 1);
 			btrfs_put_ordered_extent(ordered);
 			lock_page(page);
 			/*
@@ -1638,7 +1651,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 	if (ret)
 		return ret;
 
-	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
+	if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_RESIZE)) {
 		mnt_drop_write_file(file);
 		return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
 	}
@@ -1752,7 +1765,7 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 out_free:
 	kfree(vol_args);
 out:
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+	btrfs_exclop_finish(fs_info);
 	mnt_drop_write_file(file);
 	return ret;
 }
@@ -3126,7 +3139,7 @@ static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg)
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags))
+	if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_ADD))
 		return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
 
 	vol_args = memdup_user(arg, sizeof(*vol_args));
@@ -3143,7 +3156,7 @@ static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg)
 
 	kfree(vol_args);
 out:
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+	btrfs_exclop_finish(fs_info);
 	return ret;
 }
 
@@ -3172,7 +3185,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 		goto out;
 	}
 
-	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
+	if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_REMOVE)) {
 		ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
 		goto out;
 	}
@@ -3183,7 +3196,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 		vol_args->name[BTRFS_SUBVOL_NAME_MAX] = '\0';
 		ret = btrfs_rm_device(fs_info, vol_args->name, 0);
 	}
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+	btrfs_exclop_finish(fs_info);
 
 	if (!ret) {
 		if (vol_args->flags & BTRFS_DEVICE_SPEC_BY_ID)
@@ -3214,7 +3227,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	if (ret)
 		return ret;
 
-	if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
+	if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_REMOVE)) {
 		ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
 		goto out_drop_write;
 	}
@@ -3232,7 +3245,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 		btrfs_info(fs_info, "disk deleted %s", vol_args->name);
 	kfree(vol_args);
 out:
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+	btrfs_exclop_finish(fs_info);
 out_drop_write:
 	mnt_drop_write_file(file);
 
@@ -3462,15 +3475,12 @@ static long btrfs_ioctl_space_info(struct btrfs_fs_info *fs_info,
 		struct btrfs_space_info *tmp;
 
 		info = NULL;
-		rcu_read_lock();
-		list_for_each_entry_rcu(tmp, &fs_info->space_info,
-					list) {
+		list_for_each_entry(tmp, &fs_info->space_info, list) {
 			if (tmp->flags == types[i]) {
 				info = tmp;
 				break;
 			}
 		}
-		rcu_read_unlock();
 
 		if (!info)
 			continue;
@@ -3518,15 +3528,12 @@ static long btrfs_ioctl_space_info(struct btrfs_fs_info *fs_info,
 			break;
 
 		info = NULL;
-		rcu_read_lock();
-		list_for_each_entry_rcu(tmp, &fs_info->space_info,
-					list) {
+		list_for_each_entry(tmp, &fs_info->space_info, list) {
 			if (tmp->flags == types[i]) {
 				info = tmp;
 				break;
 			}
 		}
-		rcu_read_unlock();
 
 		if (!info)
 			continue;
@@ -3736,11 +3743,11 @@ static long btrfs_ioctl_dev_replace(struct btrfs_fs_info *fs_info,
 			ret = -EROFS;
 			goto out;
 		}
-		if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
+		if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_REPLACE)) {
 			ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
 		} else {
 			ret = btrfs_dev_replace_by_ioctl(fs_info, p);
-			clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+			btrfs_exclop_finish(fs_info);
 		}
 		break;
 	case BTRFS_IOCTL_DEV_REPLACE_CMD_STATUS:
@@ -3951,7 +3958,7 @@ static long btrfs_ioctl_balance(struct file *file, void __user *arg)
 		return ret;
 
 again:
-	if (!test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) {
+	if (btrfs_exclop_start(fs_info, BTRFS_EXCLOP_BALANCE)) {
 		mutex_lock(&fs_info->balance_mutex);
 		need_unlock = true;
 		goto locked;
@@ -3997,7 +4004,6 @@ again:
 	}
 
 locked:
-	BUG_ON(!test_bit(BTRFS_FS_EXCL_OP, &fs_info->flags));
 
 	if (arg) {
 		bargs = memdup_user(arg, sizeof(*bargs));
@@ -4052,10 +4058,10 @@ locked:
 
 do_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.
+	 * Ownership of bctl and exclusive operation 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;
 
@@ -4074,7 +4080,7 @@ out_bargs:
 out_unlock:
 	mutex_unlock(&fs_info->balance_mutex);
 	if (need_unlock)
-		clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+		btrfs_exclop_finish(fs_info);
 out:
 	mnt_drop_write_file(file);
 	return ret;
@@ -4897,7 +4903,7 @@ long btrfs_ioctl(struct file *file, unsigned int
 	case BTRFS_IOC_SYNC: {
 		int ret;
 
-		ret = btrfs_start_delalloc_roots(fs_info, -1);
+		ret = btrfs_start_delalloc_roots(fs_info, U64_MAX);
 		if (ret)
 			return ret;
 		ret = btrfs_sync_fs(inode->i_sb, 1);
diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c
index f75612e18a82fc..66e02ebdd34089 100644
--- a/fs/btrfs/locking.c
+++ b/fs/btrfs/locking.c
@@ -57,8 +57,8 @@
  * performance reasons.
  *
  *
- * Lock nesting
- * ------------
+ * Lock recursion
+ * --------------
  *
  * A write operation on a tree might indirectly start a look up on the same
  * tree.  This can happen when btrfs_cow_block locks the tree and needs to
@@ -201,7 +201,7 @@ void btrfs_set_lock_blocking_read(struct extent_buffer *eb)
 	 * lock, but it won't change to or away from us.  If we have the write
 	 * lock, we are the owner and it'll never change.
 	 */
-	if (eb->lock_nested && current->pid == eb->lock_owner)
+	if (eb->lock_recursed && current->pid == eb->lock_owner)
 		return;
 	btrfs_assert_tree_read_locked(eb);
 	atomic_inc(&eb->blocking_readers);
@@ -225,7 +225,7 @@ void btrfs_set_lock_blocking_write(struct extent_buffer *eb)
 	 * lock, but it won't change to or away from us.  If we have the write
 	 * lock, we are the owner and it'll never change.
 	 */
-	if (eb->lock_nested && current->pid == eb->lock_owner)
+	if (eb->lock_recursed && current->pid == eb->lock_owner)
 		return;
 	if (eb->blocking_writers == 0) {
 		btrfs_assert_spinning_writers_put(eb);
@@ -244,7 +244,8 @@ void btrfs_set_lock_blocking_write(struct extent_buffer *eb)
  *
  * The rwlock is held upon exit.
  */
-void btrfs_tree_read_lock(struct extent_buffer *eb)
+void __btrfs_tree_read_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest,
+			    bool recurse)
 {
 	u64 start_ns = 0;
 
@@ -263,8 +264,9 @@ again:
 			 * depends on this as it may be called on a partly
 			 * (write-)locked tree.
 			 */
-			BUG_ON(eb->lock_nested);
-			eb->lock_nested = true;
+			WARN_ON(!recurse);
+			BUG_ON(eb->lock_recursed);
+			eb->lock_recursed = true;
 			read_unlock(&eb->lock);
 			trace_btrfs_tree_read_lock(eb, start_ns);
 			return;
@@ -279,6 +281,11 @@ again:
 	trace_btrfs_tree_read_lock(eb, start_ns);
 }
 
+void btrfs_tree_read_lock(struct extent_buffer *eb)
+{
+	__btrfs_tree_read_lock(eb, BTRFS_NESTING_NORMAL, false);
+}
+
 /*
  * Lock extent buffer for read, optimistically expecting that there are no
  * contending blocking writers. If there are, don't wait.
@@ -362,11 +369,11 @@ void btrfs_tree_read_unlock(struct extent_buffer *eb)
 	/*
 	 * if we're nested, we have the write lock.  No new locking
 	 * is needed as long as we are the lock owner.
-	 * The write unlock will do a barrier for us, and the lock_nested
+	 * The write unlock will do a barrier for us, and the lock_recursed
 	 * field only matters to the lock owner.
 	 */
-	if (eb->lock_nested && current->pid == eb->lock_owner) {
-		eb->lock_nested = false;
+	if (eb->lock_recursed && current->pid == eb->lock_owner) {
+		eb->lock_recursed = false;
 		return;
 	}
 	btrfs_assert_tree_read_locked(eb);
@@ -388,11 +395,11 @@ void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb)
 	/*
 	 * if we're nested, we have the write lock.  No new locking
 	 * is needed as long as we are the lock owner.
-	 * The write unlock will do a barrier for us, and the lock_nested
+	 * The write unlock will do a barrier for us, and the lock_recursed
 	 * field only matters to the lock owner.
 	 */
-	if (eb->lock_nested && current->pid == eb->lock_owner) {
-		eb->lock_nested = false;
+	if (eb->lock_recursed && current->pid == eb->lock_owner) {
+		eb->lock_recursed = false;
 		return;
 	}
 	btrfs_assert_tree_read_locked(eb);
@@ -409,7 +416,7 @@ void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb)
  *
  * The rwlock is held for write upon exit.
  */
-void btrfs_tree_lock(struct extent_buffer *eb)
+void __btrfs_tree_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest)
 	__acquires(&eb->lock)
 {
 	u64 start_ns = 0;
@@ -434,6 +441,11 @@ again:
 	trace_btrfs_tree_lock(eb, start_ns);
 }
 
+void btrfs_tree_lock(struct extent_buffer *eb)
+{
+	__btrfs_tree_lock(eb, BTRFS_NESTING_NORMAL);
+}
+
 /*
  * Release the write lock, either blocking or spinning (ie. there's no need
  * for an explicit blocking unlock, like btrfs_tree_read_unlock_blocking).
@@ -552,13 +564,14 @@ struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root)
  *
  * Return: root extent buffer with read lock held
  */
-struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
+struct extent_buffer *__btrfs_read_lock_root_node(struct btrfs_root *root,
+						  bool recurse)
 {
 	struct extent_buffer *eb;
 
 	while (1) {
 		eb = btrfs_root_node(root);
-		btrfs_tree_read_lock(eb);
+		__btrfs_tree_read_lock(eb, BTRFS_NESTING_NORMAL, recurse);
 		if (eb == root->node)
 			break;
 		btrfs_tree_read_unlock(eb);
diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h
index d715846c10b8d3..3ea81ed3320b5f 100644
--- a/fs/btrfs/locking.h
+++ b/fs/btrfs/locking.h
@@ -16,11 +16,81 @@
 #define BTRFS_WRITE_LOCK_BLOCKING 3
 #define BTRFS_READ_LOCK_BLOCKING 4
 
+/*
+ * We are limited in number of subclasses by MAX_LOCKDEP_SUBCLASSES, which at
+ * the time of this patch is 8, which is how many we use.  Keep this in mind if
+ * you decide you want to add another subclass.
+ */
+enum btrfs_lock_nesting {
+	BTRFS_NESTING_NORMAL,
+
+	/*
+	 * When we COW a block we are holding the lock on the original block,
+	 * and since our lockdep maps are rootid+level, this confuses lockdep
+	 * when we lock the newly allocated COW'd block.  Handle this by having
+	 * a subclass for COW'ed blocks so that lockdep doesn't complain.
+	 */
+	BTRFS_NESTING_COW,
+
+	/*
+	 * Oftentimes we need to lock adjacent nodes on the same level while
+	 * still holding the lock on the original node we searched to, such as
+	 * for searching forward or for split/balance.
+	 *
+	 * Because of this we need to indicate to lockdep that this is
+	 * acceptable by having a different subclass for each of these
+	 * operations.
+	 */
+	BTRFS_NESTING_LEFT,
+	BTRFS_NESTING_RIGHT,
+
+	/*
+	 * When splitting we will be holding a lock on the left/right node when
+	 * we need to cow that node, thus we need a new set of subclasses for
+	 * these two operations.
+	 */
+	BTRFS_NESTING_LEFT_COW,
+	BTRFS_NESTING_RIGHT_COW,
+
+	/*
+	 * When splitting we may push nodes to the left or right, but still use
+	 * the subsequent nodes in our path, keeping our locks on those adjacent
+	 * blocks.  Thus when we go to allocate a new split block we've already
+	 * used up all of our available subclasses, so this subclass exists to
+	 * handle this case where we need to allocate a new split block.
+	 */
+	BTRFS_NESTING_SPLIT,
+
+	/*
+	 * When promoting a new block to a root we need to have a special
+	 * subclass so we don't confuse lockdep, as it will appear that we are
+	 * locking a higher level node before a lower level one.  Copying also
+	 * has this problem as it appears we're locking the same block again
+	 * when we make a snapshot of an existing root.
+	 */
+	BTRFS_NESTING_NEW_ROOT,
+
+	/*
+	 * We are limited to MAX_LOCKDEP_SUBLCLASSES number of subclasses, so
+	 * add this in here and add a static_assert to keep us from going over
+	 * the limit.  As of this writing we're limited to 8, and we're
+	 * definitely using 8, hence this check to keep us from messing up in
+	 * the future.
+	 */
+	BTRFS_NESTING_MAX,
+};
+
+static_assert(BTRFS_NESTING_MAX <= MAX_LOCKDEP_SUBCLASSES,
+	      "too many lock subclasses defined");
+
 struct btrfs_path;
 
+void __btrfs_tree_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest);
 void btrfs_tree_lock(struct extent_buffer *eb);
 void btrfs_tree_unlock(struct extent_buffer *eb);
 
+void __btrfs_tree_read_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest,
+			    bool recurse);
 void btrfs_tree_read_lock(struct extent_buffer *eb);
 void btrfs_tree_read_unlock(struct extent_buffer *eb);
 void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb);
@@ -29,6 +99,14 @@ void btrfs_set_lock_blocking_write(struct extent_buffer *eb);
 int btrfs_try_tree_read_lock(struct extent_buffer *eb);
 int btrfs_try_tree_write_lock(struct extent_buffer *eb);
 int btrfs_tree_read_lock_atomic(struct extent_buffer *eb);
+struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root);
+struct extent_buffer *__btrfs_read_lock_root_node(struct btrfs_root *root,
+						  bool recurse);
+
+static inline struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
+{
+	return __btrfs_read_lock_root_node(root, false);
+}
 
 #ifdef CONFIG_BTRFS_DEBUG
 static inline void btrfs_assert_tree_locked(struct extent_buffer *eb) {
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index ebac13389e7e6b..87bac9ecdf4c8d 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -212,11 +212,12 @@ static int __btrfs_add_ordered_extent(struct btrfs_inode *inode, u64 file_offset
 	refcount_set(&entry->refs, 1);
 	init_waitqueue_head(&entry->wait);
 	INIT_LIST_HEAD(&entry->list);
+	INIT_LIST_HEAD(&entry->log_list);
 	INIT_LIST_HEAD(&entry->root_extent_list);
 	INIT_LIST_HEAD(&entry->work_list);
 	init_completion(&entry->completion);
 
-	trace_btrfs_ordered_extent_add(&inode->vfs_inode, entry);
+	trace_btrfs_ordered_extent_add(inode, entry);
 
 	spin_lock_irq(&tree->lock);
 	node = tree_insert(&tree->tree, file_offset,
@@ -377,17 +378,16 @@ out:
  * test_and_set_bit on a flag in the struct btrfs_ordered_extent is used
  * to make sure this function only returns 1 once for a given ordered extent.
  */
-int btrfs_dec_test_ordered_pending(struct inode *inode,
+int btrfs_dec_test_ordered_pending(struct btrfs_inode *inode,
 				   struct btrfs_ordered_extent **cached,
 				   u64 file_offset, u64 io_size, int uptodate)
 {
-	struct btrfs_ordered_inode_tree *tree;
+	struct btrfs_ordered_inode_tree *tree = &inode->ordered_tree;
 	struct rb_node *node;
 	struct btrfs_ordered_extent *entry = NULL;
 	unsigned long flags;
 	int ret;
 
-	tree = &BTRFS_I(inode)->ordered_tree;
 	spin_lock_irqsave(&tree->lock, flags);
 	if (cached && *cached) {
 		entry = *cached;
@@ -408,7 +408,7 @@ have_entry:
 	}
 
 	if (io_size > entry->bytes_left) {
-		btrfs_crit(BTRFS_I(inode)->root->fs_info,
+		btrfs_crit(inode->root->fs_info,
 			   "bad ordered accounting left %llu size %llu",
 		       entry->bytes_left, io_size);
 	}
@@ -441,10 +441,11 @@ void btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry)
 	struct list_head *cur;
 	struct btrfs_ordered_sum *sum;
 
-	trace_btrfs_ordered_extent_put(entry->inode, entry);
+	trace_btrfs_ordered_extent_put(BTRFS_I(entry->inode), entry);
 
 	if (refcount_dec_and_test(&entry->refs)) {
 		ASSERT(list_empty(&entry->root_extent_list));
+		ASSERT(list_empty(&entry->log_list));
 		ASSERT(RB_EMPTY_NODE(&entry->rb_node));
 		if (entry->inode)
 			btrfs_add_delayed_iput(entry->inode);
@@ -462,14 +463,14 @@ void btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry)
  * remove an ordered extent from the tree.  No references are dropped
  * and waiters are woken up.
  */
-void btrfs_remove_ordered_extent(struct inode *inode,
+void btrfs_remove_ordered_extent(struct btrfs_inode *btrfs_inode,
 				 struct btrfs_ordered_extent *entry)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ordered_inode_tree *tree;
-	struct btrfs_inode *btrfs_inode = BTRFS_I(inode);
 	struct btrfs_root *root = btrfs_inode->root;
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct rb_node *node;
+	bool pending;
 
 	/* This is paired with btrfs_add_ordered_extent. */
 	spin_lock(&btrfs_inode->lock);
@@ -491,13 +492,41 @@ void btrfs_remove_ordered_extent(struct inode *inode,
 	if (tree->last == node)
 		tree->last = NULL;
 	set_bit(BTRFS_ORDERED_COMPLETE, &entry->flags);
+	pending = test_and_clear_bit(BTRFS_ORDERED_PENDING, &entry->flags);
 	spin_unlock_irq(&tree->lock);
 
+	/*
+	 * The current running transaction is waiting on us, we need to let it
+	 * know that we're complete and wake it up.
+	 */
+	if (pending) {
+		struct btrfs_transaction *trans;
+
+		/*
+		 * The checks for trans are just a formality, it should be set,
+		 * but if it isn't we don't want to deref/assert under the spin
+		 * lock, so be nice and check if trans is set, but ASSERT() so
+		 * if it isn't set a developer will notice.
+		 */
+		spin_lock(&fs_info->trans_lock);
+		trans = fs_info->running_transaction;
+		if (trans)
+			refcount_inc(&trans->use_count);
+		spin_unlock(&fs_info->trans_lock);
+
+		ASSERT(trans);
+		if (trans) {
+			if (atomic_dec_and_test(&trans->pending_ordered))
+				wake_up(&trans->pending_wait);
+			btrfs_put_transaction(trans);
+		}
+	}
+
 	spin_lock(&root->ordered_extent_lock);
 	list_del_init(&entry->root_extent_list);
 	root->nr_ordered_extents--;
 
-	trace_btrfs_ordered_extent_remove(inode, entry);
+	trace_btrfs_ordered_extent_remove(btrfs_inode, entry);
 
 	if (!root->nr_ordered_extents) {
 		spin_lock(&fs_info->ordered_root_lock);
@@ -514,7 +543,7 @@ static void btrfs_run_ordered_extent_work(struct btrfs_work *work)
 	struct btrfs_ordered_extent *ordered;
 
 	ordered = container_of(work, struct btrfs_ordered_extent, flush_work);
-	btrfs_start_ordered_extent(ordered->inode, ordered, 1);
+	btrfs_start_ordered_extent(ordered, 1);
 	complete(&ordered->completion);
 }
 
@@ -620,12 +649,11 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
  * in the extent, and it waits on the io completion code to insert
  * metadata into the btree corresponding to the extent
  */
-void btrfs_start_ordered_extent(struct inode *inode,
-				       struct btrfs_ordered_extent *entry,
-				       int wait)
+void btrfs_start_ordered_extent(struct btrfs_ordered_extent *entry, int wait)
 {
 	u64 start = entry->file_offset;
 	u64 end = start + entry->num_bytes - 1;
+	struct btrfs_inode *inode = BTRFS_I(entry->inode);
 
 	trace_btrfs_ordered_extent_start(inode, entry);
 
@@ -635,7 +663,7 @@ void btrfs_start_ordered_extent(struct inode *inode,
 	 * for the flusher thread to find them
 	 */
 	if (!test_bit(BTRFS_ORDERED_DIRECT, &entry->flags))
-		filemap_fdatawrite_range(inode->i_mapping, start, end);
+		filemap_fdatawrite_range(inode->vfs_inode.i_mapping, start, end);
 	if (wait) {
 		wait_event(entry->wait, test_bit(BTRFS_ORDERED_COMPLETE,
 						 &entry->flags));
@@ -679,7 +707,7 @@ int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
 
 	end = orig_end;
 	while (1) {
-		ordered = btrfs_lookup_first_ordered_extent(inode, end);
+		ordered = btrfs_lookup_first_ordered_extent(BTRFS_I(inode), end);
 		if (!ordered)
 			break;
 		if (ordered->file_offset > orig_end) {
@@ -690,7 +718,7 @@ int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
 			btrfs_put_ordered_extent(ordered);
 			break;
 		}
-		btrfs_start_ordered_extent(inode, ordered, 1);
+		btrfs_start_ordered_extent(ordered, 1);
 		end = ordered->file_offset;
 		/*
 		 * If the ordered extent had an error save the error but don't
@@ -775,17 +803,45 @@ out:
 }
 
 /*
+ * Adds all ordered extents to the given list. The list ends up sorted by the
+ * file_offset of the ordered extents.
+ */
+void btrfs_get_ordered_extents_for_logging(struct btrfs_inode *inode,
+					   struct list_head *list)
+{
+	struct btrfs_ordered_inode_tree *tree = &inode->ordered_tree;
+	struct rb_node *n;
+
+	ASSERT(inode_is_locked(&inode->vfs_inode));
+
+	spin_lock_irq(&tree->lock);
+	for (n = rb_first(&tree->tree); n; n = rb_next(n)) {
+		struct btrfs_ordered_extent *ordered;
+
+		ordered = rb_entry(n, struct btrfs_ordered_extent, rb_node);
+
+		if (test_bit(BTRFS_ORDERED_LOGGED, &ordered->flags))
+			continue;
+
+		ASSERT(list_empty(&ordered->log_list));
+		list_add_tail(&ordered->log_list, list);
+		refcount_inc(&ordered->refs);
+	}
+	spin_unlock_irq(&tree->lock);
+}
+
+/*
  * lookup and return any extent before 'file_offset'.  NULL is returned
  * if none is found
  */
 struct btrfs_ordered_extent *
-btrfs_lookup_first_ordered_extent(struct inode *inode, u64 file_offset)
+btrfs_lookup_first_ordered_extent(struct btrfs_inode *inode, u64 file_offset)
 {
 	struct btrfs_ordered_inode_tree *tree;
 	struct rb_node *node;
 	struct btrfs_ordered_extent *entry = NULL;
 
-	tree = &BTRFS_I(inode)->ordered_tree;
+	tree = &inode->ordered_tree;
 	spin_lock_irq(&tree->lock);
 	node = tree_search(tree, file_offset);
 	if (!node)
@@ -803,20 +859,21 @@ out:
  * try to find a checksum.  This is used because we allow pages to
  * be reclaimed before their checksum is actually put into the btree
  */
-int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
-			   u8 *sum, int len)
+int btrfs_find_ordered_sum(struct btrfs_inode *inode, u64 offset,
+			   u64 disk_bytenr, u8 *sum, int len)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_ordered_sum *ordered_sum;
 	struct btrfs_ordered_extent *ordered;
-	struct btrfs_ordered_inode_tree *tree = &BTRFS_I(inode)->ordered_tree;
+	struct btrfs_ordered_inode_tree *tree = &inode->ordered_tree;
 	unsigned long num_sectors;
 	unsigned long i;
 	u32 sectorsize = btrfs_inode_sectorsize(inode);
+	const u8 blocksize_bits = inode->vfs_inode.i_sb->s_blocksize_bits;
 	const u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
 	int index = 0;
 
-	ordered = btrfs_lookup_ordered_extent(BTRFS_I(inode), offset);
+	ordered = btrfs_lookup_ordered_extent(inode, offset);
 	if (!ordered)
 		return 0;
 
@@ -824,10 +881,8 @@ int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
 	list_for_each_entry_reverse(ordered_sum, &ordered->list, list) {
 		if (disk_bytenr >= ordered_sum->bytenr &&
 		    disk_bytenr < ordered_sum->bytenr + ordered_sum->len) {
-			i = (disk_bytenr - ordered_sum->bytenr) >>
-			    inode->i_sb->s_blocksize_bits;
-			num_sectors = ordered_sum->len >>
-				      inode->i_sb->s_blocksize_bits;
+			i = (disk_bytenr - ordered_sum->bytenr) >> blocksize_bits;
+			num_sectors = ordered_sum->len >> blocksize_bits;
 			num_sectors = min_t(int, len - index, num_sectors - i);
 			memcpy(sum + index, ordered_sum->sums + i * csum_size,
 			       num_sectors * csum_size);
@@ -883,7 +938,7 @@ void btrfs_lock_and_flush_ordered_range(struct btrfs_inode *inode, u64 start,
 			break;
 		}
 		unlock_extent_cached(&inode->io_tree, start, end, cachedp);
-		btrfs_start_ordered_extent(&inode->vfs_inode, ordered, 1);
+		btrfs_start_ordered_extent(ordered, 1);
 		btrfs_put_ordered_extent(ordered);
 	}
 }
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index d61ea9c880a327..c3a2325e64a416 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -56,6 +56,12 @@ enum {
 	BTRFS_ORDERED_TRUNCATED,
 	/* Regular IO for COW */
 	BTRFS_ORDERED_REGULAR,
+	/* Used during fsync to track already logged extents */
+	BTRFS_ORDERED_LOGGED,
+	/* We have already logged all the csums of the ordered extent */
+	BTRFS_ORDERED_LOGGED_CSUM,
+	/* We wait for this extent to complete in the current transaction */
+	BTRFS_ORDERED_PENDING,
 };
 
 struct btrfs_ordered_extent {
@@ -104,6 +110,9 @@ struct btrfs_ordered_extent {
 	/* list of checksums for insertion when the extent io is done */
 	struct list_head list;
 
+	/* used for fast fsyncs */
+	struct list_head log_list;
+
 	/* used to wait for the BTRFS_ORDERED_COMPLETE bit */
 	wait_queue_head_t wait;
 
@@ -142,9 +151,9 @@ btrfs_ordered_inode_tree_init(struct btrfs_ordered_inode_tree *t)
 }
 
 void btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry);
-void btrfs_remove_ordered_extent(struct inode *inode,
+void btrfs_remove_ordered_extent(struct btrfs_inode *btrfs_inode,
 				struct btrfs_ordered_extent *entry);
-int btrfs_dec_test_ordered_pending(struct inode *inode,
+int btrfs_dec_test_ordered_pending(struct btrfs_inode *inode,
 				   struct btrfs_ordered_extent **cached,
 				   u64 file_offset, u64 io_size, int uptodate);
 int btrfs_dec_test_first_ordered_pending(struct btrfs_inode *inode,
@@ -165,17 +174,18 @@ void btrfs_add_ordered_sum(struct btrfs_ordered_extent *entry,
 			   struct btrfs_ordered_sum *sum);
 struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct btrfs_inode *inode,
 							 u64 file_offset);
-void btrfs_start_ordered_extent(struct inode *inode,
-				struct btrfs_ordered_extent *entry, int wait);
+void btrfs_start_ordered_extent(struct btrfs_ordered_extent *entry, int wait);
 int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len);
 struct btrfs_ordered_extent *
-btrfs_lookup_first_ordered_extent(struct inode * inode, u64 file_offset);
+btrfs_lookup_first_ordered_extent(struct btrfs_inode *inode, u64 file_offset);
 struct btrfs_ordered_extent *btrfs_lookup_ordered_range(
 		struct btrfs_inode *inode,
 		u64 file_offset,
 		u64 len);
-int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
-			   u8 *sum, int len);
+void btrfs_get_ordered_extents_for_logging(struct btrfs_inode *inode,
+					   struct list_head *list);
+int btrfs_find_ordered_sum(struct btrfs_inode *inode, u64 offset,
+			   u64 disk_bytenr, u8 *sum, int len);
 u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
 			       const u64 range_start, const u64 range_len);
 void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index 80567c11ec122d..7695c4783d33b0 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -7,6 +7,44 @@
 #include "disk-io.h"
 #include "print-tree.h"
 
+struct root_name_map {
+	u64 id;
+	char name[16];
+};
+
+static const struct root_name_map root_map[] = {
+	{ BTRFS_ROOT_TREE_OBJECTID,		"ROOT_TREE"		},
+	{ BTRFS_EXTENT_TREE_OBJECTID,		"EXTENT_TREE"		},
+	{ BTRFS_CHUNK_TREE_OBJECTID,		"CHUNK_TREE"		},
+	{ BTRFS_DEV_TREE_OBJECTID,		"DEV_TREE"		},
+	{ BTRFS_FS_TREE_OBJECTID,		"FS_TREE"		},
+	{ BTRFS_CSUM_TREE_OBJECTID,		"CSUM_TREE"		},
+	{ BTRFS_TREE_LOG_OBJECTID,		"TREE_LOG"		},
+	{ BTRFS_QUOTA_TREE_OBJECTID,		"QUOTA_TREE"		},
+	{ BTRFS_UUID_TREE_OBJECTID,		"UUID_TREE"		},
+	{ BTRFS_FREE_SPACE_TREE_OBJECTID,	"FREE_SPACE_TREE"	},
+	{ BTRFS_DATA_RELOC_TREE_OBJECTID,	"DATA_RELOC_TREE"	},
+};
+
+const char *btrfs_root_name(u64 objectid, char *buf)
+{
+	int i;
+
+	if (objectid == BTRFS_TREE_RELOC_OBJECTID) {
+		snprintf(buf, BTRFS_ROOT_NAME_BUF_LEN,
+			 "TREE_RELOC offset=%llu", objectid);
+		return buf;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(root_map); i++) {
+		if (root_map[i].id == objectid)
+			return root_map[i].name;
+	}
+
+	snprintf(buf, BTRFS_ROOT_NAME_BUF_LEN, "%llu", objectid);
+	return buf;
+}
+
 static void print_chunk(struct extent_buffer *eb, struct btrfs_chunk *chunk)
 {
 	int num_stripes = btrfs_chunk_num_stripes(eb, chunk);
diff --git a/fs/btrfs/print-tree.h b/fs/btrfs/print-tree.h
index e6bb38fd75ad88..78b99385a503fb 100644
--- a/fs/btrfs/print-tree.h
+++ b/fs/btrfs/print-tree.h
@@ -6,7 +6,11 @@
 #ifndef BTRFS_PRINT_TREE_H
 #define BTRFS_PRINT_TREE_H
 
+/* Buffer size to contain tree name and possibly additional data (offset) */
+#define BTRFS_ROOT_NAME_BUF_LEN				48
+
 void btrfs_print_leaf(struct extent_buffer *l);
 void btrfs_print_tree(struct extent_buffer *c, bool follow);
+const char *btrfs_root_name(u64 objectid, char *buf);
 
 #endif
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index c0f350c3a0cf47..580899bdb9915b 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -2315,7 +2315,7 @@ static int qgroup_update_refcnt(struct btrfs_fs_info *fs_info,
  * Update qgroup rfer/excl counters.
  * Rfer update is easy, codes can explain themselves.
  *
- * Excl update is tricky, the update is split into 2 part.
+ * Excl update is tricky, the update is split into 2 parts.
  * Part 1: Possible exclusive <-> sharing detect:
  *	|	A	|	!A	|
  *  -------------------------------------
diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c
index 243a2e44526ef9..9d4f5316a7e8be 100644
--- a/fs/btrfs/reada.c
+++ b/fs/btrfs/reada.c
@@ -767,31 +767,39 @@ static void reada_start_machine_worker(struct btrfs_work *work)
 	kfree(rmw);
 }
 
-static void __reada_start_machine(struct btrfs_fs_info *fs_info)
+/* Try to start up to 10k READA requests for a group of devices */
+static int reada_start_for_fsdevs(struct btrfs_fs_devices *fs_devices)
 {
-	struct btrfs_device *device;
-	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
 	u64 enqueued;
 	u64 total = 0;
-	int i;
+	struct btrfs_device *device;
 
-again:
 	do {
 		enqueued = 0;
-		mutex_lock(&fs_devices->device_list_mutex);
 		list_for_each_entry(device, &fs_devices->devices, dev_list) {
 			if (atomic_read(&device->reada_in_flight) <
 			    MAX_IN_FLIGHT)
 				enqueued += reada_start_machine_dev(device);
 		}
-		mutex_unlock(&fs_devices->device_list_mutex);
 		total += enqueued;
 	} while (enqueued && total < 10000);
-	if (fs_devices->seed) {
-		fs_devices = fs_devices->seed;
-		goto again;
-	}
 
+	return total;
+}
+
+static void __reada_start_machine(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices, *seed_devs;
+	int i;
+	u64 enqueued = 0;
+
+	mutex_lock(&fs_devices->device_list_mutex);
+
+	enqueued += reada_start_for_fsdevs(fs_devices);
+	list_for_each_entry(seed_devs, &fs_devices->seed_list, seed_list)
+		enqueued += reada_start_for_fsdevs(seed_devs);
+
+	mutex_unlock(&fs_devices->device_list_mutex);
 	if (enqueued == 0)
 		return;
 
diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c
index 5cd02514cf4d48..99aa87c089121b 100644
--- a/fs/btrfs/reflink.c
+++ b/fs/btrfs/reflink.c
@@ -45,7 +45,7 @@ out:
 	return ret;
 }
 
-static int copy_inline_to_page(struct inode *inode,
+static int copy_inline_to_page(struct btrfs_inode *inode,
 			       const u64 file_offset,
 			       char *inline_data,
 			       const u64 size,
@@ -58,6 +58,7 @@ static int copy_inline_to_page(struct inode *inode,
 	char *data_start = inline_data + btrfs_file_extent_calc_inline_size(0);
 	struct extent_changeset *data_reserved = NULL;
 	struct page *page = NULL;
+	struct address_space *mapping = inode->vfs_inode.i_mapping;
 	int ret;
 
 	ASSERT(IS_ALIGNED(file_offset, block_size));
@@ -68,24 +69,23 @@ static int copy_inline_to_page(struct inode *inode,
 	 * reservation here. Also we must not do the reservation while holding
 	 * a transaction open, otherwise we would deadlock.
 	 */
-	ret = btrfs_delalloc_reserve_space(BTRFS_I(inode), &data_reserved,
-					   file_offset, block_size);
+	ret = btrfs_delalloc_reserve_space(inode, &data_reserved, file_offset,
+					   block_size);
 	if (ret)
 		goto out;
 
-	page = find_or_create_page(inode->i_mapping, file_offset >> PAGE_SHIFT,
-				   btrfs_alloc_write_mask(inode->i_mapping));
+	page = find_or_create_page(mapping, file_offset >> PAGE_SHIFT,
+				   btrfs_alloc_write_mask(mapping));
 	if (!page) {
 		ret = -ENOMEM;
 		goto out_unlock;
 	}
 
 	set_page_extent_mapped(page);
-	clear_extent_bit(&BTRFS_I(inode)->io_tree, file_offset, range_end,
+	clear_extent_bit(&inode->io_tree, file_offset, range_end,
 			 EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
 			 0, 0, NULL);
-	ret = btrfs_set_extent_delalloc(BTRFS_I(inode), file_offset, range_end,
-					0, NULL);
+	ret = btrfs_set_extent_delalloc(inode, file_offset, range_end, 0, NULL);
 	if (ret)
 		goto out_unlock;
 
@@ -134,9 +134,9 @@ out_unlock:
 		put_page(page);
 	}
 	if (ret)
-		btrfs_delalloc_release_space(BTRFS_I(inode), data_reserved,
-					     file_offset, block_size, true);
-	btrfs_delalloc_release_extents(BTRFS_I(inode), block_size);
+		btrfs_delalloc_release_space(inode, data_reserved, file_offset,
+					     block_size, true);
+	btrfs_delalloc_release_extents(inode, block_size);
 out:
 	extent_changeset_free(data_reserved);
 
@@ -167,8 +167,8 @@ static int clone_copy_inline_extent(struct inode *dst,
 	struct btrfs_key key;
 
 	if (new_key->offset > 0) {
-		ret = copy_inline_to_page(dst, new_key->offset, inline_data,
-					  size, datal, comp_type);
+		ret = copy_inline_to_page(BTRFS_I(dst), new_key->offset,
+					  inline_data, size, datal, comp_type);
 		goto out;
 	}
 
@@ -194,7 +194,7 @@ static int clone_copy_inline_extent(struct inode *dst,
 			 * inline extent's data to the page.
 			 */
 			ASSERT(key.offset > 0);
-			ret = copy_inline_to_page(dst, new_key->offset,
+			ret = copy_inline_to_page(BTRFS_I(dst), new_key->offset,
 						  inline_data, size, datal,
 						  comp_type);
 			goto out;
@@ -213,8 +213,8 @@ static int clone_copy_inline_extent(struct inode *dst,
 		    BTRFS_FILE_EXTENT_INLINE)
 			goto copy_inline_extent;
 
-		ret = copy_inline_to_page(dst, new_key->offset, inline_data,
-					  size, datal, comp_type);
+		ret = copy_inline_to_page(BTRFS_I(dst), new_key->offset,
+					  inline_data, size, datal, comp_type);
 		goto out;
 	}
 
@@ -231,8 +231,8 @@ copy_inline_extent:
 		 * clone. Deal with all these cases by copying the inline extent
 		 * data into the respective page at the destination inode.
 		 */
-		ret = copy_inline_to_page(dst, new_key->offset, inline_data,
-					   size, datal, comp_type);
+		ret = copy_inline_to_page(BTRFS_I(dst), new_key->offset,
+					  inline_data, size, datal, comp_type);
 		goto out;
 	}
 
@@ -439,7 +439,7 @@ process_slot:
 
 		if (type == BTRFS_FILE_EXTENT_REG ||
 		    type == BTRFS_FILE_EXTENT_PREALLOC) {
-			struct btrfs_clone_extent_info clone_info;
+			struct btrfs_replace_extent_info clone_info;
 
 			/*
 			 *    a  | --- range to clone ---|  b
@@ -462,8 +462,8 @@ process_slot:
 			clone_info.data_len = datal;
 			clone_info.file_offset = new_key.offset;
 			clone_info.extent_buf = buf;
-			clone_info.item_size = size;
-			ret = btrfs_punch_hole_range(inode, path, drop_start,
+			clone_info.is_new_extent = false;
+			ret = btrfs_replace_file_extents(inode, path, drop_start,
 					new_key.offset + datal - 1, &clone_info,
 					&trans);
 			if (ret)
@@ -520,6 +520,8 @@ process_slot:
 			ret = -EINTR;
 			goto out;
 		}
+
+		cond_resched();
 	}
 	ret = 0;
 
@@ -533,7 +535,7 @@ process_slot:
 		btrfs_release_path(path);
 		path->leave_spinning = 0;
 
-		ret = btrfs_punch_hole_range(inode, path, last_dest_end,
+		ret = btrfs_replace_file_extents(inode, path, last_dest_end,
 				destoff + len - 1, NULL, &trans);
 		if (ret)
 			goto out;
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 4ba1ab9cc76db8..3602806d71bd4f 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1206,7 +1206,8 @@ again:
 	}
 
 	if (cow) {
-		ret = btrfs_cow_block(trans, dest, eb, NULL, 0, &eb);
+		ret = btrfs_cow_block(trans, dest, eb, NULL, 0, &eb,
+				      BTRFS_NESTING_COW);
 		BUG_ON(ret);
 	}
 	btrfs_set_lock_blocking_write(eb);
@@ -1274,7 +1275,8 @@ again:
 			btrfs_tree_lock(eb);
 			if (cow) {
 				ret = btrfs_cow_block(trans, dest, eb, parent,
-						      slot, &eb);
+						      slot, &eb,
+						      BTRFS_NESTING_COW);
 				BUG_ON(ret);
 			}
 			btrfs_set_lock_blocking_write(eb);
@@ -1781,7 +1783,8 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
 	 * relocated and the block is tree root.
 	 */
 	leaf = btrfs_lock_root_node(root);
-	ret = btrfs_cow_block(trans, root, leaf, NULL, 0, &leaf);
+	ret = btrfs_cow_block(trans, root, leaf, NULL, 0, &leaf,
+			      BTRFS_NESTING_COW);
 	btrfs_tree_unlock(leaf);
 	free_extent_buffer(leaf);
 	if (ret < 0)
@@ -2308,7 +2311,7 @@ static int do_relocation(struct btrfs_trans_handle *trans,
 
 		if (!node->eb) {
 			ret = btrfs_cow_block(trans, root, eb, upper->eb,
-					      slot, &eb);
+					      slot, &eb, BTRFS_NESTING_COW);
 			btrfs_tree_unlock(eb);
 			free_extent_buffer(eb);
 			if (ret < 0) {
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index c89697486366ac..702dc5441f0397 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -512,11 +512,20 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 	if (ret && qgroup_num_bytes)
 		btrfs_qgroup_free_meta_prealloc(root, qgroup_num_bytes);
 
+	if (!ret) {
+		spin_lock(&rsv->lock);
+		rsv->qgroup_rsv_reserved += qgroup_num_bytes;
+		spin_unlock(&rsv->lock);
+	}
 	return ret;
 }
 
-void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info,
+void btrfs_subvolume_release_metadata(struct btrfs_root *root,
 				      struct btrfs_block_rsv *rsv)
 {
-	btrfs_block_rsv_release(fs_info, rsv, (u64)-1, NULL);
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	u64 qgroup_to_release;
+
+	btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release);
+	btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release);
 }
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 354ab9985a3426..cf63f1e27a2791 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -835,7 +835,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 	int success;
 	bool full_stripe_locked;
 	unsigned int nofs_flag;
-	static DEFINE_RATELIMIT_STATE(_rs, DEFAULT_RATELIMIT_INTERVAL,
+	static DEFINE_RATELIMIT_STATE(rs, DEFAULT_RATELIMIT_INTERVAL,
 				      DEFAULT_RATELIMIT_BURST);
 
 	BUG_ON(sblock_to_check->page_count < 1);
@@ -969,14 +969,14 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 		spin_lock(&sctx->stat_lock);
 		sctx->stat.read_errors++;
 		spin_unlock(&sctx->stat_lock);
-		if (__ratelimit(&_rs))
+		if (__ratelimit(&rs))
 			scrub_print_warning("i/o error", sblock_to_check);
 		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_READ_ERRS);
 	} else if (sblock_bad->checksum_error) {
 		spin_lock(&sctx->stat_lock);
 		sctx->stat.csum_errors++;
 		spin_unlock(&sctx->stat_lock);
-		if (__ratelimit(&_rs))
+		if (__ratelimit(&rs))
 			scrub_print_warning("checksum error", sblock_to_check);
 		btrfs_dev_stat_inc_and_print(dev,
 					     BTRFS_DEV_STAT_CORRUPTION_ERRS);
@@ -984,7 +984,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 		spin_lock(&sctx->stat_lock);
 		sctx->stat.verify_errors++;
 		spin_unlock(&sctx->stat_lock);
-		if (__ratelimit(&_rs))
+		if (__ratelimit(&rs))
 			scrub_print_warning("checksum/header error",
 					    sblock_to_check);
 		if (sblock_bad->generation_error)
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index d9813a5b075aca..340c76a12ce100 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -122,8 +122,6 @@ struct send_ctx {
 
 	struct file_ra_state ra;
 
-	char *read_buf;
-
 	/*
 	 * We process inodes by their increasing order, so if before an
 	 * incremental send we reverse the parent/child relationship of
@@ -278,11 +276,6 @@ enum btrfs_compare_tree_result {
 	BTRFS_COMPARE_TREE_CHANGED,
 	BTRFS_COMPARE_TREE_SAME,
 };
-typedef int (*btrfs_changed_cb_t)(struct btrfs_path *left_path,
-				  struct btrfs_path *right_path,
-				  struct btrfs_key *key,
-				  enum btrfs_compare_tree_result result,
-				  void *ctx);
 
 __cold
 static void inconsistent_snapshot_error(struct send_ctx *sctx,
@@ -584,8 +577,8 @@ static int tlv_put(struct send_ctx *sctx, u16 attr, const void *data, int len)
 		return -EOVERFLOW;
 
 	hdr = (struct btrfs_tlv_header *) (sctx->send_buf + sctx->send_size);
-	hdr->tlv_type = cpu_to_le16(attr);
-	hdr->tlv_len = cpu_to_le16(len);
+	put_unaligned_le16(attr, &hdr->tlv_type);
+	put_unaligned_le16(len, &hdr->tlv_len);
 	memcpy(hdr + 1, data, len);
 	sctx->send_size += total_len;
 
@@ -695,7 +688,7 @@ static int begin_cmd(struct send_ctx *sctx, int cmd)
 
 	sctx->send_size += sizeof(*hdr);
 	hdr = (struct btrfs_cmd_header *)sctx->send_buf;
-	hdr->cmd = cpu_to_le16(cmd);
+	put_unaligned_le16(cmd, &hdr->cmd);
 
 	return 0;
 }
@@ -707,17 +700,17 @@ static int send_cmd(struct send_ctx *sctx)
 	u32 crc;
 
 	hdr = (struct btrfs_cmd_header *)sctx->send_buf;
-	hdr->len = cpu_to_le32(sctx->send_size - sizeof(*hdr));
-	hdr->crc = 0;
+	put_unaligned_le32(sctx->send_size - sizeof(*hdr), &hdr->len);
+	put_unaligned_le32(0, &hdr->crc);
 
 	crc = btrfs_crc32c(0, (unsigned char *)sctx->send_buf, sctx->send_size);
-	hdr->crc = cpu_to_le32(crc);
+	put_unaligned_le32(crc, &hdr->crc);
 
 	ret = write_buf(sctx->send_filp, sctx->send_buf, sctx->send_size,
 					&sctx->send_off);
 
 	sctx->total_send_size += sctx->send_size;
-	sctx->cmd_send_size[le16_to_cpu(hdr->cmd)] += sctx->send_size;
+	sctx->cmd_send_size[get_unaligned_le16(&hdr->cmd)] += sctx->send_size;
 	sctx->send_size = 0;
 
 	return ret;
@@ -3813,6 +3806,72 @@ static int update_ref_path(struct send_ctx *sctx, struct recorded_ref *ref)
 }
 
 /*
+ * When processing the new references for an inode we may orphanize an existing
+ * directory inode because its old name conflicts with one of the new references
+ * of the current inode. Later, when processing another new reference of our
+ * inode, we might need to orphanize another inode, but the path we have in the
+ * reference reflects the pre-orphanization name of the directory we previously
+ * orphanized. For example:
+ *
+ * parent snapshot looks like:
+ *
+ * .                                     (ino 256)
+ * |----- f1                             (ino 257)
+ * |----- f2                             (ino 258)
+ * |----- d1/                            (ino 259)
+ *        |----- d2/                     (ino 260)
+ *
+ * send snapshot looks like:
+ *
+ * .                                     (ino 256)
+ * |----- d1                             (ino 258)
+ * |----- f2/                            (ino 259)
+ *        |----- f2_link/                (ino 260)
+ *        |       |----- f1              (ino 257)
+ *        |
+ *        |----- d2                      (ino 258)
+ *
+ * When processing inode 257 we compute the name for inode 259 as "d1", and we
+ * cache it in the name cache. Later when we start processing inode 258, when
+ * collecting all its new references we set a full path of "d1/d2" for its new
+ * reference with name "d2". When we start processing the new references we
+ * start by processing the new reference with name "d1", and this results in
+ * orphanizing inode 259, since its old reference causes a conflict. Then we
+ * move on the next new reference, with name "d2", and we find out we must
+ * orphanize inode 260, as its old reference conflicts with ours - but for the
+ * orphanization we use a source path corresponding to the path we stored in the
+ * new reference, which is "d1/d2" and not "o259-6-0/d2" - this makes the
+ * receiver fail since the path component "d1/" no longer exists, it was renamed
+ * to "o259-6-0/" when processing the previous new reference. So in this case we
+ * must recompute the path in the new reference and use it for the new
+ * orphanization operation.
+ */
+static int refresh_ref_path(struct send_ctx *sctx, struct recorded_ref *ref)
+{
+	char *name;
+	int ret;
+
+	name = kmemdup(ref->name, ref->name_len, GFP_KERNEL);
+	if (!name)
+		return -ENOMEM;
+
+	fs_path_reset(ref->full_path);
+	ret = get_cur_path(sctx, ref->dir, ref->dir_gen, ref->full_path);
+	if (ret < 0)
+		goto out;
+
+	ret = fs_path_add(ref->full_path, name, ref->name_len);
+	if (ret < 0)
+		goto out;
+
+	/* Update the reference's base name pointer. */
+	set_ref_path(ref, ref->full_path);
+out:
+	kfree(name);
+	return ret;
+}
+
+/*
  * This does all the move/link/unlink/rmdir magic.
  */
 static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
@@ -3880,52 +3939,56 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
 			goto out;
 	}
 
+	/*
+	 * Before doing any rename and link operations, do a first pass on the
+	 * new references to orphanize any unprocessed inodes that may have a
+	 * reference that conflicts with one of the new references of the current
+	 * inode. This needs to happen first because a new reference may conflict
+	 * with the old reference of a parent directory, so we must make sure
+	 * that the path used for link and rename commands don't use an
+	 * orphanized name when an ancestor was not yet orphanized.
+	 *
+	 * Example:
+	 *
+	 * Parent snapshot:
+	 *
+	 * .                                                      (ino 256)
+	 * |----- testdir/                                        (ino 259)
+	 * |          |----- a                                    (ino 257)
+	 * |
+	 * |----- b                                               (ino 258)
+	 *
+	 * Send snapshot:
+	 *
+	 * .                                                      (ino 256)
+	 * |----- testdir_2/                                      (ino 259)
+	 * |          |----- a                                    (ino 260)
+	 * |
+	 * |----- testdir                                         (ino 257)
+	 * |----- b                                               (ino 257)
+	 * |----- b2                                              (ino 258)
+	 *
+	 * Processing the new reference for inode 257 with name "b" may happen
+	 * before processing the new reference with name "testdir". If so, we
+	 * must make sure that by the time we send a link command to create the
+	 * hard link "b", inode 259 was already orphanized, since the generated
+	 * path in "valid_path" already contains the orphanized name for 259.
+	 * We are processing inode 257, so only later when processing 259 we do
+	 * the rename operation to change its temporary (orphanized) name to
+	 * "testdir_2".
+	 */
 	list_for_each_entry(cur, &sctx->new_refs, list) {
-		/*
-		 * We may have refs where the parent directory does not exist
-		 * yet. This happens if the parent directories inum is higher
-		 * than the current inum. To handle this case, we create the
-		 * parent directory out of order. But we need to check if this
-		 * did already happen before due to other refs in the same dir.
-		 */
 		ret = get_cur_inode_state(sctx, cur->dir, cur->dir_gen);
 		if (ret < 0)
 			goto out;
-		if (ret == inode_state_will_create) {
-			ret = 0;
-			/*
-			 * First check if any of the current inodes refs did
-			 * already create the dir.
-			 */
-			list_for_each_entry(cur2, &sctx->new_refs, list) {
-				if (cur == cur2)
-					break;
-				if (cur2->dir == cur->dir) {
-					ret = 1;
-					break;
-				}
-			}
-
-			/*
-			 * If that did not happen, check if a previous inode
-			 * did already create the dir.
-			 */
-			if (!ret)
-				ret = did_create_dir(sctx, cur->dir);
-			if (ret < 0)
-				goto out;
-			if (!ret) {
-				ret = send_create_inode(sctx, cur->dir);
-				if (ret < 0)
-					goto out;
-			}
-		}
+		if (ret == inode_state_will_create)
+			continue;
 
 		/*
-		 * Check if this new ref would overwrite the first ref of
-		 * another unprocessed inode. If yes, orphanize the
-		 * overwritten inode. If we find an overwritten ref that is
-		 * not the first ref, simply unlink it.
+		 * Check if this new ref would overwrite the first ref of another
+		 * unprocessed inode. If yes, orphanize the overwritten inode.
+		 * If we find an overwritten ref that is not the first ref,
+		 * simply unlink it.
 		 */
 		ret = will_overwrite_ref(sctx, cur->dir, cur->dir_gen,
 				cur->name, cur->name_len,
@@ -3942,6 +4005,12 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
 				struct name_cache_entry *nce;
 				struct waiting_dir_move *wdm;
 
+				if (orphanized_dir) {
+					ret = refresh_ref_path(sctx, cur);
+					if (ret < 0)
+						goto out;
+				}
+
 				ret = orphanize_inode(sctx, ow_inode, ow_gen,
 						cur->full_path);
 				if (ret < 0)
@@ -4004,6 +4073,49 @@ static int process_recorded_refs(struct send_ctx *sctx, int *pending_move)
 			}
 		}
 
+	}
+
+	list_for_each_entry(cur, &sctx->new_refs, list) {
+		/*
+		 * We may have refs where the parent directory does not exist
+		 * yet. This happens if the parent directories inum is higher
+		 * than the current inum. To handle this case, we create the
+		 * parent directory out of order. But we need to check if this
+		 * did already happen before due to other refs in the same dir.
+		 */
+		ret = get_cur_inode_state(sctx, cur->dir, cur->dir_gen);
+		if (ret < 0)
+			goto out;
+		if (ret == inode_state_will_create) {
+			ret = 0;
+			/*
+			 * First check if any of the current inodes refs did
+			 * already create the dir.
+			 */
+			list_for_each_entry(cur2, &sctx->new_refs, list) {
+				if (cur == cur2)
+					break;
+				if (cur2->dir == cur->dir) {
+					ret = 1;
+					break;
+				}
+			}
+
+			/*
+			 * If that did not happen, check if a previous inode
+			 * did already create the dir.
+			 */
+			if (!ret)
+				ret = did_create_dir(sctx, cur->dir);
+			if (ret < 0)
+				goto out;
+			if (!ret) {
+				ret = send_create_inode(sctx, cur->dir);
+				if (ret < 0)
+					goto out;
+			}
+		}
+
 		if (S_ISDIR(sctx->cur_inode_mode) && sctx->parent_root) {
 			ret = wait_for_dest_dir_move(sctx, cur, is_orphan);
 			if (ret < 0)
@@ -4799,7 +4911,25 @@ out:
 	return ret;
 }
 
-static ssize_t fill_read_buf(struct send_ctx *sctx, u64 offset, u32 len)
+static inline u64 max_send_read_size(const struct send_ctx *sctx)
+{
+	return sctx->send_max_size - SZ_16K;
+}
+
+static int put_data_header(struct send_ctx *sctx, u32 len)
+{
+	struct btrfs_tlv_header *hdr;
+
+	if (sctx->send_max_size - sctx->send_size < sizeof(*hdr) + len)
+		return -EOVERFLOW;
+	hdr = (struct btrfs_tlv_header *)(sctx->send_buf + sctx->send_size);
+	put_unaligned_le16(BTRFS_SEND_A_DATA, &hdr->tlv_type);
+	put_unaligned_le16(len, &hdr->tlv_len);
+	sctx->send_size += sizeof(*hdr);
+	return 0;
+}
+
+static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len)
 {
 	struct btrfs_root *root = sctx->send_root;
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -4809,21 +4939,16 @@ static ssize_t fill_read_buf(struct send_ctx *sctx, u64 offset, u32 len)
 	pgoff_t index = offset >> PAGE_SHIFT;
 	pgoff_t last_index;
 	unsigned pg_offset = offset_in_page(offset);
-	ssize_t ret = 0;
+	int ret;
+
+	ret = put_data_header(sctx, len);
+	if (ret)
+		return ret;
 
 	inode = btrfs_iget(fs_info->sb, sctx->cur_ino, root);
 	if (IS_ERR(inode))
 		return PTR_ERR(inode);
 
-	if (offset + len > i_size_read(inode)) {
-		if (offset > i_size_read(inode))
-			len = 0;
-		else
-			len = offset - i_size_read(inode);
-	}
-	if (len == 0)
-		goto out;
-
 	last_index = (offset + len - 1) >> PAGE_SHIFT;
 
 	/* initial readahead */
@@ -4864,16 +4989,16 @@ static ssize_t fill_read_buf(struct send_ctx *sctx, u64 offset, u32 len)
 		}
 
 		addr = kmap(page);
-		memcpy(sctx->read_buf + ret, addr + pg_offset, cur_len);
+		memcpy(sctx->send_buf + sctx->send_size, addr + pg_offset,
+		       cur_len);
 		kunmap(page);
 		unlock_page(page);
 		put_page(page);
 		index++;
 		pg_offset = 0;
 		len -= cur_len;
-		ret += cur_len;
+		sctx->send_size += cur_len;
 	}
-out:
 	iput(inode);
 	return ret;
 }
@@ -4887,7 +5012,6 @@ static int send_write(struct send_ctx *sctx, u64 offset, u32 len)
 	struct btrfs_fs_info *fs_info = sctx->send_root->fs_info;
 	int ret = 0;
 	struct fs_path *p;
-	ssize_t num_read = 0;
 
 	p = fs_path_alloc();
 	if (!p)
@@ -4895,13 +5019,6 @@ static int send_write(struct send_ctx *sctx, u64 offset, u32 len)
 
 	btrfs_debug(fs_info, "send_write offset=%llu, len=%d", offset, len);
 
-	num_read = fill_read_buf(sctx, offset, len);
-	if (num_read <= 0) {
-		if (num_read < 0)
-			ret = num_read;
-		goto out;
-	}
-
 	ret = begin_cmd(sctx, BTRFS_SEND_C_WRITE);
 	if (ret < 0)
 		goto out;
@@ -4912,16 +5029,16 @@ static int send_write(struct send_ctx *sctx, u64 offset, u32 len)
 
 	TLV_PUT_PATH(sctx, BTRFS_SEND_A_PATH, p);
 	TLV_PUT_U64(sctx, BTRFS_SEND_A_FILE_OFFSET, offset);
-	TLV_PUT(sctx, BTRFS_SEND_A_DATA, sctx->read_buf, num_read);
+	ret = put_file_data(sctx, offset, len);
+	if (ret < 0)
+		goto out;
 
 	ret = send_cmd(sctx);
 
 tlv_put_failure:
 out:
 	fs_path_free(p);
-	if (ret < 0)
-		return ret;
-	return num_read;
+	return ret;
 }
 
 /*
@@ -5033,8 +5150,8 @@ out:
 static int send_hole(struct send_ctx *sctx, u64 end)
 {
 	struct fs_path *p = NULL;
+	u64 read_size = max_send_read_size(sctx);
 	u64 offset = sctx->cur_inode_last_extent;
-	u64 len;
 	int ret = 0;
 
 	/*
@@ -5061,16 +5178,19 @@ static int send_hole(struct send_ctx *sctx, u64 end)
 	ret = get_cur_path(sctx, sctx->cur_ino, sctx->cur_inode_gen, p);
 	if (ret < 0)
 		goto tlv_put_failure;
-	memset(sctx->read_buf, 0, BTRFS_SEND_READ_SIZE);
 	while (offset < end) {
-		len = min_t(u64, end - offset, BTRFS_SEND_READ_SIZE);
+		u64 len = min(end - offset, read_size);
 
 		ret = begin_cmd(sctx, BTRFS_SEND_C_WRITE);
 		if (ret < 0)
 			break;
 		TLV_PUT_PATH(sctx, BTRFS_SEND_A_PATH, p);
 		TLV_PUT_U64(sctx, BTRFS_SEND_A_FILE_OFFSET, offset);
-		TLV_PUT(sctx, BTRFS_SEND_A_DATA, sctx->read_buf, len);
+		ret = put_data_header(sctx, len);
+		if (ret < 0)
+			break;
+		memset(sctx->send_buf + sctx->send_size, 0, len);
+		sctx->send_size += len;
 		ret = send_cmd(sctx);
 		if (ret < 0)
 			break;
@@ -5086,23 +5206,20 @@ static int send_extent_data(struct send_ctx *sctx,
 			    const u64 offset,
 			    const u64 len)
 {
+	u64 read_size = max_send_read_size(sctx);
 	u64 sent = 0;
 
 	if (sctx->flags & BTRFS_SEND_FLAG_NO_FILE_DATA)
 		return send_update_extent(sctx, offset, len);
 
 	while (sent < len) {
-		u64 size = len - sent;
+		u64 size = min(len - sent, read_size);
 		int ret;
 
-		if (size > BTRFS_SEND_READ_SIZE)
-			size = BTRFS_SEND_READ_SIZE;
 		ret = send_write(sctx, offset + sent, size);
 		if (ret < 0)
 			return ret;
-		if (!ret)
-			break;
-		sent += ret;
+		sent += size;
 	}
 	return 0;
 }
@@ -5402,51 +5519,29 @@ static int send_write_or_clone(struct send_ctx *sctx,
 			       struct clone_root *clone_root)
 {
 	int ret = 0;
-	struct btrfs_file_extent_item *ei;
 	u64 offset = key->offset;
-	u64 len;
-	u8 type;
+	u64 end;
 	u64 bs = sctx->send_root->fs_info->sb->s_blocksize;
 
-	ei = btrfs_item_ptr(path->nodes[0], path->slots[0],
-			struct btrfs_file_extent_item);
-	type = btrfs_file_extent_type(path->nodes[0], ei);
-	if (type == BTRFS_FILE_EXTENT_INLINE) {
-		len = btrfs_file_extent_ram_bytes(path->nodes[0], ei);
-		/*
-		 * it is possible the inline item won't cover the whole page,
-		 * but there may be items after this page.  Make
-		 * sure to send the whole thing
-		 */
-		len = PAGE_ALIGN(len);
-	} else {
-		len = btrfs_file_extent_num_bytes(path->nodes[0], ei);
-	}
-
-	if (offset >= sctx->cur_inode_size) {
-		ret = 0;
-		goto out;
-	}
-	if (offset + len > sctx->cur_inode_size)
-		len = sctx->cur_inode_size - offset;
-	if (len == 0) {
-		ret = 0;
-		goto out;
-	}
+	end = min_t(u64, btrfs_file_extent_end(path), sctx->cur_inode_size);
+	if (offset >= end)
+		return 0;
 
-	if (clone_root && IS_ALIGNED(offset + len, bs)) {
+	if (clone_root && IS_ALIGNED(end, bs)) {
+		struct btrfs_file_extent_item *ei;
 		u64 disk_byte;
 		u64 data_offset;
 
+		ei = btrfs_item_ptr(path->nodes[0], path->slots[0],
+				    struct btrfs_file_extent_item);
 		disk_byte = btrfs_file_extent_disk_bytenr(path->nodes[0], ei);
 		data_offset = btrfs_file_extent_offset(path->nodes[0], ei);
 		ret = clone_range(sctx, clone_root, disk_byte, data_offset,
-				  offset, len);
+				  offset, end - offset);
 	} else {
-		ret = send_extent_data(sctx, offset, len);
+		ret = send_extent_data(sctx, offset, end - offset);
 	}
-	sctx->cur_inode_next_write_offset = offset + len;
-out:
+	sctx->cur_inode_next_write_offset = end;
 	return ret;
 }
 
@@ -6692,8 +6787,7 @@ static int tree_compare_item(struct btrfs_path *left_path,
  * If it detects a change, it aborts immediately.
  */
 static int btrfs_compare_trees(struct btrfs_root *left_root,
-			struct btrfs_root *right_root,
-			btrfs_changed_cb_t changed_cb, void *ctx)
+			struct btrfs_root *right_root, void *ctx)
 {
 	struct btrfs_fs_info *fs_info = left_root->fs_info;
 	int ret;
@@ -6960,8 +7054,7 @@ static int send_subvol(struct send_ctx *sctx)
 		goto out;
 
 	if (sctx->parent_root) {
-		ret = btrfs_compare_trees(sctx->send_root, sctx->parent_root,
-				changed_cb, sctx);
+		ret = btrfs_compare_trees(sctx->send_root, sctx->parent_root, sctx);
 		if (ret < 0)
 			goto out;
 		ret = finish_inode_if_needed(sctx, 1);
@@ -7087,7 +7180,7 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	u32 i;
 	u64 *clone_sources_tmp = NULL;
 	int clone_sources_to_rollback = 0;
-	unsigned alloc_size;
+	size_t alloc_size;
 	int sort_clone_roots = 0;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -7169,25 +7262,20 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 		goto out;
 	}
 
-	sctx->read_buf = kvmalloc(BTRFS_SEND_READ_SIZE, GFP_KERNEL);
-	if (!sctx->read_buf) {
-		ret = -ENOMEM;
-		goto out;
-	}
-
 	sctx->pending_dir_moves = RB_ROOT;
 	sctx->waiting_dir_moves = RB_ROOT;
 	sctx->orphan_dirs = RB_ROOT;
 
-	alloc_size = sizeof(struct clone_root) * (arg->clone_sources_count + 1);
-
-	sctx->clone_roots = kzalloc(alloc_size, GFP_KERNEL);
+	sctx->clone_roots = kvcalloc(sizeof(*sctx->clone_roots),
+				     arg->clone_sources_count + 1,
+				     GFP_KERNEL);
 	if (!sctx->clone_roots) {
 		ret = -ENOMEM;
 		goto out;
 	}
 
-	alloc_size = arg->clone_sources_count * sizeof(*arg->clone_sources);
+	alloc_size = array_size(sizeof(*arg->clone_sources),
+				arg->clone_sources_count);
 
 	if (arg->clone_sources_count) {
 		clone_sources_tmp = kvmalloc(alloc_size, GFP_KERNEL);
@@ -7378,7 +7466,6 @@ out:
 
 		kvfree(sctx->clone_roots);
 		kvfree(sctx->send_buf);
-		kvfree(sctx->read_buf);
 
 		name_cache_free(sctx);
 
diff --git a/fs/btrfs/send.h b/fs/btrfs/send.h
index ead397f7034f60..de91488b7cd003 100644
--- a/fs/btrfs/send.h
+++ b/fs/btrfs/send.h
@@ -13,7 +13,6 @@
 #define BTRFS_SEND_STREAM_VERSION 1
 
 #define BTRFS_SEND_BUF_SIZE SZ_64K
-#define BTRFS_SEND_READ_SIZE (48 * SZ_1K)
 
 enum btrfs_tlv_type {
 	BTRFS_TLV_U8,
diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c
index 475968ccbd1d1b..64099565ab8f5d 100644
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -175,10 +175,8 @@ void btrfs_clear_space_info_full(struct btrfs_fs_info *info)
 	struct list_head *head = &info->space_info;
 	struct btrfs_space_info *found;
 
-	rcu_read_lock();
-	list_for_each_entry_rcu(found, head, list)
+	list_for_each_entry(found, head, list)
 		found->full = 0;
-	rcu_read_unlock();
 }
 
 static int create_space_info(struct btrfs_fs_info *info, u64 flags)
@@ -213,7 +211,7 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags)
 	if (ret)
 		return ret;
 
-	list_add_rcu(&space_info->list, &info->space_info);
+	list_add(&space_info->list, &info->space_info);
 	if (flags & BTRFS_BLOCK_GROUP_DATA)
 		info->data_sinfo = space_info;
 
@@ -290,22 +288,13 @@ struct btrfs_space_info *btrfs_find_space_info(struct btrfs_fs_info *info,
 
 	flags &= BTRFS_BLOCK_GROUP_TYPE_MASK;
 
-	rcu_read_lock();
-	list_for_each_entry_rcu(found, head, list) {
-		if (found->flags & flags) {
-			rcu_read_unlock();
+	list_for_each_entry(found, head, list) {
+		if (found->flags & flags)
 			return found;
-		}
 	}
-	rcu_read_unlock();
 	return NULL;
 }
 
-static inline u64 calc_global_rsv_need_space(struct btrfs_block_rsv *global)
-{
-	return (global->size << 1);
-}
-
 static u64 calc_available_free_space(struct btrfs_fs_info *fs_info,
 			  struct btrfs_space_info *space_info,
 			  enum btrfs_reserve_flush_enum flush)
@@ -476,28 +465,6 @@ again:
 	up_read(&info->groups_sem);
 }
 
-static void btrfs_writeback_inodes_sb_nr(struct btrfs_fs_info *fs_info,
-					 unsigned long nr_pages, int nr_items)
-{
-	struct super_block *sb = fs_info->sb;
-
-	if (down_read_trylock(&sb->s_umount)) {
-		writeback_inodes_sb_nr(sb, nr_pages, WB_REASON_FS_FREE_SPACE);
-		up_read(&sb->s_umount);
-	} else {
-		/*
-		 * We needn't worry the filesystem going from r/w to r/o though
-		 * we don't acquire ->s_umount mutex, because the filesystem
-		 * should guarantee the delalloc inodes list be empty after
-		 * the filesystem is readonly(all dirty pages are written to
-		 * the disk).
-		 */
-		btrfs_start_delalloc_roots(fs_info, nr_items);
-		if (!current->journal_info)
-			btrfs_wait_ordered_roots(fs_info, nr_items, 0, (u64)-1);
-	}
-}
-
 static inline u64 calc_reclaim_items_nr(struct btrfs_fs_info *fs_info,
 					u64 to_reclaim)
 {
@@ -516,25 +483,33 @@ static inline u64 calc_reclaim_items_nr(struct btrfs_fs_info *fs_info,
 /*
  * shrink metadata reservation for delalloc
  */
-static void shrink_delalloc(struct btrfs_fs_info *fs_info, u64 to_reclaim,
-			    u64 orig, bool wait_ordered)
+static void shrink_delalloc(struct btrfs_fs_info *fs_info,
+			    struct btrfs_space_info *space_info,
+			    u64 to_reclaim, bool wait_ordered)
 {
-	struct btrfs_space_info *space_info;
 	struct btrfs_trans_handle *trans;
 	u64 delalloc_bytes;
 	u64 dio_bytes;
-	u64 async_pages;
 	u64 items;
 	long time_left;
-	unsigned long nr_pages;
 	int loops;
 
 	/* Calc the number of the pages we need flush for space reservation */
-	items = calc_reclaim_items_nr(fs_info, to_reclaim);
-	to_reclaim = items * EXTENT_SIZE_PER_ITEM;
+	if (to_reclaim == U64_MAX) {
+		items = U64_MAX;
+	} else {
+		/*
+		 * to_reclaim is set to however much metadata we need to
+		 * reclaim, but reclaiming that much data doesn't really track
+		 * exactly, so increase the amount to reclaim by 2x in order to
+		 * make sure we're flushing enough delalloc to hopefully reclaim
+		 * some metadata reservations.
+		 */
+		items = calc_reclaim_items_nr(fs_info, to_reclaim) * 2;
+		to_reclaim = items * EXTENT_SIZE_PER_ITEM;
+	}
 
 	trans = (struct btrfs_trans_handle *)current->journal_info;
-	space_info = btrfs_find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
 
 	delalloc_bytes = percpu_counter_sum_positive(
 						&fs_info->delalloc_bytes);
@@ -557,37 +532,17 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info, u64 to_reclaim,
 
 	loops = 0;
 	while ((delalloc_bytes || dio_bytes) && loops < 3) {
-		nr_pages = min(delalloc_bytes, to_reclaim) >> PAGE_SHIFT;
-
-		/*
-		 * Triggers inode writeback for up to nr_pages. This will invoke
-		 * ->writepages callback and trigger delalloc filling
-		 *  (btrfs_run_delalloc_range()).
-		 */
-		btrfs_writeback_inodes_sb_nr(fs_info, nr_pages, items);
+		btrfs_start_delalloc_roots(fs_info, items);
 
-		/*
-		 * We need to wait for the compressed pages to start before
-		 * we continue.
-		 */
-		async_pages = atomic_read(&fs_info->async_delalloc_pages);
-		if (!async_pages)
-			goto skip_async;
-
-		/*
-		 * Calculate how many compressed pages we want to be written
-		 * before we continue. I.e if there are more async pages than we
-		 * require wait_event will wait until nr_pages are written.
-		 */
-		if (async_pages <= nr_pages)
-			async_pages = 0;
-		else
-			async_pages -= nr_pages;
+		loops++;
+		if (wait_ordered && !trans) {
+			btrfs_wait_ordered_roots(fs_info, items, 0, (u64)-1);
+		} else {
+			time_left = schedule_timeout_killable(1);
+			if (time_left)
+				break;
+		}
 
-		wait_event(fs_info->async_submit_wait,
-			   atomic_read(&fs_info->async_delalloc_pages) <=
-			   (int)async_pages);
-skip_async:
 		spin_lock(&space_info->lock);
 		if (list_empty(&space_info->tickets) &&
 		    list_empty(&space_info->priority_tickets)) {
@@ -596,14 +551,6 @@ skip_async:
 		}
 		spin_unlock(&space_info->lock);
 
-		loops++;
-		if (wait_ordered && !trans) {
-			btrfs_wait_ordered_roots(fs_info, items, 0, (u64)-1);
-		} else {
-			time_left = schedule_timeout_killable(1);
-			if (time_left)
-				break;
-		}
 		delalloc_bytes = percpu_counter_sum_positive(
 						&fs_info->delalloc_bytes);
 		dio_bytes = percpu_counter_sum_positive(&fs_info->dio_bytes);
@@ -628,8 +575,8 @@ static int may_commit_transaction(struct btrfs_fs_info *fs_info,
 	struct btrfs_block_rsv *delayed_refs_rsv = &fs_info->delayed_refs_rsv;
 	struct btrfs_block_rsv *trans_rsv = &fs_info->trans_block_rsv;
 	struct btrfs_trans_handle *trans;
-	u64 bytes_needed;
 	u64 reclaim_bytes = 0;
+	u64 bytes_needed = 0;
 	u64 cur_free_bytes = 0;
 
 	trans = (struct btrfs_trans_handle *)current->journal_info;
@@ -649,7 +596,8 @@ static int may_commit_transaction(struct btrfs_fs_info *fs_info,
 	else if (!list_empty(&space_info->tickets))
 		ticket = list_first_entry(&space_info->tickets,
 					  struct reserve_ticket, list);
-	bytes_needed = (ticket) ? ticket->bytes : 0;
+	if (ticket)
+		bytes_needed = ticket->bytes;
 
 	if (bytes_needed > cur_free_bytes)
 		bytes_needed -= cur_free_bytes;
@@ -676,8 +624,10 @@ static int may_commit_transaction(struct btrfs_fs_info *fs_info,
 		goto commit;
 
 	/*
-	 * See if there is some space in the delayed insertion reservation for
-	 * this reservation.
+	 * See if there is some space in the delayed insertion reserve for this
+	 * reservation.  If the space_info's don't match (like for DATA or
+	 * SYSTEM) then just go enospc, reclaiming this space won't recover any
+	 * space to satisfy those reservations.
 	 */
 	if (space_info != delayed_rsv->space_info)
 		goto enospc;
@@ -742,7 +692,7 @@ static void flush_space(struct btrfs_fs_info *fs_info,
 		break;
 	case FLUSH_DELALLOC:
 	case FLUSH_DELALLOC_WAIT:
-		shrink_delalloc(fs_info, num_bytes * 2, num_bytes,
+		shrink_delalloc(fs_info, space_info, num_bytes,
 				state == FLUSH_DELALLOC_WAIT);
 		break;
 	case FLUSH_DELAYED_REFS_NR:
@@ -767,7 +717,7 @@ static void flush_space(struct btrfs_fs_info *fs_info,
 			break;
 		}
 		ret = btrfs_chunk_alloc(trans,
-				btrfs_metadata_alloc_profile(fs_info),
+				btrfs_get_alloc_profile(fs_info, space_info->flags),
 				(state == ALLOC_CHUNK) ? CHUNK_ALLOC_NO_FORCE :
 					CHUNK_ALLOC_FORCE);
 		btrfs_end_transaction(trans);
@@ -1037,9 +987,132 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 	} while (flush_state <= COMMIT_TRANS);
 }
 
-void btrfs_init_async_reclaim_work(struct work_struct *work)
+/*
+ * FLUSH_DELALLOC_WAIT:
+ *   Space is freed from flushing delalloc in one of two ways.
+ *
+ *   1) compression is on and we allocate less space than we reserved
+ *   2) we are overwriting existing space
+ *
+ *   For #1 that extra space is reclaimed as soon as the delalloc pages are
+ *   COWed, by way of btrfs_add_reserved_bytes() which adds the actual extent
+ *   length to ->bytes_reserved, and subtracts the reserved space from
+ *   ->bytes_may_use.
+ *
+ *   For #2 this is trickier.  Once the ordered extent runs we will drop the
+ *   extent in the range we are overwriting, which creates a delayed ref for
+ *   that freed extent.  This however is not reclaimed until the transaction
+ *   commits, thus the next stages.
+ *
+ * RUN_DELAYED_IPUTS
+ *   If we are freeing inodes, we want to make sure all delayed iputs have
+ *   completed, because they could have been on an inode with i_nlink == 0, and
+ *   thus have been truncated and freed up space.  But again this space is not
+ *   immediately re-usable, it comes in the form of a delayed ref, which must be
+ *   run and then the transaction must be committed.
+ *
+ * FLUSH_DELAYED_REFS
+ *   The above two cases generate delayed refs that will affect
+ *   ->total_bytes_pinned.  However this counter can be inconsistent with
+ *   reality if there are outstanding delayed refs.  This is because we adjust
+ *   the counter based solely on the current set of delayed refs and disregard
+ *   any on-disk state which might include more refs.  So for example, if we
+ *   have an extent with 2 references, but we only drop 1, we'll see that there
+ *   is a negative delayed ref count for the extent and assume that the space
+ *   will be freed, and thus increase ->total_bytes_pinned.
+ *
+ *   Running the delayed refs gives us the actual real view of what will be
+ *   freed at the transaction commit time.  This stage will not actually free
+ *   space for us, it just makes sure that may_commit_transaction() has all of
+ *   the information it needs to make the right decision.
+ *
+ * COMMIT_TRANS
+ *   This is where we reclaim all of the pinned space generated by the previous
+ *   two stages.  We will not commit the transaction if we don't think we're
+ *   likely to satisfy our request, which means if our current free space +
+ *   total_bytes_pinned < reservation we will not commit.  This is why the
+ *   previous states are actually important, to make sure we know for sure
+ *   whether committing the transaction will allow us to make progress.
+ *
+ * ALLOC_CHUNK_FORCE
+ *   For data we start with alloc chunk force, however we could have been full
+ *   before, and then the transaction commit could have freed new block groups,
+ *   so if we now have space to allocate do the force chunk allocation.
+ */
+static const enum btrfs_flush_state data_flush_states[] = {
+	FLUSH_DELALLOC_WAIT,
+	RUN_DELAYED_IPUTS,
+	FLUSH_DELAYED_REFS,
+	COMMIT_TRANS,
+	ALLOC_CHUNK_FORCE,
+};
+
+static void btrfs_async_reclaim_data_space(struct work_struct *work)
+{
+	struct btrfs_fs_info *fs_info;
+	struct btrfs_space_info *space_info;
+	u64 last_tickets_id;
+	int flush_state = 0;
+
+	fs_info = container_of(work, struct btrfs_fs_info, async_data_reclaim_work);
+	space_info = fs_info->data_sinfo;
+
+	spin_lock(&space_info->lock);
+	if (list_empty(&space_info->tickets)) {
+		space_info->flush = 0;
+		spin_unlock(&space_info->lock);
+		return;
+	}
+	last_tickets_id = space_info->tickets_id;
+	spin_unlock(&space_info->lock);
+
+	while (!space_info->full) {
+		flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE);
+		spin_lock(&space_info->lock);
+		if (list_empty(&space_info->tickets)) {
+			space_info->flush = 0;
+			spin_unlock(&space_info->lock);
+			return;
+		}
+		last_tickets_id = space_info->tickets_id;
+		spin_unlock(&space_info->lock);
+	}
+
+	while (flush_state < ARRAY_SIZE(data_flush_states)) {
+		flush_space(fs_info, space_info, U64_MAX,
+			    data_flush_states[flush_state]);
+		spin_lock(&space_info->lock);
+		if (list_empty(&space_info->tickets)) {
+			space_info->flush = 0;
+			spin_unlock(&space_info->lock);
+			return;
+		}
+
+		if (last_tickets_id == space_info->tickets_id) {
+			flush_state++;
+		} else {
+			last_tickets_id = space_info->tickets_id;
+			flush_state = 0;
+		}
+
+		if (flush_state >= ARRAY_SIZE(data_flush_states)) {
+			if (space_info->full) {
+				if (maybe_fail_all_tickets(fs_info, space_info))
+					flush_state = 0;
+				else
+					space_info->flush = 0;
+			} else {
+				flush_state = 0;
+			}
+		}
+		spin_unlock(&space_info->lock);
+	}
+}
+
+void btrfs_init_async_reclaim_work(struct btrfs_fs_info *fs_info)
 {
-	INIT_WORK(work, btrfs_async_reclaim_metadata_space);
+	INIT_WORK(&fs_info->async_reclaim_work, btrfs_async_reclaim_metadata_space);
+	INIT_WORK(&fs_info->async_data_reclaim_work, btrfs_async_reclaim_data_space);
 }
 
 static const enum btrfs_flush_state priority_flush_states[] = {
@@ -1089,6 +1162,21 @@ static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info,
 	} while (flush_state < states_nr);
 }
 
+static void priority_reclaim_data_space(struct btrfs_fs_info *fs_info,
+					struct btrfs_space_info *space_info,
+					struct reserve_ticket *ticket)
+{
+	while (!space_info->full) {
+		flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE);
+		spin_lock(&space_info->lock);
+		if (ticket->bytes == 0) {
+			spin_unlock(&space_info->lock);
+			return;
+		}
+		spin_unlock(&space_info->lock);
+	}
+}
+
 static void wait_reserve_ticket(struct btrfs_fs_info *fs_info,
 				struct btrfs_space_info *space_info,
 				struct reserve_ticket *ticket)
@@ -1141,6 +1229,7 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 	int ret;
 
 	switch (flush) {
+	case BTRFS_RESERVE_FLUSH_DATA:
 	case BTRFS_RESERVE_FLUSH_ALL:
 	case BTRFS_RESERVE_FLUSH_ALL_STEAL:
 		wait_reserve_ticket(fs_info, space_info, ticket);
@@ -1155,6 +1244,9 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 						evict_flush_states,
 						ARRAY_SIZE(evict_flush_states));
 		break;
+	case BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE:
+		priority_reclaim_data_space(fs_info, space_info, ticket);
+		break;
 	default:
 		ASSERT(0);
 		break;
@@ -1214,11 +1306,11 @@ static inline bool is_normal_flushing(enum btrfs_reserve_flush_enum flush)
  * regain reservations will be made and this will fail if there is not enough
  * space already.
  */
-static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
-				    struct btrfs_space_info *space_info,
-				    u64 orig_bytes,
-				    enum btrfs_reserve_flush_enum flush)
+static int __reserve_bytes(struct btrfs_fs_info *fs_info,
+			   struct btrfs_space_info *space_info, u64 orig_bytes,
+			   enum btrfs_reserve_flush_enum flush)
 {
+	struct work_struct *async_work;
 	struct reserve_ticket ticket;
 	u64 used;
 	int ret = 0;
@@ -1227,6 +1319,11 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 	ASSERT(orig_bytes);
 	ASSERT(!current->journal_info || flush != BTRFS_RESERVE_FLUSH_ALL);
 
+	if (flush == BTRFS_RESERVE_FLUSH_DATA)
+		async_work = &fs_info->async_data_reclaim_work;
+	else
+		async_work = &fs_info->async_reclaim_work;
+
 	spin_lock(&space_info->lock);
 	ret = -ENOSPC;
 	used = btrfs_space_info_used(space_info, true);
@@ -1268,7 +1365,8 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 		init_waitqueue_head(&ticket.wait);
 		ticket.steal = (flush == BTRFS_RESERVE_FLUSH_ALL_STEAL);
 		if (flush == BTRFS_RESERVE_FLUSH_ALL ||
-		    flush == BTRFS_RESERVE_FLUSH_ALL_STEAL) {
+		    flush == BTRFS_RESERVE_FLUSH_ALL_STEAL ||
+		    flush == BTRFS_RESERVE_FLUSH_DATA) {
 			list_add_tail(&ticket.list, &space_info->tickets);
 			if (!space_info->flush) {
 				space_info->flush = 1;
@@ -1276,8 +1374,7 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 							  space_info->flags,
 							  orig_bytes, flush,
 							  "enospc");
-				queue_work(system_unbound_wq,
-					   &fs_info->async_reclaim_work);
+				queue_work(system_unbound_wq, async_work);
 			}
 		} else {
 			list_add_tail(&ticket.list,
@@ -1329,8 +1426,7 @@ int btrfs_reserve_metadata_bytes(struct btrfs_root *root,
 	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
 	int ret;
 
-	ret = __reserve_metadata_bytes(fs_info, block_rsv->space_info,
-				       orig_bytes, flush);
+	ret = __reserve_bytes(fs_info, block_rsv->space_info, orig_bytes, flush);
 	if (ret == -ENOSPC &&
 	    unlikely(root->orphan_cleanup_state == ORPHAN_CLEANUP_STARTED)) {
 		if (block_rsv != global_rsv &&
@@ -1348,3 +1444,32 @@ int btrfs_reserve_metadata_bytes(struct btrfs_root *root,
 	}
 	return ret;
 }
+
+/**
+ * btrfs_reserve_data_bytes - try to reserve data bytes for an allocation
+ * @fs_info - the filesystem
+ * @bytes - the number of bytes we need
+ * @flush - how we are allowed to flush
+ *
+ * This will reserve bytes from the data space info.  If there is not enough
+ * space then we will attempt to flush space as specified by flush.
+ */
+int btrfs_reserve_data_bytes(struct btrfs_fs_info *fs_info, u64 bytes,
+			     enum btrfs_reserve_flush_enum flush)
+{
+	struct btrfs_space_info *data_sinfo = fs_info->data_sinfo;
+	int ret;
+
+	ASSERT(flush == BTRFS_RESERVE_FLUSH_DATA ||
+	       flush == BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE);
+	ASSERT(!current->journal_info || flush != BTRFS_RESERVE_FLUSH_DATA);
+
+	ret = __reserve_bytes(fs_info, data_sinfo, bytes, flush);
+	if (ret == -ENOSPC) {
+		trace_btrfs_space_reservation(fs_info, "space_info:enospc",
+					      data_sinfo->flags, bytes, 1);
+		if (btrfs_test_opt(fs_info, ENOSPC_DEBUG))
+			btrfs_dump_space_info(fs_info, data_sinfo, bytes, 0);
+	}
+	return ret;
+}
diff --git a/fs/btrfs/space-info.h b/fs/btrfs/space-info.h
index c3c64019950aa0..5646393b928c9c 100644
--- a/fs/btrfs/space-info.h
+++ b/fs/btrfs/space-info.h
@@ -149,5 +149,7 @@ static inline void btrfs_space_info_free_bytes_may_use(
 	btrfs_try_granting_tickets(fs_info, space_info);
 	spin_unlock(&space_info->lock);
 }
+int btrfs_reserve_data_bytes(struct btrfs_fs_info *fs_info, u64 bytes,
+			     enum btrfs_reserve_flush_enum flush);
 
 #endif /* BTRFS_SPACE_INFO_H */
diff --git a/fs/btrfs/struct-funcs.c b/fs/btrfs/struct-funcs.c
index 079b059818e9a5..c46be27be700e7 100644
--- a/fs/btrfs/struct-funcs.c
+++ b/fs/btrfs/struct-funcs.c
@@ -7,16 +7,6 @@
 
 #include "ctree.h"
 
-static inline u8 get_unaligned_le8(const void *p)
-{
-       return *(u8 *)p;
-}
-
-static inline void put_unaligned_le8(u8 val, void *p)
-{
-       *(u8 *)p = val;
-}
-
 static bool check_setget_bounds(const struct extent_buffer *eb,
 				const void *ptr, unsigned off, int size)
 {
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 25967ecaaf0af9..8840a4fa81eb78 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -1871,6 +1871,7 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 		 * the filesystem is busy.
 		 */
 		cancel_work_sync(&fs_info->async_reclaim_work);
+		cancel_work_sync(&fs_info->async_data_reclaim_work);
 
 		btrfs_discard_cleanup(fs_info);
 
@@ -2163,8 +2164,7 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	u64 thresh = 0;
 	int mixed = 0;
 
-	rcu_read_lock();
-	list_for_each_entry_rcu(found, &fs_info->space_info, list) {
+	list_for_each_entry(found, &fs_info->space_info, list) {
 		if (found->flags & BTRFS_BLOCK_GROUP_DATA) {
 			int i;
 
@@ -2193,8 +2193,6 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 		total_used += found->disk_used;
 	}
 
-	rcu_read_unlock();
-
 	buf->f_blocks = div_u64(btrfs_super_total_bytes(disk_super), factor);
 	buf->f_blocks >>= bits;
 	buf->f_bfree = buf->f_blocks - (div_u64(total_used, factor) >> bits);
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index 5be30066563c83..279d9262b676d4 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -14,6 +14,7 @@
 #include "ctree.h"
 #include "discard.h"
 #include "disk-io.h"
+#include "send.h"
 #include "transaction.h"
 #include "sysfs.h"
 #include "volumes.h"
@@ -321,9 +322,17 @@ static ssize_t supported_checksums_show(struct kobject *kobj,
 }
 BTRFS_ATTR(static_feature, supported_checksums, supported_checksums_show);
 
+static ssize_t send_stream_version_show(struct kobject *kobj,
+					struct kobj_attribute *ka, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%d\n", BTRFS_SEND_STREAM_VERSION);
+}
+BTRFS_ATTR(static_feature, send_stream_version, send_stream_version_show);
+
 static struct attribute *btrfs_supported_static_feature_attrs[] = {
 	BTRFS_ATTR_PTR(static_feature, rmdir_subvol),
 	BTRFS_ATTR_PTR(static_feature, supported_checksums),
+	BTRFS_ATTR_PTR(static_feature, send_stream_version),
 	NULL
 };
 
@@ -809,6 +818,42 @@ static ssize_t btrfs_checksum_show(struct kobject *kobj,
 
 BTRFS_ATTR(, checksum, btrfs_checksum_show);
 
+static ssize_t btrfs_exclusive_operation_show(struct kobject *kobj,
+		struct kobj_attribute *a, char *buf)
+{
+	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
+	const char *str;
+
+	switch (READ_ONCE(fs_info->exclusive_operation)) {
+		case  BTRFS_EXCLOP_NONE:
+			str = "none\n";
+			break;
+		case BTRFS_EXCLOP_BALANCE:
+			str = "balance\n";
+			break;
+		case BTRFS_EXCLOP_DEV_ADD:
+			str = "device add\n";
+			break;
+		case BTRFS_EXCLOP_DEV_REMOVE:
+			str = "device remove\n";
+			break;
+		case BTRFS_EXCLOP_DEV_REPLACE:
+			str = "device replace\n";
+			break;
+		case BTRFS_EXCLOP_RESIZE:
+			str = "resize\n";
+			break;
+		case BTRFS_EXCLOP_SWAP_ACTIVATE:
+			str = "swap activate\n";
+			break;
+		default:
+			str = "UNKNOWN\n";
+			break;
+	}
+	return scnprintf(buf, PAGE_SIZE, "%s", str);
+}
+BTRFS_ATTR(, exclusive_operation, btrfs_exclusive_operation_show);
+
 static const struct attribute *btrfs_attrs[] = {
 	BTRFS_ATTR_PTR(, label),
 	BTRFS_ATTR_PTR(, nodesize),
@@ -817,6 +862,7 @@ static const struct attribute *btrfs_attrs[] = {
 	BTRFS_ATTR_PTR(, quota_override),
 	BTRFS_ATTR_PTR(, metadata_uuid),
 	BTRFS_ATTR_PTR(, checksum),
+	BTRFS_ATTR_PTR(, exclusive_operation),
 	NULL,
 };
 
@@ -935,12 +981,24 @@ void btrfs_sysfs_remove_fsid(struct btrfs_fs_devices *fs_devs)
 	}
 }
 
+static void btrfs_sysfs_remove_fs_devices(struct btrfs_fs_devices *fs_devices)
+{
+	struct btrfs_device *device;
+	struct btrfs_fs_devices *seed;
+
+	list_for_each_entry(device, &fs_devices->devices, dev_list)
+		btrfs_sysfs_remove_device(device);
+
+	list_for_each_entry(seed, &fs_devices->seed_list, seed_list) {
+		list_for_each_entry(device, &seed->devices, dev_list)
+			btrfs_sysfs_remove_device(device);
+	}
+}
+
 void btrfs_sysfs_remove_mounted(struct btrfs_fs_info *fs_info)
 {
 	struct kobject *fsid_kobj = &fs_info->fs_devices->fsid_kobj;
 
-	btrfs_reset_fs_info_ptr(fs_info);
-
 	sysfs_remove_link(fsid_kobj, "bdi");
 
 	if (fs_info->space_info_kobj) {
@@ -964,7 +1022,7 @@ void btrfs_sysfs_remove_mounted(struct btrfs_fs_info *fs_info)
 	addrm_unknown_feature_attrs(fs_info, false);
 	sysfs_remove_group(fsid_kobj, &btrfs_feature_attr_group);
 	sysfs_remove_files(fsid_kobj, btrfs_attrs);
-	btrfs_sysfs_remove_devices_dir(fs_info->fs_devices, NULL);
+	btrfs_sysfs_remove_fs_devices(fs_info->fs_devices);
 }
 
 static const char * const btrfs_feature_set_names[FEAT_MAX] = {
@@ -973,7 +1031,7 @@ static const char * const btrfs_feature_set_names[FEAT_MAX] = {
 	[FEAT_INCOMPAT]	 = "incompat",
 };
 
-const char * const btrfs_feature_set_name(enum btrfs_feature_set set)
+const char *btrfs_feature_set_name(enum btrfs_feature_set set)
 {
 	return btrfs_feature_set_names[set];
 }
@@ -1079,17 +1137,38 @@ void btrfs_sysfs_add_block_group_type(struct btrfs_block_group *cache)
 
 	rkobj->flags = cache->flags;
 	kobject_init(&rkobj->kobj, &btrfs_raid_ktype);
+
+	/*
+	 * We call this either on mount, or if we've created a block group for a
+	 * new index type while running (i.e. when restriping).  The running
+	 * case is tricky because we could race with other threads, so we need
+	 * to have this check to make sure we didn't already init the kobject.
+	 *
+	 * We don't have to protect on the free side because it only happens on
+	 * unmount.
+	 */
+	spin_lock(&space_info->lock);
+	if (space_info->block_group_kobjs[index]) {
+		spin_unlock(&space_info->lock);
+		kobject_put(&rkobj->kobj);
+		return;
+	} else {
+		space_info->block_group_kobjs[index] = &rkobj->kobj;
+	}
+	spin_unlock(&space_info->lock);
+
 	ret = kobject_add(&rkobj->kobj, &space_info->kobj, "%s",
 			  btrfs_bg_type_to_raid_name(rkobj->flags));
 	memalloc_nofs_restore(nofs_flag);
 	if (ret) {
+		spin_lock(&space_info->lock);
+		space_info->block_group_kobjs[index] = NULL;
+		spin_unlock(&space_info->lock);
 		kobject_put(&rkobj->kobj);
 		btrfs_warn(fs_info,
 			"failed to add kobject for block cache, ignoring");
 		return;
 	}
-
-	space_info->block_group_kobjs[index] = &rkobj->kobj;
 }
 
 /*
@@ -1151,52 +1230,30 @@ int btrfs_sysfs_add_space_info_type(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
-/* when one_device is NULL, it removes all device links */
-
-int btrfs_sysfs_remove_devices_dir(struct btrfs_fs_devices *fs_devices,
-		struct btrfs_device *one_device)
+void btrfs_sysfs_remove_device(struct btrfs_device *device)
 {
 	struct hd_struct *disk;
 	struct kobject *disk_kobj;
+	struct kobject *devices_kobj;
 
-	if (!fs_devices->devices_kobj)
-		return -EINVAL;
-
-	if (one_device) {
-		if (one_device->bdev) {
-			disk = one_device->bdev->bd_part;
-			disk_kobj = &part_to_dev(disk)->kobj;
-			sysfs_remove_link(fs_devices->devices_kobj,
-					  disk_kobj->name);
-		}
-
-		if (one_device->devid_kobj.state_initialized) {
-			kobject_del(&one_device->devid_kobj);
-			kobject_put(&one_device->devid_kobj);
-
-			wait_for_completion(&one_device->kobj_unregister);
-		}
+	/*
+	 * Seed fs_devices devices_kobj aren't used, fetch kobject from the
+	 * fs_info::fs_devices.
+	 */
+	devices_kobj = device->fs_info->fs_devices->devices_kobj;
+	ASSERT(devices_kobj);
 
-		return 0;
+	if (device->bdev) {
+		disk = device->bdev->bd_part;
+		disk_kobj = &part_to_dev(disk)->kobj;
+		sysfs_remove_link(devices_kobj, disk_kobj->name);
 	}
 
-	list_for_each_entry(one_device, &fs_devices->devices, dev_list) {
-
-		if (one_device->bdev) {
-			disk = one_device->bdev->bd_part;
-			disk_kobj = &part_to_dev(disk)->kobj;
-			sysfs_remove_link(fs_devices->devices_kobj,
-					  disk_kobj->name);
-		}
-		if (one_device->devid_kobj.state_initialized) {
-			kobject_del(&one_device->devid_kobj);
-			kobject_put(&one_device->devid_kobj);
-
-			wait_for_completion(&one_device->kobj_unregister);
-		}
+	if (device->devid_kobj.state_initialized) {
+		kobject_del(&device->devid_kobj);
+		kobject_put(&device->devid_kobj);
+		wait_for_completion(&device->kobj_unregister);
 	}
-
-	return 0;
 }
 
 static ssize_t btrfs_devinfo_in_fs_metadata_show(struct kobject *kobj,
@@ -1277,44 +1334,80 @@ static struct kobj_type devid_ktype = {
 	.release	= btrfs_release_devid_kobj,
 };
 
-int btrfs_sysfs_add_devices_dir(struct btrfs_fs_devices *fs_devices,
-				struct btrfs_device *one_device)
+int btrfs_sysfs_add_device(struct btrfs_device *device)
 {
-	int error = 0;
-	struct btrfs_device *dev;
+	int ret;
 	unsigned int nofs_flag;
+	struct kobject *devices_kobj;
+	struct kobject *devinfo_kobj;
 
-	nofs_flag = memalloc_nofs_save();
-	list_for_each_entry(dev, &fs_devices->devices, dev_list) {
+	/*
+	 * Make sure we use the fs_info::fs_devices to fetch the kobjects even
+	 * for the seed fs_devices
+	 */
+	devices_kobj = device->fs_info->fs_devices->devices_kobj;
+	devinfo_kobj = device->fs_info->fs_devices->devinfo_kobj;
+	ASSERT(devices_kobj);
+	ASSERT(devinfo_kobj);
 
-		if (one_device && one_device != dev)
-			continue;
+	nofs_flag = memalloc_nofs_save();
 
-		if (dev->bdev) {
-			struct hd_struct *disk;
-			struct kobject *disk_kobj;
+	if (device->bdev) {
+		struct hd_struct *disk;
+		struct kobject *disk_kobj;
 
-			disk = dev->bdev->bd_part;
-			disk_kobj = &part_to_dev(disk)->kobj;
+		disk = device->bdev->bd_part;
+		disk_kobj = &part_to_dev(disk)->kobj;
 
-			error = sysfs_create_link(fs_devices->devices_kobj,
-						  disk_kobj, disk_kobj->name);
-			if (error)
-				break;
+		ret = sysfs_create_link(devices_kobj, disk_kobj, disk_kobj->name);
+		if (ret) {
+			btrfs_warn(device->fs_info,
+				"creating sysfs device link for devid %llu failed: %d",
+				device->devid, ret);
+			goto out;
 		}
+	}
 
-		init_completion(&dev->kobj_unregister);
-		error = kobject_init_and_add(&dev->devid_kobj, &devid_ktype,
-					     fs_devices->devinfo_kobj, "%llu",
-					     dev->devid);
-		if (error) {
-			kobject_put(&dev->devid_kobj);
-			break;
-		}
+	init_completion(&device->kobj_unregister);
+	ret = kobject_init_and_add(&device->devid_kobj, &devid_ktype,
+				   devinfo_kobj, "%llu", device->devid);
+	if (ret) {
+		kobject_put(&device->devid_kobj);
+		btrfs_warn(device->fs_info,
+			   "devinfo init for devid %llu failed: %d",
+			   device->devid, ret);
 	}
+
+out:
 	memalloc_nofs_restore(nofs_flag);
+	return ret;
+}
 
-	return error;
+static int btrfs_sysfs_add_fs_devices(struct btrfs_fs_devices *fs_devices)
+{
+	int ret;
+	struct btrfs_device *device;
+	struct btrfs_fs_devices *seed;
+
+	list_for_each_entry(device, &fs_devices->devices, dev_list) {
+		ret = btrfs_sysfs_add_device(device);
+		if (ret)
+			goto fail;
+	}
+
+	list_for_each_entry(seed, &fs_devices->seed_list, seed_list) {
+		list_for_each_entry(device, &seed->devices, dev_list) {
+			ret = btrfs_sysfs_add_device(device);
+			if (ret)
+				goto fail;
+		}
+	}
+
+	return 0;
+
+fail:
+	btrfs_sysfs_remove_fs_devices(fs_devices);
+	return ret;
 }
 
 void btrfs_kobject_uevent(struct block_device *bdev, enum kobject_action action)
@@ -1328,8 +1421,8 @@ void btrfs_kobject_uevent(struct block_device *bdev, enum kobject_action action)
 			&disk_to_dev(bdev->bd_disk)->kobj);
 }
 
-void btrfs_sysfs_update_sprout_fsid(struct btrfs_fs_devices *fs_devices,
-				    const u8 *fsid)
+void btrfs_sysfs_update_sprout_fsid(struct btrfs_fs_devices *fs_devices)
+
 {
 	char fsid_buf[BTRFS_UUID_UNPARSED_SIZE];
 
@@ -1337,7 +1430,7 @@ void btrfs_sysfs_update_sprout_fsid(struct btrfs_fs_devices *fs_devices,
 	 * Sprouting changes fsid of the mounted filesystem, rename the fsid
 	 * directory
 	 */
-	snprintf(fsid_buf, BTRFS_UUID_UNPARSED_SIZE, "%pU", fsid);
+	snprintf(fsid_buf, BTRFS_UUID_UNPARSED_SIZE, "%pU", fs_devices->fsid);
 	if (kobject_rename(&fs_devices->fsid_kobj, fsid_buf))
 		btrfs_warn(fs_devices->fs_info,
 				"sysfs: failed to create fsid for sprout");
@@ -1404,15 +1497,13 @@ int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info)
 	struct btrfs_fs_devices *fs_devs = fs_info->fs_devices;
 	struct kobject *fsid_kobj = &fs_devs->fsid_kobj;
 
-	btrfs_set_fs_info_ptr(fs_info);
-
-	error = btrfs_sysfs_add_devices_dir(fs_devs, NULL);
+	error = btrfs_sysfs_add_fs_devices(fs_devs);
 	if (error)
 		return error;
 
 	error = sysfs_create_files(fsid_kobj, btrfs_attrs);
 	if (error) {
-		btrfs_sysfs_remove_devices_dir(fs_devs, NULL);
+		btrfs_sysfs_remove_fs_devices(fs_devs);
 		return error;
 	}
 
@@ -1630,12 +1721,16 @@ void btrfs_sysfs_feature_update(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_fs_devices *fs_devs;
 	struct kobject *fsid_kobj;
-	u64 features;
-	int ret;
+	u64 __maybe_unused features;
+	int __maybe_unused ret;
 
 	if (!fs_info)
 		return;
 
+	/*
+	 * See 14e46e04958df74 and e410e34fad913dd, feature bit updates are not
+	 * safe when called from some contexts (eg. balance)
+	 */
 	features = get_features(fs_info, set);
 	ASSERT(bit & supported_feature_masks[set]);
 
diff --git a/fs/btrfs/sysfs.h b/fs/btrfs/sysfs.h
index cf839c46a131bb..bacef43f72672b 100644
--- a/fs/btrfs/sysfs.h
+++ b/fs/btrfs/sysfs.h
@@ -13,15 +13,12 @@ enum btrfs_feature_set {
 };
 
 char *btrfs_printable_features(enum btrfs_feature_set set, u64 flags);
-const char * const btrfs_feature_set_name(enum btrfs_feature_set set);
-int btrfs_sysfs_add_devices_dir(struct btrfs_fs_devices *fs_devices,
-		struct btrfs_device *one_device);
-int btrfs_sysfs_remove_devices_dir(struct btrfs_fs_devices *fs_devices,
-                struct btrfs_device *one_device);
+const char *btrfs_feature_set_name(enum btrfs_feature_set set);
+int btrfs_sysfs_add_device(struct btrfs_device *device);
+void btrfs_sysfs_remove_device(struct btrfs_device *device);
 int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs);
 void btrfs_sysfs_remove_fsid(struct btrfs_fs_devices *fs_devs);
-void btrfs_sysfs_update_sprout_fsid(struct btrfs_fs_devices *fs_devices,
-				    const u8 *fsid);
+void btrfs_sysfs_update_sprout_fsid(struct btrfs_fs_devices *fs_devices);
 void btrfs_sysfs_feature_update(struct btrfs_fs_info *fs_info,
 		u64 bit, enum btrfs_feature_set set);
 void btrfs_kobject_uevent(struct block_device *bdev, enum kobject_action action);
diff --git a/fs/btrfs/tests/extent-buffer-tests.c b/fs/btrfs/tests/extent-buffer-tests.c
index a1b9f9b5978ede..df54cdfdc2508f 100644
--- a/fs/btrfs/tests/extent-buffer-tests.c
+++ b/fs/btrfs/tests/extent-buffer-tests.c
@@ -60,8 +60,7 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	key.type = BTRFS_EXTENT_CSUM_KEY;
 	key.offset = 0;
 
-	setup_items_for_insert(root, path, &key, &value_len, value_len,
-			       value_len + sizeof(struct btrfs_item), 1);
+	setup_items_for_insert(root, path, &key, &value_len, 1);
 	item = btrfs_item_nr(0);
 	write_extent_buffer(eb, value, btrfs_item_ptr_offset(eb, 0),
 			    value_len);
diff --git a/fs/btrfs/tests/inode-tests.c b/fs/btrfs/tests/inode-tests.c
index 894a63a92236e7..e6719f7db386a7 100644
--- a/fs/btrfs/tests/inode-tests.c
+++ b/fs/btrfs/tests/inode-tests.c
@@ -33,8 +33,7 @@ static void insert_extent(struct btrfs_root *root, u64 start, u64 len,
 	key.type = BTRFS_EXTENT_DATA_KEY;
 	key.offset = start;
 
-	setup_items_for_insert(root, &path, &key, &value_len, value_len,
-			       value_len + sizeof(struct btrfs_item), 1);
+	setup_items_for_insert(root, &path, &key, &value_len, 1);
 	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
 	btrfs_set_file_extent_generation(leaf, fi, 1);
 	btrfs_set_file_extent_type(leaf, fi, type);
@@ -64,8 +63,7 @@ static void insert_inode_item_key(struct btrfs_root *root)
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 
-	setup_items_for_insert(root, &path, &key, &value_len, value_len,
-			       value_len + sizeof(struct btrfs_item), 1);
+	setup_items_for_insert(root, &path, &key, &value_len, 1);
 }
 
 /*
@@ -951,7 +949,6 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize)
 	}
 
 	BTRFS_I(inode)->root = root;
-	btrfs_test_inode_set_ops(inode);
 
 	/* [BTRFS_MAX_EXTENT_SIZE] */
 	ret = btrfs_set_extent_delalloc(BTRFS_I(inode), 0,
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index d2fc292ac61b1d..52ada47aff50d7 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -292,6 +292,8 @@ loop:
 	}
 
 	cur_trans->fs_info = fs_info;
+	atomic_set(&cur_trans->pending_ordered, 0);
+	init_waitqueue_head(&cur_trans->pending_wait);
 	atomic_set(&cur_trans->num_writers, 1);
 	extwriter_counter_init(cur_trans, type);
 	init_waitqueue_head(&cur_trans->writer_wait);
@@ -1182,7 +1184,7 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans)
 
 	eb = btrfs_lock_root_node(fs_info->tree_root);
 	ret = btrfs_cow_block(trans, fs_info->tree_root, eb, NULL,
-			      0, &eb);
+			      0, &eb, BTRFS_NESTING_COW);
 	btrfs_tree_unlock(eb);
 	free_extent_buffer(eb);
 
@@ -1587,7 +1589,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	btrfs_set_root_otransid(new_root_item, trans->transid);
 
 	old = btrfs_lock_root_node(root);
-	ret = btrfs_cow_block(trans, root, old, NULL, 0, &old);
+	ret = btrfs_cow_block(trans, root, old, NULL, 0, &old,
+			      BTRFS_NESTING_COW);
 	if (ret) {
 		btrfs_tree_unlock(old);
 		free_extent_buffer(old);
@@ -2165,6 +2168,14 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 
 	btrfs_wait_delalloc_flush(trans);
 
+	/*
+	 * Wait for all ordered extents started by a fast fsync that joined this
+	 * transaction. Otherwise if this transaction commits before the ordered
+	 * extents complete we lose logged data after a power failure.
+	 */
+	wait_event(cur_trans->pending_wait,
+		   atomic_read(&cur_trans->pending_ordered) == 0);
+
 	btrfs_scrub_pause(fs_info);
 	/*
 	 * Ok now we need to make sure to block out any other joins while we
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index d60b055b8695a9..858d9153a1cd14 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -85,6 +85,13 @@ struct btrfs_transaction {
 	spinlock_t dropped_roots_lock;
 	struct btrfs_delayed_ref_root delayed_refs;
 	struct btrfs_fs_info *fs_info;
+
+	/*
+	 * Number of ordered extents the transaction must wait for before
+	 * committing. These are ordered extents started by a fast fsync.
+	 */
+	atomic_t pending_ordered;
+	wait_queue_head_t pending_wait;
 };
 
 #define __TRANS_FREEZABLE	(1U << 0)
@@ -105,6 +112,7 @@ struct btrfs_transaction {
 #define TRANS_EXTWRITERS	(__TRANS_START | __TRANS_ATTACH)
 
 #define BTRFS_SEND_TRANS_STUB	((void *)1)
+#define BTRFS_DIO_SYNC_STUB	((void *)2)
 
 struct btrfs_trans_handle {
 	u64 transid;
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 7b1fee630f9785..f0ffd5ee77bd57 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -1035,7 +1035,7 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key,
 			   int slot)
 {
 	struct btrfs_fs_info *fs_info = leaf->fs_info;
-	struct btrfs_root_item ri;
+	struct btrfs_root_item ri = { 0 };
 	const u64 valid_root_flags = BTRFS_ROOT_SUBVOL_RDONLY |
 				     BTRFS_ROOT_SUBVOL_DEAD;
 	int ret;
@@ -1044,14 +1044,21 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key,
 	if (ret < 0)
 		return ret;
 
-	if (btrfs_item_size_nr(leaf, slot) != sizeof(ri)) {
+	if (btrfs_item_size_nr(leaf, slot) != sizeof(ri) &&
+	    btrfs_item_size_nr(leaf, slot) != btrfs_legacy_root_item_size()) {
 		generic_err(leaf, slot,
-			    "invalid root item size, have %u expect %zu",
-			    btrfs_item_size_nr(leaf, slot), sizeof(ri));
+			    "invalid root item size, have %u expect %zu or %u",
+			    btrfs_item_size_nr(leaf, slot), sizeof(ri),
+			    btrfs_legacy_root_item_size());
 	}
 
+	/*
+	 * For legacy root item, the members starting at generation_v2 will be
+	 * all filled with 0.
+	 * And since we allow geneartion_v2 as 0, it will still pass the check.
+	 */
 	read_extent_buffer(leaf, &ri, btrfs_item_ptr_offset(leaf, slot),
-			   sizeof(ri));
+			   btrfs_item_size_nr(leaf, slot));
 
 	/* Generation related */
 	if (btrfs_root_generation(&ri) >
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 39da9db3527860..56cbc1706b6f7e 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -96,8 +96,6 @@ enum {
 static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 			   struct btrfs_root *root, struct btrfs_inode *inode,
 			   int inode_only,
-			   const loff_t start,
-			   const loff_t end,
 			   struct btrfs_log_ctx *ctx);
 static int link_to_fixup_dir(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root,
@@ -176,7 +174,7 @@ static int start_log_trans(struct btrfs_trans_handle *trans,
 
 	atomic_inc(&root->log_batch);
 	atomic_inc(&root->log_writers);
-	if (ctx) {
+	if (ctx && !ctx->logging_new_name) {
 		int index = root->log_transid % 2;
 		list_add_tail(&ctx->list, &root->log_ctxs[index]);
 		ctx->log_transid = root->log_transid;
@@ -215,9 +213,7 @@ static int join_running_log_trans(struct btrfs_root *root)
  */
 void btrfs_pin_log_trans(struct btrfs_root *root)
 {
-	mutex_lock(&root->log_mutex);
 	atomic_inc(&root->log_writers);
-	mutex_unlock(&root->log_mutex);
 }
 
 /*
@@ -3615,6 +3611,7 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 	 * search and this search we'll not find the key again and can just
 	 * bail.
 	 */
+search:
 	ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0);
 	if (ret != 0)
 		goto done;
@@ -3634,6 +3631,13 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 
 			if (min_key.objectid != ino || min_key.type != key_type)
 				goto done;
+
+			if (need_resched()) {
+				btrfs_release_path(path);
+				cond_resched();
+				goto search;
+			}
+
 			ret = overwrite_item(trans, log, dst_path, src, i,
 					     &min_key);
 			if (ret) {
@@ -4082,10 +4086,14 @@ static int extent_cmp(void *priv, struct list_head *a, struct list_head *b)
 static int log_extent_csums(struct btrfs_trans_handle *trans,
 			    struct btrfs_inode *inode,
 			    struct btrfs_root *log_root,
-			    const struct extent_map *em)
+			    const struct extent_map *em,
+			    struct btrfs_log_ctx *ctx)
 {
+	struct btrfs_ordered_extent *ordered;
 	u64 csum_offset;
 	u64 csum_len;
+	u64 mod_start = em->mod_start;
+	u64 mod_len = em->mod_len;
 	LIST_HEAD(ordered_sums);
 	int ret = 0;
 
@@ -4094,13 +4102,71 @@ static int log_extent_csums(struct btrfs_trans_handle *trans,
 	    em->block_start == EXTENT_MAP_HOLE)
 		return 0;
 
+	list_for_each_entry(ordered, &ctx->ordered_extents, log_list) {
+		const u64 ordered_end = ordered->file_offset + ordered->num_bytes;
+		const u64 mod_end = mod_start + mod_len;
+		struct btrfs_ordered_sum *sums;
+
+		if (mod_len == 0)
+			break;
+
+		if (ordered_end <= mod_start)
+			continue;
+		if (mod_end <= ordered->file_offset)
+			break;
+
+		/*
+		 * We are going to copy all the csums on this ordered extent, so
+		 * go ahead and adjust mod_start and mod_len in case this ordered
+		 * extent has already been logged.
+		 */
+		if (ordered->file_offset > mod_start) {
+			if (ordered_end >= mod_end)
+				mod_len = ordered->file_offset - mod_start;
+			/*
+			 * If we have this case
+			 *
+			 * |--------- logged extent ---------|
+			 *       |----- ordered extent ----|
+			 *
+			 * Just don't mess with mod_start and mod_len, we'll
+			 * just end up logging more csums than we need and it
+			 * will be ok.
+			 */
+		} else {
+			if (ordered_end < mod_end) {
+				mod_len = mod_end - ordered_end;
+				mod_start = ordered_end;
+			} else {
+				mod_len = 0;
+			}
+		}
+
+		/*
+		 * To keep us from looping for the above case of an ordered
+		 * extent that falls inside of the logged extent.
+		 */
+		if (test_and_set_bit(BTRFS_ORDERED_LOGGED_CSUM, &ordered->flags))
+			continue;
+
+		list_for_each_entry(sums, &ordered->list, list) {
+			ret = log_csums(trans, inode, log_root, sums);
+			if (ret)
+				return ret;
+		}
+	}
+
+	/* We're done, found all csums in the ordered extents. */
+	if (mod_len == 0)
+		return 0;
+
 	/* If we're compressed we have to save the entire range of csums. */
 	if (em->compress_type) {
 		csum_offset = 0;
 		csum_len = max(em->block_len, em->orig_block_len);
 	} else {
-		csum_offset = em->mod_start - em->start;
-		csum_len = em->mod_len;
+		csum_offset = mod_start - em->start;
+		csum_len = mod_len;
 	}
 
 	/* block start is already adjusted for the file extent offset. */
@@ -4140,7 +4206,7 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 	int ret;
 	int extent_inserted = 0;
 
-	ret = log_extent_csums(trans, inode, log, em);
+	ret = log_extent_csums(trans, inode, log, em, ctx);
 	if (ret)
 		return ret;
 
@@ -4342,10 +4408,10 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 				     struct btrfs_root *root,
 				     struct btrfs_inode *inode,
 				     struct btrfs_path *path,
-				     struct btrfs_log_ctx *ctx,
-				     const u64 start,
-				     const u64 end)
+				     struct btrfs_log_ctx *ctx)
 {
+	struct btrfs_ordered_extent *ordered;
+	struct btrfs_ordered_extent *tmp;
 	struct extent_map *em, *n;
 	struct list_head extents;
 	struct extent_map_tree *tree = &inode->extent_tree;
@@ -4359,23 +4425,6 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 	test_gen = root->fs_info->last_trans_committed;
 
 	list_for_each_entry_safe(em, n, &tree->modified_extents, list) {
-		/*
-		 * Skip extents outside our logging range. It's important to do
-		 * it for correctness because if we don't ignore them, we may
-		 * log them before their ordered extent completes, and therefore
-		 * we could log them without logging their respective checksums
-		 * (the checksum items are added to the csum tree at the very
-		 * end of btrfs_finish_ordered_io()). Also leave such extents
-		 * outside of our range in the list, since we may have another
-		 * ranged fsync in the near future that needs them. If an extent
-		 * outside our range corresponds to a hole, log it to avoid
-		 * leaving gaps between extents (fsck will complain when we are
-		 * not using the NO_HOLES feature).
-		 */
-		if ((em->start > end || em->start + em->len <= start) &&
-		    em->block_start != EXTENT_MAP_HOLE)
-			continue;
-
 		list_del_init(&em->list);
 		/*
 		 * Just an arbitrary number, this can be really CPU intensive
@@ -4434,8 +4483,32 @@ process:
 	btrfs_release_path(path);
 	if (!ret)
 		ret = btrfs_log_prealloc_extents(trans, inode, path);
+	if (ret)
+		return ret;
 
-	return ret;
+	/*
+	 * We have logged all extents successfully, now make sure the commit of
+	 * the current transaction waits for the ordered extents to complete
+	 * before it commits and wipes out the log trees, otherwise we would
+	 * lose data if an ordered extents completes after the transaction
+	 * commits and a power failure happens after the transaction commit.
+	 */
+	list_for_each_entry_safe(ordered, tmp, &ctx->ordered_extents, log_list) {
+		list_del_init(&ordered->log_list);
+		set_bit(BTRFS_ORDERED_LOGGED, &ordered->flags);
+
+		if (!test_bit(BTRFS_ORDERED_COMPLETE, &ordered->flags)) {
+			spin_lock_irq(&inode->ordered_tree.lock);
+			if (!test_bit(BTRFS_ORDERED_COMPLETE, &ordered->flags)) {
+				set_bit(BTRFS_ORDERED_PENDING, &ordered->flags);
+				atomic_inc(&trans->transaction->pending_ordered);
+			}
+			spin_unlock_irq(&inode->ordered_tree.lock);
+		}
+		btrfs_put_ordered_extent(ordered);
+	}
+
+	return 0;
 }
 
 static int logged_inode_size(struct btrfs_root *log, struct btrfs_inode *inode,
@@ -4841,7 +4914,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 					ret = btrfs_log_inode(trans, root,
 						      BTRFS_I(inode),
 						      LOG_OTHER_INODE_ALL,
-						      0, LLONG_MAX, ctx);
+						      ctx);
 					btrfs_add_delayed_iput(inode);
 				}
 			}
@@ -4883,7 +4956,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 		 * Check the inode's logged_trans only instead of
 		 * btrfs_inode_in_log(). This is because the last_log_commit of
 		 * the inode is not updated when we only log that it exists and
-		 * and it has the full sync bit set (see btrfs_log_inode()).
+		 * it has the full sync bit set (see btrfs_log_inode()).
 		 */
 		if (BTRFS_I(inode)->logged_trans == trans->transid) {
 			spin_unlock(&BTRFS_I(inode)->lock);
@@ -4899,7 +4972,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 		 * log with the new name before we unpin it.
 		 */
 		ret = btrfs_log_inode(trans, root, BTRFS_I(inode),
-				      LOG_OTHER_INODE, 0, LLONG_MAX, ctx);
+				      LOG_OTHER_INODE, ctx);
 		if (ret) {
 			btrfs_add_delayed_iput(inode);
 			continue;
@@ -5112,8 +5185,6 @@ next_key:
 static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 			   struct btrfs_root *root, struct btrfs_inode *inode,
 			   int inode_only,
-			   const loff_t start,
-			   const loff_t end,
 			   struct btrfs_log_ctx *ctx)
 {
 	struct btrfs_path *path;
@@ -5292,7 +5363,7 @@ log_extents:
 	}
 	if (fast_search) {
 		ret = btrfs_log_changed_extents(trans, root, inode, dst_path,
-						ctx, start, end);
+						ctx);
 		if (ret) {
 			err = ret;
 			goto out_unlock;
@@ -5301,31 +5372,8 @@ log_extents:
 		struct extent_map *em, *n;
 
 		write_lock(&em_tree->lock);
-		/*
-		 * We can't just remove every em if we're called for a ranged
-		 * fsync - that is, one that doesn't cover the whole possible
-		 * file range (0 to LLONG_MAX). This is because we can have
-		 * em's that fall outside the range we're logging and therefore
-		 * their ordered operations haven't completed yet
-		 * (btrfs_finish_ordered_io() not invoked yet). This means we
-		 * didn't get their respective file extent item in the fs/subvol
-		 * tree yet, and need to let the next fast fsync (one which
-		 * consults the list of modified extent maps) find the em so
-		 * that it logs a matching file extent item and waits for the
-		 * respective ordered operation to complete (if it's still
-		 * running).
-		 *
-		 * Removing every em outside the range we're logging would make
-		 * the next fast fsync not log their matching file extent items,
-		 * therefore making us lose data after a log replay.
-		 */
-		list_for_each_entry_safe(em, n, &em_tree->modified_extents,
-					 list) {
-			const u64 mod_end = em->mod_start + em->mod_len - 1;
-
-			if (em->mod_start >= start && mod_end <= end)
-				list_del_init(&em->list);
-		}
+		list_for_each_entry_safe(em, n, &em_tree->modified_extents, list)
+			list_del_init(&em->list);
 		write_unlock(&em_tree->lock);
 	}
 
@@ -5339,19 +5387,34 @@ log_extents:
 	}
 
 	/*
-	 * Don't update last_log_commit if we logged that an inode exists after
-	 * it was loaded to memory (full_sync bit set).
-	 * This is to prevent data loss when we do a write to the inode, then
-	 * the inode gets evicted after all delalloc was flushed, then we log
-	 * it exists (due to a rename for example) and then fsync it. This last
-	 * fsync would do nothing (not logging the extents previously written).
+	 * If we are logging that an ancestor inode exists as part of logging a
+	 * new name from a link or rename operation, don't mark the inode as
+	 * logged - otherwise if an explicit fsync is made against an ancestor,
+	 * the fsync considers the inode in the log and doesn't sync the log,
+	 * resulting in the ancestor missing after a power failure unless the
+	 * log was synced as part of an fsync against any other unrelated inode.
+	 * So keep it simple for this case and just don't flag the ancestors as
+	 * logged.
 	 */
-	spin_lock(&inode->lock);
-	inode->logged_trans = trans->transid;
-	if (inode_only != LOG_INODE_EXISTS ||
-	    !test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags))
-		inode->last_log_commit = inode->last_sub_trans;
-	spin_unlock(&inode->lock);
+	if (!ctx ||
+	    !(S_ISDIR(inode->vfs_inode.i_mode) && ctx->logging_new_name &&
+	      &inode->vfs_inode != ctx->inode)) {
+		spin_lock(&inode->lock);
+		inode->logged_trans = trans->transid;
+		/*
+		 * Don't update last_log_commit if we logged that an inode exists
+		 * after it was loaded to memory (full_sync bit set).
+		 * This is to prevent data loss when we do a write to the inode,
+		 * then the inode gets evicted after all delalloc was flushed,
+		 * then we log it exists (due to a rename for example) and then
+		 * fsync it. This last fsync would do nothing (not logging the
+		 * extents previously written).
+		 */
+		if (inode_only != LOG_INODE_EXISTS ||
+		    !test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags))
+			inode->last_log_commit = inode->last_sub_trans;
+		spin_unlock(&inode->lock);
+	}
 out_unlock:
 	mutex_unlock(&inode->log_mutex);
 
@@ -5591,7 +5654,7 @@ process_leaf:
 			if (type == BTRFS_FT_DIR || type == BTRFS_FT_SYMLINK)
 				log_mode = LOG_INODE_ALL;
 			ret = btrfs_log_inode(trans, root, BTRFS_I(di_inode),
-					      log_mode, 0, LLONG_MAX, ctx);
+					      log_mode, ctx);
 			if (!ret &&
 			    btrfs_must_commit_transaction(trans, BTRFS_I(di_inode)))
 				ret = 1;
@@ -5735,7 +5798,7 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
 			if (ctx)
 				ctx->log_new_dentries = false;
 			ret = btrfs_log_inode(trans, root, BTRFS_I(dir_inode),
-					      LOG_INODE_ALL, 0, LLONG_MAX, ctx);
+					      LOG_INODE_ALL, ctx);
 			if (!ret &&
 			    btrfs_must_commit_transaction(trans, BTRFS_I(dir_inode)))
 				ret = 1;
@@ -5786,8 +5849,7 @@ static int log_new_ancestors(struct btrfs_trans_handle *trans,
 
 		if (BTRFS_I(inode)->generation > last_committed)
 			ret = btrfs_log_inode(trans, root, BTRFS_I(inode),
-					      LOG_INODE_EXISTS,
-					      0, LLONG_MAX, ctx);
+					      LOG_INODE_EXISTS, ctx);
 		btrfs_add_delayed_iput(inode);
 		if (ret)
 			return ret;
@@ -5842,7 +5904,7 @@ static int log_new_ancestors_fast(struct btrfs_trans_handle *trans,
 
 		if (inode->generation > fs_info->last_trans_committed) {
 			ret = btrfs_log_inode(trans, root, inode,
-					LOG_INODE_EXISTS, 0, LLONG_MAX, ctx);
+					      LOG_INODE_EXISTS, ctx);
 			if (ret)
 				break;
 		}
@@ -5950,8 +6012,6 @@ out:
 static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 				  struct btrfs_inode *inode,
 				  struct dentry *parent,
-				  const loff_t start,
-				  const loff_t end,
 				  int inode_only,
 				  struct btrfs_log_ctx *ctx)
 {
@@ -6004,7 +6064,7 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 	if (ret)
 		goto end_no_trans;
 
-	ret = btrfs_log_inode(trans, root, inode, inode_only, start, end, ctx);
+	ret = btrfs_log_inode(trans, root, inode, inode_only, ctx);
 	if (ret)
 		goto end_trans;
 
@@ -6100,15 +6160,13 @@ end_no_trans:
  */
 int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
 			  struct dentry *dentry,
-			  const loff_t start,
-			  const loff_t end,
 			  struct btrfs_log_ctx *ctx)
 {
 	struct dentry *parent = dget_parent(dentry);
 	int ret;
 
 	ret = btrfs_log_inode_parent(trans, BTRFS_I(d_inode(dentry)), parent,
-				     start, end, LOG_INODE_ALL, ctx);
+				     LOG_INODE_ALL, ctx);
 	dput(parent);
 
 	return ret;
@@ -6371,26 +6429,13 @@ void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans,
 /*
  * Call this after adding a new name for a file and it will properly
  * update the log to reflect the new name.
- *
- * @ctx can not be NULL when @sync_log is false, and should be NULL when it's
- * true (because it's not used).
- *
- * Return value depends on whether @sync_log is true or false.
- * When true: returns BTRFS_NEED_TRANS_COMMIT if the transaction needs to be
- *            committed by the caller, and BTRFS_DONT_NEED_TRANS_COMMIT
- *            otherwise.
- * When false: returns BTRFS_DONT_NEED_LOG_SYNC if the caller does not need to
- *             to sync the log, BTRFS_NEED_LOG_SYNC if it needs to sync the log,
- *             or BTRFS_NEED_TRANS_COMMIT if the transaction needs to be
- *             committed (without attempting to sync the log).
  */
-int btrfs_log_new_name(struct btrfs_trans_handle *trans,
+void btrfs_log_new_name(struct btrfs_trans_handle *trans,
 			struct btrfs_inode *inode, struct btrfs_inode *old_dir,
-			struct dentry *parent,
-			bool sync_log, struct btrfs_log_ctx *ctx)
+			struct dentry *parent)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	int ret;
+	struct btrfs_log_ctx ctx;
 
 	/*
 	 * this will force the logging code to walk the dentry chain
@@ -6405,34 +6450,17 @@ int btrfs_log_new_name(struct btrfs_trans_handle *trans,
 	 */
 	if (inode->logged_trans <= fs_info->last_trans_committed &&
 	    (!old_dir || old_dir->logged_trans <= fs_info->last_trans_committed))
-		return sync_log ? BTRFS_DONT_NEED_TRANS_COMMIT :
-			BTRFS_DONT_NEED_LOG_SYNC;
-
-	if (sync_log) {
-		struct btrfs_log_ctx ctx2;
-
-		btrfs_init_log_ctx(&ctx2, &inode->vfs_inode);
-		ret = btrfs_log_inode_parent(trans, inode, parent, 0, LLONG_MAX,
-					     LOG_INODE_EXISTS, &ctx2);
-		if (ret == BTRFS_NO_LOG_SYNC)
-			return BTRFS_DONT_NEED_TRANS_COMMIT;
-		else if (ret)
-			return BTRFS_NEED_TRANS_COMMIT;
-
-		ret = btrfs_sync_log(trans, inode->root, &ctx2);
-		if (ret)
-			return BTRFS_NEED_TRANS_COMMIT;
-		return BTRFS_DONT_NEED_TRANS_COMMIT;
-	}
-
-	ASSERT(ctx);
-	ret = btrfs_log_inode_parent(trans, inode, parent, 0, LLONG_MAX,
-				     LOG_INODE_EXISTS, ctx);
-	if (ret == BTRFS_NO_LOG_SYNC)
-		return BTRFS_DONT_NEED_LOG_SYNC;
-	else if (ret)
-		return BTRFS_NEED_TRANS_COMMIT;
+		return;
 
-	return BTRFS_NEED_LOG_SYNC;
+	btrfs_init_log_ctx(&ctx, &inode->vfs_inode);
+	ctx.logging_new_name = true;
+	/*
+	 * We don't care about the return value. If we fail to log the new name
+	 * then we know the next attempt to sync the log will fallback to a full
+	 * transaction commit (due to a call to btrfs_set_log_full_commit()), so
+	 * we don't need to worry about getting a log committed that has an
+	 * inconsistent state after a rename operation.
+	 */
+	btrfs_log_inode_parent(trans, inode, parent, LOG_INODE_EXISTS, &ctx);
 }
 
diff --git a/fs/btrfs/tree-log.h b/fs/btrfs/tree-log.h
index 132e43d29034eb..731bd9c029f55f 100644
--- a/fs/btrfs/tree-log.h
+++ b/fs/btrfs/tree-log.h
@@ -16,8 +16,11 @@ struct btrfs_log_ctx {
 	int log_ret;
 	int log_transid;
 	bool log_new_dentries;
+	bool logging_new_name;
 	struct inode *inode;
 	struct list_head list;
+	/* Only used for fast fsyncs. */
+	struct list_head ordered_extents;
 };
 
 static inline void btrfs_init_log_ctx(struct btrfs_log_ctx *ctx,
@@ -26,8 +29,23 @@ static inline void btrfs_init_log_ctx(struct btrfs_log_ctx *ctx,
 	ctx->log_ret = 0;
 	ctx->log_transid = 0;
 	ctx->log_new_dentries = false;
+	ctx->logging_new_name = false;
 	ctx->inode = inode;
 	INIT_LIST_HEAD(&ctx->list);
+	INIT_LIST_HEAD(&ctx->ordered_extents);
+}
+
+static inline void btrfs_release_log_ctx_extents(struct btrfs_log_ctx *ctx)
+{
+	struct btrfs_ordered_extent *ordered;
+	struct btrfs_ordered_extent *tmp;
+
+	ASSERT(inode_is_locked(ctx->inode));
+
+	list_for_each_entry_safe(ordered, tmp, &ctx->ordered_extents, log_list) {
+		list_del_init(&ordered->log_list);
+		btrfs_put_ordered_extent(ordered);
+	}
 }
 
 static inline void btrfs_set_log_full_commit(struct btrfs_trans_handle *trans)
@@ -49,8 +67,6 @@ int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans,
 int btrfs_recover_log_trees(struct btrfs_root *tree_root);
 int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
 			  struct dentry *dentry,
-			  const loff_t start,
-			  const loff_t end,
 			  struct btrfs_log_ctx *ctx);
 int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
 				 struct btrfs_root *root,
@@ -67,16 +83,8 @@ void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
 			     int for_rename);
 void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans,
 				   struct btrfs_inode *dir);
-/* Return values for btrfs_log_new_name() */
-enum {
-	BTRFS_DONT_NEED_TRANS_COMMIT,
-	BTRFS_NEED_TRANS_COMMIT,
-	BTRFS_DONT_NEED_LOG_SYNC,
-	BTRFS_NEED_LOG_SYNC,
-};
-int btrfs_log_new_name(struct btrfs_trans_handle *trans,
+void btrfs_log_new_name(struct btrfs_trans_handle *trans,
 			struct btrfs_inode *inode, struct btrfs_inode *old_dir,
-			struct dentry *parent,
-			bool sync_log, struct btrfs_log_ctx *ctx);
+			struct dentry *parent);
 
 #endif
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 1997a7d67f22fa..58b9c419a2b6b2 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -291,8 +291,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
  *   balance_mutex
  *
  *
- * Exclusive operations, BTRFS_FS_EXCL_OP
- * ======================================
+ * Exclusive operations
+ * ====================
  *
  * Maintains the exclusivity of the following operations that apply to the
  * whole filesystem and cannot run in parallel.
@@ -318,11 +318,11 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
  * - 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.
+ * The status of exclusive operation is set and cleared atomically.
+ * During the course of Paused state, fs_info::exclusive_operation 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
+ * The exclusive status is cleared when the device operation is canceled or
  * completed.
  */
 
@@ -356,6 +356,7 @@ static struct btrfs_fs_devices *alloc_fs_devices(const u8 *fsid,
 	INIT_LIST_HEAD(&fs_devs->devices);
 	INIT_LIST_HEAD(&fs_devs->alloc_list);
 	INIT_LIST_HEAD(&fs_devs->fs_list);
+	INIT_LIST_HEAD(&fs_devs->seed_list);
 	if (fsid)
 		memcpy(fs_devs->fsid, fsid, BTRFS_FSID_SIZE);
 
@@ -406,7 +407,7 @@ void __exit btrfs_cleanup_fs_uuids(void)
  * Returned struct is not linked onto any lists and must be destroyed using
  * btrfs_free_device.
  */
-static struct btrfs_device *__alloc_device(void)
+static struct btrfs_device *__alloc_device(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_device *dev;
 
@@ -433,7 +434,8 @@ static struct btrfs_device *__alloc_device(void)
 	btrfs_device_data_ordered_init(dev);
 	INIT_RADIX_TREE(&dev->reada_zones, GFP_NOFS & ~__GFP_DIRECT_RECLAIM);
 	INIT_RADIX_TREE(&dev->reada_extents, GFP_NOFS & ~__GFP_DIRECT_RECLAIM);
-	extent_io_tree_init(NULL, &dev->alloc_state, 0, NULL);
+	extent_io_tree_init(fs_info, &dev->alloc_state,
+			    IO_TREE_DEVICE_ALLOC_STATE, NULL);
 
 	return dev;
 }
@@ -593,8 +595,6 @@ static int btrfs_free_stale_devices(const char *path,
 			btrfs_free_device(device);
 
 			ret = 0;
-			if (fs_devices->num_devices == 0)
-				break;
 		}
 		mutex_unlock(&fs_devices->device_list_mutex);
 
@@ -941,16 +941,18 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 				bdput(path_bdev);
 				mutex_unlock(&fs_devices->device_list_mutex);
 				btrfs_warn_in_rcu(device->fs_info,
-			"duplicate device fsid:devid for %pU:%llu old:%s new:%s",
-					disk_super->fsid, devid,
-					rcu_str_deref(device->name), path);
+	"duplicate device %s devid %llu generation %llu scanned by %s (%d)",
+						  path, devid, found_transid,
+						  current->comm,
+						  task_pid_nr(current));
 				return ERR_PTR(-EEXIST);
 			}
 			bdput(path_bdev);
 			btrfs_info_in_rcu(device->fs_info,
-				"device fsid %pU devid %llu moved old:%s new:%s",
-				disk_super->fsid, devid,
-				rcu_str_deref(device->name), path);
+	"devid %llu device path %s changed to %s scanned by %s (%d)",
+					  devid, rcu_str_deref(device->name),
+					  path, current->comm,
+					  task_pid_nr(current));
 		}
 
 		name = rcu_string_strdup(path, GFP_NOFS);
@@ -1035,28 +1037,21 @@ error:
 	return ERR_PTR(ret);
 }
 
-/*
- * After we have read the system tree and know devids belonging to
- * this filesystem, remove the device which does not belong there.
- */
-void btrfs_free_extra_devids(struct btrfs_fs_devices *fs_devices, int step)
+static void __btrfs_free_extra_devids(struct btrfs_fs_devices *fs_devices,
+				      int step, struct btrfs_device **latest_dev)
 {
 	struct btrfs_device *device, *next;
-	struct btrfs_device *latest_dev = NULL;
 
-	mutex_lock(&uuid_mutex);
-again:
 	/* This is the initialized path, it is safe to release the devices. */
 	list_for_each_entry_safe(device, next, &fs_devices->devices, dev_list) {
-		if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
-							&device->dev_state)) {
+		if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state)) {
 			if (!test_bit(BTRFS_DEV_STATE_REPLACE_TGT,
-			     &device->dev_state) &&
+				      &device->dev_state) &&
 			    !test_bit(BTRFS_DEV_STATE_MISSING,
 				      &device->dev_state) &&
-			     (!latest_dev ||
-			      device->generation > latest_dev->generation)) {
-				latest_dev = device;
+			    (!*latest_dev ||
+			     device->generation > (*latest_dev)->generation)) {
+				*latest_dev = device;
 			}
 			continue;
 		}
@@ -1094,10 +1089,22 @@ again:
 		btrfs_free_device(device);
 	}
 
-	if (fs_devices->seed) {
-		fs_devices = fs_devices->seed;
-		goto again;
-	}
+}
+
+/*
+ * After we have read the system tree and know devids belonging to this
+ * filesystem, remove the device which does not belong there.
+ */
+void btrfs_free_extra_devids(struct btrfs_fs_devices *fs_devices, int step)
+{
+	struct btrfs_device *latest_dev = NULL;
+	struct btrfs_fs_devices *seed_dev;
+
+	mutex_lock(&uuid_mutex);
+	__btrfs_free_extra_devids(fs_devices, step, &latest_dev);
+
+	list_for_each_entry(seed_dev, &fs_devices->seed_list, seed_list)
+		__btrfs_free_extra_devids(seed_dev, step, &latest_dev);
 
 	fs_devices->latest_bdev = latest_dev->bdev;
 
@@ -1149,47 +1156,41 @@ static void btrfs_close_one_device(struct btrfs_device *device)
 	ASSERT(atomic_read(&device->reada_in_flight) == 0);
 }
 
-static int close_fs_devices(struct btrfs_fs_devices *fs_devices)
+static void close_fs_devices(struct btrfs_fs_devices *fs_devices)
 {
 	struct btrfs_device *device, *tmp;
 
+	lockdep_assert_held(&uuid_mutex);
+
 	if (--fs_devices->opened > 0)
-		return 0;
+		return;
 
-	mutex_lock(&fs_devices->device_list_mutex);
-	list_for_each_entry_safe(device, tmp, &fs_devices->devices, dev_list) {
+	list_for_each_entry_safe(device, tmp, &fs_devices->devices, dev_list)
 		btrfs_close_one_device(device);
-	}
-	mutex_unlock(&fs_devices->device_list_mutex);
 
 	WARN_ON(fs_devices->open_devices);
 	WARN_ON(fs_devices->rw_devices);
 	fs_devices->opened = 0;
 	fs_devices->seeding = false;
-
-	return 0;
+	fs_devices->fs_info = NULL;
 }
 
-int btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
+void btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
 {
-	struct btrfs_fs_devices *seed_devices = NULL;
-	int ret;
+	LIST_HEAD(list);
+	struct btrfs_fs_devices *tmp;
 
 	mutex_lock(&uuid_mutex);
-	ret = close_fs_devices(fs_devices);
-	if (!fs_devices->opened) {
-		seed_devices = fs_devices->seed;
-		fs_devices->seed = NULL;
-	}
-	mutex_unlock(&uuid_mutex);
+	close_fs_devices(fs_devices);
+	if (!fs_devices->opened)
+		list_splice_init(&fs_devices->seed_list, &list);
 
-	while (seed_devices) {
-		fs_devices = seed_devices;
-		seed_devices = fs_devices->seed;
+	list_for_each_entry_safe(fs_devices, tmp, &list, seed_list) {
 		close_fs_devices(fs_devices);
+		list_del(&fs_devices->seed_list);
 		free_fs_devices(fs_devices);
 	}
-	return ret;
+	mutex_unlock(&uuid_mutex);
 }
 
 static int open_fs_devices(struct btrfs_fs_devices *fs_devices,
@@ -1197,17 +1198,23 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices,
 {
 	struct btrfs_device *device;
 	struct btrfs_device *latest_dev = NULL;
+	struct btrfs_device *tmp_device;
 
 	flags |= FMODE_EXCL;
 
-	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;
+	list_for_each_entry_safe(device, tmp_device, &fs_devices->devices,
+				 dev_list) {
+		int ret;
 
-		if (!latest_dev ||
-		    device->generation > latest_dev->generation)
+		ret = btrfs_open_one_device(fs_devices, device, flags, holder);
+		if (ret == 0 &&
+		    (!latest_dev || device->generation > latest_dev->generation)) {
 			latest_dev = device;
+		} else if (ret == -ENODATA) {
+			fs_devices->num_devices--;
+			list_del(&device->dev_list);
+			btrfs_free_device(device);
+		}
 	}
 	if (fs_devices->open_devices == 0)
 		return -EINVAL;
@@ -1961,16 +1968,13 @@ static struct btrfs_device * btrfs_find_next_active_device(
  * this_dev) which is active.
  */
 void __cold btrfs_assign_next_active_device(struct btrfs_device *device,
-				     struct btrfs_device *this_dev)
+					    struct btrfs_device *next_device)
 {
 	struct btrfs_fs_info *fs_info = device->fs_info;
-	struct btrfs_device *next_device;
 
-	if (this_dev)
-		next_device = this_dev;
-	else
+	if (!next_device)
 		next_device = btrfs_find_next_active_device(fs_info->fs_devices,
-								device);
+							    device);
 	ASSERT(next_device);
 
 	if (fs_info->sb->s_bdev &&
@@ -2040,7 +2044,7 @@ void btrfs_scratch_superblocks(struct btrfs_fs_info *fs_info,
 }
 
 int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
-		u64 devid)
+		    u64 devid)
 {
 	struct btrfs_device *device;
 	struct btrfs_fs_devices *cur_devices;
@@ -2144,7 +2148,7 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	if (device->bdev) {
 		cur_devices->open_devices--;
 		/* remove sysfs entry */
-		btrfs_sysfs_remove_devices_dir(fs_devices, device);
+		btrfs_sysfs_remove_device(device);
 	}
 
 	num_devices = btrfs_super_num_devices(fs_info->super_copy) - 1;
@@ -2165,14 +2169,7 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	btrfs_free_device(device);
 
 	if (cur_devices->open_devices == 0) {
-		while (fs_devices) {
-			if (fs_devices->seed == cur_devices) {
-				fs_devices->seed = cur_devices->seed;
-				break;
-			}
-			fs_devices = fs_devices->seed;
-		}
-		cur_devices->seed = NULL;
+		list_del_init(&cur_devices->seed_list);
 		close_fs_devices(cur_devices);
 		free_fs_devices(cur_devices);
 	}
@@ -2221,7 +2218,6 @@ void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_device *srcdev)
 
 void btrfs_rm_dev_replace_free_srcdev(struct btrfs_device *srcdev)
 {
-	struct btrfs_fs_info *fs_info = srcdev->fs_info;
 	struct btrfs_fs_devices *fs_devices = srcdev->fs_devices;
 
 	mutex_lock(&uuid_mutex);
@@ -2232,8 +2228,6 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_device *srcdev)
 
 	/* if this is no devs we rather delete the fs_devices */
 	if (!fs_devices->num_devices) {
-		struct btrfs_fs_devices *tmp_fs_devices;
-
 		/*
 		 * On a mounted FS, num_devices can't be zero unless it's a
 		 * seed. In case of a seed device being replaced, the replace
@@ -2242,15 +2236,7 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_device *srcdev)
 		 */
 		ASSERT(fs_devices->seeding);
 
-		tmp_fs_devices = fs_info->fs_devices;
-		while (tmp_fs_devices) {
-			if (tmp_fs_devices->seed == fs_devices) {
-				tmp_fs_devices->seed = fs_devices->seed;
-				break;
-			}
-			tmp_fs_devices = tmp_fs_devices->seed;
-		}
-		fs_devices->seed = NULL;
+		list_del_init(&fs_devices->seed_list);
 		close_fs_devices(fs_devices);
 		free_fs_devices(fs_devices);
 	}
@@ -2263,7 +2249,7 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev)
 
 	mutex_lock(&fs_devices->device_list_mutex);
 
-	btrfs_sysfs_remove_devices_dir(fs_devices, tgtdev);
+	btrfs_sysfs_remove_device(tgtdev);
 
 	if (tgtdev->bdev)
 		fs_devices->open_devices--;
@@ -2372,10 +2358,20 @@ static int btrfs_prepare_sprout(struct btrfs_fs_info *fs_info)
 	if (!fs_devices->seeding)
 		return -EINVAL;
 
+	/*
+	 * Private copy of the seed devices, anchored at
+	 * fs_info->fs_devices->seed_list
+	 */
 	seed_devices = alloc_fs_devices(NULL, NULL);
 	if (IS_ERR(seed_devices))
 		return PTR_ERR(seed_devices);
 
+	/*
+	 * It's necessary to retain a copy of the original seed fs_devices in
+	 * fs_uuids so that filesystems which have been seeded can successfully
+	 * reference the seed device from open_seed_devices. This also supports
+	 * multiple fs seed.
+	 */
 	old_devices = clone_fs_devices(fs_devices);
 	if (IS_ERR(old_devices)) {
 		kfree(seed_devices);
@@ -2396,16 +2392,12 @@ static int btrfs_prepare_sprout(struct btrfs_fs_info *fs_info)
 	list_for_each_entry(device, &seed_devices->devices, dev_list)
 		device->fs_devices = seed_devices;
 
-	mutex_lock(&fs_info->chunk_mutex);
-	list_splice_init(&fs_devices->alloc_list, &seed_devices->alloc_list);
-	mutex_unlock(&fs_info->chunk_mutex);
-
 	fs_devices->seeding = false;
 	fs_devices->num_devices = 0;
 	fs_devices->open_devices = 0;
 	fs_devices->missing_devices = 0;
 	fs_devices->rotating = false;
-	fs_devices->seed = seed_devices;
+	list_add(&seed_devices->seed_list, &fs_devices->seed_list);
 
 	generate_random_uuid(fs_devices->fsid);
 	memcpy(fs_devices->metadata_uuid, fs_devices->fsid, BTRFS_FSID_SIZE);
@@ -2508,7 +2500,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	u64 orig_super_num_devices;
 	int seeding_dev = 0;
 	int ret = 0;
-	bool unlocked = false;
+	bool locked = false;
 
 	if (sb_rdonly(sb) && !fs_devices->seeding)
 		return -EROFS;
@@ -2522,20 +2514,20 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 		seeding_dev = 1;
 		down_write(&sb->s_umount);
 		mutex_lock(&uuid_mutex);
+		locked = true;
 	}
 
-	filemap_write_and_wait(bdev->bd_inode->i_mapping);
+	sync_blockdev(bdev);
 
-	mutex_lock(&fs_devices->device_list_mutex);
-	list_for_each_entry(device, &fs_devices->devices, dev_list) {
+	rcu_read_lock();
+	list_for_each_entry_rcu(device, &fs_devices->devices, dev_list) {
 		if (device->bdev == bdev) {
 			ret = -EEXIST;
-			mutex_unlock(
-				&fs_devices->device_list_mutex);
+			rcu_read_unlock();
 			goto error;
 		}
 	}
-	mutex_unlock(&fs_devices->device_list_mutex);
+	rcu_read_unlock();
 
 	device = btrfs_alloc_device(fs_info, NULL, NULL);
 	if (IS_ERR(device)) {
@@ -2610,9 +2602,6 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	btrfs_set_super_num_devices(fs_info->super_copy,
 				    orig_super_num_devices + 1);
 
-	/* add sysfs device entry */
-	btrfs_sysfs_add_devices_dir(fs_devices, device);
-
 	/*
 	 * we've got more storage, clear any full flags on the space
 	 * infos
@@ -2620,6 +2609,10 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	btrfs_clear_space_info_full(fs_info);
 
 	mutex_unlock(&fs_info->chunk_mutex);
+
+	/* Add sysfs device entry */
+	btrfs_sysfs_add_device(device);
+
 	mutex_unlock(&fs_devices->device_list_mutex);
 
 	if (seeding_dev) {
@@ -2645,8 +2638,11 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 			goto error_sysfs;
 		}
 
-		btrfs_sysfs_update_sprout_fsid(fs_devices,
-				fs_info->fs_devices->fsid);
+		/*
+		 * fs_devices now represents the newly sprouted filesystem and
+		 * its fsid has been changed by btrfs_prepare_sprout
+		 */
+		btrfs_sysfs_update_sprout_fsid(fs_devices);
 	}
 
 	ret = btrfs_commit_transaction(trans);
@@ -2654,7 +2650,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	if (seeding_dev) {
 		mutex_unlock(&uuid_mutex);
 		up_write(&sb->s_umount);
-		unlocked = true;
+		locked = false;
 
 		if (ret) /* transaction commit */
 			return ret;
@@ -2689,7 +2685,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	return ret;
 
 error_sysfs:
-	btrfs_sysfs_remove_devices_dir(fs_devices, device);
+	btrfs_sysfs_remove_device(device);
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
 	mutex_lock(&fs_info->chunk_mutex);
 	list_del_rcu(&device->dev_list);
@@ -2715,7 +2711,7 @@ error_free_device:
 	btrfs_free_device(device);
 error:
 	blkdev_put(bdev, FMODE_EXCL);
-	if (seeding_dev && !unlocked) {
+	if (locked) {
 		mutex_unlock(&uuid_mutex);
 		up_write(&sb->s_umount);
 	}
@@ -4042,7 +4038,7 @@ int btrfs_balance(struct btrfs_fs_info *fs_info,
 
 	/*
 	 * rw_devices will not change at the moment, device add/delete/replace
-	 * are excluded by EXCL_OP
+	 * are exclusive
 	 */
 	num_devices = fs_info->fs_devices->rw_devices;
 
@@ -4178,7 +4174,7 @@ int btrfs_balance(struct btrfs_fs_info *fs_info,
 	if ((ret && ret != -ECANCELED && ret != -ENOSPC) ||
 	    balance_need_close(fs_info)) {
 		reset_balance_state(fs_info);
-		clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+		btrfs_exclop_finish(fs_info);
 	}
 
 	wake_up(&fs_info->balance_wait_q);
@@ -4189,7 +4185,7 @@ out:
 		reset_balance_state(fs_info);
 	else
 		kfree(bctl);
-	clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+	btrfs_exclop_finish(fs_info);
 
 	return ret;
 }
@@ -4291,7 +4287,7 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info)
 	 * 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))
+	if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_BALANCE))
 		btrfs_warn(fs_info,
 	"balance: cannot set exclusive op status, resume manually");
 
@@ -4373,7 +4369,7 @@ int btrfs_cancel_balance(struct btrfs_fs_info *fs_info)
 
 		if (fs_info->balance_ctl) {
 			reset_balance_state(fs_info);
-			clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
+			btrfs_exclop_finish(fs_info);
 			btrfs_info(fs_info, "balance: canceled");
 		}
 	}
@@ -6458,11 +6454,21 @@ struct btrfs_device *btrfs_find_device(struct btrfs_fs_devices *fs_devices,
 				       bool seed)
 {
 	struct btrfs_device *device;
+	struct btrfs_fs_devices *seed_devs;
 
-	while (fs_devices) {
+	if (!fsid || !memcmp(fs_devices->metadata_uuid, fsid, BTRFS_FSID_SIZE)) {
+		list_for_each_entry(device, &fs_devices->devices, dev_list) {
+			if (device->devid == devid &&
+			    (!uuid || memcmp(device->uuid, uuid,
+					     BTRFS_UUID_SIZE) == 0))
+				return device;
+		}
+	}
+
+	list_for_each_entry(seed_devs, &fs_devices->seed_list, seed_list) {
 		if (!fsid ||
-		    !memcmp(fs_devices->metadata_uuid, fsid, BTRFS_FSID_SIZE)) {
-			list_for_each_entry(device, &fs_devices->devices,
+		    !memcmp(seed_devs->metadata_uuid, fsid, BTRFS_FSID_SIZE)) {
+			list_for_each_entry(device, &seed_devs->devices,
 					    dev_list) {
 				if (device->devid == devid &&
 				    (!uuid || memcmp(device->uuid, uuid,
@@ -6470,11 +6476,8 @@ struct btrfs_device *btrfs_find_device(struct btrfs_fs_devices *fs_devices,
 					return device;
 			}
 		}
-		if (seed)
-			fs_devices = fs_devices->seed;
-		else
-			return NULL;
 	}
+
 	return NULL;
 }
 
@@ -6529,7 +6532,7 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 	if (WARN_ON(!devid && !fs_info))
 		return ERR_PTR(-EINVAL);
 
-	dev = __alloc_device();
+	dev = __alloc_device(fs_info);
 	if (IS_ERR(dev))
 		return dev;
 
@@ -6725,13 +6728,11 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info,
 	lockdep_assert_held(&uuid_mutex);
 	ASSERT(fsid);
 
-	fs_devices = fs_info->fs_devices->seed;
-	while (fs_devices) {
+	/* This will match only for multi-device seed fs */
+	list_for_each_entry(fs_devices, &fs_info->fs_devices->seed_list, seed_list)
 		if (!memcmp(fs_devices->fsid, fsid, BTRFS_FSID_SIZE))
 			return fs_devices;
 
-		fs_devices = fs_devices->seed;
-	}
 
 	fs_devices = find_fsid(fsid, NULL);
 	if (!fs_devices) {
@@ -6747,6 +6748,10 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info,
 		return fs_devices;
 	}
 
+	/*
+	 * Upon first call for a seed fs fsid, just create a private copy of the
+	 * respective fs_devices and anchor it at fs_info->fs_devices->seed_list
+	 */
 	fs_devices = clone_fs_devices(fs_devices);
 	if (IS_ERR(fs_devices))
 		return fs_devices;
@@ -6754,20 +6759,17 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info,
 	ret = open_fs_devices(fs_devices, FMODE_READ, fs_info->bdev_holder);
 	if (ret) {
 		free_fs_devices(fs_devices);
-		fs_devices = ERR_PTR(ret);
-		goto out;
+		return ERR_PTR(ret);
 	}
 
 	if (!fs_devices->seeding) {
 		close_fs_devices(fs_devices);
 		free_fs_devices(fs_devices);
-		fs_devices = ERR_PTR(-EINVAL);
-		goto out;
+		return ERR_PTR(-EINVAL);
 	}
 
-	fs_devices->seed = fs_info->fs_devices->seed;
-	fs_info->fs_devices->seed = fs_devices;
-out:
+	list_add(&fs_devices->seed_list, &fs_info->fs_devices->seed_list);
+
 	return fs_devices;
 }
 
@@ -7186,17 +7188,22 @@ error:
 
 void btrfs_init_devices_late(struct btrfs_fs_info *fs_info)
 {
-	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices, *seed_devs;
 	struct btrfs_device *device;
 
-	while (fs_devices) {
-		mutex_lock(&fs_devices->device_list_mutex);
-		list_for_each_entry(device, &fs_devices->devices, dev_list)
+	fs_devices->fs_info = fs_info;
+
+	mutex_lock(&fs_devices->device_list_mutex);
+	list_for_each_entry(device, &fs_devices->devices, dev_list)
+		device->fs_info = fs_info;
+
+	list_for_each_entry(seed_devs, &fs_devices->seed_list, seed_list) {
+		list_for_each_entry(device, &seed_devs->devices, dev_list)
 			device->fs_info = fs_info;
-		mutex_unlock(&fs_devices->device_list_mutex);
 
-		fs_devices = fs_devices->seed;
+		seed_devs->fs_info = fs_info;
 	}
+	mutex_unlock(&fs_devices->device_list_mutex);
 }
 
 static u64 btrfs_dev_stats_value(const struct extent_buffer *eb,
@@ -7222,17 +7229,53 @@ static void btrfs_set_dev_stats_value(struct extent_buffer *eb,
 			    sizeof(val));
 }
 
-int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info)
+static int btrfs_device_init_dev_stats(struct btrfs_device *device,
+				       struct btrfs_path *path)
 {
-	struct btrfs_key key;
-	struct btrfs_root *dev_root = fs_info->dev_root;
-	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+	struct btrfs_dev_stats_item *ptr;
 	struct extent_buffer *eb;
-	int slot;
-	int ret = 0;
+	struct btrfs_key key;
+	int item_size;
+	int i, ret, slot;
+
+	key.objectid = BTRFS_DEV_STATS_OBJECTID;
+	key.type = BTRFS_PERSISTENT_ITEM_KEY;
+	key.offset = device->devid;
+	ret = btrfs_search_slot(NULL, device->fs_info->dev_root, &key, path, 0, 0);
+	if (ret) {
+		for (i = 0; i < BTRFS_DEV_STAT_VALUES_MAX; i++)
+			btrfs_dev_stat_set(device, i, 0);
+		device->dev_stats_valid = 1;
+		btrfs_release_path(path);
+		return ret < 0 ? ret : 0;
+	}
+	slot = path->slots[0];
+	eb = path->nodes[0];
+	item_size = btrfs_item_size_nr(eb, slot);
+
+	ptr = btrfs_item_ptr(eb, slot, struct btrfs_dev_stats_item);
+
+	for (i = 0; i < BTRFS_DEV_STAT_VALUES_MAX; i++) {
+		if (item_size >= (1 + i) * sizeof(__le64))
+			btrfs_dev_stat_set(device, i,
+					   btrfs_dev_stats_value(eb, ptr, i));
+		else
+			btrfs_dev_stat_set(device, i, 0);
+	}
+
+	device->dev_stats_valid = 1;
+	btrfs_dev_stat_print_on_load(device);
+	btrfs_release_path(path);
+
+	return 0;
+}
+
+int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices, *seed_devs;
 	struct btrfs_device *device;
 	struct btrfs_path *path = NULL;
-	int i;
+	int ret = 0;
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -7240,43 +7283,22 @@ int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info)
 
 	mutex_lock(&fs_devices->device_list_mutex);
 	list_for_each_entry(device, &fs_devices->devices, dev_list) {
-		int item_size;
-		struct btrfs_dev_stats_item *ptr;
-
-		key.objectid = BTRFS_DEV_STATS_OBJECTID;
-		key.type = BTRFS_PERSISTENT_ITEM_KEY;
-		key.offset = device->devid;
-		ret = btrfs_search_slot(NULL, dev_root, &key, path, 0, 0);
-		if (ret) {
-			for (i = 0; i < BTRFS_DEV_STAT_VALUES_MAX; i++)
-				btrfs_dev_stat_set(device, i, 0);
-			device->dev_stats_valid = 1;
-			btrfs_release_path(path);
-			continue;
-		}
-		slot = path->slots[0];
-		eb = path->nodes[0];
-		item_size = btrfs_item_size_nr(eb, slot);
-
-		ptr = btrfs_item_ptr(eb, slot,
-				     struct btrfs_dev_stats_item);
-
-		for (i = 0; i < BTRFS_DEV_STAT_VALUES_MAX; i++) {
-			if (item_size >= (1 + i) * sizeof(__le64))
-				btrfs_dev_stat_set(device, i,
-					btrfs_dev_stats_value(eb, ptr, i));
-			else
-				btrfs_dev_stat_set(device, i, 0);
+		ret = btrfs_device_init_dev_stats(device, path);
+		if (ret)
+			goto out;
+	}
+	list_for_each_entry(seed_devs, &fs_devices->seed_list, seed_list) {
+		list_for_each_entry(device, &seed_devs->devices, dev_list) {
+			ret = btrfs_device_init_dev_stats(device, path);
+			if (ret)
+				goto out;
 		}
-
-		device->dev_stats_valid = 1;
-		btrfs_dev_stat_print_on_load(device);
-		btrfs_release_path(path);
 	}
+out:
 	mutex_unlock(&fs_devices->device_list_mutex);
 
 	btrfs_free_path(path);
-	return ret < 0 ? ret : 0;
+	return ret;
 }
 
 static int update_dev_stat_item(struct btrfs_trans_handle *trans,
@@ -7493,24 +7515,6 @@ void btrfs_commit_device_sizes(struct btrfs_transaction *trans)
 	mutex_unlock(&trans->fs_info->chunk_mutex);
 }
 
-void btrfs_set_fs_info_ptr(struct btrfs_fs_info *fs_info)
-{
-	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
-	while (fs_devices) {
-		fs_devices->fs_info = fs_info;
-		fs_devices = fs_devices->seed;
-	}
-}
-
-void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info)
-{
-	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
-	while (fs_devices) {
-		fs_devices->fs_info = NULL;
-		fs_devices = fs_devices->seed;
-	}
-}
-
 /*
  * Multiplicity factor for simple profiles: DUP, RAID1-like and RAID10.
  */
@@ -7591,8 +7595,11 @@ static int verify_one_dev_extent(struct btrfs_fs_info *fs_info,
 
 	/* It's possible this device is a dummy for seed device */
 	if (dev->disk_total_bytes == 0) {
-		dev = btrfs_find_device(fs_info->fs_devices->seed, devid, NULL,
-					NULL, false);
+		struct btrfs_fs_devices *devs;
+
+		devs = list_first_entry(&fs_info->fs_devices->seed_list,
+					struct btrfs_fs_devices, seed_list);
+		dev = btrfs_find_device(devs, devid, NULL, NULL, false);
 		if (!dev) {
 			btrfs_err(fs_info, "failed to find seed devid %llu",
 				  devid);
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index 302c9234f7d081..bf27ac07d31539 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -58,7 +58,7 @@ struct btrfs_device {
 	struct btrfs_fs_devices *fs_devices;
 	struct btrfs_fs_info *fs_info;
 
-	struct rcu_string *name;
+	struct rcu_string __rcu *name;
 
 	u64 generation;
 
@@ -246,7 +246,7 @@ struct btrfs_fs_devices {
 	 */
 	struct list_head alloc_list;
 
-	struct btrfs_fs_devices *seed;
+	struct list_head seed_list;
 	bool seeding;
 
 	int opened;
@@ -435,7 +435,7 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 struct btrfs_device *btrfs_scan_one_device(const char *path,
 					   fmode_t flags, void *holder);
 int btrfs_forget_devices(const char *path);
-int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
+void btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
 void btrfs_free_extra_devids(struct btrfs_fs_devices *fs_devices, int step);
 void btrfs_assign_next_active_device(struct btrfs_device *device,
 				     struct btrfs_device *this_dev);
@@ -569,8 +569,6 @@ static inline enum btrfs_raid_types btrfs_bg_flags_to_raid_index(u64 flags)
 void btrfs_commit_device_sizes(struct btrfs_transaction *trans);
 
 struct list_head * __attribute_const__ btrfs_get_fs_uuids(void);
-void btrfs_set_fs_info_ptr(struct btrfs_fs_info *fs_info);
-void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info);
 bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info,
 					struct btrfs_device *failing_dev);
 void btrfs_scratch_superblocks(struct btrfs_fs_info *fs_info,
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 183299892465af..abf535b036ab77 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -386,25 +386,6 @@ static void dio_bio_end_io(struct bio *bio)
 	spin_unlock_irqrestore(&dio->bio_lock, flags);
 }
 
-/**
- * dio_end_io - handle the end io action for the given bio
- * @bio: The direct io bio thats being completed
- *
- * This is meant to be called by any filesystem that uses their own dio_submit_t
- * so that the DIO specific endio actions are dealt with after the filesystem
- * has done it's completion work.
- */
-void dio_end_io(struct bio *bio)
-{
-	struct dio *dio = bio->bi_private;
-
-	if (dio->is_async)
-		dio_bio_end_aio(bio);
-	else
-		dio_bio_end_io(bio);
-}
-EXPORT_SYMBOL_GPL(dio_end_io);
-
 static inline void
 dio_bio_alloc(struct dio *dio, struct dio_submit *sdio,
 	      struct block_device *bdev,
diff --git a/include/linux/fs.h b/include/linux/fs.h
index e69b45b6cc7b5f..34ad5fe166a1dc 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -3066,8 +3066,6 @@ enum {
 	DIO_SKIP_HOLES	= 0x02,
 };
 
-void dio_end_io(struct bio *bio);
-
 ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 			     struct block_device *bdev, struct iov_iter *iter,
 			     get_block_t get_block,
diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h
index 863335ecb7e8a3..ecd24c719de4d3 100644
--- a/include/trace/events/btrfs.h
+++ b/include/trace/events/btrfs.h
@@ -79,6 +79,7 @@ struct btrfs_space_info;
 #define IO_TREE_OWNER						    \
 	EM( IO_TREE_FS_PINNED_EXTENTS, 	  "PINNED_EXTENTS")	    \
 	EM( IO_TREE_FS_EXCLUDED_EXTENTS,  "EXCLUDED_EXTENTS")	    \
+	EM( IO_TREE_BTREE_INODE_IO,	  "BTREE_INODE_IO")	    \
 	EM( IO_TREE_INODE_IO,		  "INODE_IO")		    \
 	EM( IO_TREE_INODE_IO_FAILURE,	  "INODE_IO_FAILURE")	    \
 	EM( IO_TREE_RELOC_BLOCKS,	  "RELOC_BLOCKS")	    \
@@ -510,7 +511,7 @@ DEFINE_EVENT(
 
 DECLARE_EVENT_CLASS(btrfs__ordered_extent,
 
-	TP_PROTO(const struct inode *inode,
+	TP_PROTO(const struct btrfs_inode *inode,
 		 const struct btrfs_ordered_extent *ordered),
 
 	TP_ARGS(inode, ordered),
@@ -529,8 +530,8 @@ DECLARE_EVENT_CLASS(btrfs__ordered_extent,
 		__field(	u64,  truncated_len	)
 	),
 
-	TP_fast_assign_btrfs(btrfs_sb(inode->i_sb),
-		__entry->ino 		= btrfs_ino(BTRFS_I(inode));
+	TP_fast_assign_btrfs(inode->root->fs_info,
+		__entry->ino 		= btrfs_ino(inode);
 		__entry->file_offset	= ordered->file_offset;
 		__entry->start		= ordered->disk_bytenr;
 		__entry->len		= ordered->num_bytes;
@@ -539,8 +540,7 @@ DECLARE_EVENT_CLASS(btrfs__ordered_extent,
 		__entry->flags		= ordered->flags;
 		__entry->compress_type	= ordered->compress_type;
 		__entry->refs		= refcount_read(&ordered->refs);
-		__entry->root_objectid	=
-				BTRFS_I(inode)->root->root_key.objectid;
+		__entry->root_objectid	= inode->root->root_key.objectid;
 		__entry->truncated_len	= ordered->truncated_len;
 	),
 
@@ -563,7 +563,7 @@ DECLARE_EVENT_CLASS(btrfs__ordered_extent,
 
 DEFINE_EVENT(btrfs__ordered_extent, btrfs_ordered_extent_add,
 
-	TP_PROTO(const struct inode *inode,
+	TP_PROTO(const struct btrfs_inode *inode,
 		 const struct btrfs_ordered_extent *ordered),
 
 	TP_ARGS(inode, ordered)
@@ -571,7 +571,7 @@ DEFINE_EVENT(btrfs__ordered_extent, btrfs_ordered_extent_add,
 
 DEFINE_EVENT(btrfs__ordered_extent, btrfs_ordered_extent_remove,
 
-	TP_PROTO(const struct inode *inode,
+	TP_PROTO(const struct btrfs_inode *inode,
 		 const struct btrfs_ordered_extent *ordered),
 
 	TP_ARGS(inode, ordered)
@@ -579,7 +579,7 @@ DEFINE_EVENT(btrfs__ordered_extent, btrfs_ordered_extent_remove,
 
 DEFINE_EVENT(btrfs__ordered_extent, btrfs_ordered_extent_start,
 
-	TP_PROTO(const struct inode *inode,
+	TP_PROTO(const struct btrfs_inode *inode,
 		 const struct btrfs_ordered_extent *ordered),
 
 	TP_ARGS(inode, ordered)
@@ -587,7 +587,7 @@ DEFINE_EVENT(btrfs__ordered_extent, btrfs_ordered_extent_start,
 
 DEFINE_EVENT(btrfs__ordered_extent, btrfs_ordered_extent_put,
 
-	TP_PROTO(const struct inode *inode,
+	TP_PROTO(const struct btrfs_inode *inode,
 		 const struct btrfs_ordered_extent *ordered),
 
 	TP_ARGS(inode, ordered)
@@ -1176,25 +1176,27 @@ DEFINE_EVENT(btrfs__reserved_extent,  btrfs_reserved_extent_free,
 
 TRACE_EVENT(find_free_extent,
 
-	TP_PROTO(const struct btrfs_fs_info *fs_info, u64 num_bytes,
+	TP_PROTO(const struct btrfs_root *root, u64 num_bytes,
 		 u64 empty_size, u64 data),
 
-	TP_ARGS(fs_info, num_bytes, empty_size, data),
+	TP_ARGS(root, num_bytes, empty_size, data),
 
 	TP_STRUCT__entry_btrfs(
+		__field(	u64,	root_objectid		)
 		__field(	u64,	num_bytes		)
 		__field(	u64,	empty_size		)
 		__field(	u64,	data			)
 	),
 
-	TP_fast_assign_btrfs(fs_info,
+	TP_fast_assign_btrfs(root->fs_info,
+		__entry->root_objectid	= root->root_key.objectid;
 		__entry->num_bytes	= num_bytes;
 		__entry->empty_size	= empty_size;
 		__entry->data		= data;
 	),
 
 	TP_printk_btrfs("root=%llu(%s) len=%llu empty_size=%llu flags=%llu(%s)",
-		  show_root_type(BTRFS_EXTENT_TREE_OBJECTID),
+		  show_root_type(__entry->root_objectid),
 		  __entry->num_bytes, __entry->empty_size, __entry->data,
 		  __print_flags((unsigned long)__entry->data, "|",
 				 BTRFS_GROUP_FLAGS))
diff --git a/include/uapi/linux/btrfs_tree.h b/include/uapi/linux/btrfs_tree.h
index 9ba64ca6b4ac95..6b885982ece687 100644
--- a/include/uapi/linux/btrfs_tree.h
+++ b/include/uapi/linux/btrfs_tree.h
@@ -4,6 +4,11 @@
 
 #include <linux/btrfs.h>
 #include <linux/types.h>
+#ifdef __KERNEL__
+#include <linux/stddef.h>
+#else
+#include <stddef.h>
+#endif
 
 /*
  * This header contains the structure definitions and constants used
@@ -645,6 +650,15 @@ struct btrfs_root_item {
 } __attribute__ ((__packed__));
 
 /*
+ * Btrfs root item used to be smaller than current size.  The old format ends
+ * at where member generation_v2 is.
+ */
+static inline __u32 btrfs_legacy_root_item_size(void)
+{
+	return offsetof(struct btrfs_root_item, generation_v2);
+}
+
+/*
  * this is used for both forward and backward root refs
  */
 struct btrfs_root_ref {