Merge tag 'for-5.8-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "Highlights:

   - speedup dead root detection during orphan cleanup, eg. when there
     are many deleted subvolumes waiting to be cleaned, the trees are
     now looked up in radix tree instead of a O(N^2) search

   - snapshot creation with inherited qgroup will mark the qgroup
     inconsistent, requires a rescan

   - send will emit file capabilities after chown, this produces a
     stream that does not need postprocessing to set the capabilities
     again

   - direct io ported to iomap infrastructure, cleaned up and simplified
     code, notably removing last use of struct buffer_head in btrfs code

  Core changes:

   - factor out backreference iteration, to be used by ordinary
     backreferences and relocation code

   - improved global block reserve utilization
      * better logic to serialize requests
      * increased maximum available for unlink
      * improved handling on large pages (64K)

   - direct io cleanups and fixes
      * simplify layering, where cloned bios were unnecessarily created
        for some cases
      * error handling fixes (submit, endio)
      * remove repair worker thread, used to avoid deadlocks during
        repair

   - refactored block group reading code, preparatory work for new type
     of block group storage that should improve mount time on large
     filesystems

  Cleanups:

   - cleaned up (and slightly sped up) set/get helpers for metadata data
     structure members

   - root bit REF_COWS got renamed to SHAREABLE to reflect the that the
     blocks of the tree get shared either among subvolumes or with the
     relocation trees

  Fixes:

   - when subvolume deletion fails due to ENOSPC, the filesystem is not
     turned read-only

   - device scan deals with devices from other filesystems that changed
     ownership due to overwrite (mkfs)

   - fix a race between scrub and block group removal/allocation

   - fix long standing bug of a runaway balance operation, printing the
     same line to the syslog, caused by a stale status bit on a reloc
     tree that prevented progress

   - fix corrupt log due to concurrent fsync of inodes with shared
     extents

   - fix space underflow for NODATACOW and buffered writes when it for
     some reason needs to fallback to COW mode"

* tag 'for-5.8-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (133 commits)
  btrfs: fix space_info bytes_may_use underflow during space cache writeout
  btrfs: fix space_info bytes_may_use underflow after nocow buffered write
  btrfs: fix wrong file range cleanup after an error filling dealloc range
  btrfs: remove redundant local variable in read_block_for_search
  btrfs: open code key_search
  btrfs: split btrfs_direct_IO to read and write part
  btrfs: remove BTRFS_INODE_READDIO_NEED_LOCK
  fs: remove dio_end_io()
  btrfs: switch to iomap_dio_rw() for dio
  iomap: remove lockdep_assert_held()
  iomap: add a filesystem hook for direct I/O bio submission
  fs: export generic_file_buffered_read()
  btrfs: turn space cache writeout failure messages into debug messages
  btrfs: include error on messages about failure to write space/inode caches
  btrfs: remove useless 'fail_unlock' label from btrfs_csum_file_blocks()
  btrfs: do not ignore error from btrfs_next_leaf() when inserting checksums
  btrfs: make checksum item extension more efficient
  btrfs: fix corrupt log due to concurrent fsync of inodes with shared extents
  btrfs: unexport btrfs_compress_set_level()
  btrfs: simplify iget helpers
  ...

42 files changed, 3196 insertions, 3007 deletions

diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig
index 575636f6491e..68b95ad82126 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 0cc02577577b..d888e71e66b6 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -13,6 +13,7 @@
 #include "transaction.h"
 #include "delayed-ref.h"
 #include "locking.h"
+#include "misc.h"
 
 /* Just an arbitrary number so we can be sure this happened */
 #define BACKREF_FOUND_SHARED 6
@@ -537,18 +538,13 @@ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 				const u64 *extent_item_pos, bool ignore_offset)
 {
 	struct btrfs_root *root;
-	struct btrfs_key root_key;
 	struct extent_buffer *eb;
 	int ret = 0;
 	int root_level;
 	int level = ref->level;
 	struct btrfs_key search_key = ref->key_for_search;
 
-	root_key.objectid = ref->root_id;
-	root_key.type = BTRFS_ROOT_ITEM_KEY;
-	root_key.offset = (u64)-1;
-
-	root = btrfs_get_fs_root(fs_info, &root_key, false);
+	root = btrfs_get_fs_root(fs_info, ref->root_id, false);
 	if (IS_ERR(root)) {
 		ret = PTR_ERR(root);
 		goto out_free;
@@ -2295,3 +2291,832 @@ void free_ipath(struct inode_fs_paths *ipath)
 	kvfree(ipath->fspath);
 	kfree(ipath);
 }
+
+struct btrfs_backref_iter *btrfs_backref_iter_alloc(
+		struct btrfs_fs_info *fs_info, gfp_t gfp_flag)
+{
+	struct btrfs_backref_iter *ret;
+
+	ret = kzalloc(sizeof(*ret), gfp_flag);
+	if (!ret)
+		return NULL;
+
+	ret->path = btrfs_alloc_path();
+	if (!ret) {
+		kfree(ret);
+		return NULL;
+	}
+
+	/* Current backref iterator only supports iteration in commit root */
+	ret->path->search_commit_root = 1;
+	ret->path->skip_locking = 1;
+	ret->fs_info = fs_info;
+
+	return ret;
+}
+
+int btrfs_backref_iter_start(struct btrfs_backref_iter *iter, u64 bytenr)
+{
+	struct btrfs_fs_info *fs_info = iter->fs_info;
+	struct btrfs_path *path = iter->path;
+	struct btrfs_extent_item *ei;
+	struct btrfs_key key;
+	int ret;
+
+	key.objectid = bytenr;
+	key.type = BTRFS_METADATA_ITEM_KEY;
+	key.offset = (u64)-1;
+	iter->bytenr = bytenr;
+
+	ret = btrfs_search_slot(NULL, fs_info->extent_root, &key, path, 0, 0);
+	if (ret < 0)
+		return ret;
+	if (ret == 0) {
+		ret = -EUCLEAN;
+		goto release;
+	}
+	if (path->slots[0] == 0) {
+		WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG));
+		ret = -EUCLEAN;
+		goto release;
+	}
+	path->slots[0]--;
+
+	btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+	if ((key.type != BTRFS_EXTENT_ITEM_KEY &&
+	     key.type != BTRFS_METADATA_ITEM_KEY) || key.objectid != bytenr) {
+		ret = -ENOENT;
+		goto release;
+	}
+	memcpy(&iter->cur_key, &key, sizeof(key));
+	iter->item_ptr = (u32)btrfs_item_ptr_offset(path->nodes[0],
+						    path->slots[0]);
+	iter->end_ptr = (u32)(iter->item_ptr +
+			btrfs_item_size_nr(path->nodes[0], path->slots[0]));
+	ei = btrfs_item_ptr(path->nodes[0], path->slots[0],
+			    struct btrfs_extent_item);
+
+	/*
+	 * Only support iteration on tree backref yet.
+	 *
+	 * This is an extra precaution for non skinny-metadata, where
+	 * EXTENT_ITEM is also used for tree blocks, that we can only use
+	 * extent flags to determine if it's a tree block.
+	 */
+	if (btrfs_extent_flags(path->nodes[0], ei) & BTRFS_EXTENT_FLAG_DATA) {
+		ret = -ENOTSUPP;
+		goto release;
+	}
+	iter->cur_ptr = (u32)(iter->item_ptr + sizeof(*ei));
+
+	/* If there is no inline backref, go search for keyed backref */
+	if (iter->cur_ptr >= iter->end_ptr) {
+		ret = btrfs_next_item(fs_info->extent_root, path);
+
+		/* No inline nor keyed ref */
+		if (ret > 0) {
+			ret = -ENOENT;
+			goto release;
+		}
+		if (ret < 0)
+			goto release;
+
+		btrfs_item_key_to_cpu(path->nodes[0], &iter->cur_key,
+				path->slots[0]);
+		if (iter->cur_key.objectid != bytenr ||
+		    (iter->cur_key.type != BTRFS_SHARED_BLOCK_REF_KEY &&
+		     iter->cur_key.type != BTRFS_TREE_BLOCK_REF_KEY)) {
+			ret = -ENOENT;
+			goto release;
+		}
+		iter->cur_ptr = (u32)btrfs_item_ptr_offset(path->nodes[0],
+							   path->slots[0]);
+		iter->item_ptr = iter->cur_ptr;
+		iter->end_ptr = (u32)(iter->item_ptr + btrfs_item_size_nr(
+				      path->nodes[0], path->slots[0]));
+	}
+
+	return 0;
+release:
+	btrfs_backref_iter_release(iter);
+	return ret;
+}
+
+/*
+ * Go to the next backref item of current bytenr, can be either inlined or
+ * keyed.
+ *
+ * Caller needs to check whether it's inline ref or not by iter->cur_key.
+ *
+ * Return 0 if we get next backref without problem.
+ * Return >0 if there is no extra backref for this bytenr.
+ * Return <0 if there is something wrong happened.
+ */
+int btrfs_backref_iter_next(struct btrfs_backref_iter *iter)
+{
+	struct extent_buffer *eb = btrfs_backref_get_eb(iter);
+	struct btrfs_path *path = iter->path;
+	struct btrfs_extent_inline_ref *iref;
+	int ret;
+	u32 size;
+
+	if (btrfs_backref_iter_is_inline_ref(iter)) {
+		/* We're still inside the inline refs */
+		ASSERT(iter->cur_ptr < iter->end_ptr);
+
+		if (btrfs_backref_has_tree_block_info(iter)) {
+			/* First tree block info */
+			size = sizeof(struct btrfs_tree_block_info);
+		} else {
+			/* Use inline ref type to determine the size */
+			int type;
+
+			iref = (struct btrfs_extent_inline_ref *)
+				((unsigned long)iter->cur_ptr);
+			type = btrfs_extent_inline_ref_type(eb, iref);
+
+			size = btrfs_extent_inline_ref_size(type);
+		}
+		iter->cur_ptr += size;
+		if (iter->cur_ptr < iter->end_ptr)
+			return 0;
+
+		/* All inline items iterated, fall through */
+	}
+
+	/* We're at keyed items, there is no inline item, go to the next one */
+	ret = btrfs_next_item(iter->fs_info->extent_root, iter->path);
+	if (ret)
+		return ret;
+
+	btrfs_item_key_to_cpu(path->nodes[0], &iter->cur_key, path->slots[0]);
+	if (iter->cur_key.objectid != iter->bytenr ||
+	    (iter->cur_key.type != BTRFS_TREE_BLOCK_REF_KEY &&
+	     iter->cur_key.type != BTRFS_SHARED_BLOCK_REF_KEY))
+		return 1;
+	iter->item_ptr = (u32)btrfs_item_ptr_offset(path->nodes[0],
+					path->slots[0]);
+	iter->cur_ptr = iter->item_ptr;
+	iter->end_ptr = iter->item_ptr + (u32)btrfs_item_size_nr(path->nodes[0],
+						path->slots[0]);
+	return 0;
+}
+
+void btrfs_backref_init_cache(struct btrfs_fs_info *fs_info,
+			      struct btrfs_backref_cache *cache, int is_reloc)
+{
+	int i;
+
+	cache->rb_root = RB_ROOT;
+	for (i = 0; i < BTRFS_MAX_LEVEL; i++)
+		INIT_LIST_HEAD(&cache->pending[i]);
+	INIT_LIST_HEAD(&cache->changed);
+	INIT_LIST_HEAD(&cache->detached);
+	INIT_LIST_HEAD(&cache->leaves);
+	INIT_LIST_HEAD(&cache->pending_edge);
+	INIT_LIST_HEAD(&cache->useless_node);
+	cache->fs_info = fs_info;
+	cache->is_reloc = is_reloc;
+}
+
+struct btrfs_backref_node *btrfs_backref_alloc_node(
+		struct btrfs_backref_cache *cache, u64 bytenr, int level)
+{
+	struct btrfs_backref_node *node;
+
+	ASSERT(level >= 0 && level < BTRFS_MAX_LEVEL);
+	node = kzalloc(sizeof(*node), GFP_NOFS);
+	if (!node)
+		return node;
+
+	INIT_LIST_HEAD(&node->list);
+	INIT_LIST_HEAD(&node->upper);
+	INIT_LIST_HEAD(&node->lower);
+	RB_CLEAR_NODE(&node->rb_node);
+	cache->nr_nodes++;
+	node->level = level;
+	node->bytenr = bytenr;
+
+	return node;
+}
+
+struct btrfs_backref_edge *btrfs_backref_alloc_edge(
+		struct btrfs_backref_cache *cache)
+{
+	struct btrfs_backref_edge *edge;
+
+	edge = kzalloc(sizeof(*edge), GFP_NOFS);
+	if (edge)
+		cache->nr_edges++;
+	return edge;
+}
+
+/*
+ * Drop the backref node from cache, also cleaning up all its
+ * upper edges and any uncached nodes in the path.
+ *
+ * This cleanup happens bottom up, thus the node should either
+ * be the lowest node in the cache or a detached node.
+ */
+void btrfs_backref_cleanup_node(struct btrfs_backref_cache *cache,
+				struct btrfs_backref_node *node)
+{
+	struct btrfs_backref_node *upper;
+	struct btrfs_backref_edge *edge;
+
+	if (!node)
+		return;
+
+	BUG_ON(!node->lowest && !node->detached);
+	while (!list_empty(&node->upper)) {
+		edge = list_entry(node->upper.next, struct btrfs_backref_edge,
+				  list[LOWER]);
+		upper = edge->node[UPPER];
+		list_del(&edge->list[LOWER]);
+		list_del(&edge->list[UPPER]);
+		btrfs_backref_free_edge(cache, edge);
+
+		if (RB_EMPTY_NODE(&upper->rb_node)) {
+			BUG_ON(!list_empty(&node->upper));
+			btrfs_backref_drop_node(cache, node);
+			node = upper;
+			node->lowest = 1;
+			continue;
+		}
+		/*
+		 * Add the node to leaf node list if no other child block
+		 * cached.
+		 */
+		if (list_empty(&upper->lower)) {
+			list_add_tail(&upper->lower, &cache->leaves);
+			upper->lowest = 1;
+		}
+	}
+
+	btrfs_backref_drop_node(cache, node);
+}
+
+/*
+ * Release all nodes/edges from current cache
+ */
+void btrfs_backref_release_cache(struct btrfs_backref_cache *cache)
+{
+	struct btrfs_backref_node *node;
+	int i;
+
+	while (!list_empty(&cache->detached)) {
+		node = list_entry(cache->detached.next,
+				  struct btrfs_backref_node, list);
+		btrfs_backref_cleanup_node(cache, node);
+	}
+
+	while (!list_empty(&cache->leaves)) {
+		node = list_entry(cache->leaves.next,
+				  struct btrfs_backref_node, lower);
+		btrfs_backref_cleanup_node(cache, node);
+	}
+
+	cache->last_trans = 0;
+
+	for (i = 0; i < BTRFS_MAX_LEVEL; i++)
+		ASSERT(list_empty(&cache->pending[i]));
+	ASSERT(list_empty(&cache->pending_edge));
+	ASSERT(list_empty(&cache->useless_node));
+	ASSERT(list_empty(&cache->changed));
+	ASSERT(list_empty(&cache->detached));
+	ASSERT(RB_EMPTY_ROOT(&cache->rb_root));
+	ASSERT(!cache->nr_nodes);
+	ASSERT(!cache->nr_edges);
+}
+
+/*
+ * Handle direct tree backref
+ *
+ * Direct tree backref means, the backref item shows its parent bytenr
+ * directly. This is for SHARED_BLOCK_REF backref (keyed or inlined).
+ *
+ * @ref_key:	The converted backref key.
+ *		For keyed backref, it's the item key.
+ *		For inlined backref, objectid is the bytenr,
+ *		type is btrfs_inline_ref_type, offset is
+ *		btrfs_inline_ref_offset.
+ */
+static int handle_direct_tree_backref(struct btrfs_backref_cache *cache,
+				      struct btrfs_key *ref_key,
+				      struct btrfs_backref_node *cur)
+{
+	struct btrfs_backref_edge *edge;
+	struct btrfs_backref_node *upper;
+	struct rb_node *rb_node;
+
+	ASSERT(ref_key->type == BTRFS_SHARED_BLOCK_REF_KEY);
+
+	/* Only reloc root uses backref pointing to itself */
+	if (ref_key->objectid == ref_key->offset) {
+		struct btrfs_root *root;
+
+		cur->is_reloc_root = 1;
+		/* Only reloc backref cache cares about a specific root */
+		if (cache->is_reloc) {
+			root = find_reloc_root(cache->fs_info, cur->bytenr);
+			if (WARN_ON(!root))
+				return -ENOENT;
+			cur->root = root;
+		} else {
+			/*
+			 * For generic purpose backref cache, reloc root node
+			 * is useless.
+			 */
+			list_add(&cur->list, &cache->useless_node);
+		}
+		return 0;
+	}
+
+	edge = btrfs_backref_alloc_edge(cache);
+	if (!edge)
+		return -ENOMEM;
+
+	rb_node = rb_simple_search(&cache->rb_root, ref_key->offset);
+	if (!rb_node) {
+		/* Parent node not yet cached */
+		upper = btrfs_backref_alloc_node(cache, ref_key->offset,
+					   cur->level + 1);
+		if (!upper) {
+			btrfs_backref_free_edge(cache, edge);
+			return -ENOMEM;
+		}
+
+		/*
+		 *  Backrefs for the upper level block isn't cached, add the
+		 *  block to pending list
+		 */
+		list_add_tail(&edge->list[UPPER], &cache->pending_edge);
+	} else {
+		/* Parent node already cached */
+		upper = rb_entry(rb_node, struct btrfs_backref_node, rb_node);
+		ASSERT(upper->checked);
+		INIT_LIST_HEAD(&edge->list[UPPER]);
+	}
+	btrfs_backref_link_edge(edge, cur, upper, LINK_LOWER);
+	return 0;
+}
+
+/*
+ * Handle indirect tree backref
+ *
+ * Indirect tree backref means, we only know which tree the node belongs to.
+ * We still need to do a tree search to find out the parents. This is for
+ * TREE_BLOCK_REF backref (keyed or inlined).
+ *
+ * @ref_key:	The same as @ref_key in  handle_direct_tree_backref()
+ * @tree_key:	The first key of this tree block.
+ * @path:	A clean (released) path, to avoid allocating path everytime
+ *		the function get called.
+ */
+static int handle_indirect_tree_backref(struct btrfs_backref_cache *cache,
+					struct btrfs_path *path,
+					struct btrfs_key *ref_key,
+					struct btrfs_key *tree_key,
+					struct btrfs_backref_node *cur)
+{
+	struct btrfs_fs_info *fs_info = cache->fs_info;
+	struct btrfs_backref_node *upper;
+	struct btrfs_backref_node *lower;
+	struct btrfs_backref_edge *edge;
+	struct extent_buffer *eb;
+	struct btrfs_root *root;
+	struct rb_node *rb_node;
+	int level;
+	bool need_check = true;
+	int ret;
+
+	root = btrfs_get_fs_root(fs_info, ref_key->offset, false);
+	if (IS_ERR(root))
+		return PTR_ERR(root);
+	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
+		cur->cowonly = 1;
+
+	if (btrfs_root_level(&root->root_item) == cur->level) {
+		/* Tree root */
+		ASSERT(btrfs_root_bytenr(&root->root_item) == cur->bytenr);
+		/*
+		 * For reloc backref cache, we may ignore reloc root.  But for
+		 * general purpose backref cache, we can't rely on
+		 * btrfs_should_ignore_reloc_root() as it may conflict with
+		 * current running relocation and lead to missing root.
+		 *
+		 * For general purpose backref cache, reloc root detection is
+		 * completely relying on direct backref (key->offset is parent
+		 * bytenr), thus only do such check for reloc cache.
+		 */
+		if (btrfs_should_ignore_reloc_root(root) && cache->is_reloc) {
+			btrfs_put_root(root);
+			list_add(&cur->list, &cache->useless_node);
+		} else {
+			cur->root = root;
+		}
+		return 0;
+	}
+
+	level = cur->level + 1;
+
+	/* Search the tree to find parent blocks referring to the block */
+	path->search_commit_root = 1;
+	path->skip_locking = 1;
+	path->lowest_level = level;
+	ret = btrfs_search_slot(NULL, root, tree_key, path, 0, 0);
+	path->lowest_level = 0;
+	if (ret < 0) {
+		btrfs_put_root(root);
+		return ret;
+	}
+	if (ret > 0 && path->slots[level] > 0)
+		path->slots[level]--;
+
+	eb = path->nodes[level];
+	if (btrfs_node_blockptr(eb, path->slots[level]) != cur->bytenr) {
+		btrfs_err(fs_info,
+"couldn't find block (%llu) (level %d) in tree (%llu) with key (%llu %u %llu)",
+			  cur->bytenr, level - 1, root->root_key.objectid,
+			  tree_key->objectid, tree_key->type, tree_key->offset);
+		btrfs_put_root(root);
+		ret = -ENOENT;
+		goto out;
+	}
+	lower = cur;
+
+	/* Add all nodes and edges in the path */
+	for (; level < BTRFS_MAX_LEVEL; level++) {
+		if (!path->nodes[level]) {
+			ASSERT(btrfs_root_bytenr(&root->root_item) ==
+			       lower->bytenr);
+			/* Same as previous should_ignore_reloc_root() call */
+			if (btrfs_should_ignore_reloc_root(root) &&
+			    cache->is_reloc) {
+				btrfs_put_root(root);
+				list_add(&lower->list, &cache->useless_node);
+			} else {
+				lower->root = root;
+			}
+			break;
+		}
+
+		edge = btrfs_backref_alloc_edge(cache);
+		if (!edge) {
+			btrfs_put_root(root);
+			ret = -ENOMEM;
+			goto out;
+		}
+
+		eb = path->nodes[level];
+		rb_node = rb_simple_search(&cache->rb_root, eb->start);
+		if (!rb_node) {
+			upper = btrfs_backref_alloc_node(cache, eb->start,
+							 lower->level + 1);
+			if (!upper) {
+				btrfs_put_root(root);
+				btrfs_backref_free_edge(cache, edge);
+				ret = -ENOMEM;
+				goto out;
+			}
+			upper->owner = btrfs_header_owner(eb);
+			if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
+				upper->cowonly = 1;
+
+			/*
+			 * If we know the block isn't shared we can avoid
+			 * checking its backrefs.
+			 */
+			if (btrfs_block_can_be_shared(root, eb))
+				upper->checked = 0;
+			else
+				upper->checked = 1;
+
+			/*
+			 * Add the block to pending list if we need to check its
+			 * backrefs, we only do this once while walking up a
+			 * tree as we will catch anything else later on.
+			 */
+			if (!upper->checked && need_check) {
+				need_check = false;
+				list_add_tail(&edge->list[UPPER],
+					      &cache->pending_edge);
+			} else {
+				if (upper->checked)
+					need_check = true;
+				INIT_LIST_HEAD(&edge->list[UPPER]);
+			}
+		} else {
+			upper = rb_entry(rb_node, struct btrfs_backref_node,
+					 rb_node);
+			ASSERT(upper->checked);
+			INIT_LIST_HEAD(&edge->list[UPPER]);
+			if (!upper->owner)
+				upper->owner = btrfs_header_owner(eb);
+		}
+		btrfs_backref_link_edge(edge, lower, upper, LINK_LOWER);
+
+		if (rb_node) {
+			btrfs_put_root(root);
+			break;
+		}
+		lower = upper;
+		upper = NULL;
+	}
+out:
+	btrfs_release_path(path);
+	return ret;
+}
+
+/*
+ * Add backref node @cur into @cache.
+ *
+ * NOTE: Even if the function returned 0, @cur is not yet cached as its upper
+ *	 links aren't yet bi-directional. Needs to finish such links.
+ *	 Use btrfs_backref_finish_upper_links() to finish such linkage.
+ *
+ * @path:	Released path for indirect tree backref lookup
+ * @iter:	Released backref iter for extent tree search
+ * @node_key:	The first key of the tree block
+ */
+int btrfs_backref_add_tree_node(struct btrfs_backref_cache *cache,
+				struct btrfs_path *path,
+				struct btrfs_backref_iter *iter,
+				struct btrfs_key *node_key,
+				struct btrfs_backref_node *cur)
+{
+	struct btrfs_fs_info *fs_info = cache->fs_info;
+	struct btrfs_backref_edge *edge;
+	struct btrfs_backref_node *exist;
+	int ret;
+
+	ret = btrfs_backref_iter_start(iter, cur->bytenr);
+	if (ret < 0)
+		return ret;
+	/*
+	 * We skip the first btrfs_tree_block_info, as we don't use the key
+	 * stored in it, but fetch it from the tree block
+	 */
+	if (btrfs_backref_has_tree_block_info(iter)) {
+		ret = btrfs_backref_iter_next(iter);
+		if (ret < 0)
+			goto out;
+		/* No extra backref? This means the tree block is corrupted */
+		if (ret > 0) {
+			ret = -EUCLEAN;
+			goto out;
+		}
+	}
+	WARN_ON(cur->checked);
+	if (!list_empty(&cur->upper)) {
+		/*
+		 * The backref was added previously when processing backref of
+		 * type BTRFS_TREE_BLOCK_REF_KEY
+		 */
+		ASSERT(list_is_singular(&cur->upper));
+		edge = list_entry(cur->upper.next, struct btrfs_backref_edge,
+				  list[LOWER]);
+		ASSERT(list_empty(&edge->list[UPPER]));
+		exist = edge->node[UPPER];
+		/*
+		 * Add the upper level block to pending list if we need check
+		 * its backrefs
+		 */
+		if (!exist->checked)
+			list_add_tail(&edge->list[UPPER], &cache->pending_edge);
+	} else {
+		exist = NULL;
+	}
+
+	for (; ret == 0; ret = btrfs_backref_iter_next(iter)) {
+		struct extent_buffer *eb;
+		struct btrfs_key key;
+		int type;
+
+		cond_resched();
+		eb = btrfs_backref_get_eb(iter);
+
+		key.objectid = iter->bytenr;
+		if (btrfs_backref_iter_is_inline_ref(iter)) {
+			struct btrfs_extent_inline_ref *iref;
+
+			/* Update key for inline backref */
+			iref = (struct btrfs_extent_inline_ref *)
+				((unsigned long)iter->cur_ptr);
+			type = btrfs_get_extent_inline_ref_type(eb, iref,
+							BTRFS_REF_TYPE_BLOCK);
+			if (type == BTRFS_REF_TYPE_INVALID) {
+				ret = -EUCLEAN;
+				goto out;
+			}
+			key.type = type;
+			key.offset = btrfs_extent_inline_ref_offset(eb, iref);
+		} else {
+			key.type = iter->cur_key.type;
+			key.offset = iter->cur_key.offset;
+		}
+
+		/*
+		 * Parent node found and matches current inline ref, no need to
+		 * rebuild this node for this inline ref
+		 */
+		if (exist &&
+		    ((key.type == BTRFS_TREE_BLOCK_REF_KEY &&
+		      exist->owner == key.offset) ||
+		     (key.type == BTRFS_SHARED_BLOCK_REF_KEY &&
+		      exist->bytenr == key.offset))) {
+			exist = NULL;
+			continue;
+		}
+
+		/* SHARED_BLOCK_REF means key.offset is the parent bytenr */
+		if (key.type == BTRFS_SHARED_BLOCK_REF_KEY) {
+			ret = handle_direct_tree_backref(cache, &key, cur);
+			if (ret < 0)
+				goto out;
+			continue;
+		} else if (unlikely(key.type == BTRFS_EXTENT_REF_V0_KEY)) {
+			ret = -EINVAL;
+			btrfs_print_v0_err(fs_info);
+			btrfs_handle_fs_error(fs_info, ret, NULL);
+			goto out;
+		} else if (key.type != BTRFS_TREE_BLOCK_REF_KEY) {
+			continue;
+		}
+
+		/*
+		 * key.type == BTRFS_TREE_BLOCK_REF_KEY, inline ref offset
+		 * means the root objectid. We need to search the tree to get
+		 * its parent bytenr.
+		 */
+		ret = handle_indirect_tree_backref(cache, path, &key, node_key,
+						   cur);
+		if (ret < 0)
+			goto out;
+	}
+	ret = 0;
+	cur->checked = 1;
+	WARN_ON(exist);
+out:
+	btrfs_backref_iter_release(iter);
+	return ret;
+}
+
+/*
+ * Finish the upwards linkage created by btrfs_backref_add_tree_node()
+ */
+int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache,
+				     struct btrfs_backref_node *start)
+{
+	struct list_head *useless_node = &cache->useless_node;
+	struct btrfs_backref_edge *edge;
+	struct rb_node *rb_node;
+	LIST_HEAD(pending_edge);
+
+	ASSERT(start->checked);
+
+	/* Insert this node to cache if it's not COW-only */
+	if (!start->cowonly) {
+		rb_node = rb_simple_insert(&cache->rb_root, start->bytenr,
+					   &start->rb_node);
+		if (rb_node)
+			btrfs_backref_panic(cache->fs_info, start->bytenr,
+					    -EEXIST);
+		list_add_tail(&start->lower, &cache->leaves);
+	}
+
+	/*
+	 * Use breadth first search to iterate all related edges.
+	 *
+	 * The starting points are all the edges of this node
+	 */
+	list_for_each_entry(edge, &start->upper, list[LOWER])
+		list_add_tail(&edge->list[UPPER], &pending_edge);
+
+	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]);
+		list_del_init(&edge->list[UPPER]);
+		upper = edge->node[UPPER];
+		lower = edge->node[LOWER];
+
+		/* Parent is detached, no need to keep any edges */
+		if (upper->detached) {
+			list_del(&edge->list[LOWER]);
+			btrfs_backref_free_edge(cache, edge);
+
+			/* Lower node is orphan, queue for cleanup */
+			if (list_empty(&lower->upper))
+				list_add(&lower->list, useless_node);
+			continue;
+		}
+
+		/*
+		 * All new nodes added in current build_backref_tree() haven't
+		 * been linked to the cache rb tree.
+		 * So if we have upper->rb_node populated, this means a cache
+		 * hit. We only need to link the edge, as @upper and all its
+		 * parents have already been linked.
+		 */
+		if (!RB_EMPTY_NODE(&upper->rb_node)) {
+			if (upper->lowest) {
+				list_del_init(&upper->lower);
+				upper->lowest = 0;
+			}
+
+			list_add_tail(&edge->list[UPPER], &upper->lower);
+			continue;
+		}
+
+		/* Sanity check, we shouldn't have any unchecked nodes */
+		if (!upper->checked) {
+			ASSERT(0);
+			return -EUCLEAN;
+		}
+
+		/* Sanity check, COW-only node has non-COW-only parent */
+		if (start->cowonly != upper->cowonly) {
+			ASSERT(0);
+			return -EUCLEAN;
+		}
+
+		/* Only cache non-COW-only (subvolume trees) tree blocks */
+		if (!upper->cowonly) {
+			rb_node = rb_simple_insert(&cache->rb_root, upper->bytenr,
+						   &upper->rb_node);
+			if (rb_node) {
+				btrfs_backref_panic(cache->fs_info,
+						upper->bytenr, -EEXIST);
+				return -EUCLEAN;
+			}
+		}
+
+		list_add_tail(&edge->list[UPPER], &upper->lower);
+
+		/*
+		 * Also queue all the parent edges of this uncached node
+		 * to finish the upper linkage
+		 */
+		list_for_each_entry(edge, &upper->upper, list[LOWER])
+			list_add_tail(&edge->list[UPPER], &pending_edge);
+	}
+	return 0;
+}
+
+void btrfs_backref_error_cleanup(struct btrfs_backref_cache *cache,
+				 struct btrfs_backref_node *node)
+{
+	struct btrfs_backref_node *lower;
+	struct btrfs_backref_node *upper;
+	struct btrfs_backref_edge *edge;
+
+	while (!list_empty(&cache->useless_node)) {
+		lower = list_first_entry(&cache->useless_node,
+				   struct btrfs_backref_node, list);
+		list_del_init(&lower->list);
+	}
+	while (!list_empty(&cache->pending_edge)) {
+		edge = list_first_entry(&cache->pending_edge,
+				struct btrfs_backref_edge, list[UPPER]);
+		list_del(&edge->list[UPPER]);
+		list_del(&edge->list[LOWER]);
+		lower = edge->node[LOWER];
+		upper = edge->node[UPPER];
+		btrfs_backref_free_edge(cache, edge);
+
+		/*
+		 * Lower is no longer linked to any upper backref nodes and
+		 * isn't in the cache, we can free it ourselves.
+		 */
+		if (list_empty(&lower->upper) &&
+		    RB_EMPTY_NODE(&lower->rb_node))
+			list_add(&lower->list, &cache->useless_node);
+
+		if (!RB_EMPTY_NODE(&upper->rb_node))
+			continue;
+
+		/* Add this guy's upper edges to the list to process */
+		list_for_each_entry(edge, &upper->upper, list[LOWER])
+			list_add_tail(&edge->list[UPPER],
+				      &cache->pending_edge);
+		if (list_empty(&upper->upper))
+			list_add(&upper->list, &cache->useless_node);
+	}
+
+	while (!list_empty(&cache->useless_node)) {
+		lower = list_first_entry(&cache->useless_node,
+				   struct btrfs_backref_node, list);
+		list_del_init(&lower->list);
+		if (lower == node)
+			node = NULL;
+		btrfs_backref_free_node(cache, lower);
+	}
+
+	btrfs_backref_cleanup_node(cache, node);
+	ASSERT(list_empty(&cache->useless_node) &&
+	       list_empty(&cache->pending_edge));
+}
diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h
index 723d6da99114..ff705cc564a9 100644
--- a/fs/btrfs/backref.h
+++ b/fs/btrfs/backref.h
@@ -8,6 +8,7 @@
 
 #include <linux/btrfs.h>
 #include "ulist.h"
+#include "disk-io.h"
 #include "extent_io.h"
 
 struct inode_fs_paths {
@@ -78,4 +79,300 @@ struct prelim_ref {
 	u64 wanted_disk_byte;
 };
 
+/*
+ * Iterate backrefs of one extent.
+ *
+ * Now it only supports iteration of tree block in commit root.
+ */
+struct btrfs_backref_iter {
+	u64 bytenr;
+	struct btrfs_path *path;
+	struct btrfs_fs_info *fs_info;
+	struct btrfs_key cur_key;
+	u32 item_ptr;
+	u32 cur_ptr;
+	u32 end_ptr;
+};
+
+struct btrfs_backref_iter *btrfs_backref_iter_alloc(
+		struct btrfs_fs_info *fs_info, gfp_t gfp_flag);
+
+static inline void btrfs_backref_iter_free(struct btrfs_backref_iter *iter)
+{
+	if (!iter)
+		return;
+	btrfs_free_path(iter->path);
+	kfree(iter);
+}
+
+static inline struct extent_buffer *btrfs_backref_get_eb(
+		struct btrfs_backref_iter *iter)
+{
+	if (!iter)
+		return NULL;
+	return iter->path->nodes[0];
+}
+
+/*
+ * For metadata with EXTENT_ITEM key (non-skinny) case, the first inline data
+ * is btrfs_tree_block_info, without a btrfs_extent_inline_ref header.
+ *
+ * This helper determines if that's the case.
+ */
+static inline bool btrfs_backref_has_tree_block_info(
+		struct btrfs_backref_iter *iter)
+{
+	if (iter->cur_key.type == BTRFS_EXTENT_ITEM_KEY &&
+	    iter->cur_ptr - iter->item_ptr == sizeof(struct btrfs_extent_item))
+		return true;
+	return false;
+}
+
+int btrfs_backref_iter_start(struct btrfs_backref_iter *iter, u64 bytenr);
+
+int btrfs_backref_iter_next(struct btrfs_backref_iter *iter);
+
+static inline bool btrfs_backref_iter_is_inline_ref(
+		struct btrfs_backref_iter *iter)
+{
+	if (iter->cur_key.type == BTRFS_EXTENT_ITEM_KEY ||
+	    iter->cur_key.type == BTRFS_METADATA_ITEM_KEY)
+		return true;
+	return false;
+}
+
+static inline void btrfs_backref_iter_release(struct btrfs_backref_iter *iter)
+{
+	iter->bytenr = 0;
+	iter->item_ptr = 0;
+	iter->cur_ptr = 0;
+	iter->end_ptr = 0;
+	btrfs_release_path(iter->path);
+	memset(&iter->cur_key, 0, sizeof(iter->cur_key));
+}
+
+/*
+ * Backref cache related structures
+ *
+ * The whole objective of backref_cache is to build a bi-directional map
+ * of tree blocks (represented by backref_node) and all their parents.
+ */
+
+/*
+ * Represent a tree block in the backref cache
+ */
+struct btrfs_backref_node {
+	struct {
+		struct rb_node rb_node;
+		u64 bytenr;
+	}; /* Use rb_simple_node for search/insert */
+
+	u64 new_bytenr;
+	/* Objectid of tree block owner, can be not uptodate */
+	u64 owner;
+	/* Link to pending, changed or detached list */
+	struct list_head list;
+
+	/* List of upper level edges, which link this node to its parents */
+	struct list_head upper;
+	/* List of lower level edges, which link this node to its children */
+	struct list_head lower;
+
+	/* NULL if this node is not tree root */
+	struct btrfs_root *root;
+	/* Extent buffer got by COWing the block */
+	struct extent_buffer *eb;
+	/* Level of the tree block */
+	unsigned int level:8;
+	/* Is the block in a non-shareable tree */
+	unsigned int cowonly:1;
+	/* 1 if no child node is in the cache */
+	unsigned int lowest:1;
+	/* Is the extent buffer locked */
+	unsigned int locked:1;
+	/* Has the block been processed */
+	unsigned int processed:1;
+	/* Have backrefs of this block been checked */
+	unsigned int checked:1;
+	/*
+	 * 1 if corresponding block has been COWed but some upper level block
+	 * pointers may not point to the new location
+	 */
+	unsigned int pending:1;
+	/* 1 if the backref node isn't connected to any other backref node */
+	unsigned int detached:1;
+
+	/*
+	 * For generic purpose backref cache, where we only care if it's a reloc
+	 * root, doesn't care the source subvolid.
+	 */
+	unsigned int is_reloc_root:1;
+};
+
+#define LOWER	0
+#define UPPER	1
+
+/*
+ * Represent an edge connecting upper and lower backref nodes.
+ */
+struct btrfs_backref_edge {
+	/*
+	 * list[LOWER] is linked to btrfs_backref_node::upper of lower level
+	 * node, and list[UPPER] is linked to btrfs_backref_node::lower of
+	 * upper level node.
+	 *
+	 * Also, build_backref_tree() uses list[UPPER] for pending edges, before
+	 * linking list[UPPER] to its upper level nodes.
+	 */
+	struct list_head list[2];
+
+	/* Two related nodes */
+	struct btrfs_backref_node *node[2];
+};
+
+struct btrfs_backref_cache {
+	/* Red black tree of all backref nodes in the cache */
+	struct rb_root rb_root;
+	/* For passing backref nodes to btrfs_reloc_cow_block */
+	struct btrfs_backref_node *path[BTRFS_MAX_LEVEL];
+	/*
+	 * List of blocks that have been COWed but some block pointers in upper
+	 * level blocks may not reflect the new location
+	 */
+	struct list_head pending[BTRFS_MAX_LEVEL];
+	/* List of backref nodes with no child node */
+	struct list_head leaves;
+	/* List of blocks that have been COWed in current transaction */
+	struct list_head changed;
+	/* List of detached backref node. */
+	struct list_head detached;
+
+	u64 last_trans;
+
+	int nr_nodes;
+	int nr_edges;
+
+	/* List of unchecked backref edges during backref cache build */
+	struct list_head pending_edge;
+
+	/* List of useless backref nodes during backref cache build */
+	struct list_head useless_node;
+
+	struct btrfs_fs_info *fs_info;
+
+	/*
+	 * Whether this cache is for relocation
+	 *
+	 * Reloction backref cache require more info for reloc root compared
+	 * to generic backref cache.
+	 */
+	unsigned int is_reloc;
+};
+
+void btrfs_backref_init_cache(struct btrfs_fs_info *fs_info,
+			      struct btrfs_backref_cache *cache, int is_reloc);
+struct btrfs_backref_node *btrfs_backref_alloc_node(
+		struct btrfs_backref_cache *cache, u64 bytenr, int level);
+struct btrfs_backref_edge *btrfs_backref_alloc_edge(
+		struct btrfs_backref_cache *cache);
+
+#define		LINK_LOWER	(1 << 0)
+#define		LINK_UPPER	(1 << 1)
+static inline void btrfs_backref_link_edge(struct btrfs_backref_edge *edge,
+					   struct btrfs_backref_node *lower,
+					   struct btrfs_backref_node *upper,
+					   int link_which)
+{
+	ASSERT(upper && lower && upper->level == lower->level + 1);
+	edge->node[LOWER] = lower;
+	edge->node[UPPER] = upper;
+	if (link_which & LINK_LOWER)
+		list_add_tail(&edge->list[LOWER], &lower->upper);
+	if (link_which & LINK_UPPER)
+		list_add_tail(&edge->list[UPPER], &upper->lower);
+}
+
+static inline void btrfs_backref_free_node(struct btrfs_backref_cache *cache,
+					   struct btrfs_backref_node *node)
+{
+	if (node) {
+		cache->nr_nodes--;
+		btrfs_put_root(node->root);
+		kfree(node);
+	}
+}
+
+static inline void btrfs_backref_free_edge(struct btrfs_backref_cache *cache,
+					   struct btrfs_backref_edge *edge)
+{
+	if (edge) {
+		cache->nr_edges--;
+		kfree(edge);
+	}
+}
+
+static inline void btrfs_backref_unlock_node_buffer(
+		struct btrfs_backref_node *node)
+{
+	if (node->locked) {
+		btrfs_tree_unlock(node->eb);
+		node->locked = 0;
+	}
+}
+
+static inline void btrfs_backref_drop_node_buffer(
+		struct btrfs_backref_node *node)
+{
+	if (node->eb) {
+		btrfs_backref_unlock_node_buffer(node);
+		free_extent_buffer(node->eb);
+		node->eb = NULL;
+	}
+}
+
+/*
+ * Drop the backref node from cache without cleaning up its children
+ * edges.
+ *
+ * This can only be called on node without parent edges.
+ * The children edges are still kept as is.
+ */
+static inline void btrfs_backref_drop_node(struct btrfs_backref_cache *tree,
+					   struct btrfs_backref_node *node)
+{
+	BUG_ON(!list_empty(&node->upper));
+
+	btrfs_backref_drop_node_buffer(node);
+	list_del(&node->list);
+	list_del(&node->lower);
+	if (!RB_EMPTY_NODE(&node->rb_node))
+		rb_erase(&node->rb_node, &tree->rb_root);
+	btrfs_backref_free_node(tree, node);
+}
+
+void btrfs_backref_cleanup_node(struct btrfs_backref_cache *cache,
+				struct btrfs_backref_node *node);
+
+void btrfs_backref_release_cache(struct btrfs_backref_cache *cache);
+
+static inline void btrfs_backref_panic(struct btrfs_fs_info *fs_info,
+				       u64 bytenr, int errno)
+{
+	btrfs_panic(fs_info, errno,
+		    "Inconsistency in backref cache found at offset %llu",
+		    bytenr);
+}
+
+int btrfs_backref_add_tree_node(struct btrfs_backref_cache *cache,
+				struct btrfs_path *path,
+				struct btrfs_backref_iter *iter,
+				struct btrfs_key *node_key,
+				struct btrfs_backref_node *cur);
+
+int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache,
+				     struct btrfs_backref_node *start);
+
+void btrfs_backref_error_cleanup(struct btrfs_backref_cache *cache,
+				 struct btrfs_backref_node *node);
+
 #endif
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 696f47103cfc..176e8a292fd1 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -7,7 +7,6 @@
 #include "disk-io.h"
 #include "free-space-cache.h"
 #include "free-space-tree.h"
-#include "disk-io.h"
 #include "volumes.h"
 #include "transaction.h"
 #include "ref-verify.h"
@@ -161,6 +160,8 @@ static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
 	struct rb_node *parent = NULL;
 	struct btrfs_block_group *cache;
 
+	ASSERT(block_group->length != 0);
+
 	spin_lock(&info->block_group_cache_lock);
 	p = &info->block_group_cache_tree.rb_node;
 
@@ -863,11 +864,34 @@ static void clear_incompat_bg_bits(struct btrfs_fs_info *fs_info, u64 flags)
 	}
 }
 
+static int remove_block_group_item(struct btrfs_trans_handle *trans,
+				   struct btrfs_path *path,
+				   struct btrfs_block_group *block_group)
+{
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_root *root;
+	struct btrfs_key key;
+	int ret;
+
+	root = fs_info->extent_root;
+	key.objectid = block_group->start;
+	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+	key.offset = block_group->length;
+
+	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
+	if (ret > 0)
+		ret = -ENOENT;
+	if (ret < 0)
+		return ret;
+
+	ret = btrfs_del_item(trans, root, path);
+	return ret;
+}
+
 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 			     u64 group_start, struct extent_map *em)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_root *root = fs_info->extent_root;
 	struct btrfs_path *path;
 	struct btrfs_block_group *block_group;
 	struct btrfs_free_cluster *cluster;
@@ -1065,26 +1089,25 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 
 	spin_unlock(&block_group->space_info->lock);
 
-	key.objectid = block_group->start;
-	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
-	key.offset = block_group->length;
-
 	mutex_lock(&fs_info->chunk_mutex);
 	spin_lock(&block_group->lock);
 	block_group->removed = 1;
 	/*
-	 * At this point trimming can't start on this block group, because we
-	 * removed the block group from the tree fs_info->block_group_cache_tree
-	 * so no one can't find it anymore and even if someone already got this
-	 * block group before we removed it from the rbtree, they have already
-	 * incremented block_group->trimming - if they didn't, they won't find
-	 * any free space entries because we already removed them all when we
-	 * called btrfs_remove_free_space_cache().
+	 * At this point trimming or scrub can't start on this block group,
+	 * because we removed the block group from the rbtree
+	 * fs_info->block_group_cache_tree so no one can't find it anymore and
+	 * even if someone already got this block group before we removed it
+	 * from the rbtree, they have already incremented block_group->frozen -
+	 * if they didn't, for the trimming case they won't find any free space
+	 * entries because we already removed them all when we called
+	 * btrfs_remove_free_space_cache().
 	 *
 	 * And we must not remove the extent map from the fs_info->mapping_tree
 	 * to prevent the same logical address range and physical device space
-	 * ranges from being reused for a new block group. This is because our
-	 * fs trim operation (btrfs_trim_fs() / btrfs_ioctl_fitrim()) is
+	 * ranges from being reused for a new block group. This is needed to
+	 * avoid races with trimming and scrub.
+	 *
+	 * An fs trim operation (btrfs_trim_fs() / btrfs_ioctl_fitrim()) is
 	 * completely transactionless, so while it is trimming a range the
 	 * currently running transaction might finish and a new one start,
 	 * allowing for new block groups to be created that can reuse the same
@@ -1095,7 +1118,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	 * in place until the extents have been discarded completely when
 	 * the transaction commit has completed.
 	 */
-	remove_em = (atomic_read(&block_group->trimming) == 0);
+	remove_em = (atomic_read(&block_group->frozen) == 0);
 	spin_unlock(&block_group->lock);
 
 	mutex_unlock(&fs_info->chunk_mutex);
@@ -1107,16 +1130,10 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	/* Once for the block groups rbtree */
 	btrfs_put_block_group(block_group);
 
-	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
-	if (ret > 0)
-		ret = -EIO;
+	ret = remove_block_group_item(trans, path, block_group);
 	if (ret < 0)
 		goto out;
 
-	ret = btrfs_del_item(trans, root, path);
-	if (ret)
-		goto out;
-
 	if (remove_em) {
 		struct extent_map_tree *em_tree;
 
@@ -1175,7 +1192,7 @@ struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
 	free_extent_map(em);
 
 	return btrfs_start_transaction_fallback_global_rsv(fs_info->extent_root,
-							   num_items, 1);
+							   num_items);
 }
 
 /*
@@ -1284,25 +1301,17 @@ static bool clean_pinned_extents(struct btrfs_trans_handle *trans,
 		ret = clear_extent_bits(&prev_trans->pinned_extents, start, end,
 					EXTENT_DIRTY);
 		if (ret)
-			goto err;
+			goto out;
 	}
 
 	ret = clear_extent_bits(&trans->transaction->pinned_extents, start, end,
 				EXTENT_DIRTY);
-	if (ret)
-		goto err;
+out:
 	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 	if (prev_trans)
 		btrfs_put_transaction(prev_trans);
 
-	return true;
-
-err:
-	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
-	if (prev_trans)
-		btrfs_put_transaction(prev_trans);
-	btrfs_dec_block_group_ro(bg);
-	return false;
+	return ret == 0;
 }
 
 /*
@@ -1400,8 +1409,10 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		 * We could have pending pinned extents for this block group,
 		 * just delete them, we don't care about them anymore.
 		 */
-		if (!clean_pinned_extents(trans, block_group))
+		if (!clean_pinned_extents(trans, block_group)) {
+			btrfs_dec_block_group_ro(block_group);
 			goto end_trans;
+		}
 
 		/*
 		 * At this point, the block_group is read only and should fail
@@ -1450,7 +1461,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 		/* Implicit trim during transaction commit. */
 		if (trimming)
-			btrfs_get_block_group_trimming(block_group);
+			btrfs_freeze_block_group(block_group);
 
 		/*
 		 * Btrfs_remove_chunk will abort the transaction if things go
@@ -1460,7 +1471,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 		if (ret) {
 			if (trimming)
-				btrfs_put_block_group_trimming(block_group);
+				btrfs_unfreeze_block_group(block_group);
 			goto end_trans;
 		}
 
@@ -1774,7 +1785,7 @@ static void link_block_group(struct btrfs_block_group *cache)
 }
 
 static struct btrfs_block_group *btrfs_create_block_group_cache(
-		struct btrfs_fs_info *fs_info, u64 start, u64 size)
+		struct btrfs_fs_info *fs_info, u64 start)
 {
 	struct btrfs_block_group *cache;
 
@@ -1790,7 +1801,6 @@ static struct btrfs_block_group *btrfs_create_block_group_cache(
 	}
 
 	cache->start = start;
-	cache->length = size;
 
 	cache->fs_info = fs_info;
 	cache->full_stripe_len = btrfs_full_stripe_len(fs_info, start);
@@ -1809,7 +1819,7 @@ static struct btrfs_block_group *btrfs_create_block_group_cache(
 	INIT_LIST_HEAD(&cache->dirty_list);
 	INIT_LIST_HEAD(&cache->io_list);
 	btrfs_init_free_space_ctl(cache);
-	atomic_set(&cache->trimming, 0);
+	atomic_set(&cache->frozen, 0);
 	mutex_init(&cache->free_space_lock);
 	btrfs_init_full_stripe_locks_tree(&cache->full_stripe_locks_root);
 
@@ -1870,25 +1880,44 @@ 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,
+				 struct btrfs_path *path,
+				 const struct btrfs_key *key)
+{
+	struct extent_buffer *leaf = path->nodes[0];
+	struct btrfs_block_group_item bgi;
+	int slot = path->slots[0];
+
+	cache->length = key->offset;
+
+	read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot),
+			   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,
 				struct btrfs_path *path,
 				const struct btrfs_key *key,
 				int need_clear)
 {
-	struct extent_buffer *leaf = path->nodes[0];
 	struct btrfs_block_group *cache;
 	struct btrfs_space_info *space_info;
-	struct btrfs_block_group_item bgi;
 	const bool mixed = btrfs_fs_incompat(info, MIXED_GROUPS);
-	int slot = path->slots[0];
 	int ret;
 
 	ASSERT(key->type == BTRFS_BLOCK_GROUP_ITEM_KEY);
 
-	cache = btrfs_create_block_group_cache(info, key->objectid, key->offset);
+	cache = btrfs_create_block_group_cache(info, key->objectid);
 	if (!cache)
 		return -ENOMEM;
 
+	ret = read_block_group_item(cache, path, key);
+	if (ret < 0)
+		goto error;
+
 	if (need_clear) {
 		/*
 		 * When we mount with old space cache, we need to
@@ -1903,10 +1932,6 @@ static int read_one_block_group(struct btrfs_fs_info *info,
 		if (btrfs_test_opt(info, SPACE_CACHE))
 			cache->disk_cache_state = BTRFS_DC_CLEAR;
 	}
-	read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot),
-			   sizeof(bgi));
-	cache->used = btrfs_stack_block_group_used(&bgi);
-	cache->flags = btrfs_stack_block_group_flags(&bgi);
 	if (!mixed && ((cache->flags & BTRFS_BLOCK_GROUP_METADATA) &&
 	    (cache->flags & BTRFS_BLOCK_GROUP_DATA))) {
 			btrfs_err(info,
@@ -1934,15 +1959,15 @@ static int read_one_block_group(struct btrfs_fs_info *info,
 	 * are empty, and we can just add all the space in and be done with it.
 	 * This saves us _a_lot_ of time, particularly in the full case.
 	 */
-	if (key->offset == cache->used) {
+	if (cache->length == cache->used) {
 		cache->last_byte_to_unpin = (u64)-1;
 		cache->cached = BTRFS_CACHE_FINISHED;
 		btrfs_free_excluded_extents(cache);
 	} else if (cache->used == 0) {
 		cache->last_byte_to_unpin = (u64)-1;
 		cache->cached = BTRFS_CACHE_FINISHED;
-		add_new_free_space(cache, key->objectid,
-				   key->objectid + key->offset);
+		add_new_free_space(cache, cache->start,
+				   cache->start + cache->length);
 		btrfs_free_excluded_extents(cache);
 	}
 
@@ -1952,7 +1977,7 @@ static int read_one_block_group(struct btrfs_fs_info *info,
 		goto error;
 	}
 	trace_btrfs_add_block_group(info, cache, 0);
-	btrfs_update_space_info(info, cache->flags, key->offset,
+	btrfs_update_space_info(info, cache->flags, cache->length,
 				cache->used, cache->bytes_super, &space_info);
 
 	cache->space_info = space_info;
@@ -1991,7 +2016,6 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
-	path->reada = READA_FORWARD;
 
 	cache_gen = btrfs_super_cache_generation(info->super_copy);
 	if (btrfs_test_opt(info, SPACE_CACHE) &&
@@ -2046,13 +2070,32 @@ error:
 	return ret;
 }
 
+static int insert_block_group_item(struct btrfs_trans_handle *trans,
+				   struct btrfs_block_group *block_group)
+{
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_block_group_item bgi;
+	struct btrfs_root *root;
+	struct btrfs_key key;
+
+	spin_lock(&block_group->lock);
+	btrfs_set_stack_block_group_used(&bgi, block_group->used);
+	btrfs_set_stack_block_group_chunk_objectid(&bgi,
+				BTRFS_FIRST_CHUNK_TREE_OBJECTID);
+	btrfs_set_stack_block_group_flags(&bgi, block_group->flags);
+	key.objectid = block_group->start;
+	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+	key.offset = block_group->length;
+	spin_unlock(&block_group->lock);
+
+	root = fs_info->extent_root;
+	return btrfs_insert_item(trans, root, &key, &bgi, sizeof(bgi));
+}
+
 void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_block_group *block_group;
-	struct btrfs_root *extent_root = fs_info->extent_root;
-	struct btrfs_block_group_item item;
-	struct btrfs_key key;
 	int ret = 0;
 
 	if (!trans->can_flush_pending_bgs)
@@ -2065,21 +2108,11 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 		if (ret)
 			goto next;
 
-		spin_lock(&block_group->lock);
-		btrfs_set_stack_block_group_used(&item, block_group->used);
-		btrfs_set_stack_block_group_chunk_objectid(&item,
-				BTRFS_FIRST_CHUNK_TREE_OBJECTID);
-		btrfs_set_stack_block_group_flags(&item, block_group->flags);
-		key.objectid = block_group->start;
-		key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
-		key.offset = block_group->length;
-		spin_unlock(&block_group->lock);
-
-		ret = btrfs_insert_item(trans, extent_root, &key, &item,
-					sizeof(item));
+		ret = insert_block_group_item(trans, block_group);
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
-		ret = btrfs_finish_chunk_alloc(trans, key.objectid, key.offset);
+		ret = btrfs_finish_chunk_alloc(trans, block_group->start,
+					block_group->length);
 		if (ret)
 			btrfs_abort_transaction(trans, ret);
 		add_block_group_free_space(trans, block_group);
@@ -2100,10 +2133,11 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used,
 
 	btrfs_set_log_full_commit(trans);
 
-	cache = btrfs_create_block_group_cache(fs_info, chunk_offset, size);
+	cache = btrfs_create_block_group_cache(fs_info, chunk_offset);
 	if (!cache)
 		return -ENOMEM;
 
+	cache->length = size;
 	cache->used = bytes_used;
 	cache->flags = type;
 	cache->last_byte_to_unpin = (u64)-1;
@@ -2314,13 +2348,13 @@ void btrfs_dec_block_group_ro(struct btrfs_block_group *cache)
 	spin_unlock(&sinfo->lock);
 }
 
-static int write_one_cache_group(struct btrfs_trans_handle *trans,
-				 struct btrfs_path *path,
-				 struct btrfs_block_group *cache)
+static int update_block_group_item(struct btrfs_trans_handle *trans,
+				   struct btrfs_path *path,
+				   struct btrfs_block_group *cache)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret;
-	struct btrfs_root *extent_root = fs_info->extent_root;
+	struct btrfs_root *root = fs_info->extent_root;
 	unsigned long bi;
 	struct extent_buffer *leaf;
 	struct btrfs_block_group_item bgi;
@@ -2330,7 +2364,7 @@ static int write_one_cache_group(struct btrfs_trans_handle *trans,
 	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
 	key.offset = cache->length;
 
-	ret = btrfs_search_slot(trans, extent_root, &key, path, 0, 1);
+	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
 	if (ret) {
 		if (ret > 0)
 			ret = -ENOENT;
@@ -2642,7 +2676,7 @@ again:
 			}
 		}
 		if (!ret) {
-			ret = write_one_cache_group(trans, path, cache);
+			ret = update_block_group_item(trans, path, cache);
 			/*
 			 * Our block group might still be attached to the list
 			 * of new block groups in the transaction handle of some
@@ -2791,7 +2825,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 			}
 		}
 		if (!ret) {
-			ret = write_one_cache_group(trans, path, cache);
+			ret = update_block_group_item(trans, path, cache);
 			/*
 			 * One of the free space endio workers might have
 			 * created a new block group while updating a free space
@@ -2808,7 +2842,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 			if (ret == -ENOENT) {
 				wait_event(cur_trans->writer_wait,
 				   atomic_read(&cur_trans->num_writers) == 1);
-				ret = write_one_cache_group(trans, path, cache);
+				ret = update_block_group_item(trans, path, cache);
 			}
 			if (ret)
 				btrfs_abort_transaction(trans, ret);
@@ -3384,3 +3418,44 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 	}
 	return 0;
 }
+
+void btrfs_freeze_block_group(struct btrfs_block_group *cache)
+{
+	atomic_inc(&cache->frozen);
+}
+
+void btrfs_unfreeze_block_group(struct btrfs_block_group *block_group)
+{
+	struct btrfs_fs_info *fs_info = block_group->fs_info;
+	struct extent_map_tree *em_tree;
+	struct extent_map *em;
+	bool cleanup;
+
+	spin_lock(&block_group->lock);
+	cleanup = (atomic_dec_and_test(&block_group->frozen) &&
+		   block_group->removed);
+	spin_unlock(&block_group->lock);
+
+	if (cleanup) {
+		mutex_lock(&fs_info->chunk_mutex);
+		em_tree = &fs_info->mapping_tree;
+		write_lock(&em_tree->lock);
+		em = lookup_extent_mapping(em_tree, block_group->start,
+					   1);
+		BUG_ON(!em); /* logic error, can't happen */
+		remove_extent_mapping(em_tree, em);
+		write_unlock(&em_tree->lock);
+		mutex_unlock(&fs_info->chunk_mutex);
+
+		/* once for us and once for the tree */
+		free_extent_map(em);
+		free_extent_map(em);
+
+		/*
+		 * We may have left one free space entry and other possible
+		 * tasks trimming this block group have left 1 entry each one.
+		 * Free them if any.
+		 */
+		__btrfs_remove_free_space_cache(block_group->free_space_ctl);
+	}
+}
diff --git a/fs/btrfs/block-group.h b/fs/btrfs/block-group.h
index 107bb557ca8d..b6ee70a039c7 100644
--- a/fs/btrfs/block-group.h
+++ b/fs/btrfs/block-group.h
@@ -129,8 +129,17 @@ struct btrfs_block_group {
 	/* For read-only block groups */
 	struct list_head ro_list;
 
+	/*
+	 * When non-zero it means the block group's logical address and its
+	 * device extents can not be reused for future block group allocations
+	 * until the counter goes down to 0. This is to prevent them from being
+	 * reused while some task is still using the block group after it was
+	 * deleted - we want to make sure they can only be reused for new block
+	 * groups after that task is done with the deleted block group.
+	 */
+	atomic_t frozen;
+
 	/* For discard operations */
-	atomic_t trimming;
 	struct list_head discard_list;
 	int discard_index;
 	u64 discard_eligible_time;
@@ -283,6 +292,9 @@ static inline int btrfs_block_group_done(struct btrfs_block_group *cache)
 		cache->cached == BTRFS_CACHE_ERROR;
 }
 
+void btrfs_freeze_block_group(struct btrfs_block_group *cache);
+void btrfs_unfreeze_block_group(struct btrfs_block_group *cache);
+
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
 		     u64 physical, u64 **logical, int *naddrs, int *stripe_len);
diff --git a/fs/btrfs/block-rsv.c b/fs/btrfs/block-rsv.c
index 27efec8f7c5b..7e1549a84fcc 100644
--- a/fs/btrfs/block-rsv.c
+++ b/fs/btrfs/block-rsv.c
@@ -5,6 +5,7 @@
 #include "block-rsv.h"
 #include "space-info.h"
 #include "transaction.h"
+#include "block-group.h"
 
 /*
  * HOW DO BLOCK RESERVES WORK
@@ -405,6 +406,8 @@ void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info)
 	else
 		block_rsv->full = 0;
 
+	if (block_rsv->size >= sinfo->total_bytes)
+		sinfo->force_alloc = CHUNK_ALLOC_FORCE;
 	spin_unlock(&block_rsv->lock);
 	spin_unlock(&sinfo->lock);
 }
@@ -455,7 +458,7 @@ static struct btrfs_block_rsv *get_block_rsv(
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_block_rsv *block_rsv = NULL;
 
-	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) ||
+	if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state) ||
 	    (root == fs_info->csum_root && trans->adding_csums) ||
 	    (root == fs_info->uuid_root))
 		block_rsv = trans->block_rsv;
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 27a1fefce508..aeff56a0e105 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -7,6 +7,7 @@
 #define BTRFS_INODE_H
 
 #include <linux/hash.h>
+#include <linux/refcount.h>
 #include "extent_map.h"
 #include "extent_io.h"
 #include "ordered-data.h"
@@ -27,7 +28,6 @@ enum {
 	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,
 };
@@ -293,53 +293,25 @@ static inline int btrfs_inode_in_log(struct btrfs_inode *inode, u64 generation)
 	return ret;
 }
 
-#define BTRFS_DIO_ORIG_BIO_SUBMITTED	0x1
-
 struct btrfs_dio_private {
 	struct inode *inode;
-	unsigned long flags;
 	u64 logical_offset;
 	u64 disk_bytenr;
 	u64 bytes;
-	void *private;
-
-	/* number of bios pending for this dio */
-	atomic_t pending_bios;
 
-	/* IO errors */
-	int errors;
-
-	/* orig_bio is our btrfs_io_bio */
-	struct bio *orig_bio;
+	/*
+	 * References to this structure. There is one reference per in-flight
+	 * bio plus one while we're still setting up.
+	 */
+	refcount_t refs;
 
 	/* dio_bio came from fs/direct-io.c */
 	struct bio *dio_bio;
 
-	/*
-	 * The original bio may be split to several sub-bios, this is
-	 * done during endio of sub-bios
-	 */
-	blk_status_t (*subio_endio)(struct inode *, struct btrfs_io_bio *,
-			blk_status_t);
+	/* Array of checksums */
+	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 9ab610cc9114..c6e648603f85 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -194,11 +194,9 @@ static int check_compressed_csum(struct btrfs_inode *inode,
 	for (i = 0; i < cb->nr_pages; i++) {
 		page = cb->compressed_pages[i];
 
-		crypto_shash_init(shash);
 		kaddr = kmap_atomic(page);
-		crypto_shash_update(shash, kaddr, PAGE_SIZE);
+		crypto_shash_digest(shash, kaddr, PAGE_SIZE, csum);
 		kunmap_atomic(kaddr);
-		crypto_shash_final(shash, (u8 *)&csum);
 
 		if (memcmp(&csum, cb_sum, csum_size)) {
 			btrfs_print_data_csum_error(inode, disk_start,
@@ -1142,6 +1140,22 @@ static void put_workspace(int type, struct list_head *ws)
 }
 
 /*
+ * Adjust @level according to the limits of the compression algorithm or
+ * fallback to default
+ */
+static unsigned int btrfs_compress_set_level(int type, unsigned level)
+{
+	const struct btrfs_compress_op *ops = btrfs_compress_op[type];
+
+	if (level == 0)
+		level = ops->default_level;
+	else
+		level = min(level, ops->max_level);
+
+	return level;
+}
+
+/*
  * Given an address space and start and length, compress the bytes into @pages
  * that are allocated on demand.
  *
@@ -1748,19 +1762,3 @@ unsigned int btrfs_compress_str2level(unsigned int type, const char *str)
 
 	return level;
 }
-
-/*
- * Adjust @level according to the limits of the compression algorithm or
- * fallback to default
- */
-unsigned int btrfs_compress_set_level(int type, unsigned level)
-{
-	const struct btrfs_compress_op *ops = btrfs_compress_op[type];
-
-	if (level == 0)
-		level = ops->default_level;
-	else
-		level = min(level, ops->max_level);
-
-	return level;
-}
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index d253f7aa8ed5..284a3ad31350 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -140,8 +140,6 @@ extern const struct btrfs_compress_op btrfs_zstd_compress;
 const char* btrfs_compress_type2str(enum btrfs_compression_type type);
 bool btrfs_compress_is_valid_type(const char *str, size_t len);
 
-unsigned int btrfs_compress_set_level(int type, unsigned level);
-
 int btrfs_compress_heuristic(struct inode *inode, u64 start, u64 end);
 
 #endif
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index bfedbbe2311f..3a7648bff42c 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -144,9 +144,10 @@ struct extent_buffer *btrfs_root_node(struct btrfs_root *root)
 	return eb;
 }
 
-/* cowonly root (everything not a reference counted cow subvolume), just get
- * put onto a simple dirty list.  transaction.c walks this to make sure they
- * get properly updated on disk.
+/*
+ * Cowonly root (not-shareable trees, everything not subvolume or reloc roots),
+ * just get put onto a simple dirty list.  Transaction walks this list to make
+ * sure they get properly updated on disk.
  */
 static void add_root_to_dirty_list(struct btrfs_root *root)
 {
@@ -185,9 +186,9 @@ int btrfs_copy_root(struct btrfs_trans_handle *trans,
 	int level;
 	struct btrfs_disk_key disk_key;
 
-	WARN_ON(test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
+	WARN_ON(test_bit(BTRFS_ROOT_SHAREABLE, &root->state) &&
 		trans->transid != fs_info->running_transaction->transid);
-	WARN_ON(test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
+	WARN_ON(test_bit(BTRFS_ROOT_SHAREABLE, &root->state) &&
 		trans->transid != root->last_trans);
 
 	level = btrfs_header_level(buf);
@@ -826,12 +827,11 @@ int btrfs_block_can_be_shared(struct btrfs_root *root,
 			      struct extent_buffer *buf)
 {
 	/*
-	 * Tree blocks not in reference counted trees and tree roots
-	 * are never shared. If a block was allocated after the last
-	 * snapshot and the block was not allocated by tree relocation,
-	 * we know the block is not shared.
+	 * Tree blocks not in shareable trees and tree roots are never shared.
+	 * If a block was allocated after the last snapshot and the block was
+	 * not allocated by tree relocation, we know the block is not shared.
 	 */
-	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
+	if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state) &&
 	    buf != root->node && buf != root->commit_root &&
 	    (btrfs_header_generation(buf) <=
 	     btrfs_root_last_snapshot(&root->root_item) ||
@@ -1024,9 +1024,9 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 
 	btrfs_assert_tree_locked(buf);
 
-	WARN_ON(test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
+	WARN_ON(test_bit(BTRFS_ROOT_SHAREABLE, &root->state) &&
 		trans->transid != fs_info->running_transaction->transid);
-	WARN_ON(test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
+	WARN_ON(test_bit(BTRFS_ROOT_SHAREABLE, &root->state) &&
 		trans->transid != root->last_trans);
 
 	level = btrfs_header_level(buf);
@@ -1065,7 +1065,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		return ret;
 	}
 
-	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) {
+	if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) {
 		ret = btrfs_reloc_cow_block(trans, root, buf, cow);
 		if (ret) {
 			btrfs_abort_transaction(trans, ret);
@@ -1668,15 +1668,8 @@ static noinline int generic_bin_search(struct extent_buffer *eb,
 {
 	int low = 0;
 	int high = max;
-	int mid;
 	int ret;
-	struct btrfs_disk_key *tmp = NULL;
-	struct btrfs_disk_key unaligned;
-	unsigned long offset;
-	char *kaddr = NULL;
-	unsigned long map_start = 0;
-	unsigned long map_len = 0;
-	int err;
+	const int key_size = sizeof(struct btrfs_disk_key);
 
 	if (low > high) {
 		btrfs_err(eb->fs_info,
@@ -1687,32 +1680,26 @@ static noinline int generic_bin_search(struct extent_buffer *eb,
 	}
 
 	while (low < high) {
+		unsigned long oip;
+		unsigned long offset;
+		struct btrfs_disk_key *tmp;
+		struct btrfs_disk_key unaligned;
+		int mid;
+
 		mid = (low + high) / 2;
 		offset = p + mid * item_size;
+		oip = offset_in_page(offset);
 
-		if (!kaddr || offset < map_start ||
-		    (offset + sizeof(struct btrfs_disk_key)) >
-		    map_start + map_len) {
-
-			err = map_private_extent_buffer(eb, offset,
-						sizeof(struct btrfs_disk_key),
-						&kaddr, &map_start, &map_len);
-
-			if (!err) {
-				tmp = (struct btrfs_disk_key *)(kaddr + offset -
-							map_start);
-			} else if (err == 1) {
-				read_extent_buffer(eb, &unaligned,
-						   offset, sizeof(unaligned));
-				tmp = &unaligned;
-			} else {
-				return err;
-			}
+		if (oip + key_size <= PAGE_SIZE) {
+			const unsigned long idx = offset >> PAGE_SHIFT;
+			char *kaddr = page_address(eb->pages[idx]);
 
+			tmp = (struct btrfs_disk_key *)(kaddr + oip);
 		} else {
-			tmp = (struct btrfs_disk_key *)(kaddr + offset -
-							map_start);
+			read_extent_buffer(eb, &unaligned, offset, key_size);
+			tmp = &unaligned;
 		}
+
 		ret = comp_keys(tmp, key);
 
 		if (ret < 0)
@@ -1733,9 +1720,9 @@ static noinline int generic_bin_search(struct extent_buffer *eb,
  * leaves vs nodes
  */
 int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
-		     int level, int *slot)
+		     int *slot)
 {
-	if (level == 0)
+	if (btrfs_header_level(eb) == 0)
 		return generic_bin_search(eb,
 					  offsetof(struct btrfs_leaf, items),
 					  sizeof(struct btrfs_item),
@@ -2348,16 +2335,15 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	u64 blocknr;
 	u64 gen;
-	struct extent_buffer *b = *eb_ret;
 	struct extent_buffer *tmp;
 	struct btrfs_key first_key;
 	int ret;
 	int parent_level;
 
-	blocknr = btrfs_node_blockptr(b, slot);
-	gen = btrfs_node_ptr_generation(b, slot);
-	parent_level = btrfs_header_level(b);
-	btrfs_node_key_to_cpu(b, &first_key, slot);
+	blocknr = btrfs_node_blockptr(*eb_ret, slot);
+	gen = btrfs_node_ptr_generation(*eb_ret, slot);
+	parent_level = btrfs_header_level(*eb_ret);
+	btrfs_node_key_to_cpu(*eb_ret, &first_key, slot);
 
 	tmp = find_extent_buffer(fs_info, blocknr);
 	if (tmp) {
@@ -2501,19 +2487,6 @@ done:
 	return ret;
 }
 
-static int key_search(struct extent_buffer *b, const struct btrfs_key *key,
-		      int level, int *prev_cmp, int *slot)
-{
-	if (*prev_cmp != 0) {
-		*prev_cmp = btrfs_bin_search(b, key, level, slot);
-		return *prev_cmp;
-	}
-
-	*slot = 0;
-
-	return 0;
-}
-
 int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
 		u64 iobjectid, u64 ioff, u8 key_type,
 		struct btrfs_key *found_key)
@@ -2783,9 +2756,23 @@ cow_done:
 			}
 		}
 
-		ret = key_search(b, key, level, &prev_cmp, &slot);
-		if (ret < 0)
-			goto done;
+		/*
+		 * If btrfs_bin_search returns an exact match (prev_cmp == 0)
+		 * we can safely assume the target key will always be in slot 0
+		 * on lower levels due to the invariants BTRFS' btree provides,
+		 * namely that a btrfs_key_ptr entry always points to the
+		 * lowest key in the child node, thus we can skip searching
+		 * lower levels
+		 */
+		if (prev_cmp == 0) {
+			slot = 0;
+			ret = 0;
+		} else {
+			ret = btrfs_bin_search(b, key, &slot);
+			prev_cmp = ret;
+			if (ret < 0)
+				goto done;
+		}
 
 		if (level == 0) {
 			p->slots[level] = slot;
@@ -2909,7 +2896,6 @@ int btrfs_search_old_slot(struct btrfs_root *root, const struct btrfs_key *key,
 	int level;
 	int lowest_unlock = 1;
 	u8 lowest_level = 0;
-	int prev_cmp = -1;
 
 	lowest_level = p->lowest_level;
 	WARN_ON(p->nodes[0] != NULL);
@@ -2942,12 +2928,7 @@ again:
 		 */
 		btrfs_unlock_up_safe(p, level + 1);
 
-		/*
-		 * Since we can unwind ebs we want to do a real search every
-		 * time.
-		 */
-		prev_cmp = -1;
-		ret = key_search(b, key, level, &prev_cmp, &slot);
+		ret = btrfs_bin_search(b, key, &slot);
 		if (ret < 0)
 			goto done;
 
@@ -3507,19 +3488,17 @@ static int leaf_space_used(struct extent_buffer *l, int start, int nr)
 {
 	struct btrfs_item *start_item;
 	struct btrfs_item *end_item;
-	struct btrfs_map_token token;
 	int data_len;
 	int nritems = btrfs_header_nritems(l);
 	int end = min(nritems, start + nr) - 1;
 
 	if (!nr)
 		return 0;
-	btrfs_init_map_token(&token, l);
 	start_item = btrfs_item_nr(start);
 	end_item = btrfs_item_nr(end);
-	data_len = btrfs_token_item_offset(l, start_item, &token) +
-		btrfs_token_item_size(l, start_item, &token);
-	data_len = data_len - btrfs_token_item_offset(l, end_item, &token);
+	data_len = btrfs_item_offset(l, start_item) +
+		   btrfs_item_size(l, start_item);
+	data_len = data_len - btrfs_item_offset(l, end_item);
 	data_len += sizeof(struct btrfs_item) * nr;
 	WARN_ON(data_len < 0);
 	return data_len;
@@ -3650,8 +3629,8 @@ static noinline int __push_leaf_right(struct btrfs_path *path,
 	push_space = BTRFS_LEAF_DATA_SIZE(fs_info);
 	for (i = 0; i < right_nritems; i++) {
 		item = btrfs_item_nr(i);
-		push_space -= btrfs_token_item_size(right, item, &token);
-		btrfs_set_token_item_offset(right, item, push_space, &token);
+		push_space -= btrfs_token_item_size(&token, item);
+		btrfs_set_token_item_offset(&token, item, push_space);
 	}
 
 	left_nritems -= push_items;
@@ -3859,10 +3838,9 @@ static noinline int __push_leaf_left(struct btrfs_path *path, int data_size,
 
 		item = btrfs_item_nr(i);
 
-		ioff = btrfs_token_item_offset(left, item, &token);
-		btrfs_set_token_item_offset(left, item,
-		      ioff - (BTRFS_LEAF_DATA_SIZE(fs_info) - old_left_item_size),
-		      &token);
+		ioff = btrfs_token_item_offset(&token, item);
+		btrfs_set_token_item_offset(&token, item,
+		      ioff - (BTRFS_LEAF_DATA_SIZE(fs_info) - old_left_item_size));
 	}
 	btrfs_set_header_nritems(left, old_left_nritems + push_items);
 
@@ -3892,9 +3870,8 @@ static noinline int __push_leaf_left(struct btrfs_path *path, int data_size,
 	for (i = 0; i < right_nritems; i++) {
 		item = btrfs_item_nr(i);
 
-		push_space = push_space - btrfs_token_item_size(right,
-								item, &token);
-		btrfs_set_token_item_offset(right, item, push_space, &token);
+		push_space = push_space - btrfs_token_item_size(&token, item);
+		btrfs_set_token_item_offset(&token, item, push_space);
 	}
 
 	btrfs_mark_buffer_dirty(left);
@@ -4036,9 +4013,8 @@ static noinline void copy_for_split(struct btrfs_trans_handle *trans,
 		struct btrfs_item *item = btrfs_item_nr(i);
 		u32 ioff;
 
-		ioff = btrfs_token_item_offset(right, item, &token);
-		btrfs_set_token_item_offset(right, item,
-					    ioff + rt_data_off, &token);
+		ioff = btrfs_token_item_offset(&token, item);
+		btrfs_set_token_item_offset(&token, item, ioff + rt_data_off);
 	}
 
 	btrfs_set_header_nritems(l, mid);
@@ -4541,9 +4517,8 @@ void btrfs_truncate_item(struct btrfs_path *path, u32 new_size, int from_end)
 		u32 ioff;
 		item = btrfs_item_nr(i);
 
-		ioff = btrfs_token_item_offset(leaf, item, &token);
-		btrfs_set_token_item_offset(leaf, item,
-					    ioff + size_diff, &token);
+		ioff = btrfs_token_item_offset(&token, item);
+		btrfs_set_token_item_offset(&token, item, ioff + size_diff);
 	}
 
 	/* shift the data */
@@ -4640,9 +4615,8 @@ void btrfs_extend_item(struct btrfs_path *path, u32 data_size)
 		u32 ioff;
 		item = btrfs_item_nr(i);
 
-		ioff = btrfs_token_item_offset(leaf, item, &token);
-		btrfs_set_token_item_offset(leaf, item,
-					    ioff - data_size, &token);
+		ioff = btrfs_token_item_offset(&token, item);
+		btrfs_set_token_item_offset(&token, item, ioff - data_size);
 	}
 
 	/* shift the data */
@@ -4718,9 +4692,9 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 			u32 ioff;
 
 			item = btrfs_item_nr(i);
-			ioff = btrfs_token_item_offset(leaf, item, &token);
-			btrfs_set_token_item_offset(leaf, item,
-						    ioff - total_data, &token);
+			ioff = btrfs_token_item_offset(&token, item);
+			btrfs_set_token_item_offset(&token, item,
+						    ioff - total_data);
 		}
 		/* shift the items */
 		memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr),
@@ -4739,10 +4713,9 @@ 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(leaf, item,
-					    data_end - data_size[i], &token);
+		btrfs_set_token_item_offset(&token, item, data_end - data_size[i]);
 		data_end -= data_size[i];
-		btrfs_set_token_item_size(leaf, item, data_size[i], &token);
+		btrfs_set_token_item_size(&token, item, data_size[i]);
 	}
 
 	btrfs_set_header_nritems(leaf, nritems + nr);
@@ -4930,9 +4903,8 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 			u32 ioff;
 
 			item = btrfs_item_nr(i);
-			ioff = btrfs_token_item_offset(leaf, item, &token);
-			btrfs_set_token_item_offset(leaf, item,
-						    ioff + dsize, &token);
+			ioff = btrfs_token_item_offset(&token, item);
+			btrfs_set_token_item_offset(&token, item, ioff + dsize);
 		}
 
 		memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot),
@@ -5103,7 +5075,7 @@ again:
 	while (1) {
 		nritems = btrfs_header_nritems(cur);
 		level = btrfs_header_level(cur);
-		sret = btrfs_bin_search(cur, min_key, level, &slot);
+		sret = btrfs_bin_search(cur, min_key, &slot);
 		if (sret < 0) {
 			ret = sret;
 			goto out;
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 8aa7b9dac405..161533040978 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -28,6 +28,7 @@
 #include <linux/dynamic_debug.h>
 #include <linux/refcount.h>
 #include <linux/crc32c.h>
+#include <linux/iomap.h>
 #include "extent-io-tree.h"
 #include "extent_io.h"
 #include "extent_map.h"
@@ -582,6 +583,7 @@ struct btrfs_fs_info {
 	struct btrfs_root *quota_root;
 	struct btrfs_root *uuid_root;
 	struct btrfs_root *free_space_root;
+	struct btrfs_root *data_reloc_root;
 
 	/* the log root tree is a directory of all the other log roots */
 	struct btrfs_root *log_root_tree;
@@ -758,7 +760,6 @@ struct btrfs_fs_info {
 	struct btrfs_workqueue *endio_workers;
 	struct btrfs_workqueue *endio_meta_workers;
 	struct btrfs_workqueue *endio_raid56_workers;
-	struct btrfs_workqueue *endio_repair_workers;
 	struct btrfs_workqueue *rmw_workers;
 	struct btrfs_workqueue *endio_meta_write_workers;
 	struct btrfs_workqueue *endio_write_workers;
@@ -970,7 +971,28 @@ enum {
 	 * is used to tell us when more checks are required
 	 */
 	BTRFS_ROOT_IN_TRANS_SETUP,
-	BTRFS_ROOT_REF_COWS,
+
+	/*
+	 * Set if tree blocks of this root can be shared by other roots.
+	 * Only subvolume trees and their reloc trees have this bit set.
+	 * Conflicts with TRACK_DIRTY bit.
+	 *
+	 * This affects two things:
+	 *
+	 * - How balance works
+	 *   For shareable roots, we need to use reloc tree and do path
+	 *   replacement for balance, and need various pre/post hooks for
+	 *   snapshot creation to handle them.
+	 *
+	 *   While for non-shareable trees, we just simply do a tree search
+	 *   with COW.
+	 *
+	 * - How dirty roots are tracked
+	 *   For shareable roots, btrfs_record_root_in_trans() is needed to
+	 *   track them, while non-subvolume roots have TRACK_DIRTY bit, they
+	 *   don't need to set this manually.
+	 */
+	BTRFS_ROOT_SHAREABLE,
 	BTRFS_ROOT_TRACK_DIRTY,
 	BTRFS_ROOT_IN_RADIX,
 	BTRFS_ROOT_ORPHAN_ITEM_INSERTED,
@@ -1056,7 +1078,7 @@ struct btrfs_root {
 	struct btrfs_key defrag_progress;
 	struct btrfs_key defrag_max;
 
-	/* the dirty list is only used by non-reference counted roots */
+	/* The dirty list is only used by non-shareable roots */
 	struct list_head dirty_list;
 
 	struct list_head root_list;
@@ -1146,6 +1168,9 @@ struct btrfs_root {
 	/* Record pairs of swapped blocks for qgroup */
 	struct btrfs_qgroup_swapped_blocks swapped_blocks;
 
+	/* Used only by log trees, when logging csum items */
+	struct extent_io_tree log_csum_range;
+
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 	u64 alloc_bytenr;
 #endif
@@ -1341,7 +1366,7 @@ do {                                                                   \
 	 BTRFS_INODE_ROOT_ITEM_INIT)
 
 struct btrfs_map_token {
-	const struct extent_buffer *eb;
+	struct extent_buffer *eb;
 	char *kaddr;
 	unsigned long offset;
 };
@@ -1353,7 +1378,8 @@ static inline void btrfs_init_map_token(struct btrfs_map_token *token,
 					struct extent_buffer *eb)
 {
 	token->eb = eb;
-	token->kaddr = NULL;
+	token->kaddr = page_address(eb->pages[0]);
+	token->offset = 0;
 }
 
 /* some macros to generate set/get functions for the struct fields.  This
@@ -1377,15 +1403,14 @@ static inline void btrfs_init_map_token(struct btrfs_map_token *token,
 			   sizeof(((type *)0)->member)))
 
 #define DECLARE_BTRFS_SETGET_BITS(bits)					\
-u##bits btrfs_get_token_##bits(const struct extent_buffer *eb,		\
-			       const void *ptr, unsigned long off,	\
-			       struct btrfs_map_token *token);		\
-void btrfs_set_token_##bits(struct extent_buffer *eb, const void *ptr,	\
-			    unsigned long off, u##bits val,		\
-			    struct btrfs_map_token *token);		\
+u##bits btrfs_get_token_##bits(struct btrfs_map_token *token,		\
+			       const void *ptr, unsigned long off);	\
+void btrfs_set_token_##bits(struct btrfs_map_token *token,		\
+			    const void *ptr, unsigned long off,		\
+			    u##bits val);				\
 u##bits btrfs_get_##bits(const struct extent_buffer *eb,		\
 			 const void *ptr, unsigned long off);		\
-void btrfs_set_##bits(struct extent_buffer *eb, void *ptr,		\
+void btrfs_set_##bits(const struct extent_buffer *eb, void *ptr,	\
 		      unsigned long off, u##bits val);
 
 DECLARE_BTRFS_SETGET_BITS(8)
@@ -1400,25 +1425,23 @@ static inline u##bits btrfs_##name(const struct extent_buffer *eb,	\
 	BUILD_BUG_ON(sizeof(u##bits) != sizeof(((type *)0))->member);	\
 	return btrfs_get_##bits(eb, s, offsetof(type, member));		\
 }									\
-static inline void btrfs_set_##name(struct extent_buffer *eb, type *s,	\
+static inline void btrfs_set_##name(const struct extent_buffer *eb, type *s, \
 				    u##bits val)			\
 {									\
 	BUILD_BUG_ON(sizeof(u##bits) != sizeof(((type *)0))->member);	\
 	btrfs_set_##bits(eb, s, offsetof(type, member), val);		\
 }									\
-static inline u##bits btrfs_token_##name(const struct extent_buffer *eb,\
-					 const type *s,			\
-					 struct btrfs_map_token *token)	\
+static inline u##bits btrfs_token_##name(struct btrfs_map_token *token,	\
+					 const type *s)			\
 {									\
 	BUILD_BUG_ON(sizeof(u##bits) != sizeof(((type *)0))->member);	\
-	return btrfs_get_token_##bits(eb, s, offsetof(type, member), token); \
+	return btrfs_get_token_##bits(token, s, offsetof(type, member));\
 }									\
-static inline void btrfs_set_token_##name(struct extent_buffer *eb,	\
-					  type *s, u##bits val,		\
-                                         struct btrfs_map_token *token)	\
+static inline void btrfs_set_token_##name(struct btrfs_map_token *token,\
+					  type *s, u##bits val)		\
 {									\
 	BUILD_BUG_ON(sizeof(u##bits) != sizeof(((type *)0))->member);	\
-	btrfs_set_token_##bits(eb, s, offsetof(type, member), val, token); \
+	btrfs_set_token_##bits(token, s, offsetof(type, member), val);	\
 }
 
 #define BTRFS_SETGET_HEADER_FUNCS(name, type, member, bits)		\
@@ -1428,7 +1451,7 @@ static inline u##bits btrfs_##name(const struct extent_buffer *eb)	\
 	u##bits res = le##bits##_to_cpu(p->member);			\
 	return res;							\
 }									\
-static inline void btrfs_set_##name(struct extent_buffer *eb,		\
+static inline void btrfs_set_##name(const struct extent_buffer *eb,	\
 				    u##bits val)			\
 {									\
 	type *p = page_address(eb->pages[0]);				\
@@ -1446,7 +1469,7 @@ static inline void btrfs_set_##name(type *s, u##bits val)		\
 }
 
 
-static inline u64 btrfs_device_total_bytes(struct extent_buffer *eb,
+static inline u64 btrfs_device_total_bytes(const struct extent_buffer *eb,
 					   struct btrfs_dev_item *s)
 {
 	BUILD_BUG_ON(sizeof(u64) !=
@@ -1454,7 +1477,7 @@ static inline u64 btrfs_device_total_bytes(struct extent_buffer *eb,
 	return btrfs_get_64(eb, s, offsetof(struct btrfs_dev_item,
 					    total_bytes));
 }
-static inline void btrfs_set_device_total_bytes(struct extent_buffer *eb,
+static inline void btrfs_set_device_total_bytes(const struct extent_buffer *eb,
 						struct btrfs_dev_item *s,
 						u64 val)
 {
@@ -1558,13 +1581,13 @@ static inline char *btrfs_stripe_dev_uuid_nr(struct btrfs_chunk *c, int nr)
 	return btrfs_stripe_dev_uuid(btrfs_stripe_nr(c, nr));
 }
 
-static inline u64 btrfs_stripe_offset_nr(struct extent_buffer *eb,
+static inline u64 btrfs_stripe_offset_nr(const struct extent_buffer *eb,
 					 struct btrfs_chunk *c, int nr)
 {
 	return btrfs_stripe_offset(eb, btrfs_stripe_nr(c, nr));
 }
 
-static inline u64 btrfs_stripe_devid_nr(struct extent_buffer *eb,
+static inline u64 btrfs_stripe_devid_nr(const struct extent_buffer *eb,
 					 struct btrfs_chunk *c, int nr)
 {
 	return btrfs_stripe_devid(eb, btrfs_stripe_nr(c, nr));
@@ -1644,31 +1667,21 @@ BTRFS_SETGET_FUNCS(dev_extent_chunk_objectid, struct btrfs_dev_extent,
 BTRFS_SETGET_FUNCS(dev_extent_chunk_offset, struct btrfs_dev_extent,
 		   chunk_offset, 64);
 BTRFS_SETGET_FUNCS(dev_extent_length, struct btrfs_dev_extent, length, 64);
-
-static inline unsigned long btrfs_dev_extent_chunk_tree_uuid(struct btrfs_dev_extent *dev)
-{
-	unsigned long ptr = offsetof(struct btrfs_dev_extent, chunk_tree_uuid);
-	return (unsigned long)dev + ptr;
-}
-
 BTRFS_SETGET_FUNCS(extent_refs, struct btrfs_extent_item, refs, 64);
 BTRFS_SETGET_FUNCS(extent_generation, struct btrfs_extent_item,
 		   generation, 64);
 BTRFS_SETGET_FUNCS(extent_flags, struct btrfs_extent_item, flags, 64);
 
-BTRFS_SETGET_FUNCS(extent_refs_v0, struct btrfs_extent_item_v0, refs, 32);
-
-
 BTRFS_SETGET_FUNCS(tree_block_level, struct btrfs_tree_block_info, level, 8);
 
-static inline void btrfs_tree_block_key(struct extent_buffer *eb,
+static inline void btrfs_tree_block_key(const struct extent_buffer *eb,
 					struct btrfs_tree_block_info *item,
 					struct btrfs_disk_key *key)
 {
 	read_eb_member(eb, item, struct btrfs_tree_block_info, key, key);
 }
 
-static inline void btrfs_set_tree_block_key(struct extent_buffer *eb,
+static inline void btrfs_set_tree_block_key(const struct extent_buffer *eb,
 					    struct btrfs_tree_block_info *item,
 					    struct btrfs_disk_key *key)
 {
@@ -1706,12 +1719,6 @@ static inline u32 btrfs_extent_inline_ref_size(int type)
 	return 0;
 }
 
-BTRFS_SETGET_FUNCS(ref_root_v0, struct btrfs_extent_ref_v0, root, 64);
-BTRFS_SETGET_FUNCS(ref_generation_v0, struct btrfs_extent_ref_v0,
-		   generation, 64);
-BTRFS_SETGET_FUNCS(ref_objectid_v0, struct btrfs_extent_ref_v0, objectid, 64);
-BTRFS_SETGET_FUNCS(ref_count_v0, struct btrfs_extent_ref_v0, count, 32);
-
 /* struct btrfs_node */
 BTRFS_SETGET_FUNCS(key_blockptr, struct btrfs_key_ptr, blockptr, 64);
 BTRFS_SETGET_FUNCS(key_generation, struct btrfs_key_ptr, generation, 64);
@@ -1720,7 +1727,7 @@ BTRFS_SETGET_STACK_FUNCS(stack_key_blockptr, struct btrfs_key_ptr,
 BTRFS_SETGET_STACK_FUNCS(stack_key_generation, struct btrfs_key_ptr,
 			 generation, 64);
 
-static inline u64 btrfs_node_blockptr(struct extent_buffer *eb, int nr)
+static inline u64 btrfs_node_blockptr(const struct extent_buffer *eb, int nr)
 {
 	unsigned long ptr;
 	ptr = offsetof(struct btrfs_node, ptrs) +
@@ -1728,7 +1735,7 @@ static inline u64 btrfs_node_blockptr(struct extent_buffer *eb, int nr)
 	return btrfs_key_blockptr(eb, (struct btrfs_key_ptr *)ptr);
 }
 
-static inline void btrfs_set_node_blockptr(struct extent_buffer *eb,
+static inline void btrfs_set_node_blockptr(const struct extent_buffer *eb,
 					   int nr, u64 val)
 {
 	unsigned long ptr;
@@ -1737,7 +1744,7 @@ static inline void btrfs_set_node_blockptr(struct extent_buffer *eb,
 	btrfs_set_key_blockptr(eb, (struct btrfs_key_ptr *)ptr, val);
 }
 
-static inline u64 btrfs_node_ptr_generation(struct extent_buffer *eb, int nr)
+static inline u64 btrfs_node_ptr_generation(const struct extent_buffer *eb, int nr)
 {
 	unsigned long ptr;
 	ptr = offsetof(struct btrfs_node, ptrs) +
@@ -1745,7 +1752,7 @@ static inline u64 btrfs_node_ptr_generation(struct extent_buffer *eb, int nr)
 	return btrfs_key_generation(eb, (struct btrfs_key_ptr *)ptr);
 }
 
-static inline void btrfs_set_node_ptr_generation(struct extent_buffer *eb,
+static inline void btrfs_set_node_ptr_generation(const struct extent_buffer *eb,
 						 int nr, u64 val)
 {
 	unsigned long ptr;
@@ -1763,7 +1770,7 @@ static inline unsigned long btrfs_node_key_ptr_offset(int nr)
 void btrfs_node_key(const struct extent_buffer *eb,
 		    struct btrfs_disk_key *disk_key, int nr);
 
-static inline void btrfs_set_node_key(struct extent_buffer *eb,
+static inline void btrfs_set_node_key(const struct extent_buffer *eb,
 				      struct btrfs_disk_key *disk_key, int nr)
 {
 	unsigned long ptr;
@@ -2498,8 +2505,6 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 			 struct btrfs_ref *generic_ref);
 
 int btrfs_extent_readonly(struct btrfs_fs_info *fs_info, u64 bytenr);
-void btrfs_get_block_group_trimming(struct btrfs_block_group *cache);
-void btrfs_put_block_group_trimming(struct btrfs_block_group *cache);
 void btrfs_clear_space_info_full(struct btrfs_fs_info *info);
 
 enum btrfs_reserve_flush_enum {
@@ -2512,6 +2517,7 @@ enum btrfs_reserve_flush_enum {
 	BTRFS_RESERVE_FLUSH_LIMIT,
 	BTRFS_RESERVE_FLUSH_EVICT,
 	BTRFS_RESERVE_FLUSH_ALL,
+	BTRFS_RESERVE_FLUSH_ALL_STEAL,
 };
 
 enum btrfs_flush_state {
@@ -2551,7 +2557,7 @@ void btrfs_wait_for_snapshot_creation(struct btrfs_root *root);
 
 /* ctree.c */
 int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
-		     int level, int *slot);
+		     int *slot);
 int __pure btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2);
 int btrfs_previous_item(struct btrfs_root *root,
 			struct btrfs_path *path, u64 min_objectid,
@@ -2896,10 +2902,9 @@ void btrfs_free_inode(struct inode *inode);
 int btrfs_drop_inode(struct inode *inode);
 int __init btrfs_init_cachep(void);
 void __cold btrfs_destroy_cachep(void);
-struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
+struct inode *btrfs_iget_path(struct super_block *s, u64 ino,
 			      struct btrfs_root *root, struct btrfs_path *path);
-struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
-			 struct btrfs_root *root);
+struct inode *btrfs_iget(struct super_block *s, u64 ino, struct btrfs_root *root);
 struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 				    struct page *page, size_t pg_offset,
 				    u64 start, u64 end);
@@ -2929,6 +2934,9 @@ 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);
+extern const struct iomap_ops btrfs_dio_iomap_ops;
+extern const struct iomap_dio_ops btrfs_dops;
 
 /* ioctl.c */
 long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
@@ -3381,6 +3389,9 @@ void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending,
 int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 			      struct btrfs_pending_snapshot *pending);
 int btrfs_should_cancel_balance(struct btrfs_fs_info *fs_info);
+struct btrfs_root *find_reloc_root(struct btrfs_fs_info *fs_info,
+				   u64 bytenr);
+int btrfs_should_ignore_reloc_root(struct btrfs_root *root);
 
 /* scrub.c */
 int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 7278789ff8a7..7c6f0bbb54a5 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -358,16 +358,14 @@ static int btrfs_check_super_csum(struct btrfs_fs_info *fs_info,
 	SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
 
 	shash->tfm = fs_info->csum_shash;
-	crypto_shash_init(shash);
 
 	/*
 	 * The super_block structure does not span the whole
 	 * BTRFS_SUPER_INFO_SIZE range, we expect that the unused space is
 	 * filled with zeros and is included in the checksum.
 	 */
-	crypto_shash_update(shash, raw_disk_sb + BTRFS_CSUM_SIZE,
-			    BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE);
-	crypto_shash_final(shash, result);
+	crypto_shash_digest(shash, raw_disk_sb + BTRFS_CSUM_SIZE,
+			    BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE, result);
 
 	if (memcmp(disk_sb->csum, result, btrfs_super_csum_size(disk_sb)))
 		return 1;
@@ -709,9 +707,7 @@ static void end_workqueue_bio(struct bio *bio)
 		else
 			wq = fs_info->endio_write_workers;
 	} else {
-		if (unlikely(end_io_wq->metadata == BTRFS_WQ_ENDIO_DIO_REPAIR))
-			wq = fs_info->endio_repair_workers;
-		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
+		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
 			wq = fs_info->endio_raid56_workers;
 		else if (end_io_wq->metadata)
 			wq = fs_info->endio_meta_workers;
@@ -1135,9 +1131,12 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	root->log_transid = 0;
 	root->log_transid_committed = -1;
 	root->last_log_commit = 0;
-	if (!dummy)
+	if (!dummy) {
 		extent_io_tree_init(fs_info, &root->dirty_log_pages,
 				    IO_TREE_ROOT_DIRTY_LOG_PAGES, NULL);
+		extent_io_tree_init(fs_info, &root->log_csum_range,
+				    IO_TREE_LOG_CSUM_RANGE, NULL);
+	}
 
 	memset(&root->root_key, 0, sizeof(root->root_key));
 	memset(&root->root_item, 0, sizeof(root->root_item));
@@ -1275,12 +1274,13 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
 	root->root_key.offset = BTRFS_TREE_LOG_OBJECTID;
 
 	/*
-	 * DON'T set REF_COWS for log trees
+	 * DON'T set SHAREABLE bit for log trees.
 	 *
-	 * log trees do not get reference counted because they go away
-	 * before a real commit is actually done.  They do store pointers
-	 * to file data extents, and those reference counts still get
-	 * updated (along with back refs to the log tree).
+	 * Log trees are not exposed to user space thus can't be snapshotted,
+	 * and they go away before a real commit is actually done.
+	 *
+	 * They do store pointers to file data extents, and those reference
+	 * counts still get updated (along with back refs to the log tree).
 	 */
 
 	leaf = btrfs_alloc_tree_block(trans, root, 0, BTRFS_TREE_LOG_OBJECTID,
@@ -1418,8 +1418,9 @@ static int btrfs_init_fs_root(struct btrfs_root *root)
 	if (ret)
 		goto fail;
 
-	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
-		set_bit(BTRFS_ROOT_REF_COWS, &root->state);
+	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID &&
+	    root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) {
+		set_bit(BTRFS_ROOT_SHAREABLE, &root->state);
 		btrfs_check_and_init_root_item(&root->root_item);
 	}
 
@@ -1524,6 +1525,7 @@ void btrfs_free_fs_info(struct btrfs_fs_info *fs_info)
 	btrfs_put_root(fs_info->uuid_root);
 	btrfs_put_root(fs_info->free_space_root);
 	btrfs_put_root(fs_info->fs_root);
+	btrfs_put_root(fs_info->data_reloc_root);
 	btrfs_check_leaked_roots(fs_info);
 	btrfs_extent_buffer_leak_debug_check(fs_info);
 	kfree(fs_info->super_copy);
@@ -1533,35 +1535,34 @@ void btrfs_free_fs_info(struct btrfs_fs_info *fs_info)
 
 
 struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info,
-				     struct btrfs_key *location,
-				     bool check_ref)
+				     u64 objectid, bool check_ref)
 {
 	struct btrfs_root *root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	int ret;
 
-	if (location->objectid == BTRFS_ROOT_TREE_OBJECTID)
+	if (objectid == BTRFS_ROOT_TREE_OBJECTID)
 		return btrfs_grab_root(fs_info->tree_root);
-	if (location->objectid == BTRFS_EXTENT_TREE_OBJECTID)
+	if (objectid == BTRFS_EXTENT_TREE_OBJECTID)
 		return btrfs_grab_root(fs_info->extent_root);
-	if (location->objectid == BTRFS_CHUNK_TREE_OBJECTID)
+	if (objectid == BTRFS_CHUNK_TREE_OBJECTID)
 		return btrfs_grab_root(fs_info->chunk_root);
-	if (location->objectid == BTRFS_DEV_TREE_OBJECTID)
+	if (objectid == BTRFS_DEV_TREE_OBJECTID)
 		return btrfs_grab_root(fs_info->dev_root);
-	if (location->objectid == BTRFS_CSUM_TREE_OBJECTID)
+	if (objectid == BTRFS_CSUM_TREE_OBJECTID)
 		return btrfs_grab_root(fs_info->csum_root);
-	if (location->objectid == BTRFS_QUOTA_TREE_OBJECTID)
+	if (objectid == BTRFS_QUOTA_TREE_OBJECTID)
 		return btrfs_grab_root(fs_info->quota_root) ?
 			fs_info->quota_root : ERR_PTR(-ENOENT);
-	if (location->objectid == BTRFS_UUID_TREE_OBJECTID)
+	if (objectid == BTRFS_UUID_TREE_OBJECTID)
 		return btrfs_grab_root(fs_info->uuid_root) ?
 			fs_info->uuid_root : ERR_PTR(-ENOENT);
-	if (location->objectid == BTRFS_FREE_SPACE_TREE_OBJECTID)
+	if (objectid == BTRFS_FREE_SPACE_TREE_OBJECTID)
 		return btrfs_grab_root(fs_info->free_space_root) ?
 			fs_info->free_space_root : ERR_PTR(-ENOENT);
 again:
-	root = btrfs_lookup_fs_root(fs_info, location->objectid);
+	root = btrfs_lookup_fs_root(fs_info, objectid);
 	if (root) {
 		if (check_ref && btrfs_root_refs(&root->root_item) == 0) {
 			btrfs_put_root(root);
@@ -1570,7 +1571,10 @@ again:
 		return root;
 	}
 
-	root = btrfs_read_tree_root(fs_info->tree_root, location);
+	key.objectid = objectid;
+	key.type = BTRFS_ROOT_ITEM_KEY;
+	key.offset = (u64)-1;
+	root = btrfs_read_tree_root(fs_info->tree_root, &key);
 	if (IS_ERR(root))
 		return root;
 
@@ -1590,7 +1594,7 @@ again:
 	}
 	key.objectid = BTRFS_ORPHAN_OBJECTID;
 	key.type = BTRFS_ORPHAN_ITEM_KEY;
-	key.offset = location->objectid;
+	key.offset = objectid;
 
 	ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0);
 	btrfs_free_path(path);
@@ -1940,7 +1944,6 @@ static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
 	btrfs_destroy_workqueue(fs_info->workers);
 	btrfs_destroy_workqueue(fs_info->endio_workers);
 	btrfs_destroy_workqueue(fs_info->endio_raid56_workers);
-	btrfs_destroy_workqueue(fs_info->endio_repair_workers);
 	btrfs_destroy_workqueue(fs_info->rmw_workers);
 	btrfs_destroy_workqueue(fs_info->endio_write_workers);
 	btrfs_destroy_workqueue(fs_info->endio_freespace_worker);
@@ -1981,6 +1984,7 @@ static void free_root_pointers(struct btrfs_fs_info *info, bool free_chunk_root)
 	free_root_extent_buffers(info->quota_root);
 	free_root_extent_buffers(info->uuid_root);
 	free_root_extent_buffers(info->fs_root);
+	free_root_extent_buffers(info->data_reloc_root);
 	if (free_chunk_root)
 		free_root_extent_buffers(info->chunk_root);
 	free_root_extent_buffers(info->free_space_root);
@@ -1993,6 +1997,7 @@ void btrfs_put_root(struct btrfs_root *root)
 
 	if (refcount_dec_and_test(&root->refs)) {
 		WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
+		WARN_ON(test_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state));
 		if (root->anon_dev)
 			free_anon_bdev(root->anon_dev);
 		btrfs_drew_lock_destroy(&root->snapshot_lock);
@@ -2143,8 +2148,6 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 	fs_info->endio_raid56_workers =
 		btrfs_alloc_workqueue(fs_info, "endio-raid56", flags,
 				      max_active, 4);
-	fs_info->endio_repair_workers =
-		btrfs_alloc_workqueue(fs_info, "endio-repair", flags, 1, 0);
 	fs_info->rmw_workers =
 		btrfs_alloc_workqueue(fs_info, "rmw", flags, max_active, 2);
 	fs_info->endio_write_workers =
@@ -2168,7 +2171,6 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 	      fs_info->flush_workers &&
 	      fs_info->endio_workers && fs_info->endio_meta_workers &&
 	      fs_info->endio_meta_write_workers &&
-	      fs_info->endio_repair_workers &&
 	      fs_info->endio_write_workers && fs_info->endio_raid56_workers &&
 	      fs_info->endio_freespace_worker && fs_info->rmw_workers &&
 	      fs_info->caching_workers && fs_info->readahead_workers &&
@@ -2290,6 +2292,19 @@ static int btrfs_read_roots(struct btrfs_fs_info *fs_info)
 	set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
 	fs_info->csum_root = root;
 
+	/*
+	 * This tree can share blocks with some other fs tree during relocation
+	 * and we need a proper setup by btrfs_get_fs_root
+	 */
+	root = btrfs_get_fs_root(tree_root->fs_info,
+				 BTRFS_DATA_RELOC_TREE_OBJECTID, true);
+	if (IS_ERR(root)) {
+		ret = PTR_ERR(root);
+		goto out;
+	}
+	set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
+	fs_info->data_reloc_root = root;
+
 	location.objectid = BTRFS_QUOTA_TREE_OBJECTID;
 	root = btrfs_read_tree_root(tree_root, &location);
 	if (!IS_ERR(root)) {
@@ -2827,7 +2842,6 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
 	u64 generation;
 	u64 features;
 	u16 csum_type;
-	struct btrfs_key location;
 	struct btrfs_super_block *disk_super;
 	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
 	struct btrfs_root *tree_root;
@@ -3241,11 +3255,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
 		}
 	}
 
-	location.objectid = BTRFS_FS_TREE_OBJECTID;
-	location.type = BTRFS_ROOT_ITEM_KEY;
-	location.offset = 0;
-
-	fs_info->fs_root = btrfs_get_fs_root(fs_info, &location, true);
+	fs_info->fs_root = btrfs_get_fs_root(fs_info, BTRFS_FS_TREE_OBJECTID, true);
 	if (IS_ERR(fs_info->fs_root)) {
 		err = PTR_ERR(fs_info->fs_root);
 		btrfs_warn(fs_info, "failed to read fs tree: %d", err);
@@ -3508,10 +3518,9 @@ static int write_dev_supers(struct btrfs_device *device,
 
 		btrfs_set_super_bytenr(sb, bytenr);
 
-		crypto_shash_init(shash);
-		crypto_shash_update(shash, (const char *)sb + BTRFS_CSUM_SIZE,
-				    BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE);
-		crypto_shash_final(shash, sb->csum);
+		crypto_shash_digest(shash, (const char *)sb + BTRFS_CSUM_SIZE,
+				    BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE,
+				    sb->csum);
 
 		page = find_or_create_page(mapping, bytenr >> PAGE_SHIFT,
 					   GFP_NOFS);
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index cd629113f61c..bf43245406c4 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -25,7 +25,6 @@ enum btrfs_wq_endio_type {
 	BTRFS_WQ_ENDIO_METADATA,
 	BTRFS_WQ_ENDIO_FREE_SPACE,
 	BTRFS_WQ_ENDIO_RAID56,
-	BTRFS_WQ_ENDIO_DIO_REPAIR,
 };
 
 static inline u64 btrfs_sb_offset(int mirror)
@@ -67,8 +66,7 @@ int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info,
 void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info);
 
 struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info,
-				     struct btrfs_key *key,
-				     bool check_ref);
+				     u64 objectid, bool check_ref);
 
 void btrfs_free_fs_info(struct btrfs_fs_info *fs_info);
 int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info);
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index 2bb25d2dc44b..1a8d419d9e1f 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -64,24 +64,15 @@ struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
 	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
 	struct btrfs_root *root;
 	struct inode *inode;
-	struct btrfs_key key;
 
 	if (objectid < BTRFS_FIRST_FREE_OBJECTID)
 		return ERR_PTR(-ESTALE);
 
-	key.objectid = root_objectid;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	root = btrfs_get_fs_root(fs_info, &key, true);
+	root = btrfs_get_fs_root(fs_info, root_objectid, true);
 	if (IS_ERR(root))
 		return ERR_CAST(root);
 
-	key.objectid = objectid;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-
-	inode = btrfs_iget(sb, &key, root);
+	inode = btrfs_iget(sb, objectid, root);
 	btrfs_put_root(root);
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
@@ -200,9 +191,7 @@ struct dentry *btrfs_get_parent(struct dentry *child)
 					found_key.offset, 0, 0);
 	}
 
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-	return d_obtain_alias(btrfs_iget(fs_info->sb, &key, root));
+	return d_obtain_alias(btrfs_iget(fs_info->sb, key.objectid, root));
 fail:
 	btrfs_free_path(path);
 	return ERR_PTR(ret);
diff --git a/fs/btrfs/extent-io-tree.h b/fs/btrfs/extent-io-tree.h
index b4a7bad3e82e..b6561455b3c4 100644
--- a/fs/btrfs/extent-io-tree.h
+++ b/fs/btrfs/extent-io-tree.h
@@ -44,6 +44,7 @@ enum {
 	IO_TREE_TRANS_DIRTY_PAGES,
 	IO_TREE_ROOT_DIRTY_LOG_PAGES,
 	IO_TREE_INODE_FILE_EXTENT,
+	IO_TREE_LOG_CSUM_RANGE,
 	IO_TREE_SELFTEST,
 };
 
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 54a64d1e18c6..c0bc35f932bf 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -2114,22 +2114,6 @@ static u64 find_middle(struct rb_root *root)
 }
 #endif
 
-static inline u64 heads_to_leaves(struct btrfs_fs_info *fs_info, u64 heads)
-{
-	u64 num_bytes;
-
-	num_bytes = heads * (sizeof(struct btrfs_extent_item) +
-			     sizeof(struct btrfs_extent_inline_ref));
-	if (!btrfs_fs_incompat(fs_info, SKINNY_METADATA))
-		num_bytes += heads * sizeof(struct btrfs_tree_block_info);
-
-	/*
-	 * We don't ever fill up leaves all the way so multiply by 2 just to be
-	 * closer to what we're really going to want to use.
-	 */
-	return div_u64(num_bytes, BTRFS_LEAF_DATA_SIZE(fs_info));
-}
-
 /*
  * Takes the number of bytes to be csumm'ed and figures out how many leaves it
  * would require to store the csums for that many bytes.
@@ -2442,7 +2426,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
 	nritems = btrfs_header_nritems(buf);
 	level = btrfs_header_level(buf);
 
-	if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state) && level == 0)
+	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state) && level == 0)
 		return 0;
 
 	if (full_backref)
@@ -2932,7 +2916,7 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 						   &trimmed);
 
 		list_del_init(&block_group->bg_list);
-		btrfs_put_block_group_trimming(block_group);
+		btrfs_unfreeze_block_group(block_group);
 		btrfs_put_block_group(block_group);
 
 		if (ret) {
@@ -3369,6 +3353,7 @@ static struct btrfs_block_group *btrfs_lock_cluster(
 		   struct btrfs_block_group *block_group,
 		   struct btrfs_free_cluster *cluster,
 		   int delalloc)
+	__acquires(&cluster->refill_lock)
 {
 	struct btrfs_block_group *used_bg = NULL;
 
@@ -5501,8 +5486,6 @@ out:
 	 */
 	if (!for_reloc && !root_dropped)
 		btrfs_add_dead_root(root);
-	if (err && err != -EAGAIN)
-		btrfs_handle_fs_error(fs_info, err, NULL);
 	return err;
 }
 
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index e12eb32d9e17..68c96057ad2d 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2333,7 +2333,7 @@ int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
 	return 0;
 }
 
-int btrfs_repair_eb_io_failure(struct extent_buffer *eb, int mirror_num)
+int btrfs_repair_eb_io_failure(const struct extent_buffer *eb, int mirror_num)
 {
 	struct btrfs_fs_info *fs_info = eb->fs_info;
 	u64 start = eb->start;
@@ -2537,8 +2537,9 @@ int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
 	return 0;
 }
 
-bool btrfs_check_repairable(struct inode *inode, unsigned failed_bio_pages,
-			   struct io_failure_record *failrec, int failed_mirror)
+static bool btrfs_check_repairable(struct inode *inode, bool needs_validation,
+				   struct io_failure_record *failrec,
+				   int failed_mirror)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	int num_copies;
@@ -2561,7 +2562,7 @@ bool btrfs_check_repairable(struct inode *inode, unsigned failed_bio_pages,
 	 *	a) deliver good data to the caller
 	 *	b) correct the bad sectors on disk
 	 */
-	if (failed_bio_pages > 1) {
+	if (needs_validation) {
 		/*
 		 * to fulfill b), we need to know the exact failing sectors, as
 		 * we don't want to rewrite any more than the failed ones. thus,
@@ -2600,94 +2601,115 @@ bool btrfs_check_repairable(struct inode *inode, unsigned failed_bio_pages,
 	return true;
 }
 
-
-struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
-				    struct io_failure_record *failrec,
-				    struct page *page, int pg_offset, int icsum,
-				    bio_end_io_t *endio_func, void *data)
+static bool btrfs_io_needs_validation(struct inode *inode, struct bio *bio)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct bio *bio;
-	struct btrfs_io_bio *btrfs_failed_bio;
-	struct btrfs_io_bio *btrfs_bio;
+	u64 len = 0;
+	const u32 blocksize = inode->i_sb->s_blocksize;
 
-	bio = btrfs_io_bio_alloc(1);
-	bio->bi_end_io = endio_func;
-	bio->bi_iter.bi_sector = failrec->logical >> 9;
-	bio->bi_iter.bi_size = 0;
-	bio->bi_private = data;
+	/*
+	 * If bi_status is BLK_STS_OK, then this was a checksum error, not an
+	 * I/O error. In this case, we already know exactly which sector was
+	 * bad, so we don't need to validate.
+	 */
+	if (bio->bi_status == BLK_STS_OK)
+		return false;
 
-	btrfs_failed_bio = btrfs_io_bio(failed_bio);
-	if (btrfs_failed_bio->csum) {
-		u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
+	/*
+	 * We need to validate each sector individually if the failed I/O was
+	 * for multiple sectors.
+	 *
+	 * There are a few possible bios that can end up here:
+	 * 1. A buffered read bio, which is not cloned.
+	 * 2. A direct I/O read bio, which is cloned.
+	 * 3. A (buffered or direct) repair bio, which is not cloned.
+	 *
+	 * For cloned bios (case 2), we can get the size from
+	 * btrfs_io_bio->iter; for non-cloned bios (cases 1 and 3), we can get
+	 * it from the bvecs.
+	 */
+	if (bio_flagged(bio, BIO_CLONED)) {
+		if (btrfs_io_bio(bio)->iter.bi_size > blocksize)
+			return true;
+	} else {
+		struct bio_vec *bvec;
+		int i;
 
-		btrfs_bio = btrfs_io_bio(bio);
-		btrfs_bio->csum = btrfs_bio->csum_inline;
-		icsum *= csum_size;
-		memcpy(btrfs_bio->csum, btrfs_failed_bio->csum + icsum,
-		       csum_size);
+		bio_for_each_bvec_all(bvec, bio, i) {
+			len += bvec->bv_len;
+			if (len > blocksize)
+				return true;
+		}
 	}
-
-	bio_add_page(bio, page, failrec->len, pg_offset);
-
-	return bio;
+	return false;
 }
 
-/*
- * This is a generic handler for readpage errors. If other copies exist, read
- * those and write back good data to the failed position. Does not investigate
- * in remapping the failed extent elsewhere, hoping the device will be smart
- * enough to do this as needed
- */
-static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
-			      struct page *page, u64 start, u64 end,
-			      int failed_mirror)
+blk_status_t btrfs_submit_read_repair(struct inode *inode,
+				      struct bio *failed_bio, u64 phy_offset,
+				      struct page *page, unsigned int pgoff,
+				      u64 start, u64 end, int failed_mirror,
+				      submit_bio_hook_t *submit_bio_hook)
 {
 	struct io_failure_record *failrec;
-	struct inode *inode = page->mapping->host;
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
 	struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
-	struct bio *bio;
-	int read_mode = 0;
+	struct btrfs_io_bio *failed_io_bio = btrfs_io_bio(failed_bio);
+	const int icsum = phy_offset >> inode->i_sb->s_blocksize_bits;
+	bool need_validation;
+	struct bio *repair_bio;
+	struct btrfs_io_bio *repair_io_bio;
 	blk_status_t status;
 	int ret;
-	unsigned failed_bio_pages = failed_bio->bi_iter.bi_size >> PAGE_SHIFT;
+
+	btrfs_debug(fs_info,
+		   "repair read error: read error at %llu", start);
 
 	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
 
 	ret = btrfs_get_io_failure_record(inode, start, end, &failrec);
 	if (ret)
-		return ret;
+		return errno_to_blk_status(ret);
+
+	need_validation = btrfs_io_needs_validation(inode, failed_bio);
 
-	if (!btrfs_check_repairable(inode, failed_bio_pages, failrec,
+	if (!btrfs_check_repairable(inode, need_validation, failrec,
 				    failed_mirror)) {
 		free_io_failure(failure_tree, tree, failrec);
-		return -EIO;
+		return BLK_STS_IOERR;
 	}
 
-	if (failed_bio_pages > 1)
-		read_mode |= REQ_FAILFAST_DEV;
+	repair_bio = btrfs_io_bio_alloc(1);
+	repair_io_bio = btrfs_io_bio(repair_bio);
+	repair_bio->bi_opf = REQ_OP_READ;
+	if (need_validation)
+		repair_bio->bi_opf |= REQ_FAILFAST_DEV;
+	repair_bio->bi_end_io = failed_bio->bi_end_io;
+	repair_bio->bi_iter.bi_sector = failrec->logical >> 9;
+	repair_bio->bi_private = failed_bio->bi_private;
 
-	phy_offset >>= inode->i_sb->s_blocksize_bits;
-	bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page,
-				      start - page_offset(page),
-				      (int)phy_offset, failed_bio->bi_end_io,
-				      NULL);
-	bio->bi_opf = REQ_OP_READ | read_mode;
+	if (failed_io_bio->csum) {
+		const u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
+
+		repair_io_bio->csum = repair_io_bio->csum_inline;
+		memcpy(repair_io_bio->csum,
+		       failed_io_bio->csum + csum_size * icsum, csum_size);
+	}
+
+	bio_add_page(repair_bio, page, failrec->len, pgoff);
+	repair_io_bio->logical = failrec->start;
+	repair_io_bio->iter = repair_bio->bi_iter;
 
 	btrfs_debug(btrfs_sb(inode->i_sb),
-		"Repair Read Error: submitting new read[%#x] to this_mirror=%d, in_validation=%d",
-		read_mode, failrec->this_mirror, failrec->in_validation);
+"repair read error: submitting new read to mirror %d, in_validation=%d",
+		    failrec->this_mirror, failrec->in_validation);
 
-	status = tree->ops->submit_bio_hook(tree->private_data, bio, failrec->this_mirror,
-					 failrec->bio_flags);
+	status = submit_bio_hook(inode, repair_bio, failrec->this_mirror,
+				 failrec->bio_flags);
 	if (status) {
 		free_io_failure(failure_tree, tree, failrec);
-		bio_put(bio);
-		ret = blk_status_to_errno(status);
+		bio_put(repair_bio);
 	}
-
-	return ret;
+	return status;
 }
 
 /* lots and lots of room for performance fixes in the end_bio funcs */
@@ -2859,9 +2881,10 @@ static void end_bio_extent_readpage(struct bio *bio)
 			 * If it can't handle the error it will return -EIO and
 			 * we remain responsible for that page.
 			 */
-			ret = bio_readpage_error(bio, offset, page, start, end,
-						 mirror);
-			if (ret == 0) {
+			if (!btrfs_submit_read_repair(inode, bio, offset, page,
+						start - page_offset(page),
+						start, end, mirror,
+						tree->ops->submit_bio_hook)) {
 				uptodate = !bio->bi_status;
 				offset += len;
 				continue;
@@ -4862,7 +4885,7 @@ static void __free_extent_buffer(struct extent_buffer *eb)
 	kmem_cache_free(extent_buffer_cache, eb);
 }
 
-int extent_buffer_under_io(struct extent_buffer *eb)
+int extent_buffer_under_io(const struct extent_buffer *eb)
 {
 	return (atomic_read(&eb->io_pages) ||
 		test_bit(EXTENT_BUFFER_WRITEBACK, &eb->bflags) ||
@@ -4967,7 +4990,7 @@ __alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start,
 	return eb;
 }
 
-struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src)
+struct extent_buffer *btrfs_clone_extent_buffer(const struct extent_buffer *src)
 {
 	int i;
 	struct page *p;
@@ -5373,7 +5396,7 @@ void free_extent_buffer_stale(struct extent_buffer *eb)
 	release_extent_buffer(eb);
 }
 
-void clear_extent_buffer_dirty(struct extent_buffer *eb)
+void clear_extent_buffer_dirty(const struct extent_buffer *eb)
 {
 	int i;
 	int num_pages;
@@ -5571,8 +5594,7 @@ void read_extent_buffer(const struct extent_buffer *eb, void *dstv,
 	struct page *page;
 	char *kaddr;
 	char *dst = (char *)dstv;
-	size_t start_offset = offset_in_page(eb->start);
-	unsigned long i = (start_offset + start) >> PAGE_SHIFT;
+	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",
@@ -5581,7 +5603,7 @@ void read_extent_buffer(const struct extent_buffer *eb, void *dstv,
 		return;
 	}
 
-	offset = offset_in_page(start_offset + start);
+	offset = offset_in_page(start);
 
 	while (len > 0) {
 		page = eb->pages[i];
@@ -5606,14 +5628,13 @@ int read_extent_buffer_to_user(const struct extent_buffer *eb,
 	struct page *page;
 	char *kaddr;
 	char __user *dst = (char __user *)dstv;
-	size_t start_offset = offset_in_page(eb->start);
-	unsigned long i = (start_offset + start) >> PAGE_SHIFT;
+	unsigned long i = start >> PAGE_SHIFT;
 	int ret = 0;
 
 	WARN_ON(start > eb->len);
 	WARN_ON(start + len > eb->start + eb->len);
 
-	offset = offset_in_page(start_offset + start);
+	offset = offset_in_page(start);
 
 	while (len > 0) {
 		page = eb->pages[i];
@@ -5634,48 +5655,6 @@ int read_extent_buffer_to_user(const struct extent_buffer *eb,
 	return ret;
 }
 
-/*
- * return 0 if the item is found within a page.
- * return 1 if the item spans two pages.
- * return -EINVAL otherwise.
- */
-int map_private_extent_buffer(const struct extent_buffer *eb,
-			      unsigned long start, unsigned long min_len,
-			      char **map, unsigned long *map_start,
-			      unsigned long *map_len)
-{
-	size_t offset;
-	char *kaddr;
-	struct page *p;
-	size_t start_offset = offset_in_page(eb->start);
-	unsigned long i = (start_offset + start) >> PAGE_SHIFT;
-	unsigned long end_i = (start_offset + start + min_len - 1) >>
-		PAGE_SHIFT;
-
-	if (start + min_len > eb->len) {
-		WARN(1, KERN_ERR "btrfs bad mapping eb start %llu len %lu, wanted %lu %lu\n",
-		       eb->start, eb->len, start, min_len);
-		return -EINVAL;
-	}
-
-	if (i != end_i)
-		return 1;
-
-	if (i == 0) {
-		offset = start_offset;
-		*map_start = 0;
-	} else {
-		offset = 0;
-		*map_start = ((u64)i << PAGE_SHIFT) - start_offset;
-	}
-
-	p = eb->pages[i];
-	kaddr = page_address(p);
-	*map = kaddr + offset;
-	*map_len = PAGE_SIZE - offset;
-	return 0;
-}
-
 int memcmp_extent_buffer(const struct extent_buffer *eb, const void *ptrv,
 			 unsigned long start, unsigned long len)
 {
@@ -5684,14 +5663,13 @@ int memcmp_extent_buffer(const struct extent_buffer *eb, const void *ptrv,
 	struct page *page;
 	char *kaddr;
 	char *ptr = (char *)ptrv;
-	size_t start_offset = offset_in_page(eb->start);
-	unsigned long i = (start_offset + start) >> PAGE_SHIFT;
+	unsigned long i = start >> PAGE_SHIFT;
 	int ret = 0;
 
 	WARN_ON(start > eb->len);
 	WARN_ON(start + len > eb->start + eb->len);
 
-	offset = offset_in_page(start_offset + start);
+	offset = offset_in_page(start);
 
 	while (len > 0) {
 		page = eb->pages[i];
@@ -5711,7 +5689,7 @@ int memcmp_extent_buffer(const struct extent_buffer *eb, const void *ptrv,
 	return ret;
 }
 
-void write_extent_buffer_chunk_tree_uuid(struct extent_buffer *eb,
+void write_extent_buffer_chunk_tree_uuid(const struct extent_buffer *eb,
 		const void *srcv)
 {
 	char *kaddr;
@@ -5722,7 +5700,7 @@ void write_extent_buffer_chunk_tree_uuid(struct extent_buffer *eb,
 			BTRFS_FSID_SIZE);
 }
 
-void write_extent_buffer_fsid(struct extent_buffer *eb, const void *srcv)
+void write_extent_buffer_fsid(const struct extent_buffer *eb, const void *srcv)
 {
 	char *kaddr;
 
@@ -5732,7 +5710,7 @@ void write_extent_buffer_fsid(struct extent_buffer *eb, const void *srcv)
 			BTRFS_FSID_SIZE);
 }
 
-void write_extent_buffer(struct extent_buffer *eb, const void *srcv,
+void write_extent_buffer(const struct extent_buffer *eb, const void *srcv,
 			 unsigned long start, unsigned long len)
 {
 	size_t cur;
@@ -5740,13 +5718,12 @@ void write_extent_buffer(struct extent_buffer *eb, const void *srcv,
 	struct page *page;
 	char *kaddr;
 	char *src = (char *)srcv;
-	size_t start_offset = offset_in_page(eb->start);
-	unsigned long i = (start_offset + start) >> PAGE_SHIFT;
+	unsigned long i = start >> PAGE_SHIFT;
 
 	WARN_ON(start > eb->len);
 	WARN_ON(start + len > eb->start + eb->len);
 
-	offset = offset_in_page(start_offset + start);
+	offset = offset_in_page(start);
 
 	while (len > 0) {
 		page = eb->pages[i];
@@ -5763,20 +5740,19 @@ void write_extent_buffer(struct extent_buffer *eb, const void *srcv,
 	}
 }
 
-void memzero_extent_buffer(struct extent_buffer *eb, unsigned long start,
+void memzero_extent_buffer(const struct extent_buffer *eb, unsigned long start,
 		unsigned long len)
 {
 	size_t cur;
 	size_t offset;
 	struct page *page;
 	char *kaddr;
-	size_t start_offset = offset_in_page(eb->start);
-	unsigned long i = (start_offset + start) >> PAGE_SHIFT;
+	unsigned long i = start >> PAGE_SHIFT;
 
 	WARN_ON(start > eb->len);
 	WARN_ON(start + len > eb->start + eb->len);
 
-	offset = offset_in_page(start_offset + start);
+	offset = offset_in_page(start);
 
 	while (len > 0) {
 		page = eb->pages[i];
@@ -5792,8 +5768,8 @@ void memzero_extent_buffer(struct extent_buffer *eb, unsigned long start,
 	}
 }
 
-void copy_extent_buffer_full(struct extent_buffer *dst,
-			     struct extent_buffer *src)
+void copy_extent_buffer_full(const struct extent_buffer *dst,
+			     const struct extent_buffer *src)
 {
 	int i;
 	int num_pages;
@@ -5806,7 +5782,8 @@ void copy_extent_buffer_full(struct extent_buffer *dst,
 				page_address(src->pages[i]));
 }
 
-void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
+void copy_extent_buffer(const struct extent_buffer *dst,
+			const struct extent_buffer *src,
 			unsigned long dst_offset, unsigned long src_offset,
 			unsigned long len)
 {
@@ -5815,12 +5792,11 @@ void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
 	size_t offset;
 	struct page *page;
 	char *kaddr;
-	size_t start_offset = offset_in_page(dst->start);
-	unsigned long i = (start_offset + dst_offset) >> PAGE_SHIFT;
+	unsigned long i = dst_offset >> PAGE_SHIFT;
 
 	WARN_ON(src->len != dst_len);
 
-	offset = offset_in_page(start_offset + dst_offset);
+	offset = offset_in_page(dst_offset);
 
 	while (len > 0) {
 		page = dst->pages[i];
@@ -5851,12 +5827,11 @@ void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
  * This helper hides the ugliness of finding the byte in an extent buffer which
  * contains a given bit.
  */
-static inline void eb_bitmap_offset(struct extent_buffer *eb,
+static inline void eb_bitmap_offset(const struct extent_buffer *eb,
 				    unsigned long start, unsigned long nr,
 				    unsigned long *page_index,
 				    size_t *page_offset)
 {
-	size_t start_offset = offset_in_page(eb->start);
 	size_t byte_offset = BIT_BYTE(nr);
 	size_t offset;
 
@@ -5865,7 +5840,7 @@ static inline void eb_bitmap_offset(struct extent_buffer *eb,
 	 * the bitmap item in the extent buffer + the offset of the byte in the
 	 * bitmap item.
 	 */
-	offset = start_offset + start + byte_offset;
+	offset = start + byte_offset;
 
 	*page_index = offset >> PAGE_SHIFT;
 	*page_offset = offset_in_page(offset);
@@ -5877,7 +5852,7 @@ static inline void eb_bitmap_offset(struct extent_buffer *eb,
  * @start: offset of the bitmap item in the extent buffer
  * @nr: bit number to test
  */
-int extent_buffer_test_bit(struct extent_buffer *eb, unsigned long start,
+int extent_buffer_test_bit(const struct extent_buffer *eb, unsigned long start,
 			   unsigned long nr)
 {
 	u8 *kaddr;
@@ -5899,7 +5874,7 @@ int extent_buffer_test_bit(struct extent_buffer *eb, unsigned long start,
  * @pos: bit number of the first bit
  * @len: number of bits to set
  */
-void extent_buffer_bitmap_set(struct extent_buffer *eb, unsigned long start,
+void extent_buffer_bitmap_set(const struct extent_buffer *eb, unsigned long start,
 			      unsigned long pos, unsigned long len)
 {
 	u8 *kaddr;
@@ -5941,8 +5916,9 @@ void extent_buffer_bitmap_set(struct extent_buffer *eb, unsigned long start,
  * @pos: bit number of the first bit
  * @len: number of bits to clear
  */
-void extent_buffer_bitmap_clear(struct extent_buffer *eb, unsigned long start,
-				unsigned long pos, unsigned long len)
+void extent_buffer_bitmap_clear(const struct extent_buffer *eb,
+				unsigned long start, unsigned long pos,
+				unsigned long len)
 {
 	u8 *kaddr;
 	struct page *page;
@@ -6003,14 +5979,14 @@ static void copy_pages(struct page *dst_page, struct page *src_page,
 		memcpy(dst_kaddr + dst_off, src_kaddr + src_off, len);
 }
 
-void memcpy_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
-			   unsigned long src_offset, unsigned long len)
+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;
-	size_t start_offset = offset_in_page(dst->start);
 	unsigned long dst_i;
 	unsigned long src_i;
 
@@ -6028,11 +6004,11 @@ void memcpy_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
 	}
 
 	while (len > 0) {
-		dst_off_in_page = offset_in_page(start_offset + dst_offset);
-		src_off_in_page = offset_in_page(start_offset + src_offset);
+		dst_off_in_page = offset_in_page(dst_offset);
+		src_off_in_page = offset_in_page(src_offset);
 
-		dst_i = (start_offset + dst_offset) >> PAGE_SHIFT;
-		src_i = (start_offset + src_offset) >> PAGE_SHIFT;
+		dst_i = dst_offset >> PAGE_SHIFT;
+		src_i = src_offset >> PAGE_SHIFT;
 
 		cur = min(len, (unsigned long)(PAGE_SIZE -
 					       src_off_in_page));
@@ -6048,8 +6024,9 @@ void memcpy_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
 	}
 }
 
-void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
-			   unsigned long src_offset, unsigned long len)
+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;
@@ -6057,7 +6034,6 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
 	size_t src_off_in_page;
 	unsigned long dst_end = dst_offset + len - 1;
 	unsigned long src_end = src_offset + len - 1;
-	size_t start_offset = offset_in_page(dst->start);
 	unsigned long dst_i;
 	unsigned long src_i;
 
@@ -6078,11 +6054,11 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
 		return;
 	}
 	while (len > 0) {
-		dst_i = (start_offset + dst_end) >> PAGE_SHIFT;
-		src_i = (start_offset + src_end) >> PAGE_SHIFT;
+		dst_i = dst_end >> PAGE_SHIFT;
+		src_i = src_end >> PAGE_SHIFT;
 
-		dst_off_in_page = offset_in_page(start_offset + dst_end);
-		src_off_in_page = offset_in_page(start_offset + src_end);
+		dst_off_in_page = offset_in_page(dst_end);
+		src_off_in_page = offset_in_page(src_end);
 
 		cur = min_t(unsigned long, len, src_off_in_page + 1);
 		cur = min(cur, dst_off_in_page + 1);
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 25594e09fdcd..602bf3af9fb4 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -66,6 +66,10 @@ struct btrfs_io_bio;
 struct io_failure_record;
 struct extent_io_tree;
 
+typedef blk_status_t (submit_bio_hook_t)(struct inode *inode, struct bio *bio,
+					 int mirror_num,
+					 unsigned long bio_flags);
+
 typedef blk_status_t (extent_submit_bio_start_t)(void *private_data,
 		struct bio *bio, u64 bio_offset);
 
@@ -74,8 +78,7 @@ struct extent_io_ops {
 	 * The following callbacks must be always defined, the function
 	 * pointer will be called unconditionally.
 	 */
-	blk_status_t (*submit_bio_hook)(struct inode *inode, struct bio *bio,
-					int mirror_num, unsigned long bio_flags);
+	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);
@@ -209,7 +212,7 @@ struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
 						  u64 start, unsigned long len);
 struct extent_buffer *alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
 						u64 start);
-struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src);
+struct extent_buffer *btrfs_clone_extent_buffer(const struct extent_buffer *src);
 struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
 					 u64 start);
 void free_extent_buffer(struct extent_buffer *eb);
@@ -227,7 +230,7 @@ static inline int num_extent_pages(const struct extent_buffer *eb)
 	       (eb->start >> PAGE_SHIFT);
 }
 
-static inline int extent_buffer_uptodate(struct extent_buffer *eb)
+static inline int extent_buffer_uptodate(const struct extent_buffer *eb)
 {
 	return test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 }
@@ -240,37 +243,37 @@ void read_extent_buffer(const struct extent_buffer *eb, void *dst,
 int read_extent_buffer_to_user(const struct extent_buffer *eb,
 			       void __user *dst, unsigned long start,
 			       unsigned long len);
-void write_extent_buffer_fsid(struct extent_buffer *eb, const void *src);
-void write_extent_buffer_chunk_tree_uuid(struct extent_buffer *eb,
+void write_extent_buffer_fsid(const struct extent_buffer *eb, const void *src);
+void write_extent_buffer_chunk_tree_uuid(const struct extent_buffer *eb,
 		const void *src);
-void write_extent_buffer(struct extent_buffer *eb, const void *src,
+void write_extent_buffer(const struct extent_buffer *eb, const void *src,
 			 unsigned long start, unsigned long len);
-void copy_extent_buffer_full(struct extent_buffer *dst,
-			     struct extent_buffer *src);
-void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
+void copy_extent_buffer_full(const struct extent_buffer *dst,
+			     const struct extent_buffer *src);
+void copy_extent_buffer(const struct extent_buffer *dst,
+			const struct extent_buffer *src,
 			unsigned long dst_offset, unsigned long src_offset,
 			unsigned long len);
-void memcpy_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
-			   unsigned long src_offset, unsigned long len);
-void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
-			   unsigned long src_offset, unsigned long len);
-void memzero_extent_buffer(struct extent_buffer *eb, unsigned long start,
+void memcpy_extent_buffer(const struct extent_buffer *dst,
+			  unsigned long dst_offset, unsigned long src_offset,
+			  unsigned long len);
+void memmove_extent_buffer(const struct extent_buffer *dst,
+			   unsigned long dst_offset, unsigned long src_offset,
+			   unsigned long len);
+void memzero_extent_buffer(const struct extent_buffer *eb, unsigned long start,
 			   unsigned long len);
-int extent_buffer_test_bit(struct extent_buffer *eb, unsigned long start,
+int extent_buffer_test_bit(const struct extent_buffer *eb, unsigned long start,
 			   unsigned long pos);
-void extent_buffer_bitmap_set(struct extent_buffer *eb, unsigned long start,
+void extent_buffer_bitmap_set(const struct extent_buffer *eb, unsigned long start,
 			      unsigned long pos, unsigned long len);
-void extent_buffer_bitmap_clear(struct extent_buffer *eb, unsigned long start,
-				unsigned long pos, unsigned long len);
-void clear_extent_buffer_dirty(struct extent_buffer *eb);
+void extent_buffer_bitmap_clear(const struct extent_buffer *eb,
+				unsigned long start, unsigned long pos,
+				unsigned long len);
+void clear_extent_buffer_dirty(const struct extent_buffer *eb);
 bool set_extent_buffer_dirty(struct extent_buffer *eb);
 void set_extent_buffer_uptodate(struct extent_buffer *eb);
 void clear_extent_buffer_uptodate(struct extent_buffer *eb);
-int extent_buffer_under_io(struct extent_buffer *eb);
-int map_private_extent_buffer(const struct extent_buffer *eb,
-			      unsigned long offset, unsigned long min_len,
-			      char **map, unsigned long *map_start,
-			      unsigned long *map_len);
+int extent_buffer_under_io(const struct extent_buffer *eb);
 void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
 void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
 void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
@@ -289,7 +292,7 @@ int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
 		      u64 length, u64 logical, struct page *page,
 		      unsigned int pg_offset, int mirror_num);
 void end_extent_writepage(struct page *page, int err, u64 start, u64 end);
-int btrfs_repair_eb_io_failure(struct extent_buffer *eb, int mirror_num);
+int btrfs_repair_eb_io_failure(const struct extent_buffer *eb, int mirror_num);
 
 /*
  * When IO fails, either with EIO or csum verification fails, we
@@ -311,12 +314,12 @@ struct io_failure_record {
 };
 
 
-bool btrfs_check_repairable(struct inode *inode, unsigned failed_bio_pages,
-			    struct io_failure_record *failrec, int fail_mirror);
-struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
-				    struct io_failure_record *failrec,
-				    struct page *page, int pg_offset, int icsum,
-				    bio_end_io_t *endio_func, void *data);
+blk_status_t btrfs_submit_read_repair(struct inode *inode,
+				      struct bio *failed_bio, u64 phy_offset,
+				      struct page *page, unsigned int pgoff,
+				      u64 start, u64 end, int failed_mirror,
+				      submit_bio_hook_t *submit_bio_hook);
+
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 bool find_lock_delalloc_range(struct inode *inode,
 			     struct page *locked_page, u64 *start,
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index b618ad5339ba..706a3128e192 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -242,11 +242,13 @@ int btrfs_lookup_file_extent(struct btrfs_trans_handle *trans,
 /**
  * btrfs_lookup_bio_sums - Look up checksums for a bio.
  * @inode: inode that the bio is for.
- * @bio: bio embedded in btrfs_io_bio.
+ * @bio: bio to look up.
  * @offset: Unless (u64)-1, look up checksums for this offset in the file.
  *          If (u64)-1, use the page offsets from the bio instead.
- * @dst: Buffer of size btrfs_super_csum_size() used to return checksum. If
- *       NULL, the checksum is returned in btrfs_io_bio(bio)->csum instead.
+ * @dst: Buffer of size nblocks * btrfs_super_csum_size() used to return
+ *       checksum (nblocks = bio->bi_iter.bi_size / fs_info->sectorsize). If
+ *       NULL, the checksum buffer is allocated and returned in
+ *       btrfs_io_bio(bio)->csum instead.
  *
  * Return: BLK_STS_RESOURCE if allocating memory fails, BLK_STS_OK otherwise.
  */
@@ -256,7 +258,6 @@ blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct bio_vec bvec;
 	struct bvec_iter iter;
-	struct btrfs_io_bio *btrfs_bio = btrfs_io_bio(bio);
 	struct btrfs_csum_item *item = NULL;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct btrfs_path *path;
@@ -277,6 +278,8 @@ blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
 
 	nblocks = bio->bi_iter.bi_size >> inode->i_sb->s_blocksize_bits;
 	if (!dst) {
+		struct btrfs_io_bio *btrfs_bio = btrfs_io_bio(bio);
+
 		if (nblocks * csum_size > BTRFS_BIO_INLINE_CSUM_SIZE) {
 			btrfs_bio->csum = kmalloc_array(nblocks, csum_size,
 							GFP_NOFS);
@@ -598,13 +601,12 @@ blk_status_t btrfs_csum_one_bio(struct inode *inode, struct bio *bio,
 				index = 0;
 			}
 
-			crypto_shash_init(shash);
 			data = kmap_atomic(bvec.bv_page);
-			crypto_shash_update(shash, data + bvec.bv_offset
+			crypto_shash_digest(shash, data + bvec.bv_offset
 					    + (i * fs_info->sectorsize),
-					    fs_info->sectorsize);
+					    fs_info->sectorsize,
+					    sums->sums + index);
 			kunmap_atomic(data);
-			crypto_shash_final(shash, (char *)(sums->sums + index));
 			index += csum_size;
 			offset += fs_info->sectorsize;
 			this_sum_bytes += fs_info->sectorsize;
@@ -869,7 +871,7 @@ again:
 	}
 	ret = PTR_ERR(item);
 	if (ret != -EFBIG && ret != -ENOENT)
-		goto fail_unlock;
+		goto out;
 
 	if (ret == -EFBIG) {
 		u32 item_size;
@@ -887,10 +889,12 @@ again:
 		nritems = btrfs_header_nritems(path->nodes[0]);
 		if (!nritems || (path->slots[0] >= nritems - 1)) {
 			ret = btrfs_next_leaf(root, path);
-			if (ret == 1)
+			if (ret < 0) {
+				goto out;
+			} else if (ret > 0) {
 				found_next = 1;
-			if (ret != 0)
 				goto insert;
+			}
 			slot = path->slots[0];
 		}
 		btrfs_item_key_to_cpu(path->nodes[0], &found_key, slot);
@@ -905,14 +909,27 @@ again:
 	}
 
 	/*
-	 * at this point, we know the tree has an item, but it isn't big
-	 * enough yet to put our csum in.  Grow it
+	 * At this point, we know the tree has a checksum item that ends at an
+	 * offset matching the start of the checksum range we want to insert.
+	 * We try to extend that item as much as possible and then add as many
+	 * checksums to it as they fit.
+	 *
+	 * First check if the leaf has enough free space for at least one
+	 * checksum. If it has go directly to the item extension code, otherwise
+	 * release the path and do a search for insertion before the extension.
 	 */
+	if (btrfs_leaf_free_space(leaf) >= csum_size) {
+		btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
+		csum_offset = (bytenr - found_key.offset) >>
+			fs_info->sb->s_blocksize_bits;
+		goto extend_csum;
+	}
+
 	btrfs_release_path(path);
 	ret = btrfs_search_slot(trans, root, &file_key, path,
 				csum_size, 1);
 	if (ret < 0)
-		goto fail_unlock;
+		goto out;
 
 	if (ret > 0) {
 		if (path->slots[0] == 0)
@@ -931,19 +948,13 @@ again:
 		goto insert;
 	}
 
+extend_csum:
 	if (csum_offset == btrfs_item_size_nr(leaf, path->slots[0]) /
 	    csum_size) {
 		int extend_nr;
 		u64 tmp;
 		u32 diff;
-		u32 free_space;
-
-		if (btrfs_leaf_free_space(leaf) <
-				 sizeof(struct btrfs_item) + csum_size * 2)
-			goto insert;
 
-		free_space = btrfs_leaf_free_space(leaf) -
-					 sizeof(struct btrfs_item) - csum_size;
 		tmp = sums->len - total_bytes;
 		tmp >>= fs_info->sb->s_blocksize_bits;
 		WARN_ON(tmp < 1);
@@ -954,7 +965,7 @@ again:
 			   MAX_CSUM_ITEMS(fs_info, csum_size) * csum_size);
 
 		diff = diff - btrfs_item_size_nr(leaf, path->slots[0]);
-		diff = min(free_space, diff);
+		diff = min_t(u32, btrfs_leaf_free_space(leaf), diff);
 		diff /= csum_size;
 		diff *= csum_size;
 
@@ -985,9 +996,9 @@ insert:
 				      ins_size);
 	path->leave_spinning = 0;
 	if (ret < 0)
-		goto fail_unlock;
+		goto out;
 	if (WARN_ON(ret != 0))
-		goto fail_unlock;
+		goto out;
 	leaf = path->nodes[0];
 csum:
 	item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_csum_item);
@@ -1017,9 +1028,6 @@ found:
 out:
 	btrfs_free_path(path);
 	return ret;
-
-fail_unlock:
-	goto out;
 }
 
 void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 719e68ab552c..fde125616687 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -275,26 +275,18 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_root *inode_root;
 	struct inode *inode;
-	struct btrfs_key key;
 	struct btrfs_ioctl_defrag_range_args range;
 	int num_defrag;
 	int ret;
 
 	/* get the inode */
-	key.objectid = defrag->root;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	inode_root = btrfs_get_fs_root(fs_info, &key, true);
+	inode_root = btrfs_get_fs_root(fs_info, defrag->root, true);
 	if (IS_ERR(inode_root)) {
 		ret = PTR_ERR(inode_root);
 		goto cleanup;
 	}
 
-	key.objectid = defrag->ino;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-	inode = btrfs_iget(fs_info->sb, &key, inode_root);
+	inode = btrfs_iget(fs_info->sb, defrag->ino, inode_root);
 	btrfs_put_root(inode_root);
 	if (IS_ERR(inode)) {
 		ret = PTR_ERR(inode);
@@ -775,7 +767,7 @@ int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
 	if (start >= BTRFS_I(inode)->disk_i_size && !replace_extent)
 		modify_tree = 0;
 
-	update_refs = (test_bit(BTRFS_ROOT_REF_COWS, &root->state) ||
+	update_refs = (test_bit(BTRFS_ROOT_SHAREABLE, &root->state) ||
 		       root == fs_info->tree_root);
 	while (1) {
 		recow = 0;
@@ -1817,21 +1809,61 @@ again:
 	return num_written ? num_written : ret;
 }
 
-static ssize_t __btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
+static ssize_t check_direct_IO(struct btrfs_fs_info *fs_info,
+                               const struct iov_iter *iter, loff_t offset)
+{
+        const unsigned int blocksize_mask = fs_info->sectorsize - 1;
+
+        if (offset & blocksize_mask)
+                return -EINVAL;
+
+        if (iov_iter_alignment(iter) & blocksize_mask)
+                return -EINVAL;
+
+	return 0;
+}
+
+static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
-	loff_t pos;
-	ssize_t written;
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	loff_t pos = iocb->ki_pos;
+	ssize_t written = 0;
 	ssize_t written_buffered;
 	loff_t endbyte;
 	int err;
+	size_t count = 0;
+	bool relock = false;
 
-	written = generic_file_direct_write(iocb, from);
+	if (check_direct_IO(fs_info, from, pos))
+		goto buffered;
+
+	count = iov_iter_count(from);
+	/*
+	 * If the write DIO is beyond the EOF, we need update the isize, but it
+	 * is protected by i_mutex. So we can not unlock the i_mutex at this
+	 * case.
+	 */
+	if (pos + count <= inode->i_size) {
+		inode_unlock(inode);
+		relock = true;
+	} else if (iocb->ki_flags & IOCB_NOWAIT) {
+		return -EAGAIN;
+	}
+
+	down_read(&BTRFS_I(inode)->dio_sem);
+	written = iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops, &btrfs_dops,
+			       is_sync_kiocb(iocb));
+	up_read(&BTRFS_I(inode)->dio_sem);
+
+	if (relock)
+		inode_lock(inode);
 
 	if (written < 0 || !iov_iter_count(from))
 		return written;
 
+buffered:
 	pos = iocb->ki_pos;
 	written_buffered = btrfs_buffered_write(iocb, from);
 	if (written_buffered < 0) {
@@ -1970,7 +2002,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 		atomic_inc(&BTRFS_I(inode)->sync_writers);
 
 	if (iocb->ki_flags & IOCB_DIRECT) {
-		num_written = __btrfs_direct_write(iocb, from);
+		num_written = btrfs_direct_write(iocb, from);
 	} else {
 		num_written = btrfs_buffered_write(iocb, from);
 		if (num_written > 0)
@@ -3484,9 +3516,54 @@ static int btrfs_file_open(struct inode *inode, struct file *filp)
 	return generic_file_open(inode, filp);
 }
 
+static int check_direct_read(struct btrfs_fs_info *fs_info,
+                               const struct iov_iter *iter, loff_t offset)
+{
+	int ret;
+	int i, seg;
+
+	ret = check_direct_IO(fs_info, iter, offset);
+	if (ret < 0)
+		return ret;
+
+	for (seg = 0; seg < iter->nr_segs; seg++)
+		for (i = seg + 1; i < iter->nr_segs; i++)
+			if (iter->iov[seg].iov_base == iter->iov[i].iov_base)
+				return -EINVAL;
+	return 0;
+}
+
+static ssize_t btrfs_direct_read(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	ssize_t ret;
+
+	if (check_direct_read(btrfs_sb(inode->i_sb), to, iocb->ki_pos))
+		return 0;
+
+	inode_lock_shared(inode);
+        ret = iomap_dio_rw(iocb, to, &btrfs_dio_iomap_ops, &btrfs_dops,
+			   is_sync_kiocb(iocb));
+	inode_unlock_shared(inode);
+	return ret;
+}
+
+static ssize_t btrfs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	ssize_t ret = 0;
+
+	if (iocb->ki_flags & IOCB_DIRECT) {
+		ret = btrfs_direct_read(iocb, to);
+		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,
 	.mmap		= btrfs_file_mmap,
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 3613da065a73..55955bd424d7 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -82,7 +82,7 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
 	 * sure NOFS is set to keep us from deadlocking.
 	 */
 	nofs_flag = memalloc_nofs_save();
-	inode = btrfs_iget_path(fs_info->sb, &location, root, path);
+	inode = btrfs_iget_path(fs_info->sb, location.objectid, root, path);
 	btrfs_release_path(path);
 	memalloc_nofs_restore(nofs_flag);
 	if (IS_ERR(inode))
@@ -1190,13 +1190,10 @@ out:
 	if (ret) {
 		invalidate_inode_pages2(inode->i_mapping);
 		BTRFS_I(inode)->generation = 0;
-		if (block_group) {
-#ifdef CONFIG_BTRFS_DEBUG
-			btrfs_err(root->fs_info,
-				  "failed to write free space cache for block group %llu",
-				  block_group->start);
-#endif
-		}
+		if (block_group)
+			btrfs_debug(root->fs_info,
+	  "failed to write free space cache for block group %llu error %d",
+				  block_group->start, ret);
 	}
 	btrfs_update_inode(trans, root, inode);
 
@@ -1415,11 +1412,9 @@ int btrfs_write_out_cache(struct btrfs_trans_handle *trans,
 	ret = __btrfs_write_out_cache(fs_info->tree_root, inode, ctl,
 				block_group, &block_group->io_ctl, trans);
 	if (ret) {
-#ifdef CONFIG_BTRFS_DEBUG
-		btrfs_err(fs_info,
-			  "failed to write free space cache for block group %llu",
-			  block_group->start);
-#endif
+		btrfs_debug(fs_info,
+	  "failed to write free space cache for block group %llu error %d",
+			  block_group->start, ret);
 		spin_lock(&block_group->lock);
 		block_group->disk_cache_state = BTRFS_DC_ERROR;
 		spin_unlock(&block_group->lock);
@@ -3762,46 +3757,6 @@ out:
 	return ret;
 }
 
-void btrfs_get_block_group_trimming(struct btrfs_block_group *cache)
-{
-	atomic_inc(&cache->trimming);
-}
-
-void btrfs_put_block_group_trimming(struct btrfs_block_group *block_group)
-{
-	struct btrfs_fs_info *fs_info = block_group->fs_info;
-	struct extent_map_tree *em_tree;
-	struct extent_map *em;
-	bool cleanup;
-
-	spin_lock(&block_group->lock);
-	cleanup = (atomic_dec_and_test(&block_group->trimming) &&
-		   block_group->removed);
-	spin_unlock(&block_group->lock);
-
-	if (cleanup) {
-		mutex_lock(&fs_info->chunk_mutex);
-		em_tree = &fs_info->mapping_tree;
-		write_lock(&em_tree->lock);
-		em = lookup_extent_mapping(em_tree, block_group->start,
-					   1);
-		BUG_ON(!em); /* logic error, can't happen */
-		remove_extent_mapping(em_tree, em);
-		write_unlock(&em_tree->lock);
-		mutex_unlock(&fs_info->chunk_mutex);
-
-		/* once for us and once for the tree */
-		free_extent_map(em);
-		free_extent_map(em);
-
-		/*
-		 * We've left one free space entry and other tasks trimming
-		 * this block group have left 1 entry each one. Free them.
-		 */
-		__btrfs_remove_free_space_cache(block_group->free_space_ctl);
-	}
-}
-
 int btrfs_trim_block_group(struct btrfs_block_group *block_group,
 			   u64 *trimmed, u64 start, u64 end, u64 minlen)
 {
@@ -3816,7 +3771,7 @@ int btrfs_trim_block_group(struct btrfs_block_group *block_group,
 		spin_unlock(&block_group->lock);
 		return 0;
 	}
-	btrfs_get_block_group_trimming(block_group);
+	btrfs_freeze_block_group(block_group);
 	spin_unlock(&block_group->lock);
 
 	ret = trim_no_bitmap(block_group, trimmed, start, end, minlen, false);
@@ -3829,7 +3784,7 @@ int btrfs_trim_block_group(struct btrfs_block_group *block_group,
 	if (rem)
 		reset_trimming_bitmap(ctl, offset_to_bitmap(ctl, end));
 out:
-	btrfs_put_block_group_trimming(block_group);
+	btrfs_unfreeze_block_group(block_group);
 	return ret;
 }
 
@@ -3846,11 +3801,11 @@ int btrfs_trim_block_group_extents(struct btrfs_block_group *block_group,
 		spin_unlock(&block_group->lock);
 		return 0;
 	}
-	btrfs_get_block_group_trimming(block_group);
+	btrfs_freeze_block_group(block_group);
 	spin_unlock(&block_group->lock);
 
 	ret = trim_no_bitmap(block_group, trimmed, start, end, minlen, async);
-	btrfs_put_block_group_trimming(block_group);
+	btrfs_unfreeze_block_group(block_group);
 
 	return ret;
 }
@@ -3868,13 +3823,13 @@ int btrfs_trim_block_group_bitmaps(struct btrfs_block_group *block_group,
 		spin_unlock(&block_group->lock);
 		return 0;
 	}
-	btrfs_get_block_group_trimming(block_group);
+	btrfs_freeze_block_group(block_group);
 	spin_unlock(&block_group->lock);
 
 	ret = trim_bitmaps(block_group, trimmed, start, end, minlen, maxlen,
 			   async);
 
-	btrfs_put_block_group_trimming(block_group);
+	btrfs_unfreeze_block_group(block_group);
 
 	return ret;
 }
@@ -4035,11 +3990,9 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
 		if (release_metadata)
 			btrfs_delalloc_release_metadata(BTRFS_I(inode),
 					inode->i_size, true);
-#ifdef CONFIG_BTRFS_DEBUG
-		btrfs_err(fs_info,
-			  "failed to write free ino cache for root %llu",
-			  root->root_key.objectid);
-#endif
+		btrfs_debug(fs_info,
+			  "failed to write free ino cache for root %llu error %d",
+			  root->root_key.objectid, ret);
 	}
 
 	return ret;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 8b3489f229c7..768c8be4c765 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5,7 +5,6 @@
 
 #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>
@@ -49,17 +48,18 @@
 #include "qgroup.h"
 #include "delalloc-space.h"
 #include "block-group.h"
+#include "space-info.h"
 
 struct btrfs_iget_args {
-	struct btrfs_key *location;
+	u64 ino;
 	struct btrfs_root *root;
 };
 
 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;
 };
 
 static const struct inode_operations btrfs_dir_inode_operations;
@@ -1142,7 +1142,7 @@ out_unlock:
 	 */
 	if (extent_reserved) {
 		extent_clear_unlock_delalloc(inode, start,
-					     start + cur_alloc_size,
+					     start + cur_alloc_size - 1,
 					     locked_page,
 					     clear_bits,
 					     page_ops);
@@ -1355,6 +1355,66 @@ static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info,
 	return 1;
 }
 
+static int fallback_to_cow(struct inode *inode, struct page *locked_page,
+			   const u64 start, const u64 end,
+			   int *page_started, unsigned long *nr_written)
+{
+	const bool is_space_ino = btrfs_is_free_space_inode(BTRFS_I(inode));
+	const u64 range_bytes = end + 1 - start;
+	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
+	u64 range_start = start;
+	u64 count;
+
+	/*
+	 * If EXTENT_NORESERVE is set it means that when the buffered write was
+	 * made we had not enough available data space and therefore we did not
+	 * reserve data space for it, since we though we could do NOCOW for the
+	 * respective file range (either there is prealloc extent or the inode
+	 * has the NOCOW bit set).
+	 *
+	 * However when we need to fallback to COW mode (because for example the
+	 * block group for the corresponding extent was turned to RO mode by a
+	 * scrub or relocation) we need to do the following:
+	 *
+	 * 1) We increment the bytes_may_use counter of the data space info.
+	 *    If COW succeeds, it allocates a new data extent and after doing
+	 *    that it decrements the space info's bytes_may_use counter and
+	 *    increments its bytes_reserved counter by the same amount (we do
+	 *    this at btrfs_add_reserved_bytes()). So we need to increment the
+	 *    bytes_may_use counter to compensate (when space is reserved at
+	 *    buffered write time, the bytes_may_use counter is incremented);
+	 *
+	 * 2) We clear the EXTENT_NORESERVE bit from the range. We do this so
+	 *    that if the COW path fails for any reason, it decrements (through
+	 *    extent_clear_unlock_delalloc()) the bytes_may_use counter of the
+	 *    data space info, which we incremented in the step above.
+	 *
+	 * If we need to fallback to cow and the inode corresponds to a free
+	 * space cache inode, we must also increment bytes_may_use of the data
+	 * space_info for the same reason. Space caches always get a prealloc
+	 * extent for them, however scrub or balance may have set the block
+	 * group that contains that extent to RO mode.
+	 */
+	count = count_range_bits(io_tree, &range_start, end, range_bytes,
+				 EXTENT_NORESERVE, 0);
+	if (count > 0 || is_space_ino) {
+		const u64 bytes = is_space_ino ? range_bytes : count;
+		struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+		struct btrfs_space_info *sinfo = fs_info->data_sinfo;
+
+		spin_lock(&sinfo->lock);
+		btrfs_space_info_update_bytes_may_use(fs_info, sinfo, bytes);
+		spin_unlock(&sinfo->lock);
+
+		if (count > 0)
+			clear_extent_bit(io_tree, start, end, EXTENT_NORESERVE,
+					 0, 0, NULL);
+	}
+
+	return cow_file_range(inode, locked_page, start, end, page_started,
+			      nr_written, 1);
+}
+
 /*
  * when nowcow writeback call back.  This checks for snapshots or COW copies
  * of the extents that exist in the file, and COWs the file as required.
@@ -1602,9 +1662,9 @@ out_check:
 		 * NOCOW, following one which needs to be COW'ed
 		 */
 		if (cow_start != (u64)-1) {
-			ret = cow_file_range(inode, locked_page,
-					     cow_start, found_key.offset - 1,
-					     page_started, nr_written, 1);
+			ret = fallback_to_cow(inode, locked_page, cow_start,
+					      found_key.offset - 1,
+					      page_started, nr_written);
 			if (ret) {
 				if (nocow)
 					btrfs_dec_nocow_writers(fs_info,
@@ -1693,8 +1753,8 @@ out_check:
 
 	if (cow_start != (u64)-1) {
 		cur_offset = end;
-		ret = cow_file_range(inode, locked_page, cow_start, end,
-				     page_started, nr_written, 1);
+		ret = fallback_to_cow(inode, locked_page, cow_start, end,
+				      page_started, nr_written);
 		if (ret)
 			goto error;
 	}
@@ -2726,10 +2786,9 @@ void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start,
 	btrfs_queue_work(wq, &ordered_extent->work);
 }
 
-static int __readpage_endio_check(struct inode *inode,
-				  struct btrfs_io_bio *io_bio,
-				  int icsum, struct page *page,
-				  int pgoff, u64 start, size_t len)
+static int check_data_csum(struct inode *inode, struct btrfs_io_bio *io_bio,
+			   int icsum, struct page *page, int pgoff, u64 start,
+			   size_t len)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
@@ -2743,9 +2802,7 @@ static int __readpage_endio_check(struct inode *inode,
 	kaddr = kmap_atomic(page);
 	shash->tfm = fs_info->csum_shash;
 
-	crypto_shash_init(shash);
-	crypto_shash_update(shash, kaddr + pgoff, len);
-	crypto_shash_final(shash, csum);
+	crypto_shash_digest(shash, kaddr + pgoff, len, csum);
 
 	if (memcmp(csum, csum_expected, csum_size))
 		goto zeroit;
@@ -2790,8 +2847,8 @@ static int btrfs_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
 	}
 
 	phy_offset >>= inode->i_sb->s_blocksize_bits;
-	return __readpage_endio_check(inode, io_bio, phy_offset, page, offset,
-				      start, (size_t)(end - start + 1));
+	return check_data_csum(inode, io_bio, phy_offset, page, offset, start,
+			       (size_t)(end - start + 1));
 }
 
 /*
@@ -2981,7 +3038,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 		found_key.objectid = found_key.offset;
 		found_key.type = BTRFS_INODE_ITEM_KEY;
 		found_key.offset = 0;
-		inode = btrfs_iget(fs_info->sb, &found_key, root);
+		inode = btrfs_iget(fs_info->sb, last_objectid, root);
 		ret = PTR_ERR_OR_ZERO(inode);
 		if (ret && ret != -ENOENT)
 			goto out;
@@ -3000,18 +3057,16 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 			 * orphan must not get deleted.
 			 * find_dead_roots already ran before us, so if this
 			 * is a snapshot deletion, we should find the root
-			 * in the dead_roots list
+			 * in the fs_roots radix tree.
 			 */
-			spin_lock(&fs_info->trans_lock);
-			list_for_each_entry(dead_root, &fs_info->dead_roots,
-					    root_list) {
-				if (dead_root->root_key.objectid ==
-				    found_key.objectid) {
-					is_dead_root = 1;
-					break;
-				}
-			}
-			spin_unlock(&fs_info->trans_lock);
+
+			spin_lock(&fs_info->fs_roots_radix_lock);
+			dead_root = radix_tree_lookup(&fs_info->fs_roots_radix,
+							 (unsigned long)found_key.objectid);
+			if (dead_root && btrfs_root_refs(&dead_root->root_item) == 0)
+				is_dead_root = 1;
+			spin_unlock(&fs_info->fs_roots_radix_lock);
+
 			if (is_dead_root) {
 				/* prevent this orphan from being found again */
 				key.offset = found_key.objectid - 1;
@@ -3357,43 +3412,40 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
 
 	btrfs_init_map_token(&token, leaf);
 
-	btrfs_set_token_inode_uid(leaf, item, i_uid_read(inode), &token);
-	btrfs_set_token_inode_gid(leaf, item, i_gid_read(inode), &token);
-	btrfs_set_token_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size,
-				   &token);
-	btrfs_set_token_inode_mode(leaf, item, inode->i_mode, &token);
-	btrfs_set_token_inode_nlink(leaf, item, inode->i_nlink, &token);
-
-	btrfs_set_token_timespec_sec(leaf, &item->atime,
-				     inode->i_atime.tv_sec, &token);
-	btrfs_set_token_timespec_nsec(leaf, &item->atime,
-				      inode->i_atime.tv_nsec, &token);
-
-	btrfs_set_token_timespec_sec(leaf, &item->mtime,
-				     inode->i_mtime.tv_sec, &token);
-	btrfs_set_token_timespec_nsec(leaf, &item->mtime,
-				      inode->i_mtime.tv_nsec, &token);
-
-	btrfs_set_token_timespec_sec(leaf, &item->ctime,
-				     inode->i_ctime.tv_sec, &token);
-	btrfs_set_token_timespec_nsec(leaf, &item->ctime,
-				      inode->i_ctime.tv_nsec, &token);
-
-	btrfs_set_token_timespec_sec(leaf, &item->otime,
-				     BTRFS_I(inode)->i_otime.tv_sec, &token);
-	btrfs_set_token_timespec_nsec(leaf, &item->otime,
-				      BTRFS_I(inode)->i_otime.tv_nsec, &token);
-
-	btrfs_set_token_inode_nbytes(leaf, item, inode_get_bytes(inode),
-				     &token);
-	btrfs_set_token_inode_generation(leaf, item, BTRFS_I(inode)->generation,
-					 &token);
-	btrfs_set_token_inode_sequence(leaf, item, inode_peek_iversion(inode),
-				       &token);
-	btrfs_set_token_inode_transid(leaf, item, trans->transid, &token);
-	btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token);
-	btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token);
-	btrfs_set_token_inode_block_group(leaf, item, 0, &token);
+	btrfs_set_token_inode_uid(&token, item, i_uid_read(inode));
+	btrfs_set_token_inode_gid(&token, item, i_gid_read(inode));
+	btrfs_set_token_inode_size(&token, item, BTRFS_I(inode)->disk_i_size);
+	btrfs_set_token_inode_mode(&token, item, inode->i_mode);
+	btrfs_set_token_inode_nlink(&token, item, inode->i_nlink);
+
+	btrfs_set_token_timespec_sec(&token, &item->atime,
+				     inode->i_atime.tv_sec);
+	btrfs_set_token_timespec_nsec(&token, &item->atime,
+				      inode->i_atime.tv_nsec);
+
+	btrfs_set_token_timespec_sec(&token, &item->mtime,
+				     inode->i_mtime.tv_sec);
+	btrfs_set_token_timespec_nsec(&token, &item->mtime,
+				      inode->i_mtime.tv_nsec);
+
+	btrfs_set_token_timespec_sec(&token, &item->ctime,
+				     inode->i_ctime.tv_sec);
+	btrfs_set_token_timespec_nsec(&token, &item->ctime,
+				      inode->i_ctime.tv_nsec);
+
+	btrfs_set_token_timespec_sec(&token, &item->otime,
+				     BTRFS_I(inode)->i_otime.tv_sec);
+	btrfs_set_token_timespec_nsec(&token, &item->otime,
+				      BTRFS_I(inode)->i_otime.tv_nsec);
+
+	btrfs_set_token_inode_nbytes(&token, item, inode_get_bytes(inode));
+	btrfs_set_token_inode_generation(&token, item,
+					 BTRFS_I(inode)->generation);
+	btrfs_set_token_inode_sequence(&token, item, inode_peek_iversion(inode));
+	btrfs_set_token_inode_transid(&token, item, trans->transid);
+	btrfs_set_token_inode_rdev(&token, item, inode->i_rdev);
+	btrfs_set_token_inode_flags(&token, item, BTRFS_I(inode)->flags);
+	btrfs_set_token_inode_block_group(&token, item, 0);
 }
 
 /*
@@ -3618,7 +3670,7 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir)
 	 * 1 for the inode ref
 	 * 1 for the inode
 	 */
-	return btrfs_start_transaction_fallback_global_rsv(root, 5, 5);
+	return btrfs_start_transaction_fallback_global_rsv(root, 5);
 }
 
 static int btrfs_unlink(struct inode *dir, struct dentry *dentry)
@@ -4108,11 +4160,12 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 	BUG_ON(new_size > 0 && min_type != BTRFS_EXTENT_DATA_KEY);
 
 	/*
-	 * for non-free space inodes and ref cows, we want to back off from
-	 * time to time
+	 * For non-free space inodes and non-shareable roots, we want to back
+	 * off from time to time.  This means all inodes in subvolume roots,
+	 * reloc roots, and data reloc roots.
 	 */
 	if (!btrfs_is_free_space_inode(BTRFS_I(inode)) &&
-	    test_bit(BTRFS_ROOT_REF_COWS, &root->state))
+	    test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
 		be_nice = true;
 
 	path = btrfs_alloc_path();
@@ -4120,20 +4173,19 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 		return -ENOMEM;
 	path->reada = READA_BACK;
 
-	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
+	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
 		lock_extent_bits(&BTRFS_I(inode)->io_tree, lock_start, (u64)-1,
 				 &cached_state);
 
-	/*
-	 * We want to drop from the next block forward in case this new size is
-	 * not block aligned since we will be keeping the last block of the
-	 * extent just the way it is.
-	 */
-	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) ||
-	    root == fs_info->tree_root)
+		/*
+		 * We want to drop from the next block forward in case this
+		 * new size is not block aligned since we will be keeping the
+		 * last block of the extent just the way it is.
+		 */
 		btrfs_drop_extent_cache(BTRFS_I(inode), ALIGN(new_size,
 					fs_info->sectorsize),
 					(u64)-1, 0);
+	}
 
 	/*
 	 * This function is also used to drop the items in the log tree before
@@ -4241,7 +4293,7 @@ search_again:
 							 extent_num_bytes);
 				num_dec = (orig_num_bytes -
 					   extent_num_bytes);
-				if (test_bit(BTRFS_ROOT_REF_COWS,
+				if (test_bit(BTRFS_ROOT_SHAREABLE,
 					     &root->state) &&
 				    extent_start != 0)
 					inode_sub_bytes(inode, num_dec);
@@ -4257,7 +4309,7 @@ search_again:
 				num_dec = btrfs_file_extent_num_bytes(leaf, fi);
 				if (extent_start != 0) {
 					found_extent = 1;
-					if (test_bit(BTRFS_ROOT_REF_COWS,
+					if (test_bit(BTRFS_ROOT_SHAREABLE,
 						     &root->state))
 						inode_sub_bytes(inode, num_dec);
 				}
@@ -4293,7 +4345,7 @@ search_again:
 				clear_len = fs_info->sectorsize;
 			}
 
-			if (test_bit(BTRFS_ROOT_REF_COWS, &root->state))
+			if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
 				inode_sub_bytes(inode, item_end + 1 - new_size);
 		}
 delete:
@@ -4334,8 +4386,7 @@ delete:
 		should_throttle = false;
 
 		if (found_extent &&
-		    (test_bit(BTRFS_ROOT_REF_COWS, &root->state) ||
-		     root == fs_info->tree_root)) {
+		    root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
 			struct btrfs_ref ref = { 0 };
 
 			bytes_deleted += extent_num_bytes;
@@ -4759,10 +4810,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 
 		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) {
@@ -5154,7 +5202,7 @@ static int fixup_tree_root_location(struct btrfs_fs_info *fs_info,
 
 	btrfs_release_path(path);
 
-	new_root = btrfs_get_fs_root(fs_info, location, true);
+	new_root = btrfs_get_fs_root(fs_info, location->objectid, true);
 	if (IS_ERR(new_root)) {
 		err = PTR_ERR(new_root);
 		goto out;
@@ -5232,9 +5280,11 @@ static void inode_tree_del(struct inode *inode)
 static int btrfs_init_locked_inode(struct inode *inode, void *p)
 {
 	struct btrfs_iget_args *args = p;
-	inode->i_ino = args->location->objectid;
-	memcpy(&BTRFS_I(inode)->location, args->location,
-	       sizeof(*args->location));
+
+	inode->i_ino = args->ino;
+	BTRFS_I(inode)->location.objectid = args->ino;
+	BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY;
+	BTRFS_I(inode)->location.offset = 0;
 	BTRFS_I(inode)->root = btrfs_grab_root(args->root);
 	BUG_ON(args->root && !BTRFS_I(inode)->root);
 	return 0;
@@ -5243,19 +5293,19 @@ static int btrfs_init_locked_inode(struct inode *inode, void *p)
 static int btrfs_find_actor(struct inode *inode, void *opaque)
 {
 	struct btrfs_iget_args *args = opaque;
-	return args->location->objectid == BTRFS_I(inode)->location.objectid &&
+
+	return args->ino == BTRFS_I(inode)->location.objectid &&
 		args->root == BTRFS_I(inode)->root;
 }
 
-static struct inode *btrfs_iget_locked(struct super_block *s,
-				       struct btrfs_key *location,
+static struct inode *btrfs_iget_locked(struct super_block *s, u64 ino,
 				       struct btrfs_root *root)
 {
 	struct inode *inode;
 	struct btrfs_iget_args args;
-	unsigned long hashval = btrfs_inode_hash(location->objectid, root);
+	unsigned long hashval = btrfs_inode_hash(ino, root);
 
-	args.location = location;
+	args.ino = ino;
 	args.root = root;
 
 	inode = iget5_locked(s, hashval, btrfs_find_actor,
@@ -5265,17 +5315,17 @@ static struct inode *btrfs_iget_locked(struct super_block *s,
 }
 
 /*
- * Get an inode object given its location and corresponding root.
+ * Get an inode object given its inode number and corresponding root.
  * Path can be preallocated to prevent recursing back to iget through
  * allocator. NULL is also valid but may require an additional allocation
  * later.
  */
-struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
+struct inode *btrfs_iget_path(struct super_block *s, u64 ino,
 			      struct btrfs_root *root, struct btrfs_path *path)
 {
 	struct inode *inode;
 
-	inode = btrfs_iget_locked(s, location, root);
+	inode = btrfs_iget_locked(s, ino, root);
 	if (!inode)
 		return ERR_PTR(-ENOMEM);
 
@@ -5302,10 +5352,9 @@ struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
 	return inode;
 }
 
-struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
-			 struct btrfs_root *root)
+struct inode *btrfs_iget(struct super_block *s, u64 ino, struct btrfs_root *root)
 {
-	return btrfs_iget_path(s, location, root, NULL);
+	return btrfs_iget_path(s, ino, root, NULL);
 }
 
 static struct inode *new_simple_dir(struct super_block *s,
@@ -5374,7 +5423,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 		return ERR_PTR(ret);
 
 	if (location.type == BTRFS_INODE_ITEM_KEY) {
-		inode = btrfs_iget(dir->i_sb, &location, root);
+		inode = btrfs_iget(dir->i_sb, location.objectid, root);
 		if (IS_ERR(inode))
 			return inode;
 
@@ -5398,7 +5447,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 		else
 			inode = new_simple_dir(dir->i_sb, &location, sub_root);
 	} else {
-		inode = btrfs_iget(dir->i_sb, &location, sub_root);
+		inode = btrfs_iget(dir->i_sb, location.objectid, sub_root);
 	}
 	if (root != sub_root)
 		btrfs_put_root(sub_root);
@@ -5779,7 +5828,8 @@ int btrfs_set_inode_index(struct btrfs_inode *dir, u64 *index)
 static int btrfs_insert_inode_locked(struct inode *inode)
 {
 	struct btrfs_iget_args args;
-	args.location = &BTRFS_I(inode)->location;
+
+	args.ino = BTRFS_I(inode)->location.objectid;
 	args.root = BTRFS_I(inode)->root;
 
 	return insert_inode_locked4(inode,
@@ -6991,7 +7041,7 @@ 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;
@@ -7129,30 +7179,7 @@ static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len,
 }
 
 
-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)
@@ -7214,7 +7241,6 @@ 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(inode, start, len);
 	if (IS_ERR(em)) {
@@ -7225,64 +7251,73 @@ 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 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;
 
-	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);
+
+	dio_data = kzalloc(sizeof(*dio_data), GFP_NOFS);
+	if (!dio_data)
+		return -ENOMEM;
+
+	dio_data->length = length;
+	if (write) {
+		dio_data->reserve = round_up(length, fs_info->sectorsize);
+		ret = btrfs_delalloc_reserve_space(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;
 	}
@@ -7314,36 +7349,48 @@ 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);
 
 	return 0;
@@ -7352,370 +7399,152 @@ 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(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 inline blk_status_t submit_dio_repair_bio(struct inode *inode,
-						 struct bio *bio,
-						 int mirror_num)
+static int btrfs_dio_iomap_end(struct inode *inode, loff_t pos, loff_t length,
+		ssize_t written, unsigned flags, struct iomap *iomap)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	blk_status_t ret;
+	int ret = 0;
+	struct btrfs_dio_data *dio_data = iomap->private;
+	size_t submitted = dio_data->submitted;
+	const bool write = !!(flags & IOMAP_WRITE);
 
-	BUG_ON(bio_op(bio) == REQ_OP_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;
+	}
 
-	ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DIO_REPAIR);
-	if (ret)
-		return ret;
+	if (submitted < length) {
+		pos += submitted;
+		length -= submitted;
+		if (write)
+			__endio_write_update_ordered(inode, pos, length, false);
+		else
+			unlock_extent(&BTRFS_I(inode)->io_tree, pos,
+				      pos + length - 1);
+		ret = -ENOTBLK;
+	}
 
-	ret = btrfs_map_bio(fs_info, bio, mirror_num);
+	if (write) {
+		if (dio_data->reserve)
+			btrfs_delalloc_release_space(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:
+	kfree(dio_data);
+	iomap->private = NULL;
 
 	return ret;
 }
 
-static int btrfs_check_dio_repairable(struct inode *inode,
-				      struct bio *failed_bio,
-				      struct io_failure_record *failrec,
-				      int failed_mirror)
+static void btrfs_dio_private_put(struct btrfs_dio_private *dip)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	int num_copies;
-
-	num_copies = btrfs_num_copies(fs_info, failrec->logical, failrec->len);
-	if (num_copies == 1) {
-		/*
-		 * we only have a single copy of the data, so don't bother with
-		 * all the retry and error correction code that follows. no
-		 * matter what the error is, it is very likely to persist.
-		 */
-		btrfs_debug(fs_info,
-			"Check DIO Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d",
-			num_copies, failrec->this_mirror, failed_mirror);
-		return 0;
-	}
-
-	failrec->failed_mirror = failed_mirror;
-	failrec->this_mirror++;
-	if (failrec->this_mirror == failed_mirror)
-		failrec->this_mirror++;
+	/*
+	 * This implies a barrier so that stores to dio_bio->bi_status before
+	 * this and loads of dio_bio->bi_status after this are fully ordered.
+	 */
+	if (!refcount_dec_and_test(&dip->refs))
+		return;
 
-	if (failrec->this_mirror > num_copies) {
-		btrfs_debug(fs_info,
-			"Check DIO Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d",
-			num_copies, failrec->this_mirror, failed_mirror);
-		return 0;
+	if (bio_op(dip->dio_bio) == REQ_OP_WRITE) {
+		__endio_write_update_ordered(dip->inode, dip->logical_offset,
+					     dip->bytes,
+					     !dip->dio_bio->bi_status);
+	} else {
+		unlock_extent(&BTRFS_I(dip->inode)->io_tree,
+			      dip->logical_offset,
+			      dip->logical_offset + dip->bytes - 1);
 	}
 
-	return 1;
+	bio_endio(dip->dio_bio);
+	kfree(dip);
 }
 
-static blk_status_t dio_read_error(struct inode *inode, struct bio *failed_bio,
-				   struct page *page, unsigned int pgoff,
-				   u64 start, u64 end, int failed_mirror,
-				   bio_end_io_t *repair_endio, void *repair_arg)
+static blk_status_t submit_dio_repair_bio(struct inode *inode, struct bio *bio,
+					  int mirror_num,
+					  unsigned long bio_flags)
 {
-	struct io_failure_record *failrec;
-	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
-	struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
-	struct bio *bio;
-	int isector;
-	unsigned int read_mode = 0;
-	int segs;
-	int ret;
-	blk_status_t status;
-	struct bio_vec bvec;
+	struct btrfs_dio_private *dip = bio->bi_private;
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	blk_status_t ret;
 
-	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
+	BUG_ON(bio_op(bio) == REQ_OP_WRITE);
 
-	ret = btrfs_get_io_failure_record(inode, start, end, &failrec);
+	ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA);
 	if (ret)
-		return errno_to_blk_status(ret);
-
-	ret = btrfs_check_dio_repairable(inode, failed_bio, failrec,
-					 failed_mirror);
-	if (!ret) {
-		free_io_failure(failure_tree, io_tree, failrec);
-		return BLK_STS_IOERR;
-	}
-
-	segs = bio_segments(failed_bio);
-	bio_get_first_bvec(failed_bio, &bvec);
-	if (segs > 1 ||
-	    (bvec.bv_len > btrfs_inode_sectorsize(inode)))
-		read_mode |= REQ_FAILFAST_DEV;
-
-	isector = start - btrfs_io_bio(failed_bio)->logical;
-	isector >>= inode->i_sb->s_blocksize_bits;
-	bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page,
-				pgoff, isector, repair_endio, repair_arg);
-	bio->bi_opf = REQ_OP_READ | read_mode;
-
-	btrfs_debug(BTRFS_I(inode)->root->fs_info,
-		    "repair DIO read error: submitting new dio read[%#x] to this_mirror=%d, in_validation=%d",
-		    read_mode, failrec->this_mirror, failrec->in_validation);
-
-	status = submit_dio_repair_bio(inode, bio, failrec->this_mirror);
-	if (status) {
-		free_io_failure(failure_tree, io_tree, failrec);
-		bio_put(bio);
-	}
-
-	return status;
-}
-
-struct btrfs_retry_complete {
-	struct completion done;
-	struct inode *inode;
-	u64 start;
-	int uptodate;
-};
+		return ret;
 
-static void btrfs_retry_endio_nocsum(struct bio *bio)
-{
-	struct btrfs_retry_complete *done = bio->bi_private;
-	struct inode *inode = done->inode;
-	struct bio_vec *bvec;
-	struct extent_io_tree *io_tree, *failure_tree;
-	struct bvec_iter_all iter_all;
-
-	if (bio->bi_status)
-		goto end;
-
-	ASSERT(bio->bi_vcnt == 1);
-	io_tree = &BTRFS_I(inode)->io_tree;
-	failure_tree = &BTRFS_I(inode)->io_failure_tree;
-	ASSERT(bio_first_bvec_all(bio)->bv_len == btrfs_inode_sectorsize(inode));
-
-	done->uptodate = 1;
-	ASSERT(!bio_flagged(bio, BIO_CLONED));
-	bio_for_each_segment_all(bvec, bio, iter_all)
-		clean_io_failure(BTRFS_I(inode)->root->fs_info, failure_tree,
-				 io_tree, done->start, bvec->bv_page,
-				 btrfs_ino(BTRFS_I(inode)), 0);
-end:
-	complete(&done->done);
-	bio_put(bio);
+	refcount_inc(&dip->refs);
+	ret = btrfs_map_bio(fs_info, bio, mirror_num);
+	if (ret)
+		refcount_dec(&dip->refs);
+	return ret;
 }
 
-static blk_status_t __btrfs_correct_data_nocsum(struct inode *inode,
-						struct btrfs_io_bio *io_bio)
+static blk_status_t btrfs_check_read_dio_bio(struct inode *inode,
+					     struct btrfs_io_bio *io_bio,
+					     const bool uptodate)
 {
-	struct btrfs_fs_info *fs_info;
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+	const u32 sectorsize = fs_info->sectorsize;
+	struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
+	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
+	const bool csum = !(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM);
 	struct bio_vec bvec;
 	struct bvec_iter iter;
-	struct btrfs_retry_complete done;
-	u64 start;
-	unsigned int pgoff;
-	u32 sectorsize;
-	int nr_sectors;
-	blk_status_t ret;
+	u64 start = io_bio->logical;
+	int icsum = 0;
 	blk_status_t err = BLK_STS_OK;
 
-	fs_info = BTRFS_I(inode)->root->fs_info;
-	sectorsize = fs_info->sectorsize;
-
-	start = io_bio->logical;
-	done.inode = inode;
-	io_bio->bio.bi_iter = io_bio->iter;
+	__bio_for_each_segment(bvec, &io_bio->bio, iter, io_bio->iter) {
+		unsigned int i, nr_sectors, pgoff;
 
-	bio_for_each_segment(bvec, &io_bio->bio, iter) {
 		nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec.bv_len);
 		pgoff = bvec.bv_offset;
-
-next_block_or_try_again:
-		done.uptodate = 0;
-		done.start = start;
-		init_completion(&done.done);
-
-		ret = dio_read_error(inode, &io_bio->bio, bvec.bv_page,
-				pgoff, start, start + sectorsize - 1,
-				io_bio->mirror_num,
-				btrfs_retry_endio_nocsum, &done);
-		if (ret) {
-			err = ret;
-			goto next;
-		}
-
-		wait_for_completion_io(&done.done);
-
-		if (!done.uptodate) {
-			/* We might have another mirror, so try again */
-			goto next_block_or_try_again;
-		}
-
-next:
-		start += sectorsize;
-
-		nr_sectors--;
-		if (nr_sectors) {
-			pgoff += sectorsize;
+		for (i = 0; i < nr_sectors; i++) {
 			ASSERT(pgoff < PAGE_SIZE);
-			goto next_block_or_try_again;
-		}
-	}
-
-	return err;
-}
-
-static void btrfs_retry_endio(struct bio *bio)
-{
-	struct btrfs_retry_complete *done = bio->bi_private;
-	struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
-	struct extent_io_tree *io_tree, *failure_tree;
-	struct inode *inode = done->inode;
-	struct bio_vec *bvec;
-	int uptodate;
-	int ret;
-	int i = 0;
-	struct bvec_iter_all iter_all;
-
-	if (bio->bi_status)
-		goto end;
-
-	uptodate = 1;
-
-	ASSERT(bio->bi_vcnt == 1);
-	ASSERT(bio_first_bvec_all(bio)->bv_len == btrfs_inode_sectorsize(done->inode));
-
-	io_tree = &BTRFS_I(inode)->io_tree;
-	failure_tree = &BTRFS_I(inode)->io_failure_tree;
-
-	ASSERT(!bio_flagged(bio, BIO_CLONED));
-	bio_for_each_segment_all(bvec, bio, iter_all) {
-		ret = __readpage_endio_check(inode, io_bio, i, bvec->bv_page,
-					     bvec->bv_offset, done->start,
-					     bvec->bv_len);
-		if (!ret)
-			clean_io_failure(BTRFS_I(inode)->root->fs_info,
-					 failure_tree, io_tree, done->start,
-					 bvec->bv_page,
-					 btrfs_ino(BTRFS_I(inode)),
-					 bvec->bv_offset);
-		else
-			uptodate = 0;
-		i++;
-	}
-
-	done->uptodate = uptodate;
-end:
-	complete(&done->done);
-	bio_put(bio);
-}
-
-static blk_status_t __btrfs_subio_endio_read(struct inode *inode,
-		struct btrfs_io_bio *io_bio, blk_status_t err)
-{
-	struct btrfs_fs_info *fs_info;
-	struct bio_vec bvec;
-	struct bvec_iter iter;
-	struct btrfs_retry_complete done;
-	u64 start;
-	u64 offset = 0;
-	u32 sectorsize;
-	int nr_sectors;
-	unsigned int pgoff;
-	int csum_pos;
-	bool uptodate = (err == 0);
-	int ret;
-	blk_status_t status;
-
-	fs_info = BTRFS_I(inode)->root->fs_info;
-	sectorsize = fs_info->sectorsize;
-
-	err = BLK_STS_OK;
-	start = io_bio->logical;
-	done.inode = inode;
-	io_bio->bio.bi_iter = io_bio->iter;
-
-	bio_for_each_segment(bvec, &io_bio->bio, iter) {
-		nr_sectors = BTRFS_BYTES_TO_BLKS(fs_info, bvec.bv_len);
-
-		pgoff = bvec.bv_offset;
-next_block:
-		if (uptodate) {
-			csum_pos = BTRFS_BYTES_TO_BLKS(fs_info, offset);
-			ret = __readpage_endio_check(inode, io_bio, csum_pos,
-					bvec.bv_page, pgoff, start, sectorsize);
-			if (likely(!ret))
-				goto next;
-		}
-try_again:
-		done.uptodate = 0;
-		done.start = start;
-		init_completion(&done.done);
-
-		status = dio_read_error(inode, &io_bio->bio, bvec.bv_page,
-					pgoff, start, start + sectorsize - 1,
-					io_bio->mirror_num, btrfs_retry_endio,
-					&done);
-		if (status) {
-			err = status;
-			goto next;
-		}
-
-		wait_for_completion_io(&done.done);
-
-		if (!done.uptodate) {
-			/* We might have another mirror, so try again */
-			goto try_again;
-		}
-next:
-		offset += sectorsize;
-		start += sectorsize;
-
-		ASSERT(nr_sectors);
-
-		nr_sectors--;
-		if (nr_sectors) {
+			if (uptodate &&
+			    (!csum || !check_data_csum(inode, io_bio, icsum,
+						       bvec.bv_page, pgoff,
+						       start, sectorsize))) {
+				clean_io_failure(fs_info, failure_tree, io_tree,
+						 start, bvec.bv_page,
+						 btrfs_ino(BTRFS_I(inode)),
+						 pgoff);
+			} else {
+				blk_status_t status;
+
+				status = btrfs_submit_read_repair(inode,
+							&io_bio->bio,
+							start - io_bio->logical,
+							bvec.bv_page, pgoff,
+							start,
+							start + sectorsize - 1,
+							io_bio->mirror_num,
+							submit_dio_repair_bio);
+				if (status)
+					err = status;
+			}
+			start += sectorsize;
+			icsum++;
 			pgoff += sectorsize;
-			ASSERT(pgoff < PAGE_SIZE);
-			goto next_block;
 		}
 	}
-
 	return err;
 }
 
-static blk_status_t btrfs_subio_endio_read(struct inode *inode,
-		struct btrfs_io_bio *io_bio, blk_status_t err)
-{
-	bool skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
-
-	if (skip_csum) {
-		if (unlikely(err))
-			return __btrfs_correct_data_nocsum(inode, io_bio);
-		else
-			return BLK_STS_OK;
-	} else {
-		return __btrfs_subio_endio_read(inode, io_bio, err);
-	}
-}
-
-static void btrfs_endio_direct_read(struct bio *bio)
-{
-	struct btrfs_dio_private *dip = bio->bi_private;
-	struct inode *inode = dip->inode;
-	struct bio *dio_bio;
-	struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
-	blk_status_t err = bio->bi_status;
-
-	if (dip->flags & BTRFS_DIO_ORIG_BIO_SUBMITTED)
-		err = btrfs_subio_endio_read(inode, io_bio, err);
-
-	unlock_extent(&BTRFS_I(inode)->io_tree, dip->logical_offset,
-		      dip->logical_offset + dip->bytes - 1);
-	dio_bio = dip->dio_bio;
-
-	kfree(dip);
-
-	dio_bio->bi_status = err;
-	dio_end_io(dio_bio);
-	btrfs_io_bio_free_csum(io_bio);
-	bio_put(bio);
-}
-
 static void __endio_write_update_ordered(struct inode *inode,
 					 const u64 offset, const u64 bytes,
 					 const bool uptodate)
@@ -7759,21 +7588,6 @@ static void __endio_write_update_ordered(struct inode *inode,
 	}
 }
 
-static void btrfs_endio_direct_write(struct bio *bio)
-{
-	struct btrfs_dio_private *dip = bio->bi_private;
-	struct bio *dio_bio = dip->dio_bio;
-
-	__endio_write_update_ordered(dip->inode, dip->logical_offset,
-				     dip->bytes, !bio->bi_status);
-
-	kfree(dip);
-
-	dio_bio->bi_status = bio->bi_status;
-	dio_end_io(dio_bio);
-	bio_put(bio);
-}
-
 static blk_status_t btrfs_submit_bio_start_direct_io(void *private_data,
 				    struct bio *bio, u64 offset)
 {
@@ -7797,64 +7611,16 @@ static void btrfs_end_dio_bio(struct bio *bio)
 			   (unsigned long long)bio->bi_iter.bi_sector,
 			   bio->bi_iter.bi_size, err);
 
-	if (dip->subio_endio)
-		err = dip->subio_endio(dip->inode, btrfs_io_bio(bio), err);
-
-	if (err) {
-		/*
-		 * We want to perceive the errors flag being set before
-		 * decrementing the reference count. We don't need a barrier
-		 * since atomic operations with a return value are fully
-		 * ordered as per atomic_t.txt
-		 */
-		dip->errors = 1;
+	if (bio_op(bio) == REQ_OP_READ) {
+		err = btrfs_check_read_dio_bio(dip->inode, btrfs_io_bio(bio),
+					       !err);
 	}
 
-	/* if there are more bios still pending for this dio, just exit */
-	if (!atomic_dec_and_test(&dip->pending_bios))
-		goto out;
+	if (err)
+		dip->dio_bio->bi_status = err;
 
-	if (dip->errors) {
-		bio_io_error(dip->orig_bio);
-	} else {
-		dip->dio_bio->bi_status = BLK_STS_OK;
-		bio_endio(dip->orig_bio);
-	}
-out:
 	bio_put(bio);
-}
-
-static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode,
-						 struct btrfs_dio_private *dip,
-						 struct bio *bio,
-						 u64 file_offset)
-{
-	struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
-	struct btrfs_io_bio *orig_io_bio = btrfs_io_bio(dip->orig_bio);
-	u16 csum_size;
-	blk_status_t ret;
-
-	/*
-	 * We load all the csum data we need when we submit
-	 * the first bio to reduce the csum tree search and
-	 * contention.
-	 */
-	if (dip->logical_offset == file_offset) {
-		ret = btrfs_lookup_bio_sums(inode, dip->orig_bio, file_offset,
-					    NULL);
-		if (ret)
-			return ret;
-	}
-
-	if (bio == dip->orig_bio)
-		return 0;
-
-	file_offset -= dip->logical_offset;
-	file_offset >>= inode->i_sb->s_blocksize_bits;
-	csum_size = btrfs_super_csum_size(btrfs_sb(inode->i_sb)->super_copy);
-	io_bio->csum = orig_io_bio->csum + csum_size * file_offset;
-
-	return 0;
+	btrfs_dio_private_put(dip);
 }
 
 static inline blk_status_t btrfs_submit_dio_bio(struct bio *bio,
@@ -7892,10 +7658,12 @@ static inline blk_status_t btrfs_submit_dio_bio(struct bio *bio,
 		if (ret)
 			goto err;
 	} else {
-		ret = btrfs_lookup_and_bind_dio_csum(inode, dip, bio,
-						     file_offset);
-		if (ret)
-			goto err;
+		u64 csum_offset;
+
+		csum_offset = file_offset - dip->logical_offset;
+		csum_offset >>= inode->i_sb->s_blocksize_bits;
+		csum_offset *= btrfs_super_csum_size(fs_info->super_copy);
+		btrfs_io_bio(bio)->csum = dip->csums + csum_offset;
 	}
 map:
 	ret = btrfs_map_bio(fs_info, bio, 0);
@@ -7903,14 +7671,53 @@ err:
 	return ret;
 }
 
-static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip)
+/*
+ * If this succeeds, the btrfs_dio_private is responsible for cleaning up locked
+ * or ordered extents whether or not we submit any bios.
+ */
+static struct btrfs_dio_private *btrfs_create_dio_private(struct bio *dio_bio,
+							  struct inode *inode,
+							  loff_t file_offset)
 {
-	struct inode *inode = dip->inode;
+	const bool write = (bio_op(dio_bio) == REQ_OP_WRITE);
+	const bool csum = !(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM);
+	size_t dip_size;
+	struct btrfs_dio_private *dip;
+
+	dip_size = sizeof(*dip);
+	if (!write && csum) {
+		struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+		const u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
+		size_t nblocks;
+
+		nblocks = dio_bio->bi_iter.bi_size >> inode->i_sb->s_blocksize_bits;
+		dip_size += csum_size * nblocks;
+	}
+
+	dip = kzalloc(dip_size, GFP_NOFS);
+	if (!dip)
+		return NULL;
+
+	dip->inode = inode;
+	dip->logical_offset = file_offset;
+	dip->bytes = dio_bio->bi_iter.bi_size;
+	dip->disk_bytenr = (u64)dio_bio->bi_iter.bi_sector << 9;
+	dip->dio_bio = dio_bio;
+	refcount_set(&dip->refs, 1);
+	return dip;
+}
+
+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);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	const bool raid56 = (btrfs_data_alloc_profile(fs_info) &
+			     BTRFS_BLOCK_GROUP_RAID56_MASK);
+	struct btrfs_dio_private *dip;
 	struct bio *bio;
-	struct bio *orig_bio = dip->orig_bio;
-	u64 start_sector = orig_bio->bi_iter.bi_sector;
-	u64 file_offset = dip->logical_offset;
+	u64 start_sector;
 	int async_submit = 0;
 	u64 submit_len;
 	int clone_offset = 0;
@@ -7918,330 +7725,108 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip)
 	int ret;
 	blk_status_t status;
 	struct btrfs_io_geometry geom;
+	struct btrfs_dio_data *dio_data = iomap->private;
 
-	submit_len = orig_bio->bi_iter.bi_size;
-	ret = btrfs_get_io_geometry(fs_info, btrfs_op(orig_bio),
-				    start_sector << 9, submit_len, &geom);
-	if (ret)
-		return -EIO;
+	dip = btrfs_create_dio_private(dio_bio, inode, file_offset);
+	if (!dip) {
+		if (!write) {
+			unlock_extent(&BTRFS_I(inode)->io_tree, file_offset,
+				file_offset + dio_bio->bi_iter.bi_size - 1);
+		}
+		dio_bio->bi_status = BLK_STS_RESOURCE;
+		bio_endio(dio_bio);
+		return BLK_QC_T_NONE;
+	}
 
-	if (geom.len >= submit_len) {
-		bio = orig_bio;
-		dip->flags |= BTRFS_DIO_ORIG_BIO_SUBMITTED;
-		goto submit;
+	if (!write && csum) {
+		/*
+		 * Load the csums up front to reduce csum tree searches and
+		 * contention when submitting bios.
+		 */
+		status = btrfs_lookup_bio_sums(inode, dio_bio, file_offset,
+					       dip->csums);
+		if (status != BLK_STS_OK)
+			goto out_err;
 	}
 
-	/* async crcs make it difficult to collect full stripe writes. */
-	if (btrfs_data_alloc_profile(fs_info) & BTRFS_BLOCK_GROUP_RAID56_MASK)
-		async_submit = 0;
-	else
-		async_submit = 1;
+	start_sector = dio_bio->bi_iter.bi_sector;
+	submit_len = dio_bio->bi_iter.bi_size;
 
-	/* bio split */
-	ASSERT(geom.len <= INT_MAX);
-	atomic_inc(&dip->pending_bios);
 	do {
+		ret = btrfs_get_io_geometry(fs_info, btrfs_op(dio_bio),
+					    start_sector << 9, submit_len,
+					    &geom);
+		if (ret) {
+			status = errno_to_blk_status(ret);
+			goto out_err;
+		}
+		ASSERT(geom.len <= INT_MAX);
+
 		clone_len = min_t(int, submit_len, geom.len);
 
 		/*
 		 * This will never fail as it's passing GPF_NOFS and
 		 * the allocation is backed by btrfs_bioset.
 		 */
-		bio = btrfs_bio_clone_partial(orig_bio, clone_offset,
-					      clone_len);
+		bio = btrfs_bio_clone_partial(dio_bio, clone_offset, clone_len);
 		bio->bi_private = dip;
 		bio->bi_end_io = btrfs_end_dio_bio;
 		btrfs_io_bio(bio)->logical = file_offset;
 
 		ASSERT(submit_len >= clone_len);
 		submit_len -= clone_len;
-		if (submit_len == 0)
-			break;
 
 		/*
 		 * Increase the count before we submit the bio so we know
 		 * the end IO handler won't happen before we increase the
 		 * count. Otherwise, the dip might get freed before we're
 		 * done setting it up.
+		 *
+		 * We transfer the initial reference to the last bio, so we
+		 * don't need to increment the reference count for the last one.
 		 */
-		atomic_inc(&dip->pending_bios);
+		if (submit_len > 0) {
+			refcount_inc(&dip->refs);
+			/*
+			 * If we are submitting more than one bio, submit them
+			 * all asynchronously. The exception is RAID 5 or 6, as
+			 * asynchronous checksums make it difficult to collect
+			 * full stripe writes.
+			 */
+			if (!raid56)
+				async_submit = 1;
+		}
 
 		status = btrfs_submit_dio_bio(bio, inode, file_offset,
 						async_submit);
 		if (status) {
 			bio_put(bio);
-			atomic_dec(&dip->pending_bios);
+			if (submit_len > 0)
+				refcount_dec(&dip->refs);
 			goto out_err;
 		}
 
+		dio_data->submitted += clone_len;
 		clone_offset += clone_len;
 		start_sector += clone_len >> 9;
 		file_offset += clone_len;
-
-		ret = btrfs_get_io_geometry(fs_info, btrfs_op(orig_bio),
-				      start_sector << 9, submit_len, &geom);
-		if (ret)
-			goto out_err;
 	} while (submit_len > 0);
+	return BLK_QC_T_NONE;
 
-submit:
-	status = btrfs_submit_dio_bio(bio, inode, file_offset, async_submit);
-	if (!status)
-		return 0;
-
-	bio_put(bio);
 out_err:
-	dip->errors = 1;
-	/*
-	 * Before atomic variable goto zero, we must  make sure dip->errors is
-	 * perceived to be set. This ordering is ensured by the fact that an
-	 * atomic operations with a return value are fully ordered as per
-	 * atomic_t.txt
-	 */
-	if (atomic_dec_and_test(&dip->pending_bios))
-		bio_io_error(dip->orig_bio);
-
-	/* bio_end_io() will handle error, so we needn't return it */
-	return 0;
-}
-
-static void btrfs_submit_direct(struct bio *dio_bio, struct inode *inode,
-				loff_t file_offset)
-{
-	struct btrfs_dio_private *dip = NULL;
-	struct bio *bio = NULL;
-	struct btrfs_io_bio *io_bio;
-	bool write = (bio_op(dio_bio) == REQ_OP_WRITE);
-	int ret = 0;
-
-	bio = btrfs_bio_clone(dio_bio);
-
-	dip = kzalloc(sizeof(*dip), GFP_NOFS);
-	if (!dip) {
-		ret = -ENOMEM;
-		goto free_ordered;
-	}
-
-	dip->private = dio_bio->bi_private;
-	dip->inode = inode;
-	dip->logical_offset = file_offset;
-	dip->bytes = dio_bio->bi_iter.bi_size;
-	dip->disk_bytenr = (u64)dio_bio->bi_iter.bi_sector << 9;
-	bio->bi_private = dip;
-	dip->orig_bio = bio;
-	dip->dio_bio = dio_bio;
-	atomic_set(&dip->pending_bios, 0);
-	io_bio = btrfs_io_bio(bio);
-	io_bio->logical = file_offset;
-
-	if (write) {
-		bio->bi_end_io = btrfs_endio_direct_write;
-	} else {
-		bio->bi_end_io = btrfs_endio_direct_read;
-		dip->subio_endio = btrfs_subio_endio_read;
-	}
-
-	/*
-	 * Reset the range for unsubmitted ordered extents (to a 0 length range)
-	 * even if we fail to submit a bio, because in such case we do the
-	 * corresponding error handling below and it must not be done a second
-	 * time by btrfs_direct_IO().
-	 */
-	if (write) {
-		struct btrfs_dio_data *dio_data = current->journal_info;
-
-		dio_data->unsubmitted_oe_range_end = dip->logical_offset +
-			dip->bytes;
-		dio_data->unsubmitted_oe_range_start =
-			dio_data->unsubmitted_oe_range_end;
-	}
-
-	ret = btrfs_submit_direct_hook(dip);
-	if (!ret)
-		return;
-
-	btrfs_io_bio_free_csum(io_bio);
-
-free_ordered:
-	/*
-	 * If we arrived here it means either we failed to submit the dip
-	 * or we either failed to clone the dio_bio or failed to allocate the
-	 * dip. If we cloned the dio_bio and allocated the dip, we can just
-	 * call bio_endio against our io_bio so that we get proper resource
-	 * cleanup if we fail to submit the dip, otherwise, we must do the
-	 * same as btrfs_endio_direct_[write|read] because we can't call these
-	 * callbacks - they require an allocated dip and a clone of dio_bio.
-	 */
-	if (bio && dip) {
-		bio_io_error(bio);
-		/*
-		 * The end io callbacks free our dip, do the final put on bio
-		 * and all the cleanup and final put for dio_bio (through
-		 * dio_end_io()).
-		 */
-		dip = NULL;
-		bio = NULL;
-	} else {
-		if (write)
-			__endio_write_update_ordered(inode,
-						file_offset,
-						dio_bio->bi_iter.bi_size,
-						false);
-		else
-			unlock_extent(&BTRFS_I(inode)->io_tree, file_offset,
-			      file_offset + dio_bio->bi_iter.bi_size - 1);
-
-		dio_bio->bi_status = BLK_STS_IOERR;
-		/*
-		 * Releases and cleans up our dio_bio, no need to bio_put()
-		 * nor bio_endio()/bio_io_error() against dio_bio.
-		 */
-		dio_end_io(dio_bio);
-	}
-	if (bio)
-		bio_put(bio);
-	kfree(dip);
+	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,
-			       const struct iov_iter *iter, loff_t offset)
-{
-	int seg;
-	int i;
-	unsigned int blocksize_mask = fs_info->sectorsize - 1;
-	ssize_t retval = -EINVAL;
-
-	if (offset & blocksize_mask)
-		goto out;
-
-	if (iov_iter_alignment(iter) & blocksize_mask)
-		goto out;
-
-	/* If this is a write we don't need to check anymore */
-	if (iov_iter_rw(iter) != READ || !iter_is_iovec(iter))
-		return 0;
-	/*
-	 * Check to make sure we don't have duplicate iov_base's in this
-	 * iovec, if so return EINVAL, otherwise we'll get csum errors
-	 * when reading back.
-	 */
-	for (seg = 0; seg < iter->nr_segs; seg++) {
-		for (i = seg + 1; i < iter->nr_segs; i++) {
-			if (iter->iov[seg].iov_base == iter->iov[i].iov_base)
-				goto out;
-		}
-	}
-	retval = 0;
-out:
-	return retval;
-}
-
-static 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
-		 * the isize, but it is protected by i_mutex. So we can
-		 * not unlock the i_mutex at this case.
-		 */
-		if (offset + count <= inode->i_size) {
-			dio_data.overwrite = 1;
-			inode_unlock(inode);
-			relock = true;
-		} else if (iocb->ki_flags & IOCB_NOWAIT) {
-			ret = -EAGAIN;
-			goto out;
-		}
-		ret = btrfs_delalloc_reserve_space(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) {
-		up_read(&BTRFS_I(inode)->dio_sem);
-		current->journal_info = NULL;
-		if (ret < 0 && ret != -EIOCBQUEUED) {
-			if (dio_data.reserve)
-				btrfs_delalloc_release_space(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(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(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);
+const struct iomap_ops btrfs_dio_iomap_ops = {
+	.iomap_begin            = btrfs_dio_iomap_begin,
+	.iomap_end              = btrfs_dio_iomap_end,
+};
 
-	extent_changeset_free(data_reserved);
-	return ret;
-}
+const struct iomap_dio_ops btrfs_dops = {
+	.submit_io		= btrfs_submit_direct,
+};
 
 #define BTRFS_FIEMAP_FLAGS	(FIEMAP_FLAG_SYNC)
 
@@ -10539,7 +10124,7 @@ 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 40b729dce91c..168deb8ef68a 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -660,7 +660,7 @@ static noinline int create_subvol(struct inode *dir,
 		goto fail;
 
 	key.offset = (u64)-1;
-	new_root = btrfs_get_fs_root(fs_info, &key, true);
+	new_root = btrfs_get_fs_root(fs_info, objectid, true);
 	if (IS_ERR(new_root)) {
 		ret = PTR_ERR(new_root);
 		btrfs_abort_transaction(trans, ret);
@@ -748,9 +748,8 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	struct btrfs_pending_snapshot *pending_snapshot;
 	struct btrfs_trans_handle *trans;
 	int ret;
-	bool snapshot_force_cow = false;
 
-	if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
+	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
 		return -EINVAL;
 
 	if (atomic_read(&root->nr_swapfiles)) {
@@ -771,27 +770,6 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 		goto free_pending;
 	}
 
-	/*
-	 * Force new buffered writes to reserve space even when NOCOW is
-	 * possible. This is to avoid later writeback (running dealloc) to
-	 * fallback to COW mode and unexpectedly fail with ENOSPC.
-	 */
-	btrfs_drew_read_lock(&root->snapshot_lock);
-
-	ret = btrfs_start_delalloc_snapshot(root);
-	if (ret)
-		goto dec_and_free;
-
-	/*
-	 * All previous writes have started writeback in NOCOW mode, so now
-	 * we force future writes to fallback to COW mode during snapshot
-	 * creation.
-	 */
-	atomic_inc(&root->snapshot_force_cow);
-	snapshot_force_cow = true;
-
-	btrfs_wait_ordered_extents(root, U64_MAX, 0, (u64)-1);
-
 	btrfs_init_block_rsv(&pending_snapshot->block_rsv,
 			     BTRFS_BLOCK_RSV_TEMP);
 	/*
@@ -806,7 +784,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 					&pending_snapshot->block_rsv, 8,
 					false);
 	if (ret)
-		goto dec_and_free;
+		goto free_pending;
 
 	pending_snapshot->dentry = dentry;
 	pending_snapshot->root = root;
@@ -848,11 +826,6 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 fail:
 	btrfs_put_root(pending_snapshot->snap);
 	btrfs_subvolume_release_metadata(fs_info, &pending_snapshot->block_rsv);
-dec_and_free:
-	if (snapshot_force_cow)
-		atomic_dec(&root->snapshot_force_cow);
-	btrfs_drew_read_unlock(&root->snapshot_lock);
-
 free_pending:
 	kfree(pending_snapshot->root_item);
 	btrfs_free_path(pending_snapshot->path);
@@ -983,6 +956,45 @@ out_unlock:
 	return error;
 }
 
+static noinline int btrfs_mksnapshot(const struct path *parent,
+				   const char *name, int namelen,
+				   struct btrfs_root *root,
+				   bool readonly,
+				   struct btrfs_qgroup_inherit *inherit)
+{
+	int ret;
+	bool snapshot_force_cow = false;
+
+	/*
+	 * Force new buffered writes to reserve space even when NOCOW is
+	 * possible. This is to avoid later writeback (running dealloc) to
+	 * fallback to COW mode and unexpectedly fail with ENOSPC.
+	 */
+	btrfs_drew_read_lock(&root->snapshot_lock);
+
+	ret = btrfs_start_delalloc_snapshot(root);
+	if (ret)
+		goto out;
+
+	/*
+	 * All previous writes have started writeback in NOCOW mode, so now
+	 * we force future writes to fallback to COW mode during snapshot
+	 * creation.
+	 */
+	atomic_inc(&root->snapshot_force_cow);
+	snapshot_force_cow = true;
+
+	btrfs_wait_ordered_extents(root, U64_MAX, 0, (u64)-1);
+
+	ret = btrfs_mksubvol(parent, name, namelen,
+			     root, readonly, inherit);
+out:
+	if (snapshot_force_cow)
+		atomic_dec(&root->snapshot_force_cow);
+	btrfs_drew_read_unlock(&root->snapshot_lock);
+	return ret;
+}
+
 /*
  * When we're defragging a range, we don't want to kick it off again
  * if it is really just waiting for delalloc to send it down.
@@ -1762,7 +1774,7 @@ static noinline int __btrfs_ioctl_snap_create(struct file *file,
 			 */
 			ret = -EPERM;
 		} else {
-			ret = btrfs_mksubvol(&file->f_path, name, namelen,
+			ret = btrfs_mksnapshot(&file->f_path, name, namelen,
 					     BTRFS_I(src_inode)->root,
 					     readonly, inherit);
 		}
@@ -2127,10 +2139,7 @@ static noinline int search_ioctl(struct inode *inode,
 		/* search the root of the inode that was passed */
 		root = btrfs_grab_root(BTRFS_I(inode)->root);
 	} else {
-		key.objectid = sk->tree_id;
-		key.type = BTRFS_ROOT_ITEM_KEY;
-		key.offset = (u64)-1;
-		root = btrfs_get_fs_root(info, &key, true);
+		root = btrfs_get_fs_root(info, sk->tree_id, true);
 		if (IS_ERR(root)) {
 			btrfs_free_path(path);
 			return PTR_ERR(root);
@@ -2263,10 +2272,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 
 	ptr = &name[BTRFS_INO_LOOKUP_PATH_MAX - 1];
 
-	key.objectid = tree_id;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-	root = btrfs_get_fs_root(info, &key, true);
+	root = btrfs_get_fs_root(info, tree_id, true);
 	if (IS_ERR(root)) {
 		ret = PTR_ERR(root);
 		root = NULL;
@@ -2359,10 +2365,7 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 	if (dirid != upper_limit.objectid) {
 		ptr = &args->path[BTRFS_INO_LOOKUP_USER_PATH_MAX - 1];
 
-		key.objectid = treeid;
-		key.type = BTRFS_ROOT_ITEM_KEY;
-		key.offset = (u64)-1;
-		root = btrfs_get_fs_root(fs_info, &key, true);
+		root = btrfs_get_fs_root(fs_info, treeid, true);
 		if (IS_ERR(root)) {
 			ret = PTR_ERR(root);
 			goto out;
@@ -2421,7 +2424,7 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 				goto out_put;
 			}
 
-			temp_inode = btrfs_iget(sb, &key2, root);
+			temp_inode = btrfs_iget(sb, key2.objectid, root);
 			if (IS_ERR(temp_inode)) {
 				ret = PTR_ERR(temp_inode);
 				goto out_put;
@@ -2608,9 +2611,7 @@ static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp)
 
 	/* Get root_item of inode's subvolume */
 	key.objectid = BTRFS_I(inode)->root->root_key.objectid;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-	root = btrfs_get_fs_root(fs_info, &key, true);
+	root = btrfs_get_fs_root(fs_info, key.objectid, true);
 	if (IS_ERR(root)) {
 		ret = PTR_ERR(root);
 		goto out_free;
@@ -3278,7 +3279,6 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
 	struct btrfs_dir_item *di;
 	struct btrfs_trans_handle *trans;
 	struct btrfs_path *path = NULL;
-	struct btrfs_key location;
 	struct btrfs_disk_key disk_key;
 	u64 objectid = 0;
 	u64 dir_id;
@@ -3299,11 +3299,7 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
 	if (!objectid)
 		objectid = BTRFS_FS_TREE_OBJECTID;
 
-	location.objectid = objectid;
-	location.type = BTRFS_ROOT_ITEM_KEY;
-	location.offset = (u64)-1;
-
-	new_root = btrfs_get_fs_root(fs_info, &location, true);
+	new_root = btrfs_get_fs_root(fs_info, objectid, true);
 	if (IS_ERR(new_root)) {
 		ret = PTR_ERR(new_root);
 		goto out;
diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c
index fb647d8cf527..f75612e18a82 100644
--- a/fs/btrfs/locking.c
+++ b/fs/btrfs/locking.c
@@ -410,6 +410,7 @@ 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)
+	__acquires(&eb->lock)
 {
 	u64 start_ns = 0;
 
diff --git a/fs/btrfs/misc.h b/fs/btrfs/misc.h
index 72bab64ecf60..6461ebc3a1c1 100644
--- a/fs/btrfs/misc.h
+++ b/fs/btrfs/misc.h
@@ -6,6 +6,7 @@
 #include <linux/sched.h>
 #include <linux/wait.h>
 #include <asm/div64.h>
+#include <linux/rbtree.h>
 
 #define in_range(b, first, len) ((b) >= (first) && (b) < (first) + (len))
 
@@ -58,4 +59,57 @@ static inline bool has_single_bit_set(u64 n)
 	return is_power_of_two_u64(n);
 }
 
+/*
+ * Simple bytenr based rb_tree relate structures
+ *
+ * Any structure wants to use bytenr as single search index should have their
+ * structure start with these members.
+ */
+struct rb_simple_node {
+	struct rb_node rb_node;
+	u64 bytenr;
+};
+
+static inline struct rb_node *rb_simple_search(struct rb_root *root, u64 bytenr)
+{
+	struct rb_node *node = root->rb_node;
+	struct rb_simple_node *entry;
+
+	while (node) {
+		entry = rb_entry(node, struct rb_simple_node, rb_node);
+
+		if (bytenr < entry->bytenr)
+			node = node->rb_left;
+		else if (bytenr > entry->bytenr)
+			node = node->rb_right;
+		else
+			return node;
+	}
+	return NULL;
+}
+
+static inline struct rb_node *rb_simple_insert(struct rb_root *root, u64 bytenr,
+					       struct rb_node *node)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct rb_simple_node *entry;
+
+	while (*p) {
+		parent = *p;
+		entry = rb_entry(parent, struct rb_simple_node, rb_node);
+
+		if (bytenr < entry->bytenr)
+			p = &(*p)->rb_left;
+		else if (bytenr > entry->bytenr)
+			p = &(*p)->rb_right;
+		else
+			return parent;
+	}
+
+	rb_link_node(node, parent, p);
+	rb_insert_color(node, root);
+	return NULL;
+}
+
 #endif
diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
index ff1ff90e48b1..2dcb1cb21634 100644
--- a/fs/btrfs/props.c
+++ b/fs/btrfs/props.c
@@ -408,19 +408,14 @@ int btrfs_subvol_inherit_props(struct btrfs_trans_handle *trans,
 			       struct btrfs_root *parent_root)
 {
 	struct super_block *sb = root->fs_info->sb;
-	struct btrfs_key key;
 	struct inode *parent_inode, *child_inode;
 	int ret;
 
-	key.objectid = BTRFS_FIRST_FREE_OBJECTID;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-
-	parent_inode = btrfs_iget(sb, &key, parent_root);
+	parent_inode = btrfs_iget(sb, BTRFS_FIRST_FREE_OBJECTID, parent_root);
 	if (IS_ERR(parent_inode))
 		return PTR_ERR(parent_inode);
 
-	child_inode = btrfs_iget(sb, &key, root);
+	child_inode = btrfs_iget(sb, BTRFS_FIRST_FREE_OBJECTID, root);
 	if (IS_ERR(child_inode)) {
 		iput(parent_inode);
 		return PTR_ERR(child_inode);
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index c3888fb367e7..5bd4089ad0e1 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -2622,6 +2622,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
 	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *srcgroup;
 	struct btrfs_qgroup *dstgroup;
+	bool need_rescan = false;
 	u32 level_size = 0;
 	u64 nums;
 
@@ -2765,6 +2766,13 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
 				goto unlock;
 		}
 		++i_qgroups;
+
+		/*
+		 * If we're doing a snapshot, and adding the snapshot to a new
+		 * qgroup, the numbers are guaranteed to be incorrect.
+		 */
+		if (srcid)
+			need_rescan = true;
 	}
 
 	for (i = 0; i <  inherit->num_ref_copies; ++i, i_qgroups += 2) {
@@ -2784,6 +2792,9 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
 
 		dst->rfer = src->rfer - level_size;
 		dst->rfer_cmpr = src->rfer_cmpr - level_size;
+
+		/* Manually tweaking numbers certainly needs a rescan */
+		need_rescan = true;
 	}
 	for (i = 0; i <  inherit->num_excl_copies; ++i, i_qgroups += 2) {
 		struct btrfs_qgroup *src;
@@ -2802,6 +2813,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
 
 		dst->excl = src->excl + level_size;
 		dst->excl_cmpr = src->excl_cmpr + level_size;
+		need_rescan = true;
 	}
 
 unlock:
@@ -2809,6 +2821,8 @@ unlock:
 out:
 	if (!committing)
 		mutex_unlock(&fs_info->qgroup_ioctl_lock);
+	if (need_rescan)
+		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
 	return ret;
 }
 
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 03bc7134e8cb..3bbae80c752f 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -24,6 +24,7 @@
 #include "delalloc-space.h"
 #include "block-group.h"
 #include "backref.h"
+#include "misc.h"
 
 /*
  * Relocation overview
@@ -72,100 +73,15 @@
  * The entry point of relocation is relocate_block_group() function.
  */
 
-/*
- * backref_node, mapping_node and tree_block start with this
- */
-struct tree_entry {
-	struct rb_node rb_node;
-	u64 bytenr;
-};
-
-/*
- * present a tree block in the backref cache
- */
-struct backref_node {
-	struct rb_node rb_node;
-	u64 bytenr;
-
-	u64 new_bytenr;
-	/* objectid of tree block owner, can be not uptodate */
-	u64 owner;
-	/* link to pending, changed or detached list */
-	struct list_head list;
-	/* list of upper level blocks reference this block */
-	struct list_head upper;
-	/* list of child blocks in the cache */
-	struct list_head lower;
-	/* NULL if this node is not tree root */
-	struct btrfs_root *root;
-	/* extent buffer got by COW the block */
-	struct extent_buffer *eb;
-	/* level of tree block */
-	unsigned int level:8;
-	/* is the block in non-reference counted tree */
-	unsigned int cowonly:1;
-	/* 1 if no child node in the cache */
-	unsigned int lowest:1;
-	/* is the extent buffer locked */
-	unsigned int locked:1;
-	/* has the block been processed */
-	unsigned int processed:1;
-	/* have backrefs of this block been checked */
-	unsigned int checked:1;
-	/*
-	 * 1 if corresponding block has been cowed but some upper
-	 * level block pointers may not point to the new location
-	 */
-	unsigned int pending:1;
-	/*
-	 * 1 if the backref node isn't connected to any other
-	 * backref node.
-	 */
-	unsigned int detached:1;
-};
-
-/*
- * present a block pointer in the backref cache
- */
-struct backref_edge {
-	struct list_head list[2];
-	struct backref_node *node[2];
-};
-
-#define LOWER	0
-#define UPPER	1
 #define RELOCATION_RESERVED_NODES	256
-
-struct backref_cache {
-	/* red black tree of all backref nodes in the cache */
-	struct rb_root rb_root;
-	/* for passing backref nodes to btrfs_reloc_cow_block */
-	struct backref_node *path[BTRFS_MAX_LEVEL];
-	/*
-	 * list of blocks that have been cowed but some block
-	 * pointers in upper level blocks may not reflect the
-	 * new location
-	 */
-	struct list_head pending[BTRFS_MAX_LEVEL];
-	/* list of backref nodes with no child node */
-	struct list_head leaves;
-	/* list of blocks that have been cowed in current transaction */
-	struct list_head changed;
-	/* list of detached backref node. */
-	struct list_head detached;
-
-	u64 last_trans;
-
-	int nr_nodes;
-	int nr_edges;
-};
-
 /*
  * map address of tree root to tree
  */
 struct mapping_node {
-	struct rb_node rb_node;
-	u64 bytenr;
+	struct {
+		struct rb_node rb_node;
+		u64 bytenr;
+	}; /* Use rb_simle_node for search/insert */
 	void *data;
 };
 
@@ -178,8 +94,10 @@ struct mapping_tree {
  * present a tree block to process
  */
 struct tree_block {
-	struct rb_node rb_node;
-	u64 bytenr;
+	struct {
+		struct rb_node rb_node;
+		u64 bytenr;
+	}; /* Use rb_simple_node for search/insert */
 	struct btrfs_key key;
 	unsigned int level:8;
 	unsigned int key_ready:1;
@@ -204,7 +122,7 @@ struct reloc_control {
 
 	struct btrfs_block_rsv *block_rsv;
 
-	struct backref_cache backref_cache;
+	struct btrfs_backref_cache backref_cache;
 
 	struct file_extent_cluster cluster;
 	/* tree blocks have been processed */
@@ -235,168 +153,41 @@ struct reloc_control {
 #define MOVE_DATA_EXTENTS	0
 #define UPDATE_DATA_PTRS	1
 
-static void remove_backref_node(struct backref_cache *cache,
-				struct backref_node *node);
-static void __mark_block_processed(struct reloc_control *rc,
-				   struct backref_node *node);
-
-static void mapping_tree_init(struct mapping_tree *tree)
-{
-	tree->rb_root = RB_ROOT;
-	spin_lock_init(&tree->lock);
-}
-
-static void backref_cache_init(struct backref_cache *cache)
-{
-	int i;
-	cache->rb_root = RB_ROOT;
-	for (i = 0; i < BTRFS_MAX_LEVEL; i++)
-		INIT_LIST_HEAD(&cache->pending[i]);
-	INIT_LIST_HEAD(&cache->changed);
-	INIT_LIST_HEAD(&cache->detached);
-	INIT_LIST_HEAD(&cache->leaves);
-}
-
-static void backref_cache_cleanup(struct backref_cache *cache)
-{
-	struct backref_node *node;
-	int i;
-
-	while (!list_empty(&cache->detached)) {
-		node = list_entry(cache->detached.next,
-				  struct backref_node, list);
-		remove_backref_node(cache, node);
-	}
-
-	while (!list_empty(&cache->leaves)) {
-		node = list_entry(cache->leaves.next,
-				  struct backref_node, lower);
-		remove_backref_node(cache, node);
-	}
-
-	cache->last_trans = 0;
-
-	for (i = 0; i < BTRFS_MAX_LEVEL; i++)
-		ASSERT(list_empty(&cache->pending[i]));
-	ASSERT(list_empty(&cache->changed));
-	ASSERT(list_empty(&cache->detached));
-	ASSERT(RB_EMPTY_ROOT(&cache->rb_root));
-	ASSERT(!cache->nr_nodes);
-	ASSERT(!cache->nr_edges);
-}
-
-static struct backref_node *alloc_backref_node(struct backref_cache *cache)
-{
-	struct backref_node *node;
-
-	node = kzalloc(sizeof(*node), GFP_NOFS);
-	if (node) {
-		INIT_LIST_HEAD(&node->list);
-		INIT_LIST_HEAD(&node->upper);
-		INIT_LIST_HEAD(&node->lower);
-		RB_CLEAR_NODE(&node->rb_node);
-		cache->nr_nodes++;
-	}
-	return node;
-}
-
-static void free_backref_node(struct backref_cache *cache,
-			      struct backref_node *node)
-{
-	if (node) {
-		cache->nr_nodes--;
-		btrfs_put_root(node->root);
-		kfree(node);
-	}
-}
-
-static struct backref_edge *alloc_backref_edge(struct backref_cache *cache)
-{
-	struct backref_edge *edge;
-
-	edge = kzalloc(sizeof(*edge), GFP_NOFS);
-	if (edge)
-		cache->nr_edges++;
-	return edge;
-}
-
-static void free_backref_edge(struct backref_cache *cache,
-			      struct backref_edge *edge)
+static void mark_block_processed(struct reloc_control *rc,
+				 struct btrfs_backref_node *node)
 {
-	if (edge) {
-		cache->nr_edges--;
-		kfree(edge);
-	}
-}
+	u32 blocksize;
 
-static struct rb_node *tree_insert(struct rb_root *root, u64 bytenr,
-				   struct rb_node *node)
-{
-	struct rb_node **p = &root->rb_node;
-	struct rb_node *parent = NULL;
-	struct tree_entry *entry;
-
-	while (*p) {
-		parent = *p;
-		entry = rb_entry(parent, struct tree_entry, rb_node);
-
-		if (bytenr < entry->bytenr)
-			p = &(*p)->rb_left;
-		else if (bytenr > entry->bytenr)
-			p = &(*p)->rb_right;
-		else
-			return parent;
+	if (node->level == 0 ||
+	    in_range(node->bytenr, rc->block_group->start,
+		     rc->block_group->length)) {
+		blocksize = rc->extent_root->fs_info->nodesize;
+		set_extent_bits(&rc->processed_blocks, node->bytenr,
+				node->bytenr + blocksize - 1, EXTENT_DIRTY);
 	}
-
-	rb_link_node(node, parent, p);
-	rb_insert_color(node, root);
-	return NULL;
+	node->processed = 1;
 }
 
-static struct rb_node *tree_search(struct rb_root *root, u64 bytenr)
-{
-	struct rb_node *n = root->rb_node;
-	struct tree_entry *entry;
-
-	while (n) {
-		entry = rb_entry(n, struct tree_entry, rb_node);
 
-		if (bytenr < entry->bytenr)
-			n = n->rb_left;
-		else if (bytenr > entry->bytenr)
-			n = n->rb_right;
-		else
-			return n;
-	}
-	return NULL;
-}
-
-static void backref_tree_panic(struct rb_node *rb_node, int errno, u64 bytenr)
+static void mapping_tree_init(struct mapping_tree *tree)
 {
-
-	struct btrfs_fs_info *fs_info = NULL;
-	struct backref_node *bnode = rb_entry(rb_node, struct backref_node,
-					      rb_node);
-	if (bnode->root)
-		fs_info = bnode->root->fs_info;
-	btrfs_panic(fs_info, errno,
-		    "Inconsistency in backref cache found at offset %llu",
-		    bytenr);
+	tree->rb_root = RB_ROOT;
+	spin_lock_init(&tree->lock);
 }
 
 /*
  * walk up backref nodes until reach node presents tree root
  */
-static struct backref_node *walk_up_backref(struct backref_node *node,
-					    struct backref_edge *edges[],
-					    int *index)
+static struct btrfs_backref_node *walk_up_backref(
+		struct btrfs_backref_node *node,
+		struct btrfs_backref_edge *edges[], int *index)
 {
-	struct backref_edge *edge;
+	struct btrfs_backref_edge *edge;
 	int idx = *index;
 
 	while (!list_empty(&node->upper)) {
 		edge = list_entry(node->upper.next,
-				  struct backref_edge, list[LOWER]);
+				  struct btrfs_backref_edge, list[LOWER]);
 		edges[idx++] = edge;
 		node = edge->node[UPPER];
 	}
@@ -408,11 +199,11 @@ static struct backref_node *walk_up_backref(struct backref_node *node,
 /*
  * walk down backref nodes to find start of next reference path
  */
-static struct backref_node *walk_down_backref(struct backref_edge *edges[],
-					      int *index)
+static struct btrfs_backref_node *walk_down_backref(
+		struct btrfs_backref_edge *edges[], int *index)
 {
-	struct backref_edge *edge;
-	struct backref_node *lower;
+	struct btrfs_backref_edge *edge;
+	struct btrfs_backref_node *lower;
 	int idx = *index;
 
 	while (idx > 0) {
@@ -423,7 +214,7 @@ static struct backref_node *walk_down_backref(struct backref_edge *edges[],
 			continue;
 		}
 		edge = list_entry(edge->list[LOWER].next,
-				  struct backref_edge, list[LOWER]);
+				  struct btrfs_backref_edge, list[LOWER]);
 		edges[idx - 1] = edge;
 		*index = idx;
 		return edge->node[UPPER];
@@ -432,95 +223,24 @@ static struct backref_node *walk_down_backref(struct backref_edge *edges[],
 	return NULL;
 }
 
-static void unlock_node_buffer(struct backref_node *node)
-{
-	if (node->locked) {
-		btrfs_tree_unlock(node->eb);
-		node->locked = 0;
-	}
-}
-
-static void drop_node_buffer(struct backref_node *node)
-{
-	if (node->eb) {
-		unlock_node_buffer(node);
-		free_extent_buffer(node->eb);
-		node->eb = NULL;
-	}
-}
-
-static void drop_backref_node(struct backref_cache *tree,
-			      struct backref_node *node)
-{
-	BUG_ON(!list_empty(&node->upper));
-
-	drop_node_buffer(node);
-	list_del(&node->list);
-	list_del(&node->lower);
-	if (!RB_EMPTY_NODE(&node->rb_node))
-		rb_erase(&node->rb_node, &tree->rb_root);
-	free_backref_node(tree, node);
-}
-
-/*
- * remove a backref node from the backref cache
- */
-static void remove_backref_node(struct backref_cache *cache,
-				struct backref_node *node)
-{
-	struct backref_node *upper;
-	struct backref_edge *edge;
-
-	if (!node)
-		return;
-
-	BUG_ON(!node->lowest && !node->detached);
-	while (!list_empty(&node->upper)) {
-		edge = list_entry(node->upper.next, struct backref_edge,
-				  list[LOWER]);
-		upper = edge->node[UPPER];
-		list_del(&edge->list[LOWER]);
-		list_del(&edge->list[UPPER]);
-		free_backref_edge(cache, edge);
-
-		if (RB_EMPTY_NODE(&upper->rb_node)) {
-			BUG_ON(!list_empty(&node->upper));
-			drop_backref_node(cache, node);
-			node = upper;
-			node->lowest = 1;
-			continue;
-		}
-		/*
-		 * add the node to leaf node list if no other
-		 * child block cached.
-		 */
-		if (list_empty(&upper->lower)) {
-			list_add_tail(&upper->lower, &cache->leaves);
-			upper->lowest = 1;
-		}
-	}
-
-	drop_backref_node(cache, node);
-}
-
-static void update_backref_node(struct backref_cache *cache,
-				struct backref_node *node, u64 bytenr)
+static void update_backref_node(struct btrfs_backref_cache *cache,
+				struct btrfs_backref_node *node, u64 bytenr)
 {
 	struct rb_node *rb_node;
 	rb_erase(&node->rb_node, &cache->rb_root);
 	node->bytenr = bytenr;
-	rb_node = tree_insert(&cache->rb_root, node->bytenr, &node->rb_node);
+	rb_node = rb_simple_insert(&cache->rb_root, node->bytenr, &node->rb_node);
 	if (rb_node)
-		backref_tree_panic(rb_node, -EEXIST, bytenr);
+		btrfs_backref_panic(cache->fs_info, bytenr, -EEXIST);
 }
 
 /*
  * update backref cache after a transaction commit
  */
 static int update_backref_cache(struct btrfs_trans_handle *trans,
-				struct backref_cache *cache)
+				struct btrfs_backref_cache *cache)
 {
-	struct backref_node *node;
+	struct btrfs_backref_node *node;
 	int level = 0;
 
 	if (cache->last_trans == 0) {
@@ -538,13 +258,13 @@ static int update_backref_cache(struct btrfs_trans_handle *trans,
 	 */
 	while (!list_empty(&cache->detached)) {
 		node = list_entry(cache->detached.next,
-				  struct backref_node, list);
-		remove_backref_node(cache, node);
+				  struct btrfs_backref_node, list);
+		btrfs_backref_cleanup_node(cache, node);
 	}
 
 	while (!list_empty(&cache->changed)) {
 		node = list_entry(cache->changed.next,
-				  struct backref_node, list);
+				  struct btrfs_backref_node, list);
 		list_del_init(&node->list);
 		BUG_ON(node->pending);
 		update_backref_node(cache, node, node->new_bytenr);
@@ -585,7 +305,8 @@ static bool reloc_root_is_dead(struct btrfs_root *root)
  *
  * Reloc tree after swap is considered dead, thus not considered as valid.
  * This is enough for most callers, as they don't distinguish dead reloc root
- * from no reloc root.  But should_ignore_root() below is a special case.
+ * from no reloc root.  But btrfs_should_ignore_reloc_root() below is a
+ * special case.
  */
 static bool have_reloc_root(struct btrfs_root *root)
 {
@@ -596,11 +317,11 @@ static bool have_reloc_root(struct btrfs_root *root)
 	return true;
 }
 
-static int should_ignore_root(struct btrfs_root *root)
+int btrfs_should_ignore_reloc_root(struct btrfs_root *root)
 {
 	struct btrfs_root *reloc_root;
 
-	if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
+	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
 		return 0;
 
 	/* This root has been merged with its reloc tree, we can ignore it */
@@ -622,18 +343,20 @@ static int should_ignore_root(struct btrfs_root *root)
 	 */
 	return 1;
 }
+
 /*
  * find reloc tree by address of tree root
  */
-static struct btrfs_root *find_reloc_root(struct reloc_control *rc,
-					  u64 bytenr)
+struct btrfs_root *find_reloc_root(struct btrfs_fs_info *fs_info, u64 bytenr)
 {
+	struct reloc_control *rc = fs_info->reloc_ctl;
 	struct rb_node *rb_node;
 	struct mapping_node *node;
 	struct btrfs_root *root = NULL;
 
+	ASSERT(rc);
 	spin_lock(&rc->reloc_root_tree.lock);
-	rb_node = tree_search(&rc->reloc_root_tree.rb_root, bytenr);
+	rb_node = rb_simple_search(&rc->reloc_root_tree.rb_root, bytenr);
 	if (rb_node) {
 		node = rb_entry(rb_node, struct mapping_node, rb_node);
 		root = (struct btrfs_root *)node->data;
@@ -642,594 +365,165 @@ static struct btrfs_root *find_reloc_root(struct reloc_control *rc,
 	return btrfs_grab_root(root);
 }
 
-static struct btrfs_root *read_fs_root(struct btrfs_fs_info *fs_info,
-					u64 root_objectid)
+/*
+ * For useless nodes, do two major clean ups:
+ *
+ * - Cleanup the children edges and nodes
+ *   If child node is also orphan (no parent) during cleanup, then the child
+ *   node will also be cleaned up.
+ *
+ * - Freeing up leaves (level 0), keeps nodes detached
+ *   For nodes, the node is still cached as "detached"
+ *
+ * Return false if @node is not in the @useless_nodes list.
+ * Return true if @node is in the @useless_nodes list.
+ */
+static bool handle_useless_nodes(struct reloc_control *rc,
+				 struct btrfs_backref_node *node)
 {
-	struct btrfs_key key;
+	struct btrfs_backref_cache *cache = &rc->backref_cache;
+	struct list_head *useless_node = &cache->useless_node;
+	bool ret = false;
 
-	key.objectid = root_objectid;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
+	while (!list_empty(useless_node)) {
+		struct btrfs_backref_node *cur;
 
-	return btrfs_get_fs_root(fs_info, &key, false);
-}
+		cur = list_first_entry(useless_node, struct btrfs_backref_node,
+				 list);
+		list_del_init(&cur->list);
 
-static noinline_for_stack
-int find_inline_backref(struct extent_buffer *leaf, int slot,
-			unsigned long *ptr, unsigned long *end)
-{
-	struct btrfs_key key;
-	struct btrfs_extent_item *ei;
-	struct btrfs_tree_block_info *bi;
-	u32 item_size;
+		/* Only tree root nodes can be added to @useless_nodes */
+		ASSERT(list_empty(&cur->upper));
 
-	btrfs_item_key_to_cpu(leaf, &key, slot);
+		if (cur == node)
+			ret = true;
 
-	item_size = btrfs_item_size_nr(leaf, slot);
-	if (item_size < sizeof(*ei)) {
-		btrfs_print_v0_err(leaf->fs_info);
-		btrfs_handle_fs_error(leaf->fs_info, -EINVAL, NULL);
-		return 1;
-	}
-	ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
-	WARN_ON(!(btrfs_extent_flags(leaf, ei) &
-		  BTRFS_EXTENT_FLAG_TREE_BLOCK));
+		/* The node is the lowest node */
+		if (cur->lowest) {
+			list_del_init(&cur->lower);
+			cur->lowest = 0;
+		}
 
-	if (key.type == BTRFS_EXTENT_ITEM_KEY &&
-	    item_size <= sizeof(*ei) + sizeof(*bi)) {
-		WARN_ON(item_size < sizeof(*ei) + sizeof(*bi));
-		return 1;
-	}
-	if (key.type == BTRFS_METADATA_ITEM_KEY &&
-	    item_size <= sizeof(*ei)) {
-		WARN_ON(item_size < sizeof(*ei));
-		return 1;
-	}
+		/* Cleanup the lower edges */
+		while (!list_empty(&cur->lower)) {
+			struct btrfs_backref_edge *edge;
+			struct btrfs_backref_node *lower;
 
-	if (key.type == BTRFS_EXTENT_ITEM_KEY) {
-		bi = (struct btrfs_tree_block_info *)(ei + 1);
-		*ptr = (unsigned long)(bi + 1);
-	} else {
-		*ptr = (unsigned long)(ei + 1);
+			edge = list_entry(cur->lower.next,
+					struct btrfs_backref_edge, list[UPPER]);
+			list_del(&edge->list[UPPER]);
+			list_del(&edge->list[LOWER]);
+			lower = edge->node[LOWER];
+			btrfs_backref_free_edge(cache, edge);
+
+			/* Child node is also orphan, queue for cleanup */
+			if (list_empty(&lower->upper))
+				list_add(&lower->list, useless_node);
+		}
+		/* Mark this block processed for relocation */
+		mark_block_processed(rc, cur);
+
+		/*
+		 * Backref nodes for tree leaves are deleted from the cache.
+		 * Backref nodes for upper level tree blocks are left in the
+		 * cache to avoid unnecessary backref lookup.
+		 */
+		if (cur->level > 0) {
+			list_add(&cur->list, &cache->detached);
+			cur->detached = 1;
+		} else {
+			rb_erase(&cur->rb_node, &cache->rb_root);
+			btrfs_backref_free_node(cache, cur);
+		}
 	}
-	*end = (unsigned long)ei + item_size;
-	return 0;
+	return ret;
 }
 
 /*
- * build backref tree for a given tree block. root of the backref tree
- * corresponds the tree block, leaves of the backref tree correspond
- * roots of b-trees that reference the tree block.
+ * Build backref tree for a given tree block. Root of the backref tree
+ * corresponds the tree block, leaves of the backref tree correspond roots of
+ * b-trees that reference the tree block.
  *
- * the basic idea of this function is check backrefs of a given block
- * to find upper level blocks that reference the block, and then check
- * backrefs of these upper level blocks recursively. the recursion stop
- * when tree root is reached or backrefs for the block is cached.
+ * The basic idea of this function is check backrefs of a given block to find
+ * upper level blocks that reference the block, and then check backrefs of
+ * these upper level blocks recursively. The recursion stops when tree root is
+ * reached or backrefs for the block is cached.
  *
- * NOTE: if we find backrefs for a block are cached, we know backrefs
- * for all upper level blocks that directly/indirectly reference the
- * block are also cached.
+ * NOTE: if we find that backrefs for a block are cached, we know backrefs for
+ * all upper level blocks that directly/indirectly reference the block are also
+ * cached.
  */
-static noinline_for_stack
-struct backref_node *build_backref_tree(struct reloc_control *rc,
-					struct btrfs_key *node_key,
-					int level, u64 bytenr)
+static noinline_for_stack struct btrfs_backref_node *build_backref_tree(
+			struct reloc_control *rc, struct btrfs_key *node_key,
+			int level, u64 bytenr)
 {
-	struct backref_cache *cache = &rc->backref_cache;
-	struct btrfs_path *path1; /* For searching extent root */
-	struct btrfs_path *path2; /* For searching parent of TREE_BLOCK_REF */
-	struct extent_buffer *eb;
-	struct btrfs_root *root;
-	struct backref_node *cur;
-	struct backref_node *upper;
-	struct backref_node *lower;
-	struct backref_node *node = NULL;
-	struct backref_node *exist = NULL;
-	struct backref_edge *edge;
-	struct rb_node *rb_node;
-	struct btrfs_key key;
-	unsigned long end;
-	unsigned long ptr;
-	LIST_HEAD(list); /* Pending edge list, upper node needs to be checked */
-	LIST_HEAD(useless);
-	int cowonly;
+	struct btrfs_backref_iter *iter;
+	struct btrfs_backref_cache *cache = &rc->backref_cache;
+	/* For searching parent of TREE_BLOCK_REF */
+	struct btrfs_path *path;
+	struct btrfs_backref_node *cur;
+	struct btrfs_backref_node *node = NULL;
+	struct btrfs_backref_edge *edge;
 	int ret;
 	int err = 0;
-	bool need_check = true;
 
-	path1 = btrfs_alloc_path();
-	path2 = btrfs_alloc_path();
-	if (!path1 || !path2) {
+	iter = btrfs_backref_iter_alloc(rc->extent_root->fs_info, GFP_NOFS);
+	if (!iter)
+		return ERR_PTR(-ENOMEM);
+	path = btrfs_alloc_path();
+	if (!path) {
 		err = -ENOMEM;
 		goto out;
 	}
 
-	node = alloc_backref_node(cache);
+	node = btrfs_backref_alloc_node(cache, bytenr, level);
 	if (!node) {
 		err = -ENOMEM;
 		goto out;
 	}
 
-	node->bytenr = bytenr;
-	node->level = level;
 	node->lowest = 1;
 	cur = node;
-again:
-	end = 0;
-	ptr = 0;
-	key.objectid = cur->bytenr;
-	key.type = BTRFS_METADATA_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	path1->search_commit_root = 1;
-	path1->skip_locking = 1;
-	ret = btrfs_search_slot(NULL, rc->extent_root, &key, path1,
-				0, 0);
-	if (ret < 0) {
-		err = ret;
-		goto out;
-	}
-	ASSERT(ret);
-	ASSERT(path1->slots[0]);
-
-	path1->slots[0]--;
 
-	WARN_ON(cur->checked);
-	if (!list_empty(&cur->upper)) {
-		/*
-		 * the backref was added previously when processing
-		 * backref of type BTRFS_TREE_BLOCK_REF_KEY
-		 */
-		ASSERT(list_is_singular(&cur->upper));
-		edge = list_entry(cur->upper.next, struct backref_edge,
-				  list[LOWER]);
-		ASSERT(list_empty(&edge->list[UPPER]));
-		exist = edge->node[UPPER];
-		/*
-		 * add the upper level block to pending list if we need
-		 * check its backrefs
-		 */
-		if (!exist->checked)
-			list_add_tail(&edge->list[UPPER], &list);
-	} else {
-		exist = NULL;
-	}
-
-	while (1) {
-		cond_resched();
-		eb = path1->nodes[0];
-
-		if (ptr >= end) {
-			if (path1->slots[0] >= btrfs_header_nritems(eb)) {
-				ret = btrfs_next_leaf(rc->extent_root, path1);
-				if (ret < 0) {
-					err = ret;
-					goto out;
-				}
-				if (ret > 0)
-					break;
-				eb = path1->nodes[0];
-			}
-
-			btrfs_item_key_to_cpu(eb, &key, path1->slots[0]);
-			if (key.objectid != cur->bytenr) {
-				WARN_ON(exist);
-				break;
-			}
-
-			if (key.type == BTRFS_EXTENT_ITEM_KEY ||
-			    key.type == BTRFS_METADATA_ITEM_KEY) {
-				ret = find_inline_backref(eb, path1->slots[0],
-							  &ptr, &end);
-				if (ret)
-					goto next;
-			}
-		}
-
-		if (ptr < end) {
-			/* update key for inline back ref */
-			struct btrfs_extent_inline_ref *iref;
-			int type;
-			iref = (struct btrfs_extent_inline_ref *)ptr;
-			type = btrfs_get_extent_inline_ref_type(eb, iref,
-							BTRFS_REF_TYPE_BLOCK);
-			if (type == BTRFS_REF_TYPE_INVALID) {
-				err = -EUCLEAN;
-				goto out;
-			}
-			key.type = type;
-			key.offset = btrfs_extent_inline_ref_offset(eb, iref);
-
-			WARN_ON(key.type != BTRFS_TREE_BLOCK_REF_KEY &&
-				key.type != BTRFS_SHARED_BLOCK_REF_KEY);
-		}
-
-		/*
-		 * Parent node found and matches current inline ref, no need to
-		 * rebuild this node for this inline ref.
-		 */
-		if (exist &&
-		    ((key.type == BTRFS_TREE_BLOCK_REF_KEY &&
-		      exist->owner == key.offset) ||
-		     (key.type == BTRFS_SHARED_BLOCK_REF_KEY &&
-		      exist->bytenr == key.offset))) {
-			exist = NULL;
-			goto next;
-		}
-
-		/* SHARED_BLOCK_REF means key.offset is the parent bytenr */
-		if (key.type == BTRFS_SHARED_BLOCK_REF_KEY) {
-			if (key.objectid == key.offset) {
-				/*
-				 * Only root blocks of reloc trees use backref
-				 * pointing to itself.
-				 */
-				root = find_reloc_root(rc, cur->bytenr);
-				ASSERT(root);
-				cur->root = root;
-				break;
-			}
-
-			edge = alloc_backref_edge(cache);
-			if (!edge) {
-				err = -ENOMEM;
-				goto out;
-			}
-			rb_node = tree_search(&cache->rb_root, key.offset);
-			if (!rb_node) {
-				upper = alloc_backref_node(cache);
-				if (!upper) {
-					free_backref_edge(cache, edge);
-					err = -ENOMEM;
-					goto out;
-				}
-				upper->bytenr = key.offset;
-				upper->level = cur->level + 1;
-				/*
-				 *  backrefs for the upper level block isn't
-				 *  cached, add the block to pending list
-				 */
-				list_add_tail(&edge->list[UPPER], &list);
-			} else {
-				upper = rb_entry(rb_node, struct backref_node,
-						 rb_node);
-				ASSERT(upper->checked);
-				INIT_LIST_HEAD(&edge->list[UPPER]);
-			}
-			list_add_tail(&edge->list[LOWER], &cur->upper);
-			edge->node[LOWER] = cur;
-			edge->node[UPPER] = upper;
-
-			goto next;
-		} else if (unlikely(key.type == BTRFS_EXTENT_REF_V0_KEY)) {
-			err = -EINVAL;
-			btrfs_print_v0_err(rc->extent_root->fs_info);
-			btrfs_handle_fs_error(rc->extent_root->fs_info, err,
-					      NULL);
-			goto out;
-		} else if (key.type != BTRFS_TREE_BLOCK_REF_KEY) {
-			goto next;
-		}
-
-		/*
-		 * key.type == BTRFS_TREE_BLOCK_REF_KEY, inline ref offset
-		 * means the root objectid. We need to search the tree to get
-		 * its parent bytenr.
-		 */
-		root = read_fs_root(rc->extent_root->fs_info, key.offset);
-		if (IS_ERR(root)) {
-			err = PTR_ERR(root);
-			goto out;
-		}
-
-		if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
-			cur->cowonly = 1;
-
-		if (btrfs_root_level(&root->root_item) == cur->level) {
-			/* tree root */
-			ASSERT(btrfs_root_bytenr(&root->root_item) ==
-			       cur->bytenr);
-			if (should_ignore_root(root)) {
-				btrfs_put_root(root);
-				list_add(&cur->list, &useless);
-			} else {
-				cur->root = root;
-			}
-			break;
-		}
-
-		level = cur->level + 1;
-
-		/* Search the tree to find parent blocks referring the block. */
-		path2->search_commit_root = 1;
-		path2->skip_locking = 1;
-		path2->lowest_level = level;
-		ret = btrfs_search_slot(NULL, root, node_key, path2, 0, 0);
-		path2->lowest_level = 0;
+	/* Breadth-first search to build backref cache */
+	do {
+		ret = btrfs_backref_add_tree_node(cache, path, iter, node_key,
+						  cur);
 		if (ret < 0) {
-			btrfs_put_root(root);
 			err = ret;
 			goto out;
 		}
-		if (ret > 0 && path2->slots[level] > 0)
-			path2->slots[level]--;
-
-		eb = path2->nodes[level];
-		if (btrfs_node_blockptr(eb, path2->slots[level]) !=
-		    cur->bytenr) {
-			btrfs_err(root->fs_info,
-	"couldn't find block (%llu) (level %d) in tree (%llu) with key (%llu %u %llu)",
-				  cur->bytenr, level - 1,
-				  root->root_key.objectid,
-				  node_key->objectid, node_key->type,
-				  node_key->offset);
-			btrfs_put_root(root);
-			err = -ENOENT;
-			goto out;
-		}
-		lower = cur;
-		need_check = true;
-
-		/* Add all nodes and edges in the path */
-		for (; level < BTRFS_MAX_LEVEL; level++) {
-			if (!path2->nodes[level]) {
-				ASSERT(btrfs_root_bytenr(&root->root_item) ==
-				       lower->bytenr);
-				if (should_ignore_root(root)) {
-					btrfs_put_root(root);
-					list_add(&lower->list, &useless);
-				} else {
-					lower->root = root;
-				}
-				break;
-			}
-
-			edge = alloc_backref_edge(cache);
-			if (!edge) {
-				btrfs_put_root(root);
-				err = -ENOMEM;
-				goto out;
-			}
-
-			eb = path2->nodes[level];
-			rb_node = tree_search(&cache->rb_root, eb->start);
-			if (!rb_node) {
-				upper = alloc_backref_node(cache);
-				if (!upper) {
-					btrfs_put_root(root);
-					free_backref_edge(cache, edge);
-					err = -ENOMEM;
-					goto out;
-				}
-				upper->bytenr = eb->start;
-				upper->owner = btrfs_header_owner(eb);
-				upper->level = lower->level + 1;
-				if (!test_bit(BTRFS_ROOT_REF_COWS,
-					      &root->state))
-					upper->cowonly = 1;
-
-				/*
-				 * if we know the block isn't shared
-				 * we can void checking its backrefs.
-				 */
-				if (btrfs_block_can_be_shared(root, eb))
-					upper->checked = 0;
-				else
-					upper->checked = 1;
-
-				/*
-				 * add the block to pending list if we
-				 * need check its backrefs, we only do this once
-				 * while walking up a tree as we will catch
-				 * anything else later on.
-				 */
-				if (!upper->checked && need_check) {
-					need_check = false;
-					list_add_tail(&edge->list[UPPER],
-						      &list);
-				} else {
-					if (upper->checked)
-						need_check = true;
-					INIT_LIST_HEAD(&edge->list[UPPER]);
-				}
-			} else {
-				upper = rb_entry(rb_node, struct backref_node,
-						 rb_node);
-				ASSERT(upper->checked);
-				INIT_LIST_HEAD(&edge->list[UPPER]);
-				if (!upper->owner)
-					upper->owner = btrfs_header_owner(eb);
-			}
-			list_add_tail(&edge->list[LOWER], &lower->upper);
-			edge->node[LOWER] = lower;
-			edge->node[UPPER] = upper;
-
-			if (rb_node) {
-				btrfs_put_root(root);
-				break;
-			}
-			lower = upper;
-			upper = NULL;
-		}
-		btrfs_release_path(path2);
-next:
-		if (ptr < end) {
-			ptr += btrfs_extent_inline_ref_size(key.type);
-			if (ptr >= end) {
-				WARN_ON(ptr > end);
-				ptr = 0;
-				end = 0;
-			}
-		}
-		if (ptr >= end)
-			path1->slots[0]++;
-	}
-	btrfs_release_path(path1);
-
-	cur->checked = 1;
-	WARN_ON(exist);
-
-	/* the pending list isn't empty, take the first block to process */
-	if (!list_empty(&list)) {
-		edge = list_entry(list.next, struct backref_edge, list[UPPER]);
-		list_del_init(&edge->list[UPPER]);
-		cur = edge->node[UPPER];
-		goto again;
-	}
-
-	/*
-	 * everything goes well, connect backref nodes and insert backref nodes
-	 * into the cache.
-	 */
-	ASSERT(node->checked);
-	cowonly = node->cowonly;
-	if (!cowonly) {
-		rb_node = tree_insert(&cache->rb_root, node->bytenr,
-				      &node->rb_node);
-		if (rb_node)
-			backref_tree_panic(rb_node, -EEXIST, node->bytenr);
-		list_add_tail(&node->lower, &cache->leaves);
-	}
-
-	list_for_each_entry(edge, &node->upper, list[LOWER])
-		list_add_tail(&edge->list[UPPER], &list);
-
-	while (!list_empty(&list)) {
-		edge = list_entry(list.next, struct backref_edge, list[UPPER]);
-		list_del_init(&edge->list[UPPER]);
-		upper = edge->node[UPPER];
-		if (upper->detached) {
-			list_del(&edge->list[LOWER]);
-			lower = edge->node[LOWER];
-			free_backref_edge(cache, edge);
-			if (list_empty(&lower->upper))
-				list_add(&lower->list, &useless);
-			continue;
-		}
-
-		if (!RB_EMPTY_NODE(&upper->rb_node)) {
-			if (upper->lowest) {
-				list_del_init(&upper->lower);
-				upper->lowest = 0;
-			}
-
-			list_add_tail(&edge->list[UPPER], &upper->lower);
-			continue;
-		}
-
-		if (!upper->checked) {
-			/*
-			 * Still want to blow up for developers since this is a
-			 * logic bug.
-			 */
-			ASSERT(0);
-			err = -EINVAL;
-			goto out;
-		}
-		if (cowonly != upper->cowonly) {
-			ASSERT(0);
-			err = -EINVAL;
-			goto out;
-		}
-
-		if (!cowonly) {
-			rb_node = tree_insert(&cache->rb_root, upper->bytenr,
-					      &upper->rb_node);
-			if (rb_node)
-				backref_tree_panic(rb_node, -EEXIST,
-						   upper->bytenr);
+		edge = list_first_entry_or_null(&cache->pending_edge,
+				struct btrfs_backref_edge, list[UPPER]);
+		/*
+		 * The pending list isn't empty, take the first block to
+		 * process
+		 */
+		if (edge) {
+			list_del_init(&edge->list[UPPER]);
+			cur = edge->node[UPPER];
 		}
+	} while (edge);
 
-		list_add_tail(&edge->list[UPPER], &upper->lower);
-
-		list_for_each_entry(edge, &upper->upper, list[LOWER])
-			list_add_tail(&edge->list[UPPER], &list);
+	/* Finish the upper linkage of newly added edges/nodes */
+	ret = btrfs_backref_finish_upper_links(cache, node);
+	if (ret < 0) {
+		err = ret;
+		goto out;
 	}
-	/*
-	 * process useless backref nodes. backref nodes for tree leaves
-	 * are deleted from the cache. backref nodes for upper level
-	 * tree blocks are left in the cache to avoid unnecessary backref
-	 * lookup.
-	 */
-	while (!list_empty(&useless)) {
-		upper = list_entry(useless.next, struct backref_node, list);
-		list_del_init(&upper->list);
-		ASSERT(list_empty(&upper->upper));
-		if (upper == node)
-			node = NULL;
-		if (upper->lowest) {
-			list_del_init(&upper->lower);
-			upper->lowest = 0;
-		}
-		while (!list_empty(&upper->lower)) {
-			edge = list_entry(upper->lower.next,
-					  struct backref_edge, list[UPPER]);
-			list_del(&edge->list[UPPER]);
-			list_del(&edge->list[LOWER]);
-			lower = edge->node[LOWER];
-			free_backref_edge(cache, edge);
 
-			if (list_empty(&lower->upper))
-				list_add(&lower->list, &useless);
-		}
-		__mark_block_processed(rc, upper);
-		if (upper->level > 0) {
-			list_add(&upper->list, &cache->detached);
-			upper->detached = 1;
-		} else {
-			rb_erase(&upper->rb_node, &cache->rb_root);
-			free_backref_node(cache, upper);
-		}
-	}
+	if (handle_useless_nodes(rc, node))
+		node = NULL;
 out:
-	btrfs_free_path(path1);
-	btrfs_free_path(path2);
+	btrfs_backref_iter_free(iter);
+	btrfs_free_path(path);
 	if (err) {
-		while (!list_empty(&useless)) {
-			lower = list_entry(useless.next,
-					   struct backref_node, list);
-			list_del_init(&lower->list);
-		}
-		while (!list_empty(&list)) {
-			edge = list_first_entry(&list, struct backref_edge,
-						list[UPPER]);
-			list_del(&edge->list[UPPER]);
-			list_del(&edge->list[LOWER]);
-			lower = edge->node[LOWER];
-			upper = edge->node[UPPER];
-			free_backref_edge(cache, edge);
-
-			/*
-			 * Lower is no longer linked to any upper backref nodes
-			 * and isn't in the cache, we can free it ourselves.
-			 */
-			if (list_empty(&lower->upper) &&
-			    RB_EMPTY_NODE(&lower->rb_node))
-				list_add(&lower->list, &useless);
-
-			if (!RB_EMPTY_NODE(&upper->rb_node))
-				continue;
-
-			/* Add this guy's upper edges to the list to process */
-			list_for_each_entry(edge, &upper->upper, list[LOWER])
-				list_add_tail(&edge->list[UPPER], &list);
-			if (list_empty(&upper->upper))
-				list_add(&upper->list, &useless);
-		}
-
-		while (!list_empty(&useless)) {
-			lower = list_entry(useless.next,
-					   struct backref_node, list);
-			list_del_init(&lower->list);
-			if (lower == node)
-				node = NULL;
-			free_backref_node(cache, lower);
-		}
-
-		remove_backref_node(cache, node);
+		btrfs_backref_error_cleanup(cache, node);
 		return ERR_PTR(err);
 	}
 	ASSERT(!node || !node->detached);
+	ASSERT(list_empty(&cache->useless_node) &&
+	       list_empty(&cache->pending_edge));
 	return node;
 }
 
@@ -1244,19 +538,19 @@ static int clone_backref_node(struct btrfs_trans_handle *trans,
 			      struct btrfs_root *dest)
 {
 	struct btrfs_root *reloc_root = src->reloc_root;
-	struct backref_cache *cache = &rc->backref_cache;
-	struct backref_node *node = NULL;
-	struct backref_node *new_node;
-	struct backref_edge *edge;
-	struct backref_edge *new_edge;
+	struct btrfs_backref_cache *cache = &rc->backref_cache;
+	struct btrfs_backref_node *node = NULL;
+	struct btrfs_backref_node *new_node;
+	struct btrfs_backref_edge *edge;
+	struct btrfs_backref_edge *new_edge;
 	struct rb_node *rb_node;
 
 	if (cache->last_trans > 0)
 		update_backref_cache(trans, cache);
 
-	rb_node = tree_search(&cache->rb_root, src->commit_root->start);
+	rb_node = rb_simple_search(&cache->rb_root, src->commit_root->start);
 	if (rb_node) {
-		node = rb_entry(rb_node, struct backref_node, rb_node);
+		node = rb_entry(rb_node, struct btrfs_backref_node, rb_node);
 		if (node->detached)
 			node = NULL;
 		else
@@ -1264,10 +558,10 @@ static int clone_backref_node(struct btrfs_trans_handle *trans,
 	}
 
 	if (!node) {
-		rb_node = tree_search(&cache->rb_root,
-				      reloc_root->commit_root->start);
+		rb_node = rb_simple_search(&cache->rb_root,
+					   reloc_root->commit_root->start);
 		if (rb_node) {
-			node = rb_entry(rb_node, struct backref_node,
+			node = rb_entry(rb_node, struct btrfs_backref_node,
 					rb_node);
 			BUG_ON(node->detached);
 		}
@@ -1276,12 +570,11 @@ static int clone_backref_node(struct btrfs_trans_handle *trans,
 	if (!node)
 		return 0;
 
-	new_node = alloc_backref_node(cache);
+	new_node = btrfs_backref_alloc_node(cache, dest->node->start,
+					    node->level);
 	if (!new_node)
 		return -ENOMEM;
 
-	new_node->bytenr = dest->node->start;
-	new_node->level = node->level;
 	new_node->lowest = node->lowest;
 	new_node->checked = 1;
 	new_node->root = btrfs_grab_root(dest);
@@ -1289,23 +582,21 @@ static int clone_backref_node(struct btrfs_trans_handle *trans,
 
 	if (!node->lowest) {
 		list_for_each_entry(edge, &node->lower, list[UPPER]) {
-			new_edge = alloc_backref_edge(cache);
+			new_edge = btrfs_backref_alloc_edge(cache);
 			if (!new_edge)
 				goto fail;
 
-			new_edge->node[UPPER] = new_node;
-			new_edge->node[LOWER] = edge->node[LOWER];
-			list_add_tail(&new_edge->list[UPPER],
-				      &new_node->lower);
+			btrfs_backref_link_edge(new_edge, edge->node[LOWER],
+						new_node, LINK_UPPER);
 		}
 	} else {
 		list_add_tail(&new_node->lower, &cache->leaves);
 	}
 
-	rb_node = tree_insert(&cache->rb_root, new_node->bytenr,
-			      &new_node->rb_node);
+	rb_node = rb_simple_insert(&cache->rb_root, new_node->bytenr,
+				   &new_node->rb_node);
 	if (rb_node)
-		backref_tree_panic(rb_node, -EEXIST, new_node->bytenr);
+		btrfs_backref_panic(trans->fs_info, new_node->bytenr, -EEXIST);
 
 	if (!new_node->lowest) {
 		list_for_each_entry(new_edge, &new_node->lower, list[UPPER]) {
@@ -1317,11 +608,11 @@ static int clone_backref_node(struct btrfs_trans_handle *trans,
 fail:
 	while (!list_empty(&new_node->lower)) {
 		new_edge = list_entry(new_node->lower.next,
-				      struct backref_edge, list[UPPER]);
+				      struct btrfs_backref_edge, list[UPPER]);
 		list_del(&new_edge->list[UPPER]);
-		free_backref_edge(cache, new_edge);
+		btrfs_backref_free_edge(cache, new_edge);
 	}
-	free_backref_node(cache, new_node);
+	btrfs_backref_free_node(cache, new_node);
 	return -ENOMEM;
 }
 
@@ -1343,8 +634,8 @@ static int __must_check __add_reloc_root(struct btrfs_root *root)
 	node->data = root;
 
 	spin_lock(&rc->reloc_root_tree.lock);
-	rb_node = tree_insert(&rc->reloc_root_tree.rb_root,
-			      node->bytenr, &node->rb_node);
+	rb_node = rb_simple_insert(&rc->reloc_root_tree.rb_root,
+				   node->bytenr, &node->rb_node);
 	spin_unlock(&rc->reloc_root_tree.lock);
 	if (rb_node) {
 		btrfs_panic(fs_info, -EEXIST,
@@ -1370,8 +661,8 @@ static void __del_reloc_root(struct btrfs_root *root)
 
 	if (rc && root->node) {
 		spin_lock(&rc->reloc_root_tree.lock);
-		rb_node = tree_search(&rc->reloc_root_tree.rb_root,
-				      root->commit_root->start);
+		rb_node = rb_simple_search(&rc->reloc_root_tree.rb_root,
+					   root->commit_root->start);
 		if (rb_node) {
 			node = rb_entry(rb_node, struct mapping_node, rb_node);
 			rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root);
@@ -1414,8 +705,8 @@ static int __update_reloc_root(struct btrfs_root *root)
 	struct reloc_control *rc = fs_info->reloc_ctl;
 
 	spin_lock(&rc->reloc_root_tree.lock);
-	rb_node = tree_search(&rc->reloc_root_tree.rb_root,
-			      root->commit_root->start);
+	rb_node = rb_simple_search(&rc->reloc_root_tree.rb_root,
+				   root->commit_root->start);
 	if (rb_node) {
 		node = rb_entry(rb_node, struct mapping_node, rb_node);
 		rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root);
@@ -1428,11 +719,11 @@ static int __update_reloc_root(struct btrfs_root *root)
 
 	spin_lock(&rc->reloc_root_tree.lock);
 	node->bytenr = root->node->start;
-	rb_node = tree_insert(&rc->reloc_root_tree.rb_root,
-			      node->bytenr, &node->rb_node);
+	rb_node = rb_simple_insert(&rc->reloc_root_tree.rb_root,
+				   node->bytenr, &node->rb_node);
 	spin_unlock(&rc->reloc_root_tree.lock);
 	if (rb_node)
-		backref_tree_panic(rb_node, -EEXIST, node->bytenr);
+		btrfs_backref_panic(fs_info, node->bytenr, -EEXIST);
 	return 0;
 }
 
@@ -1505,7 +796,7 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans,
 
 	reloc_root = btrfs_read_tree_root(fs_info->tree_root, &root_key);
 	BUG_ON(IS_ERR(reloc_root));
-	set_bit(BTRFS_ROOT_REF_COWS, &reloc_root->state);
+	set_bit(BTRFS_ROOT_SHAREABLE, &reloc_root->state);
 	reloc_root->last_trans = trans->transid;
 	return reloc_root;
 }
@@ -1679,14 +970,6 @@ again:
 	return NULL;
 }
 
-static int in_block_group(u64 bytenr, struct btrfs_block_group *block_group)
-{
-	if (bytenr >= block_group->start &&
-	    bytenr < block_group->start + block_group->length)
-		return 1;
-	return 0;
-}
-
 /*
  * get new location of data
  */
@@ -1784,7 +1067,8 @@ int replace_file_extents(struct btrfs_trans_handle *trans,
 		num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
 		if (bytenr == 0)
 			continue;
-		if (!in_block_group(bytenr, rc->block_group))
+		if (!in_range(bytenr, rc->block_group->start,
+			      rc->block_group->length))
 			continue;
 
 		/*
@@ -1940,7 +1224,7 @@ again:
 		level = btrfs_header_level(parent);
 		BUG_ON(level < lowest_level);
 
-		ret = btrfs_bin_search(parent, &key, level, &slot);
+		ret = btrfs_bin_search(parent, &key, &slot);
 		if (ret < 0)
 			break;
 		if (ret && slot > 0)
@@ -2560,7 +1844,8 @@ again:
 					struct btrfs_root, root_list);
 		list_del_init(&reloc_root->root_list);
 
-		root = read_fs_root(fs_info, reloc_root->root_key.offset);
+		root = btrfs_get_fs_root(fs_info, reloc_root->root_key.offset,
+				false);
 		BUG_ON(IS_ERR(root));
 		BUG_ON(root->reloc_root != reloc_root);
 
@@ -2588,13 +1873,10 @@ again:
 static noinline_for_stack
 void free_reloc_roots(struct list_head *list)
 {
-	struct btrfs_root *reloc_root;
+	struct btrfs_root *reloc_root, *tmp;
 
-	while (!list_empty(list)) {
-		reloc_root = list_entry(list->next, struct btrfs_root,
-					root_list);
+	list_for_each_entry_safe(reloc_root, tmp, list, root_list)
 		__del_reloc_root(reloc_root);
-	}
 }
 
 static noinline_for_stack
@@ -2624,12 +1906,11 @@ again:
 		reloc_root = list_entry(reloc_roots.next,
 					struct btrfs_root, root_list);
 
+		root = btrfs_get_fs_root(fs_info, reloc_root->root_key.offset,
+					 false);
 		if (btrfs_root_refs(&reloc_root->root_item) > 0) {
-			root = read_fs_root(fs_info,
-					    reloc_root->root_key.offset);
 			BUG_ON(IS_ERR(root));
 			BUG_ON(root->reloc_root != reloc_root);
-
 			ret = merge_reloc_root(rc, root);
 			btrfs_put_root(root);
 			if (ret) {
@@ -2639,6 +1920,16 @@ again:
 				goto out;
 			}
 		} else {
+			if (!IS_ERR(root)) {
+				if (root->reloc_root == reloc_root) {
+					root->reloc_root = NULL;
+					btrfs_put_root(reloc_root);
+				}
+				clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE,
+					  &root->state);
+				btrfs_put_root(root);
+			}
+
 			list_del_init(&reloc_root->root_list);
 			/* Don't forget to queue this reloc root for cleanup */
 			list_add_tail(&reloc_root->reloc_dirty_list,
@@ -2653,15 +1944,13 @@ again:
 out:
 	if (ret) {
 		btrfs_handle_fs_error(fs_info, ret, NULL);
-		if (!list_empty(&reloc_roots))
-			free_reloc_roots(&reloc_roots);
+		free_reloc_roots(&reloc_roots);
 
 		/* new reloc root may be added */
 		mutex_lock(&fs_info->reloc_mutex);
 		list_splice_init(&rc->reloc_roots, &reloc_roots);
 		mutex_unlock(&fs_info->reloc_mutex);
-		if (!list_empty(&reloc_roots))
-			free_reloc_roots(&reloc_roots);
+		free_reloc_roots(&reloc_roots);
 	}
 
 	/*
@@ -2702,7 +1991,7 @@ static int record_reloc_root_in_trans(struct btrfs_trans_handle *trans,
 	if (reloc_root->last_trans == trans->transid)
 		return 0;
 
-	root = read_fs_root(fs_info, reloc_root->root_key.offset);
+	root = btrfs_get_fs_root(fs_info, reloc_root->root_key.offset, false);
 	BUG_ON(IS_ERR(root));
 	BUG_ON(root->reloc_root != reloc_root);
 	ret = btrfs_record_root_in_trans(trans, root);
@@ -2714,10 +2003,10 @@ static int record_reloc_root_in_trans(struct btrfs_trans_handle *trans,
 static noinline_for_stack
 struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,
 				     struct reloc_control *rc,
-				     struct backref_node *node,
-				     struct backref_edge *edges[])
+				     struct btrfs_backref_node *node,
+				     struct btrfs_backref_edge *edges[])
 {
-	struct backref_node *next;
+	struct btrfs_backref_node *next;
 	struct btrfs_root *root;
 	int index = 0;
 
@@ -2727,7 +2016,7 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,
 		next = walk_up_backref(next, edges, &index);
 		root = next->root;
 		BUG_ON(!root);
-		BUG_ON(!test_bit(BTRFS_ROOT_REF_COWS, &root->state));
+		BUG_ON(!test_bit(BTRFS_ROOT_SHAREABLE, &root->state));
 
 		if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) {
 			record_reloc_root_in_trans(trans, root);
@@ -2746,7 +2035,7 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,
 			ASSERT(next->root);
 			list_add_tail(&next->list,
 				      &rc->backref_cache.changed);
-			__mark_block_processed(rc, next);
+			mark_block_processed(rc, next);
 			break;
 		}
 
@@ -2771,18 +2060,21 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,
 }
 
 /*
- * select a tree root for relocation. return NULL if the block
- * is reference counted. we should use do_relocation() in this
- * case. return a tree root pointer if the block isn't reference
- * counted. return -ENOENT if the block is root of reloc tree.
+ * Select a tree root for relocation.
+ *
+ * Return NULL if the block is not shareable. We should use do_relocation() in
+ * this case.
+ *
+ * Return a tree root pointer if the block is shareable.
+ * Return -ENOENT if the block is root of reloc tree.
  */
 static noinline_for_stack
-struct btrfs_root *select_one_root(struct backref_node *node)
+struct btrfs_root *select_one_root(struct btrfs_backref_node *node)
 {
-	struct backref_node *next;
+	struct btrfs_backref_node *next;
 	struct btrfs_root *root;
 	struct btrfs_root *fs_root = NULL;
-	struct backref_edge *edges[BTRFS_MAX_LEVEL - 1];
+	struct btrfs_backref_edge *edges[BTRFS_MAX_LEVEL - 1];
 	int index = 0;
 
 	next = node;
@@ -2792,8 +2084,8 @@ struct btrfs_root *select_one_root(struct backref_node *node)
 		root = next->root;
 		BUG_ON(!root);
 
-		/* no other choice for non-references counted tree */
-		if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
+		/* No other choice for non-shareable tree */
+		if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
 			return root;
 
 		if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID)
@@ -2814,12 +2106,12 @@ struct btrfs_root *select_one_root(struct backref_node *node)
 
 static noinline_for_stack
 u64 calcu_metadata_size(struct reloc_control *rc,
-			struct backref_node *node, int reserve)
+			struct btrfs_backref_node *node, int reserve)
 {
 	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	struct backref_node *next = node;
-	struct backref_edge *edge;
-	struct backref_edge *edges[BTRFS_MAX_LEVEL - 1];
+	struct btrfs_backref_node *next = node;
+	struct btrfs_backref_edge *edge;
+	struct btrfs_backref_edge *edges[BTRFS_MAX_LEVEL - 1];
 	u64 num_bytes = 0;
 	int index = 0;
 
@@ -2837,7 +2129,7 @@ u64 calcu_metadata_size(struct reloc_control *rc,
 				break;
 
 			edge = list_entry(next->upper.next,
-					  struct backref_edge, list[LOWER]);
+					struct btrfs_backref_edge, list[LOWER]);
 			edges[index++] = edge;
 			next = edge->node[UPPER];
 		}
@@ -2848,7 +2140,7 @@ u64 calcu_metadata_size(struct reloc_control *rc,
 
 static int reserve_metadata_space(struct btrfs_trans_handle *trans,
 				  struct reloc_control *rc,
-				  struct backref_node *node)
+				  struct btrfs_backref_node *node)
 {
 	struct btrfs_root *root = rc->extent_root;
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -2896,14 +2188,14 @@ static int reserve_metadata_space(struct btrfs_trans_handle *trans,
  */
 static int do_relocation(struct btrfs_trans_handle *trans,
 			 struct reloc_control *rc,
-			 struct backref_node *node,
+			 struct btrfs_backref_node *node,
 			 struct btrfs_key *key,
 			 struct btrfs_path *path, int lowest)
 {
 	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	struct backref_node *upper;
-	struct backref_edge *edge;
-	struct backref_edge *edges[BTRFS_MAX_LEVEL - 1];
+	struct btrfs_backref_node *upper;
+	struct btrfs_backref_edge *edge;
+	struct btrfs_backref_edge *edges[BTRFS_MAX_LEVEL - 1];
 	struct btrfs_root *root;
 	struct extent_buffer *eb;
 	u32 blocksize;
@@ -2929,8 +2221,7 @@ static int do_relocation(struct btrfs_trans_handle *trans,
 
 		if (upper->eb && !upper->locked) {
 			if (!lowest) {
-				ret = btrfs_bin_search(upper->eb, key,
-						       upper->level, &slot);
+				ret = btrfs_bin_search(upper->eb, key, &slot);
 				if (ret < 0) {
 					err = ret;
 					goto next;
@@ -2940,7 +2231,7 @@ static int do_relocation(struct btrfs_trans_handle *trans,
 				if (node->eb->start == bytenr)
 					goto next;
 			}
-			drop_node_buffer(upper);
+			btrfs_backref_drop_node_buffer(upper);
 		}
 
 		if (!upper->eb) {
@@ -2968,8 +2259,7 @@ static int do_relocation(struct btrfs_trans_handle *trans,
 			slot = path->slots[upper->level];
 			btrfs_release_path(path);
 		} else {
-			ret = btrfs_bin_search(upper->eb, key, upper->level,
-					       &slot);
+			ret = btrfs_bin_search(upper->eb, key, &slot);
 			if (ret < 0) {
 				err = ret;
 				goto next;
@@ -3039,15 +2329,15 @@ static int do_relocation(struct btrfs_trans_handle *trans,
 		}
 next:
 		if (!upper->pending)
-			drop_node_buffer(upper);
+			btrfs_backref_drop_node_buffer(upper);
 		else
-			unlock_node_buffer(upper);
+			btrfs_backref_unlock_node_buffer(upper);
 		if (err)
 			break;
 	}
 
 	if (!err && node->pending) {
-		drop_node_buffer(node);
+		btrfs_backref_drop_node_buffer(node);
 		list_move_tail(&node->list, &rc->backref_cache.changed);
 		node->pending = 0;
 	}
@@ -3059,7 +2349,7 @@ next:
 
 static int link_to_upper(struct btrfs_trans_handle *trans,
 			 struct reloc_control *rc,
-			 struct backref_node *node,
+			 struct btrfs_backref_node *node,
 			 struct btrfs_path *path)
 {
 	struct btrfs_key key;
@@ -3073,15 +2363,15 @@ static int finish_pending_nodes(struct btrfs_trans_handle *trans,
 				struct btrfs_path *path, int err)
 {
 	LIST_HEAD(list);
-	struct backref_cache *cache = &rc->backref_cache;
-	struct backref_node *node;
+	struct btrfs_backref_cache *cache = &rc->backref_cache;
+	struct btrfs_backref_node *node;
 	int level;
 	int ret;
 
 	for (level = 0; level < BTRFS_MAX_LEVEL; level++) {
 		while (!list_empty(&cache->pending[level])) {
 			node = list_entry(cache->pending[level].next,
-					  struct backref_node, list);
+					  struct btrfs_backref_node, list);
 			list_move_tail(&node->list, &list);
 			BUG_ON(!node->pending);
 
@@ -3096,35 +2386,16 @@ static int finish_pending_nodes(struct btrfs_trans_handle *trans,
 	return err;
 }
 
-static void mark_block_processed(struct reloc_control *rc,
-				 u64 bytenr, u32 blocksize)
-{
-	set_extent_bits(&rc->processed_blocks, bytenr, bytenr + blocksize - 1,
-			EXTENT_DIRTY);
-}
-
-static void __mark_block_processed(struct reloc_control *rc,
-				   struct backref_node *node)
-{
-	u32 blocksize;
-	if (node->level == 0 ||
-	    in_block_group(node->bytenr, rc->block_group)) {
-		blocksize = rc->extent_root->fs_info->nodesize;
-		mark_block_processed(rc, node->bytenr, blocksize);
-	}
-	node->processed = 1;
-}
-
 /*
  * mark a block and all blocks directly/indirectly reference the block
  * as processed.
  */
 static void update_processed_blocks(struct reloc_control *rc,
-				    struct backref_node *node)
+				    struct btrfs_backref_node *node)
 {
-	struct backref_node *next = node;
-	struct backref_edge *edge;
-	struct backref_edge *edges[BTRFS_MAX_LEVEL - 1];
+	struct btrfs_backref_node *next = node;
+	struct btrfs_backref_edge *edge;
+	struct btrfs_backref_edge *edges[BTRFS_MAX_LEVEL - 1];
 	int index = 0;
 
 	while (next) {
@@ -3133,13 +2404,13 @@ static void update_processed_blocks(struct reloc_control *rc,
 			if (next->processed)
 				break;
 
-			__mark_block_processed(rc, next);
+			mark_block_processed(rc, next);
 
 			if (list_empty(&next->upper))
 				break;
 
 			edge = list_entry(next->upper.next,
-					  struct backref_edge, list[LOWER]);
+					struct btrfs_backref_edge, list[LOWER]);
 			edges[index++] = edge;
 			next = edge->node[UPPER];
 		}
@@ -3184,7 +2455,7 @@ static int get_tree_block_key(struct btrfs_fs_info *fs_info,
  */
 static int relocate_tree_block(struct btrfs_trans_handle *trans,
 				struct reloc_control *rc,
-				struct backref_node *node,
+				struct btrfs_backref_node *node,
 				struct btrfs_key *key,
 				struct btrfs_path *path)
 {
@@ -3210,7 +2481,7 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans,
 	}
 
 	if (root) {
-		if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) {
+		if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) {
 			BUG_ON(node->new_bytenr);
 			BUG_ON(!list_empty(&node->list));
 			btrfs_record_root_in_trans(trans, root);
@@ -3234,7 +2505,7 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans,
 	}
 out:
 	if (ret || node->level == 0 || node->cowonly)
-		remove_backref_node(&rc->backref_cache, node);
+		btrfs_backref_cleanup_node(&rc->backref_cache, node);
 	return ret;
 }
 
@@ -3246,7 +2517,7 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
 			 struct reloc_control *rc, struct rb_root *blocks)
 {
 	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	struct backref_node *node;
+	struct btrfs_backref_node *node;
 	struct btrfs_path *path;
 	struct tree_block *block;
 	struct tree_block *next;
@@ -3613,9 +2884,10 @@ static int add_tree_block(struct reloc_control *rc,
 	block->level = level;
 	block->key_ready = 0;
 
-	rb_node = tree_insert(blocks, block->bytenr, &block->rb_node);
+	rb_node = rb_simple_insert(blocks, block->bytenr, &block->rb_node);
 	if (rb_node)
-		backref_tree_panic(rb_node, -EEXIST, block->bytenr);
+		btrfs_backref_panic(rc->extent_root->fs_info, block->bytenr,
+				    -EEXIST);
 
 	return 0;
 }
@@ -3636,7 +2908,7 @@ static int __add_tree_block(struct reloc_control *rc,
 	if (tree_block_processed(bytenr, rc))
 		return 0;
 
-	if (tree_search(blocks, bytenr))
+	if (rb_simple_search(blocks, bytenr))
 		return 0;
 
 	path = btrfs_alloc_path();
@@ -3698,7 +2970,6 @@ static int delete_block_group_cache(struct btrfs_fs_info *fs_info,
 				    struct inode *inode,
 				    u64 ino)
 {
-	struct btrfs_key key;
 	struct btrfs_root *root = fs_info->tree_root;
 	struct btrfs_trans_handle *trans;
 	int ret = 0;
@@ -3706,11 +2977,7 @@ static int delete_block_group_cache(struct btrfs_fs_info *fs_info,
 	if (inode)
 		goto truncate;
 
-	key.objectid = ino;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-
-	inode = btrfs_iget(fs_info->sb, &key, root);
+	inode = btrfs_iget(fs_info->sb, ino, root);
 	if (IS_ERR(inode))
 		return -ENOENT;
 
@@ -4122,7 +3389,7 @@ restart:
 	rc->create_reloc_tree = 0;
 	set_reloc_control(rc);
 
-	backref_cache_cleanup(&rc->backref_cache);
+	btrfs_backref_release_cache(&rc->backref_cache);
 	btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1, NULL);
 
 	/*
@@ -4198,14 +3465,10 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info,
 	struct inode *inode = NULL;
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root;
-	struct btrfs_key key;
 	u64 objectid;
 	int err = 0;
 
-	root = read_fs_root(fs_info, BTRFS_DATA_RELOC_TREE_OBJECTID);
-	if (IS_ERR(root))
-		return ERR_CAST(root);
-
+	root = btrfs_grab_root(fs_info->data_reloc_root);
 	trans = btrfs_start_transaction(root, 6);
 	if (IS_ERR(trans)) {
 		btrfs_put_root(root);
@@ -4219,10 +3482,7 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info,
 	err = __insert_orphan_inode(trans, root, objectid);
 	BUG_ON(err);
 
-	key.objectid = objectid;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-	inode = btrfs_iget(fs_info->sb, &key, root);
+	inode = btrfs_iget(fs_info->sb, objectid, root);
 	BUG_ON(IS_ERR(inode));
 	BTRFS_I(inode)->index_cnt = group->start;
 
@@ -4249,7 +3509,7 @@ static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info)
 
 	INIT_LIST_HEAD(&rc->reloc_roots);
 	INIT_LIST_HEAD(&rc->dirty_subvol_roots);
-	backref_cache_init(&rc->backref_cache);
+	btrfs_backref_init_cache(fs_info, &rc->backref_cache, 1);
 	mapping_tree_init(&rc->reloc_root_tree);
 	extent_io_tree_init(fs_info, &rc->processed_blocks,
 			    IO_TREE_RELOC_BLOCKS, NULL);
@@ -4494,12 +3754,12 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 			goto out;
 		}
 
-		set_bit(BTRFS_ROOT_REF_COWS, &reloc_root->state);
+		set_bit(BTRFS_ROOT_SHAREABLE, &reloc_root->state);
 		list_add(&reloc_root->root_list, &reloc_roots);
 
 		if (btrfs_root_refs(&reloc_root->root_item) > 0) {
-			fs_root = read_fs_root(fs_info,
-					       reloc_root->root_key.offset);
+			fs_root = btrfs_get_fs_root(fs_info,
+					reloc_root->root_key.offset, false);
 			if (IS_ERR(fs_root)) {
 				ret = PTR_ERR(fs_root);
 				if (ret != -ENOENT) {
@@ -4555,7 +3815,8 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 			continue;
 		}
 
-		fs_root = read_fs_root(fs_info, reloc_root->root_key.offset);
+		fs_root = btrfs_get_fs_root(fs_info, reloc_root->root_key.offset,
+					    false);
 		if (IS_ERR(fs_root)) {
 			err = PTR_ERR(fs_root);
 			list_add_tail(&reloc_root->root_list, &reloc_roots);
@@ -4591,20 +3852,16 @@ out_unset:
 	unset_reloc_control(rc);
 	free_reloc_control(rc);
 out:
-	if (!list_empty(&reloc_roots))
-		free_reloc_roots(&reloc_roots);
+	free_reloc_roots(&reloc_roots);
 
 	btrfs_free_path(path);
 
 	if (err == 0) {
 		/* cleanup orphan inode in data relocation tree */
-		fs_root = read_fs_root(fs_info, BTRFS_DATA_RELOC_TREE_OBJECTID);
-		if (IS_ERR(fs_root)) {
-			err = PTR_ERR(fs_root);
-		} else {
-			err = btrfs_orphan_cleanup(fs_root);
-			btrfs_put_root(fs_root);
-		}
+		fs_root = btrfs_grab_root(fs_info->data_reloc_root);
+		ASSERT(fs_root);
+		err = btrfs_orphan_cleanup(fs_root);
+		btrfs_put_root(fs_root);
 	}
 	return err;
 }
@@ -4666,7 +3923,7 @@ int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct reloc_control *rc;
-	struct backref_node *node;
+	struct btrfs_backref_node *node;
 	int first_cow = 0;
 	int level;
 	int ret = 0;
@@ -4691,7 +3948,7 @@ int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans,
 		BUG_ON(node->bytenr != buf->start &&
 		       node->new_bytenr != buf->start);
 
-		drop_node_buffer(node);
+		btrfs_backref_drop_node_buffer(node);
 		atomic_inc(&cow->refs);
 		node->eb = cow;
 		node->new_bytenr = cow->start;
@@ -4703,7 +3960,7 @@ int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans,
 		}
 
 		if (first_cow)
-			__mark_block_processed(rc, node);
+			mark_block_processed(rc, node);
 
 		if (first_cow && level > 0)
 			rc->nodes_relocated += buf->len;
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 668f22844017..c89697486366 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -210,7 +210,6 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 	struct extent_buffer *leaf;
 	struct btrfs_path *path;
 	struct btrfs_key key;
-	struct btrfs_key root_key;
 	struct btrfs_root *root;
 	int err = 0;
 	int ret;
@@ -223,10 +222,9 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 	key.type = BTRFS_ORPHAN_ITEM_KEY;
 	key.offset = 0;
 
-	root_key.type = BTRFS_ROOT_ITEM_KEY;
-	root_key.offset = (u64)-1;
-
 	while (1) {
+		u64 root_objectid;
+
 		ret = btrfs_search_slot(NULL, tree_root, &key, path, 0, 0);
 		if (ret < 0) {
 			err = ret;
@@ -250,10 +248,10 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 		    key.type != BTRFS_ORPHAN_ITEM_KEY)
 			break;
 
-		root_key.objectid = key.offset;
+		root_objectid = key.offset;
 		key.offset++;
 
-		root = btrfs_get_fs_root(fs_info, &root_key, false);
+		root = btrfs_get_fs_root(fs_info, root_objectid, false);
 		err = PTR_ERR_OR_ZERO(root);
 		if (err && err != -ENOENT) {
 			break;
@@ -270,7 +268,7 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 				break;
 			}
 			err = btrfs_del_orphan_item(trans, tree_root,
-						    root_key.objectid);
+						    root_objectid);
 			btrfs_end_transaction(trans);
 			if (err) {
 				btrfs_handle_fs_error(fs_info, err,
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index adaf8ab694d5..016a025e36c7 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -647,13 +647,9 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
 	struct btrfs_fs_info *fs_info = swarn->dev->fs_info;
 	struct inode_fs_paths *ipath = NULL;
 	struct btrfs_root *local_root;
-	struct btrfs_key root_key;
 	struct btrfs_key key;
 
-	root_key.objectid = root;
-	root_key.type = BTRFS_ROOT_ITEM_KEY;
-	root_key.offset = (u64)-1;
-	local_root = btrfs_get_fs_root(fs_info, &root_key, true);
+	local_root = btrfs_get_fs_root(fs_info, root, true);
 	if (IS_ERR(local_root)) {
 		ret = PTR_ERR(local_root);
 		goto err;
@@ -3046,7 +3042,8 @@ out:
 static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 					   struct map_lookup *map,
 					   struct btrfs_device *scrub_dev,
-					   int num, u64 base, u64 length)
+					   int num, u64 base, u64 length,
+					   struct btrfs_block_group *cache)
 {
 	struct btrfs_path *path, *ppath;
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
@@ -3284,6 +3281,20 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 				break;
 			}
 
+			/*
+			 * If our block group was removed in the meanwhile, just
+			 * stop scrubbing since there is no point in continuing.
+			 * Continuing would prevent reusing its device extents
+			 * for new block groups for a long time.
+			 */
+			spin_lock(&cache->lock);
+			if (cache->removed) {
+				spin_unlock(&cache->lock);
+				ret = 0;
+				goto out;
+			}
+			spin_unlock(&cache->lock);
+
 			extent = btrfs_item_ptr(l, slot,
 						struct btrfs_extent_item);
 			flags = btrfs_extent_flags(l, extent);
@@ -3328,13 +3339,14 @@ again:
 						   &extent_dev,
 						   &extent_mirror_num);
 
-			ret = btrfs_lookup_csums_range(csum_root,
-						       extent_logical,
-						       extent_logical +
-						       extent_len - 1,
-						       &sctx->csum_list, 1);
-			if (ret)
-				goto out;
+			if (flags & BTRFS_EXTENT_FLAG_DATA) {
+				ret = btrfs_lookup_csums_range(csum_root,
+						extent_logical,
+						extent_logical + extent_len - 1,
+						&sctx->csum_list, 1);
+				if (ret)
+					goto out;
+			}
 
 			ret = scrub_extent(sctx, map, extent_logical, extent_len,
 					   extent_physical, extent_dev, flags,
@@ -3457,7 +3469,7 @@ static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx,
 		if (map->stripes[i].dev->bdev == scrub_dev->bdev &&
 		    map->stripes[i].physical == dev_offset) {
 			ret = scrub_stripe(sctx, map, scrub_dev, i,
-					   chunk_offset, length);
+					   chunk_offset, length, cache);
 			if (ret)
 				goto out;
 		}
@@ -3555,6 +3567,23 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			goto skip;
 
 		/*
+		 * Make sure that while we are scrubbing the corresponding block
+		 * group doesn't get its logical address and its device extents
+		 * reused for another block group, which can possibly be of a
+		 * different type and different profile. We do this to prevent
+		 * false error detections and crashes due to bogus attempts to
+		 * repair extents.
+		 */
+		spin_lock(&cache->lock);
+		if (cache->removed) {
+			spin_unlock(&cache->lock);
+			btrfs_put_block_group(cache);
+			goto skip;
+		}
+		btrfs_freeze_block_group(cache);
+		spin_unlock(&cache->lock);
+
+		/*
 		 * we need call btrfs_inc_block_group_ro() with scrubs_paused,
 		 * to avoid deadlock caused by:
 		 * btrfs_inc_block_group_ro()
@@ -3609,6 +3638,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		} else {
 			btrfs_warn(fs_info,
 				   "failed setting block group ro: %d", ret);
+			btrfs_unfreeze_block_group(cache);
 			btrfs_put_block_group(cache);
 			scrub_pause_off(fs_info);
 			break;
@@ -3695,6 +3725,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			spin_unlock(&cache->lock);
 		}
 
+		btrfs_unfreeze_block_group(cache);
 		btrfs_put_block_group(cache);
 		if (ret)
 			break;
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 6a92ecf9eaa2..d9813a5b075a 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -23,6 +23,7 @@
 #include "btrfs_inode.h"
 #include "transaction.h"
 #include "compression.h"
+#include "xattr.h"
 
 /*
  * Maximum number of references an extent can have in order for us to attempt to
@@ -4545,6 +4546,10 @@ static int __process_new_xattr(int num, struct btrfs_key *di_key,
 	struct fs_path *p;
 	struct posix_acl_xattr_header dummy_acl;
 
+	/* Capabilities are emitted by finish_inode_if_needed */
+	if (!strncmp(name, XATTR_NAME_CAPS, name_len))
+		return 0;
+
 	p = fs_path_alloc();
 	if (!p)
 		return -ENOMEM;
@@ -4801,17 +4806,12 @@ static ssize_t fill_read_buf(struct send_ctx *sctx, u64 offset, u32 len)
 	struct inode *inode;
 	struct page *page;
 	char *addr;
-	struct btrfs_key key;
 	pgoff_t index = offset >> PAGE_SHIFT;
 	pgoff_t last_index;
 	unsigned pg_offset = offset_in_page(offset);
 	ssize_t ret = 0;
 
-	key.objectid = sctx->cur_ino;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-
-	inode = btrfs_iget(fs_info->sb, &key, root);
+	inode = btrfs_iget(fs_info->sb, sctx->cur_ino, root);
 	if (IS_ERR(inode))
 		return PTR_ERR(inode);
 
@@ -5107,6 +5107,64 @@ static int send_extent_data(struct send_ctx *sctx,
 	return 0;
 }
 
+/*
+ * Search for a capability xattr related to sctx->cur_ino. If the capability is
+ * found, call send_set_xattr function to emit it.
+ *
+ * Return 0 if there isn't a capability, or when the capability was emitted
+ * successfully, or < 0 if an error occurred.
+ */
+static int send_capabilities(struct send_ctx *sctx)
+{
+	struct fs_path *fspath = NULL;
+	struct btrfs_path *path;
+	struct btrfs_dir_item *di;
+	struct extent_buffer *leaf;
+	unsigned long data_ptr;
+	char *buf = NULL;
+	int buf_len;
+	int ret = 0;
+
+	path = alloc_path_for_send();
+	if (!path)
+		return -ENOMEM;
+
+	di = btrfs_lookup_xattr(NULL, sctx->send_root, path, sctx->cur_ino,
+				XATTR_NAME_CAPS, strlen(XATTR_NAME_CAPS), 0);
+	if (!di) {
+		/* There is no xattr for this inode */
+		goto out;
+	} else if (IS_ERR(di)) {
+		ret = PTR_ERR(di);
+		goto out;
+	}
+
+	leaf = path->nodes[0];
+	buf_len = btrfs_dir_data_len(leaf, di);
+
+	fspath = fs_path_alloc();
+	buf = kmalloc(buf_len, GFP_KERNEL);
+	if (!fspath || !buf) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = get_cur_path(sctx, sctx->cur_ino, sctx->cur_inode_gen, fspath);
+	if (ret < 0)
+		goto out;
+
+	data_ptr = (unsigned long)(di + 1) + btrfs_dir_name_len(leaf, di);
+	read_extent_buffer(leaf, buf, data_ptr, buf_len);
+
+	ret = send_set_xattr(sctx, fspath, XATTR_NAME_CAPS,
+			strlen(XATTR_NAME_CAPS), buf, buf_len);
+out:
+	kfree(buf);
+	fs_path_free(fspath);
+	btrfs_free_path(path);
+	return ret;
+}
+
 static int clone_range(struct send_ctx *sctx,
 		       struct clone_root *clone_root,
 		       const u64 disk_byte,
@@ -5972,6 +6030,10 @@ static int finish_inode_if_needed(struct send_ctx *sctx, int at_end)
 			goto out;
 	}
 
+	ret = send_capabilities(sctx);
+	if (ret < 0)
+		goto out;
+
 	/*
 	 * If other directory inodes depended on our current directory
 	 * inode's move/rename, now do their move/rename operations.
@@ -7021,7 +7083,6 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	struct btrfs_root *send_root = BTRFS_I(file_inode(mnt_file))->root;
 	struct btrfs_fs_info *fs_info = send_root->fs_info;
 	struct btrfs_root *clone_root;
-	struct btrfs_key key;
 	struct send_ctx *sctx = NULL;
 	u32 i;
 	u64 *clone_sources_tmp = NULL;
@@ -7143,11 +7204,8 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 		}
 
 		for (i = 0; i < arg->clone_sources_count; i++) {
-			key.objectid = clone_sources_tmp[i];
-			key.type = BTRFS_ROOT_ITEM_KEY;
-			key.offset = (u64)-1;
-
-			clone_root = btrfs_get_fs_root(fs_info, &key, true);
+			clone_root = btrfs_get_fs_root(fs_info,
+						clone_sources_tmp[i], true);
 			if (IS_ERR(clone_root)) {
 				ret = PTR_ERR(clone_root);
 				goto out;
@@ -7178,11 +7236,8 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	}
 
 	if (arg->parent_root) {
-		key.objectid = arg->parent_root;
-		key.type = BTRFS_ROOT_ITEM_KEY;
-		key.offset = (u64)-1;
-
-		sctx->parent_root = btrfs_get_fs_root(fs_info, &key, true);
+		sctx->parent_root = btrfs_get_fs_root(fs_info, arg->parent_root,
+						      true);
 		if (IS_ERR(sctx->parent_root)) {
 			ret = PTR_ERR(sctx->parent_root);
 			goto out;
diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c
index ff17a4420358..41ee88633769 100644
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -626,6 +626,7 @@ static int may_commit_transaction(struct btrfs_fs_info *fs_info,
 	struct reserve_ticket *ticket = NULL;
 	struct btrfs_block_rsv *delayed_rsv = &fs_info->delayed_block_rsv;
 	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;
@@ -688,6 +689,11 @@ static int may_commit_transaction(struct btrfs_fs_info *fs_info,
 	spin_lock(&delayed_refs_rsv->lock);
 	reclaim_bytes += delayed_refs_rsv->reserved;
 	spin_unlock(&delayed_refs_rsv->lock);
+
+	spin_lock(&trans_rsv->lock);
+	reclaim_bytes += trans_rsv->reserved;
+	spin_unlock(&trans_rsv->lock);
+
 	if (reclaim_bytes >= bytes_needed)
 		goto commit;
 	bytes_needed -= reclaim_bytes;
@@ -856,6 +862,34 @@ static inline int need_do_async_reclaim(struct btrfs_fs_info *fs_info,
 		!test_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state));
 }
 
+static bool steal_from_global_rsv(struct btrfs_fs_info *fs_info,
+				  struct btrfs_space_info *space_info,
+				  struct reserve_ticket *ticket)
+{
+	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
+	u64 min_bytes;
+
+	if (global_rsv->space_info != space_info)
+		return false;
+
+	spin_lock(&global_rsv->lock);
+	min_bytes = div_factor(global_rsv->size, 1);
+	if (global_rsv->reserved < min_bytes + ticket->bytes) {
+		spin_unlock(&global_rsv->lock);
+		return false;
+	}
+	global_rsv->reserved -= ticket->bytes;
+	ticket->bytes = 0;
+	list_del_init(&ticket->list);
+	wake_up(&ticket->wait);
+	space_info->tickets_id++;
+	if (global_rsv->reserved < global_rsv->size)
+		global_rsv->full = 0;
+	spin_unlock(&global_rsv->lock);
+
+	return true;
+}
+
 /*
  * maybe_fail_all_tickets - we've exhausted our flushing, start failing tickets
  * @fs_info - fs_info for this fs
@@ -888,6 +922,10 @@ static bool maybe_fail_all_tickets(struct btrfs_fs_info *fs_info,
 		ticket = list_first_entry(&space_info->tickets,
 					  struct reserve_ticket, list);
 
+		if (ticket->steal &&
+		    steal_from_global_rsv(fs_info, space_info, ticket))
+			return true;
+
 		/*
 		 * may_commit_transaction will avoid committing the transaction
 		 * if it doesn't feel like the space reclaimed by the commit
@@ -1104,6 +1142,7 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 
 	switch (flush) {
 	case BTRFS_RESERVE_FLUSH_ALL:
+	case BTRFS_RESERVE_FLUSH_ALL_STEAL:
 		wait_reserve_ticket(fs_info, space_info, ticket);
 		break;
 	case BTRFS_RESERVE_FLUSH_LIMIT:
@@ -1125,11 +1164,17 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 	ret = ticket->error;
 	if (ticket->bytes || ticket->error) {
 		/*
-		 * Need to delete here for priority tickets. For regular tickets
-		 * either the async reclaim job deletes the ticket from the list
-		 * or we delete it ourselves at wait_reserve_ticket().
+		 * We were a priority ticket, so we need to delete ourselves
+		 * from the list.  Because we could have other priority tickets
+		 * behind us that require less space, run
+		 * btrfs_try_granting_tickets() to see if their reservations can
+		 * now be made.
 		 */
-		remove_ticket(space_info, ticket);
+		if (!list_empty(&ticket->list)) {
+			remove_ticket(space_info, ticket);
+			btrfs_try_granting_tickets(fs_info, space_info);
+		}
+
 		if (!ret)
 			ret = -ENOSPC;
 	}
@@ -1145,6 +1190,16 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 	return ret;
 }
 
+/*
+ * This returns true if this flush state will go through the ordinary flushing
+ * code.
+ */
+static inline bool is_normal_flushing(enum btrfs_reserve_flush_enum flush)
+{
+	return	(flush == BTRFS_RESERVE_FLUSH_ALL) ||
+		(flush == BTRFS_RESERVE_FLUSH_ALL_STEAL);
+}
+
 /**
  * reserve_metadata_bytes - try to reserve bytes from the block_rsv's space
  * @root - the root we're allocating for
@@ -1175,8 +1230,17 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 	spin_lock(&space_info->lock);
 	ret = -ENOSPC;
 	used = btrfs_space_info_used(space_info, true);
-	pending_tickets = !list_empty(&space_info->tickets) ||
-		!list_empty(&space_info->priority_tickets);
+
+	/*
+	 * We don't want NO_FLUSH allocations to jump everybody, they can
+	 * generally handle ENOSPC in a different way, so treat them the same as
+	 * normal flushers when it comes to skipping pending tickets.
+	 */
+	if (is_normal_flushing(flush) || (flush == BTRFS_RESERVE_NO_FLUSH))
+		pending_tickets = !list_empty(&space_info->tickets) ||
+			!list_empty(&space_info->priority_tickets);
+	else
+		pending_tickets = !list_empty(&space_info->priority_tickets);
 
 	/*
 	 * Carry on if we have enough space (short-circuit) OR call
@@ -1198,12 +1262,13 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 	 * the list and we will do our own flushing further down.
 	 */
 	if (ret && flush != BTRFS_RESERVE_NO_FLUSH) {
-		ASSERT(space_info->reclaim_size >= 0);
 		ticket.bytes = orig_bytes;
 		ticket.error = 0;
 		space_info->reclaim_size += ticket.bytes;
 		init_waitqueue_head(&ticket.wait);
-		if (flush == BTRFS_RESERVE_FLUSH_ALL) {
+		ticket.steal = (flush == BTRFS_RESERVE_FLUSH_ALL_STEAL);
+		if (flush == BTRFS_RESERVE_FLUSH_ALL ||
+		    flush == BTRFS_RESERVE_FLUSH_ALL_STEAL) {
 			list_add_tail(&ticket.list, &space_info->tickets);
 			if (!space_info->flush) {
 				space_info->flush = 1;
diff --git a/fs/btrfs/space-info.h b/fs/btrfs/space-info.h
index 0a5001ef1481..c3c64019950a 100644
--- a/fs/btrfs/space-info.h
+++ b/fs/btrfs/space-info.h
@@ -78,6 +78,7 @@ struct btrfs_space_info {
 struct reserve_ticket {
 	u64 bytes;
 	int error;
+	bool steal;
 	struct list_head list;
 	wait_queue_head_t wait;
 };
diff --git a/fs/btrfs/struct-funcs.c b/fs/btrfs/struct-funcs.c
index 73f7987143df..079b059818e9 100644
--- a/fs/btrfs/struct-funcs.c
+++ b/fs/btrfs/struct-funcs.c
@@ -17,151 +17,152 @@ 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)
+{
+	const unsigned long member_offset = (unsigned long)ptr + off;
+
+	if (member_offset > eb->len) {
+		btrfs_warn(eb->fs_info,
+	"bad eb member start: ptr 0x%lx start %llu member offset %lu size %d",
+			(unsigned long)ptr, eb->start, member_offset, size);
+		return false;
+	}
+	if (member_offset + size > eb->len) {
+		btrfs_warn(eb->fs_info,
+	"bad eb member end: ptr 0x%lx start %llu member offset %lu size %d",
+			(unsigned long)ptr, eb->start, member_offset, size);
+		return false;
+	}
+
+	return true;
+}
+
 /*
- * this is some deeply nasty code.
+ * Macro templates that define helpers to read/write extent buffer data of a
+ * given size, that are also used via ctree.h for access to item members by
+ * specialized helpers.
  *
- * The end result is that anyone who #includes ctree.h gets a
- * declaration for the btrfs_set_foo functions and btrfs_foo functions,
- * which are wrappers of btrfs_set_token_#bits functions and
- * btrfs_get_token_#bits functions, which are defined in this file.
+ * Generic helpers:
+ * - btrfs_set_8 (for 8/16/32/64)
+ * - btrfs_get_8 (for 8/16/32/64)
  *
- * These setget functions do all the extent_buffer related mapping
- * required to efficiently read and write specific fields in the extent
- * buffers.  Every pointer to metadata items in btrfs is really just
- * an unsigned long offset into the extent buffer which has been
- * cast to a specific type.  This gives us all the gcc type checking.
+ * Generic helpers with a token (cached address of the most recently accessed
+ * page):
+ * - btrfs_set_token_8 (for 8/16/32/64)
+ * - btrfs_get_token_8 (for 8/16/32/64)
  *
- * The extent buffer api is used to do the page spanning work required to
- * have a metadata blocksize different from the page size.
+ * The set/get functions handle data spanning two pages transparently, in case
+ * metadata block size is larger than page.  Every pointer to metadata items is
+ * an offset into the extent buffer page array, cast to a specific type.  This
+ * gives us all the type checking.
  *
- * There are 2 variants defined, one with a token pointer and one without.
+ * The extent buffer pages stored in the array pages do not form a contiguous
+ * phyusical range, but the API functions assume the linear offset to the range
+ * from 0 to metadata node size.
  */
 
 #define DEFINE_BTRFS_SETGET_BITS(bits)					\
-u##bits btrfs_get_token_##bits(const struct extent_buffer *eb,		\
-			       const void *ptr, unsigned long off,	\
-			       struct btrfs_map_token *token)		\
+u##bits btrfs_get_token_##bits(struct btrfs_map_token *token,		\
+			       const void *ptr, unsigned long off)	\
 {									\
-	unsigned long part_offset = (unsigned long)ptr;			\
-	unsigned long offset = part_offset + off;			\
-	void *p;							\
-	int err;							\
-	char *kaddr;							\
-	unsigned long map_start;					\
-	unsigned long map_len;						\
-	int size = sizeof(u##bits);					\
-	u##bits res;							\
+	const unsigned long member_offset = (unsigned long)ptr + off;	\
+	const unsigned long idx = member_offset >> PAGE_SHIFT;		\
+	const unsigned long oip = offset_in_page(member_offset);	\
+	const int size = sizeof(u##bits);				\
+	u8 lebytes[sizeof(u##bits)];					\
+	const int part = PAGE_SIZE - oip;				\
 									\
 	ASSERT(token);							\
-	ASSERT(token->eb == eb);					\
-									\
-	if (token->kaddr && token->offset <= offset &&			\
-	   (token->offset + PAGE_SIZE >= offset + size)) {	\
-		kaddr = token->kaddr;					\
-		p = kaddr + part_offset - token->offset;		\
-		res = get_unaligned_le##bits(p + off);			\
-		return res;						\
+	ASSERT(token->kaddr);						\
+	ASSERT(check_setget_bounds(token->eb, ptr, off, size));		\
+	if (token->offset <= member_offset &&				\
+	    member_offset + size <= token->offset + PAGE_SIZE) {	\
+		return get_unaligned_le##bits(token->kaddr + oip);	\
 	}								\
-	err = map_private_extent_buffer(eb, offset, size,		\
-					&kaddr, &map_start, &map_len);	\
-	if (err) {							\
-		__le##bits leres;					\
+	token->kaddr = page_address(token->eb->pages[idx]);		\
+	token->offset = idx << PAGE_SHIFT;				\
+	if (oip + size <= PAGE_SIZE)					\
+		return get_unaligned_le##bits(token->kaddr + oip);	\
 									\
-		read_extent_buffer(eb, &leres, offset, size);		\
-		return le##bits##_to_cpu(leres);			\
-	}								\
-	p = kaddr + part_offset - map_start;				\
-	res = get_unaligned_le##bits(p + off);				\
-	token->kaddr = kaddr;						\
-	token->offset = map_start;					\
-	return res;							\
+	memcpy(lebytes, token->kaddr + oip, part);			\
+	token->kaddr = page_address(token->eb->pages[idx + 1]);		\
+	token->offset = (idx + 1) << PAGE_SHIFT;			\
+	memcpy(lebytes + part, token->kaddr, size - part);		\
+	return get_unaligned_le##bits(lebytes);				\
 }									\
 u##bits btrfs_get_##bits(const struct extent_buffer *eb,		\
 			 const void *ptr, unsigned long off)		\
 {									\
-	unsigned long part_offset = (unsigned long)ptr;			\
-	unsigned long offset = part_offset + off;			\
-	void *p;							\
-	int err;							\
-	char *kaddr;							\
-	unsigned long map_start;					\
-	unsigned long map_len;						\
-	int size = sizeof(u##bits);					\
-	u##bits res;							\
+	const unsigned long member_offset = (unsigned long)ptr + off;	\
+	const unsigned long oip = offset_in_page(member_offset);	\
+	const unsigned long idx = member_offset >> PAGE_SHIFT;		\
+	char *kaddr = page_address(eb->pages[idx]);			\
+	const int size = sizeof(u##bits);				\
+	const int part = PAGE_SIZE - oip;				\
+	u8 lebytes[sizeof(u##bits)];					\
 									\
-	err = map_private_extent_buffer(eb, offset, size,		\
-					&kaddr, &map_start, &map_len);	\
-	if (err) {							\
-		__le##bits leres;					\
+	ASSERT(check_setget_bounds(eb, ptr, off, size));		\
+	if (oip + size <= PAGE_SIZE)					\
+		return get_unaligned_le##bits(kaddr + oip);		\
 									\
-		read_extent_buffer(eb, &leres, offset, size);		\
-		return le##bits##_to_cpu(leres);			\
-	}								\
-	p = kaddr + part_offset - map_start;				\
-	res = get_unaligned_le##bits(p + off);				\
-	return res;							\
+	memcpy(lebytes, kaddr + oip, part);				\
+	kaddr = page_address(eb->pages[idx + 1]);			\
+	memcpy(lebytes + part, kaddr, size - part);			\
+	return get_unaligned_le##bits(lebytes);				\
 }									\
-void btrfs_set_token_##bits(struct extent_buffer *eb,			\
+void btrfs_set_token_##bits(struct btrfs_map_token *token,		\
 			    const void *ptr, unsigned long off,		\
-			    u##bits val,				\
-			    struct btrfs_map_token *token)		\
+			    u##bits val)				\
 {									\
-	unsigned long part_offset = (unsigned long)ptr;			\
-	unsigned long offset = part_offset + off;			\
-	void *p;							\
-	int err;							\
-	char *kaddr;							\
-	unsigned long map_start;					\
-	unsigned long map_len;						\
-	int size = sizeof(u##bits);					\
+	const unsigned long member_offset = (unsigned long)ptr + off;	\
+	const unsigned long idx = member_offset >> PAGE_SHIFT;		\
+	const unsigned long oip = offset_in_page(member_offset);	\
+	const int size = sizeof(u##bits);				\
+	u8 lebytes[sizeof(u##bits)];					\
+	const int part = PAGE_SIZE - oip;				\
 									\
 	ASSERT(token);							\
-	ASSERT(token->eb == eb);					\
-									\
-	if (token->kaddr && token->offset <= offset &&			\
-	   (token->offset + PAGE_SIZE >= offset + size)) {	\
-		kaddr = token->kaddr;					\
-		p = kaddr + part_offset - token->offset;		\
-		put_unaligned_le##bits(val, p + off);			\
+	ASSERT(token->kaddr);						\
+	ASSERT(check_setget_bounds(token->eb, ptr, off, size));		\
+	if (token->offset <= member_offset &&				\
+	    member_offset + size <= token->offset + PAGE_SIZE) {	\
+		put_unaligned_le##bits(val, token->kaddr + oip);	\
 		return;							\
 	}								\
-	err = map_private_extent_buffer(eb, offset, size,		\
-			&kaddr, &map_start, &map_len);			\
-	if (err) {							\
-		__le##bits val2;					\
-									\
-		val2 = cpu_to_le##bits(val);				\
-		write_extent_buffer(eb, &val2, offset, size);		\
+	token->kaddr = page_address(token->eb->pages[idx]);		\
+	token->offset = idx << PAGE_SHIFT;				\
+	if (oip + size <= PAGE_SIZE) {					\
+		put_unaligned_le##bits(val, token->kaddr + oip);	\
 		return;							\
 	}								\
-	p = kaddr + part_offset - map_start;				\
-	put_unaligned_le##bits(val, p + off);				\
-	token->kaddr = kaddr;						\
-	token->offset = map_start;					\
+	put_unaligned_le##bits(val, lebytes);				\
+	memcpy(token->kaddr + oip, lebytes, part);			\
+	token->kaddr = page_address(token->eb->pages[idx + 1]);		\
+	token->offset = (idx + 1) << PAGE_SHIFT;			\
+	memcpy(token->kaddr, lebytes + part, size - part);		\
 }									\
-void btrfs_set_##bits(struct extent_buffer *eb, void *ptr,		\
+void btrfs_set_##bits(const struct extent_buffer *eb, void *ptr,	\
 		      unsigned long off, u##bits val)			\
 {									\
-	unsigned long part_offset = (unsigned long)ptr;			\
-	unsigned long offset = part_offset + off;			\
-	void *p;							\
-	int err;							\
-	char *kaddr;							\
-	unsigned long map_start;					\
-	unsigned long map_len;						\
-	int size = sizeof(u##bits);					\
-									\
-	err = map_private_extent_buffer(eb, offset, size,		\
-			&kaddr, &map_start, &map_len);			\
-	if (err) {							\
-		__le##bits val2;					\
+	const unsigned long member_offset = (unsigned long)ptr + off;	\
+	const unsigned long oip = offset_in_page(member_offset);	\
+	const unsigned long idx = member_offset >> PAGE_SHIFT;		\
+	char *kaddr = page_address(eb->pages[idx]);			\
+	const int size = sizeof(u##bits);				\
+	const int part = PAGE_SIZE - oip;				\
+	u8 lebytes[sizeof(u##bits)];					\
 									\
-		val2 = cpu_to_le##bits(val);				\
-		write_extent_buffer(eb, &val2, offset, size);		\
+	ASSERT(check_setget_bounds(eb, ptr, off, size));		\
+	if (oip + size <= PAGE_SIZE) {					\
+		put_unaligned_le##bits(val, kaddr + oip);		\
 		return;							\
 	}								\
-	p = kaddr + part_offset - map_start;				\
-	put_unaligned_le##bits(val, p + off);				\
+									\
+	put_unaligned_le##bits(val, lebytes);				\
+	memcpy(kaddr + oip, lebytes, part);				\
+	kaddr = page_address(eb->pages[idx + 1]);			\
+	memcpy(kaddr, lebytes + part, size - part);			\
 }
 
 DEFINE_BTRFS_SETGET_BITS(8)
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 7932d8d07cff..bc73fd670702 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -72,23 +72,32 @@ const char * __attribute_const__ btrfs_decode_error(int errno)
 	char *errstr = "unknown";
 
 	switch (errno) {
-	case -EIO:
+	case -ENOENT:		/* -2 */
+		errstr = "No such entry";
+		break;
+	case -EIO:		/* -5 */
 		errstr = "IO failure";
 		break;
-	case -ENOMEM:
+	case -ENOMEM:		/* -12*/
 		errstr = "Out of memory";
 		break;
-	case -EROFS:
-		errstr = "Readonly filesystem";
-		break;
-	case -EEXIST:
+	case -EEXIST:		/* -17 */
 		errstr = "Object already exists";
 		break;
-	case -ENOSPC:
+	case -ENOSPC:		/* -28 */
 		errstr = "No space left";
 		break;
-	case -ENOENT:
-		errstr = "No such entry";
+	case -EROFS:		/* -30 */
+		errstr = "Readonly filesystem";
+		break;
+	case -EOPNOTSUPP:	/* -95 */
+		errstr = "Operation not supported";
+		break;
+	case -EUCLEAN:		/* -117 */
+		errstr = "Filesystem corrupted";
+		break;
+	case -EDQUOT:		/* -122 */
+		errstr = "Quota exceeded";
 		break;
 	}
 
@@ -1093,10 +1102,7 @@ char *btrfs_get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
 		dirid = btrfs_root_ref_dirid(path->nodes[0], root_ref);
 		btrfs_release_path(path);
 
-		key.objectid = subvol_objectid;
-		key.type = BTRFS_ROOT_ITEM_KEY;
-		key.offset = (u64)-1;
-		fs_root = btrfs_get_fs_root(fs_info, &key, true);
+		fs_root = btrfs_get_fs_root(fs_info, subvol_objectid, true);
 		if (IS_ERR(fs_root)) {
 			ret = PTR_ERR(fs_root);
 			fs_root = NULL;
@@ -1211,7 +1217,6 @@ static int btrfs_fill_super(struct super_block *sb,
 {
 	struct inode *inode;
 	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
-	struct btrfs_key key;
 	int err;
 
 	sb->s_maxbytes = MAX_LFS_FILESIZE;
@@ -1239,10 +1244,7 @@ static int btrfs_fill_super(struct super_block *sb,
 		return err;
 	}
 
-	key.objectid = BTRFS_FIRST_FREE_OBJECTID;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-	inode = btrfs_iget(sb, &key, fs_info->fs_root);
+	inode = btrfs_iget(sb, BTRFS_FIRST_FREE_OBJECTID, fs_info->fs_root);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
 		goto fail_close;
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 2d5498136e5e..b359d4b17658 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -21,6 +21,7 @@
 #include "dev-replace.h"
 #include "qgroup.h"
 #include "block-group.h"
+#include "space-info.h"
 
 #define BTRFS_ROOT_TRANS_TAG 0
 
@@ -141,7 +142,7 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction)
 						 struct btrfs_block_group,
 						 bg_list);
 			list_del_init(&cache->bg_list);
-			btrfs_put_block_group_trimming(cache);
+			btrfs_unfreeze_block_group(cache);
 			btrfs_put_block_group(cache);
 		}
 		WARN_ON(!list_empty(&transaction->dev_update_list));
@@ -348,10 +349,10 @@ loop:
 }
 
 /*
- * this does all the record keeping required to make sure that a reference
- * counted root is properly recorded in a given transaction.  This is required
- * to make sure the old root from before we joined the transaction is deleted
- * when the transaction commits
+ * This does all the record keeping required to make sure that a shareable root
+ * is properly recorded in a given transaction.  This is required to make sure
+ * the old root from before we joined the transaction is deleted when the
+ * transaction commits.
  */
 static int record_root_in_trans(struct btrfs_trans_handle *trans,
 			       struct btrfs_root *root,
@@ -359,7 +360,7 @@ static int record_root_in_trans(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 
-	if ((test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
+	if ((test_bit(BTRFS_ROOT_SHAREABLE, &root->state) &&
 	    root->last_trans < trans->transid) || force) {
 		WARN_ON(root == fs_info->extent_root);
 		WARN_ON(!force && root->commit_root != root->node);
@@ -438,7 +439,7 @@ int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 
-	if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
+	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
 		return 0;
 
 	/*
@@ -503,7 +504,7 @@ static inline bool need_reserve_reloc_root(struct btrfs_root *root)
 	struct btrfs_fs_info *fs_info = root->fs_info;
 
 	if (!fs_info->reloc_ctl ||
-	    !test_bit(BTRFS_ROOT_REF_COWS, &root->state) ||
+	    !test_bit(BTRFS_ROOT_SHAREABLE, &root->state) ||
 	    root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID ||
 	    root->reloc_root)
 		return false;
@@ -523,6 +524,7 @@ start_transaction(struct btrfs_root *root, unsigned int num_items,
 	u64 num_bytes = 0;
 	u64 qgroup_reserved = 0;
 	bool reloc_reserved = false;
+	bool do_chunk_alloc = false;
 	int ret;
 
 	/* Send isn't supposed to start transactions. */
@@ -563,7 +565,8 @@ start_transaction(struct btrfs_root *root, unsigned int num_items,
 		 * refill that amount for whatever is missing in the reserve.
 		 */
 		num_bytes = btrfs_calc_insert_metadata_size(fs_info, num_items);
-		if (delayed_refs_rsv->full == 0) {
+		if (flush == BTRFS_RESERVE_FLUSH_ALL &&
+		    delayed_refs_rsv->full == 0) {
 			delayed_refs_bytes = num_bytes;
 			num_bytes <<= 1;
 		}
@@ -584,6 +587,9 @@ start_transaction(struct btrfs_root *root, unsigned int num_items,
 							  delayed_refs_bytes);
 			num_bytes -= delayed_refs_bytes;
 		}
+
+		if (rsv->space_info->force_alloc)
+			do_chunk_alloc = true;
 	} else if (num_items == 0 && flush == BTRFS_RESERVE_FLUSH_ALL &&
 		   !delayed_refs_rsv->full) {
 		/*
@@ -666,6 +672,19 @@ got_it:
 		current->journal_info = h;
 
 	/*
+	 * If the space_info is marked ALLOC_FORCE then we'll get upgraded to
+	 * ALLOC_FORCE the first run through, and then we won't allocate for
+	 * anybody else who races in later.  We don't care about the return
+	 * value here.
+	 */
+	if (do_chunk_alloc && num_bytes) {
+		u64 flags = h->block_rsv->space_info->flags;
+
+		btrfs_chunk_alloc(h, btrfs_get_alloc_profile(fs_info, flags),
+				  CHUNK_ALLOC_NO_FORCE);
+	}
+
+	/*
 	 * btrfs_record_root_in_trans() needs to alloc new extents, and may
 	 * call btrfs_join_transaction() while we're also starting a
 	 * transaction.
@@ -699,43 +718,10 @@ struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root,
 
 struct btrfs_trans_handle *btrfs_start_transaction_fallback_global_rsv(
 					struct btrfs_root *root,
-					unsigned int num_items,
-					int min_factor)
+					unsigned int num_items)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct btrfs_trans_handle *trans;
-	u64 num_bytes;
-	int ret;
-
-	/*
-	 * We have two callers: unlink and block group removal.  The
-	 * former should succeed even if we will temporarily exceed
-	 * quota and the latter operates on the extent root so
-	 * qgroup enforcement is ignored anyway.
-	 */
-	trans = start_transaction(root, num_items, TRANS_START,
-				  BTRFS_RESERVE_FLUSH_ALL, false);
-	if (!IS_ERR(trans) || PTR_ERR(trans) != -ENOSPC)
-		return trans;
-
-	trans = btrfs_start_transaction(root, 0);
-	if (IS_ERR(trans))
-		return trans;
-
-	num_bytes = btrfs_calc_insert_metadata_size(fs_info, num_items);
-	ret = btrfs_cond_migrate_bytes(fs_info, &fs_info->trans_block_rsv,
-				       num_bytes, min_factor);
-	if (ret) {
-		btrfs_end_transaction(trans);
-		return ERR_PTR(ret);
-	}
-
-	trans->block_rsv = &fs_info->trans_block_rsv;
-	trans->bytes_reserved = num_bytes;
-	trace_btrfs_space_reservation(fs_info, "transaction",
-				      trans->transid, num_bytes, 1);
-
-	return trans;
+	return start_transaction(root, num_items, TRANS_START,
+				 BTRFS_RESERVE_FLUSH_ALL_STEAL, false);
 }
 
 struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root)
@@ -1644,7 +1630,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	}
 
 	key.offset = (u64)-1;
-	pending->snap = btrfs_get_fs_root(fs_info, &key, true);
+	pending->snap = btrfs_get_fs_root(fs_info, objectid, true);
 	if (IS_ERR(pending->snap)) {
 		ret = PTR_ERR(pending->snap);
 		btrfs_abort_transaction(trans, ret);
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 31ae8d273065..bf102e64bfb2 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -193,8 +193,7 @@ struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root,
 						   unsigned int num_items);
 struct btrfs_trans_handle *btrfs_start_transaction_fallback_global_rsv(
 					struct btrfs_root *root,
-					unsigned int num_items,
-					int min_factor);
+					unsigned int num_items);
 struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root);
 struct btrfs_trans_handle *btrfs_join_transaction_spacecache(struct btrfs_root *root);
 struct btrfs_trans_handle *btrfs_join_transaction_nostart(struct btrfs_root *root);
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index a92f8a6dd192..517b44300a05 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -957,10 +957,6 @@ static int check_dev_item(struct extent_buffer *leaf,
 	return 0;
 }
 
-/* Inode item error output has the same format as dir_item_err() */
-#define inode_item_err(eb, slot, fmt, ...)			\
-	dir_item_err(eb, slot, fmt, __VA_ARGS__)
-
 static int check_inode_item(struct extent_buffer *leaf,
 			    struct btrfs_key *key, int slot)
 {
diff --git a/fs/btrfs/tree-defrag.c b/fs/btrfs/tree-defrag.c
index 5f9e2dd413af..16c3a6d2586d 100644
--- a/fs/btrfs/tree-defrag.c
+++ b/fs/btrfs/tree-defrag.c
@@ -35,7 +35,7 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
+	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
 		goto out;
 
 	path = btrfs_alloc_path();
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 02ebdd9edc19..920cee312f4e 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -505,13 +505,8 @@ insert:
 			 */
 			if (S_ISREG(btrfs_inode_mode(eb, src_item)) &&
 			    S_ISREG(btrfs_inode_mode(dst_eb, dst_item)) &&
-			    ino_size != 0) {
-				struct btrfs_map_token token;
-
-				btrfs_init_map_token(&token, dst_eb);
-				btrfs_set_token_inode_size(dst_eb, dst_item,
-							   ino_size, &token);
-			}
+			    ino_size != 0)
+				btrfs_set_inode_size(dst_eb, dst_item, ino_size);
 			goto no_copy;
 		}
 
@@ -555,13 +550,9 @@ no_copy:
 static noinline struct inode *read_one_inode(struct btrfs_root *root,
 					     u64 objectid)
 {
-	struct btrfs_key key;
 	struct inode *inode;
 
-	key.objectid = objectid;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-	inode = btrfs_iget(root->fs_info->sb, &key, root);
+	inode = btrfs_iget(root->fs_info->sb, objectid, root);
 	if (IS_ERR(inode))
 		inode = NULL;
 	return inode;
@@ -3299,6 +3290,7 @@ static void free_log_tree(struct btrfs_trans_handle *trans,
 
 	clear_extent_bits(&log->dirty_log_pages, 0, (u64)-1,
 			  EXTENT_DIRTY | EXTENT_NEW | EXTENT_NEED_WAIT);
+	extent_io_tree_release(&log->log_csum_range);
 	btrfs_put_root(log);
 }
 
@@ -3816,8 +3808,7 @@ static int drop_objectid_items(struct btrfs_trans_handle *trans,
 
 		found_key.offset = 0;
 		found_key.type = 0;
-		ret = btrfs_bin_search(path->nodes[0], &found_key, 0,
-				       &start_slot);
+		ret = btrfs_bin_search(path->nodes[0], &found_key, &start_slot);
 		if (ret < 0)
 			break;
 
@@ -3853,44 +3844,41 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
 		 * just to say 'this inode exists' and a logging
 		 * to say 'update this inode with these values'
 		 */
-		btrfs_set_token_inode_generation(leaf, item, 0, &token);
-		btrfs_set_token_inode_size(leaf, item, logged_isize, &token);
+		btrfs_set_token_inode_generation(&token, item, 0);
+		btrfs_set_token_inode_size(&token, item, logged_isize);
 	} else {
-		btrfs_set_token_inode_generation(leaf, item,
-						 BTRFS_I(inode)->generation,
-						 &token);
-		btrfs_set_token_inode_size(leaf, item, inode->i_size, &token);
-	}
-
-	btrfs_set_token_inode_uid(leaf, item, i_uid_read(inode), &token);
-	btrfs_set_token_inode_gid(leaf, item, i_gid_read(inode), &token);
-	btrfs_set_token_inode_mode(leaf, item, inode->i_mode, &token);
-	btrfs_set_token_inode_nlink(leaf, item, inode->i_nlink, &token);
-
-	btrfs_set_token_timespec_sec(leaf, &item->atime,
-				     inode->i_atime.tv_sec, &token);
-	btrfs_set_token_timespec_nsec(leaf, &item->atime,
-				      inode->i_atime.tv_nsec, &token);
-
-	btrfs_set_token_timespec_sec(leaf, &item->mtime,
-				     inode->i_mtime.tv_sec, &token);
-	btrfs_set_token_timespec_nsec(leaf, &item->mtime,
-				      inode->i_mtime.tv_nsec, &token);
-
-	btrfs_set_token_timespec_sec(leaf, &item->ctime,
-				     inode->i_ctime.tv_sec, &token);
-	btrfs_set_token_timespec_nsec(leaf, &item->ctime,
-				      inode->i_ctime.tv_nsec, &token);
-
-	btrfs_set_token_inode_nbytes(leaf, item, inode_get_bytes(inode),
-				     &token);
-
-	btrfs_set_token_inode_sequence(leaf, item,
-				       inode_peek_iversion(inode), &token);
-	btrfs_set_token_inode_transid(leaf, item, trans->transid, &token);
-	btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token);
-	btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token);
-	btrfs_set_token_inode_block_group(leaf, item, 0, &token);
+		btrfs_set_token_inode_generation(&token, item,
+						 BTRFS_I(inode)->generation);
+		btrfs_set_token_inode_size(&token, item, inode->i_size);
+	}
+
+	btrfs_set_token_inode_uid(&token, item, i_uid_read(inode));
+	btrfs_set_token_inode_gid(&token, item, i_gid_read(inode));
+	btrfs_set_token_inode_mode(&token, item, inode->i_mode);
+	btrfs_set_token_inode_nlink(&token, item, inode->i_nlink);
+
+	btrfs_set_token_timespec_sec(&token, &item->atime,
+				     inode->i_atime.tv_sec);
+	btrfs_set_token_timespec_nsec(&token, &item->atime,
+				      inode->i_atime.tv_nsec);
+
+	btrfs_set_token_timespec_sec(&token, &item->mtime,
+				     inode->i_mtime.tv_sec);
+	btrfs_set_token_timespec_nsec(&token, &item->mtime,
+				      inode->i_mtime.tv_nsec);
+
+	btrfs_set_token_timespec_sec(&token, &item->ctime,
+				     inode->i_ctime.tv_sec);
+	btrfs_set_token_timespec_nsec(&token, &item->ctime,
+				      inode->i_ctime.tv_nsec);
+
+	btrfs_set_token_inode_nbytes(&token, item, inode_get_bytes(inode));
+
+	btrfs_set_token_inode_sequence(&token, item, inode_peek_iversion(inode));
+	btrfs_set_token_inode_transid(&token, item, trans->transid);
+	btrfs_set_token_inode_rdev(&token, item, inode->i_rdev);
+	btrfs_set_token_inode_flags(&token, item, BTRFS_I(inode)->flags);
+	btrfs_set_token_inode_block_group(&token, item, 0);
 }
 
 static int log_inode_item(struct btrfs_trans_handle *trans,
@@ -3916,9 +3904,21 @@ static int log_csums(struct btrfs_trans_handle *trans,
 		     struct btrfs_root *log_root,
 		     struct btrfs_ordered_sum *sums)
 {
+	const u64 lock_end = sums->bytenr + sums->len - 1;
+	struct extent_state *cached_state = NULL;
 	int ret;
 
 	/*
+	 * Serialize logging for checksums. This is to avoid racing with the
+	 * same checksum being logged by another task that is logging another
+	 * file which happens to refer to the same extent as well. Such races
+	 * can leave checksum items in the log with overlapping ranges.
+	 */
+	ret = lock_extent_bits(&log_root->log_csum_range, sums->bytenr,
+			       lock_end, &cached_state);
+	if (ret)
+		return ret;
+	/*
 	 * Due to extent cloning, we might have logged a csum item that covers a
 	 * subrange of a cloned extent, and later we can end up logging a csum
 	 * item for a larger subrange of the same extent or the entire range.
@@ -3928,10 +3928,13 @@ static int log_csums(struct btrfs_trans_handle *trans,
 	 * trim and adjust) any existing csum items in the log for this range.
 	 */
 	ret = btrfs_del_csums(trans, log_root, sums->bytenr, sums->len);
-	if (ret)
-		return ret;
+	if (!ret)
+		ret = btrfs_csum_file_blocks(trans, log_root, sums);
 
-	return btrfs_csum_file_blocks(trans, log_root, sums);
+	unlock_extent_cached(&log_root->log_csum_range, sums->bytenr, lock_end,
+			     &cached_state);
+
+	return ret;
 }
 
 static noinline int copy_items(struct btrfs_trans_handle *trans,
@@ -4164,43 +4167,35 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 	fi = btrfs_item_ptr(leaf, path->slots[0],
 			    struct btrfs_file_extent_item);
 
-	btrfs_set_token_file_extent_generation(leaf, fi, trans->transid,
-					       &token);
+	btrfs_set_token_file_extent_generation(&token, fi, trans->transid);
 	if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
-		btrfs_set_token_file_extent_type(leaf, fi,
-						 BTRFS_FILE_EXTENT_PREALLOC,
-						 &token);
+		btrfs_set_token_file_extent_type(&token, fi,
+						 BTRFS_FILE_EXTENT_PREALLOC);
 	else
-		btrfs_set_token_file_extent_type(leaf, fi,
-						 BTRFS_FILE_EXTENT_REG,
-						 &token);
+		btrfs_set_token_file_extent_type(&token, fi,
+						 BTRFS_FILE_EXTENT_REG);
 
 	block_len = max(em->block_len, em->orig_block_len);
 	if (em->compress_type != BTRFS_COMPRESS_NONE) {
-		btrfs_set_token_file_extent_disk_bytenr(leaf, fi,
-							em->block_start,
-							&token);
-		btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, block_len,
-							   &token);
+		btrfs_set_token_file_extent_disk_bytenr(&token, fi,
+							em->block_start);
+		btrfs_set_token_file_extent_disk_num_bytes(&token, fi, block_len);
 	} else if (em->block_start < EXTENT_MAP_LAST_BYTE) {
-		btrfs_set_token_file_extent_disk_bytenr(leaf, fi,
+		btrfs_set_token_file_extent_disk_bytenr(&token, fi,
 							em->block_start -
-							extent_offset, &token);
-		btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, block_len,
-							   &token);
+							extent_offset);
+		btrfs_set_token_file_extent_disk_num_bytes(&token, fi, block_len);
 	} else {
-		btrfs_set_token_file_extent_disk_bytenr(leaf, fi, 0, &token);
-		btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, 0,
-							   &token);
-	}
-
-	btrfs_set_token_file_extent_offset(leaf, fi, extent_offset, &token);
-	btrfs_set_token_file_extent_num_bytes(leaf, fi, em->len, &token);
-	btrfs_set_token_file_extent_ram_bytes(leaf, fi, em->ram_bytes, &token);
-	btrfs_set_token_file_extent_compression(leaf, fi, em->compress_type,
-						&token);
-	btrfs_set_token_file_extent_encryption(leaf, fi, 0, &token);
-	btrfs_set_token_file_extent_other_encoding(leaf, fi, 0, &token);
+		btrfs_set_token_file_extent_disk_bytenr(&token, fi, 0);
+		btrfs_set_token_file_extent_disk_num_bytes(&token, fi, 0);
+	}
+
+	btrfs_set_token_file_extent_offset(&token, fi, extent_offset);
+	btrfs_set_token_file_extent_num_bytes(&token, fi, em->len);
+	btrfs_set_token_file_extent_ram_bytes(&token, fi, em->ram_bytes);
+	btrfs_set_token_file_extent_compression(&token, fi, em->compress_type);
+	btrfs_set_token_file_extent_encryption(&token, fi, 0);
+	btrfs_set_token_file_extent_other_encoding(&token, fi, 0);
 	btrfs_mark_buffer_dirty(leaf);
 
 	btrfs_release_path(path);
@@ -4336,12 +4331,9 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans,
 			}
 		}
 	}
-	if (ins_nr > 0) {
+	if (ins_nr > 0)
 		ret = copy_items(trans, inode, dst_path, path,
 				 start_slot, ins_nr, 1, 0);
-		if (ret > 0)
-			ret = 0;
-	}
 out:
 	btrfs_release_path(path);
 	btrfs_free_path(dst_path);
@@ -4835,10 +4827,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 
 		btrfs_release_path(path);
 
-		key.objectid = ino;
-		key.type = BTRFS_INODE_ITEM_KEY;
-		key.offset = 0;
-		inode = btrfs_iget(fs_info->sb, &key, root);
+		inode = btrfs_iget(fs_info->sb, ino, root);
 		/*
 		 * If the other inode that had a conflicting dir entry was
 		 * deleted in the current transaction, we need to log its parent
@@ -4847,8 +4836,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 		if (IS_ERR(inode)) {
 			ret = PTR_ERR(inode);
 			if (ret == -ENOENT) {
-				key.objectid = parent;
-				inode = btrfs_iget(fs_info->sb, &key, root);
+				inode = btrfs_iget(fs_info->sb, parent, root);
 				if (IS_ERR(inode)) {
 					ret = PTR_ERR(inode);
 				} else {
@@ -5587,7 +5575,7 @@ process_leaf:
 				continue;
 
 			btrfs_release_path(path);
-			di_inode = btrfs_iget(fs_info->sb, &di_key, root);
+			di_inode = btrfs_iget(fs_info->sb, di_key.objectid, root);
 			if (IS_ERR(di_inode)) {
 				ret = PTR_ERR(di_inode);
 				goto next_dir_inode;
@@ -5713,7 +5701,8 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
 				cur_offset = item_size;
 			}
 
-			dir_inode = btrfs_iget(fs_info->sb, &inode_key, root);
+			dir_inode = btrfs_iget(fs_info->sb, inode_key.objectid,
+					       root);
 			/*
 			 * If the parent inode was deleted, return an error to
 			 * fallback to a transaction commit. This is to prevent
@@ -5780,14 +5769,17 @@ static int log_new_ancestors(struct btrfs_trans_handle *trans,
 		int slot = path->slots[0];
 		struct btrfs_key search_key;
 		struct inode *inode;
+		u64 ino;
 		int ret = 0;
 
 		btrfs_release_path(path);
 
+		ino = found_key.offset;
+
 		search_key.objectid = found_key.offset;
 		search_key.type = BTRFS_INODE_ITEM_KEY;
 		search_key.offset = 0;
-		inode = btrfs_iget(fs_info->sb, &search_key, root);
+		inode = btrfs_iget(fs_info->sb, ino, root);
 		if (IS_ERR(inode))
 			return PTR_ERR(inode);
 
@@ -6132,7 +6124,6 @@ int btrfs_recover_log_trees(struct btrfs_root *log_root_tree)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_key key;
 	struct btrfs_key found_key;
-	struct btrfs_key tmp_key;
 	struct btrfs_root *log;
 	struct btrfs_fs_info *fs_info = log_root_tree->fs_info;
 	struct walk_control wc = {
@@ -6194,11 +6185,8 @@ again:
 			goto error;
 		}
 
-		tmp_key.objectid = found_key.offset;
-		tmp_key.type = BTRFS_ROOT_ITEM_KEY;
-		tmp_key.offset = (u64)-1;
-
-		wc.replay_dest = btrfs_get_fs_root(fs_info, &tmp_key, true);
+		wc.replay_dest = btrfs_get_fs_root(fs_info, found_key.offset,
+						   true);
 		if (IS_ERR(wc.replay_dest)) {
 			ret = PTR_ERR(wc.replay_dest);
 
diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c
index 76671a6bcb61..28525ad7ff8c 100644
--- a/fs/btrfs/uuid-tree.c
+++ b/fs/btrfs/uuid-tree.c
@@ -257,7 +257,6 @@ out:
 static int btrfs_check_uuid_tree_entry(struct btrfs_fs_info *fs_info,
 				       u8 *uuid, u8 type, u64 subvolid)
 {
-	struct btrfs_key key;
 	int ret = 0;
 	struct btrfs_root *subvol_root;
 
@@ -265,10 +264,7 @@ static int btrfs_check_uuid_tree_entry(struct btrfs_fs_info *fs_info,
 	    type != BTRFS_UUID_KEY_RECEIVED_SUBVOL)
 		goto out;
 
-	key.objectid = subvolid;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-	subvol_root = btrfs_get_fs_root(fs_info, &key, true);
+	subvol_root = btrfs_get_fs_root(fs_info, subvolid, true);
 	if (IS_ERR(subvol_root)) {
 		ret = PTR_ERR(subvol_root);
 		if (ret == -ENOENT)
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index c1909e5f4506..0d6e785bcb98 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -280,10 +280,9 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
  * ============
  *
  * uuid_mutex
- *   volume_mutex
- *     device_list_mutex
- *       chunk_mutex
- *     balance_mutex
+ *   device_list_mutex
+ *     chunk_mutex
+ *   balance_mutex
  *
  *
  * Exclusive operations, BTRFS_FS_EXCL_OP
@@ -1042,6 +1041,8 @@ again:
 							&device->dev_state)) {
 			if (!test_bit(BTRFS_DEV_STATE_REPLACE_TGT,
 			     &device->dev_state) &&
+			    !test_bit(BTRFS_DEV_STATE_MISSING,
+				      &device->dev_state) &&
 			     (!latest_dev ||
 			      device->generation > latest_dev->generation)) {
 				latest_dev = device;
@@ -1185,7 +1186,6 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices,
 {
 	struct btrfs_device *device;
 	struct btrfs_device *latest_dev = NULL;
-	int ret = 0;
 
 	flags |= FMODE_EXCL;
 
@@ -1198,16 +1198,15 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices,
 		    device->generation > latest_dev->generation)
 			latest_dev = device;
 	}
-	if (fs_devices->open_devices == 0) {
-		ret = -EINVAL;
-		goto out;
-	}
+	if (fs_devices->open_devices == 0)
+		return -EINVAL;
+
 	fs_devices->opened = 1;
 	fs_devices->latest_bdev = latest_dev->bdev;
 	fs_devices->total_rw_bytes = 0;
 	fs_devices->chunk_alloc_policy = BTRFS_CHUNK_ALLOC_REGULAR;
-out:
-	return ret;
+
+	return 0;
 }
 
 static int devid_cmp(void *priv, struct list_head *a, struct list_head *b)
@@ -1251,49 +1250,48 @@ void btrfs_release_disk_super(struct btrfs_super_block *super)
 	put_page(page);
 }
 
-static int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
-				 struct page **page,
-				 struct btrfs_super_block **disk_super)
+static struct btrfs_super_block *btrfs_read_disk_super(struct block_device *bdev,
+						       u64 bytenr)
 {
+	struct btrfs_super_block *disk_super;
+	struct page *page;
 	void *p;
 	pgoff_t index;
 
 	/* make sure our super fits in the device */
 	if (bytenr + PAGE_SIZE >= i_size_read(bdev->bd_inode))
-		return 1;
+		return ERR_PTR(-EINVAL);
 
 	/* make sure our super fits in the page */
-	if (sizeof(**disk_super) > PAGE_SIZE)
-		return 1;
+	if (sizeof(*disk_super) > PAGE_SIZE)
+		return ERR_PTR(-EINVAL);
 
 	/* make sure our super doesn't straddle pages on disk */
 	index = bytenr >> PAGE_SHIFT;
-	if ((bytenr + sizeof(**disk_super) - 1) >> PAGE_SHIFT != index)
-		return 1;
+	if ((bytenr + sizeof(*disk_super) - 1) >> PAGE_SHIFT != index)
+		return ERR_PTR(-EINVAL);
 
 	/* pull in the page with our super */
-	*page = read_cache_page_gfp(bdev->bd_inode->i_mapping,
-				   index, GFP_KERNEL);
+	page = read_cache_page_gfp(bdev->bd_inode->i_mapping, index, GFP_KERNEL);
 
-	if (IS_ERR(*page))
-		return 1;
+	if (IS_ERR(page))
+		return ERR_CAST(page);
 
-	p = page_address(*page);
+	p = page_address(page);
 
 	/* align our pointer to the offset of the super block */
-	*disk_super = p + offset_in_page(bytenr);
+	disk_super = p + offset_in_page(bytenr);
 
-	if (btrfs_super_bytenr(*disk_super) != bytenr ||
-	    btrfs_super_magic(*disk_super) != BTRFS_MAGIC) {
+	if (btrfs_super_bytenr(disk_super) != bytenr ||
+	    btrfs_super_magic(disk_super) != BTRFS_MAGIC) {
 		btrfs_release_disk_super(p);
-		return 1;
+		return ERR_PTR(-EINVAL);
 	}
 
-	if ((*disk_super)->label[0] &&
-		(*disk_super)->label[BTRFS_LABEL_SIZE - 1])
-		(*disk_super)->label[BTRFS_LABEL_SIZE - 1] = '\0';
+	if (disk_super->label[0] && disk_super->label[BTRFS_LABEL_SIZE - 1])
+		disk_super->label[BTRFS_LABEL_SIZE - 1] = 0;
 
-	return 0;
+	return disk_super;
 }
 
 int btrfs_forget_devices(const char *path)
@@ -1319,7 +1317,6 @@ struct btrfs_device *btrfs_scan_one_device(const char *path, fmode_t flags,
 	bool new_device_added = false;
 	struct btrfs_device *device = NULL;
 	struct block_device *bdev;
-	struct page *page;
 	u64 bytenr;
 
 	lockdep_assert_held(&uuid_mutex);
@@ -1337,8 +1334,9 @@ struct btrfs_device *btrfs_scan_one_device(const char *path, fmode_t flags,
 	if (IS_ERR(bdev))
 		return ERR_CAST(bdev);
 
-	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super)) {
-		device = ERR_PTR(-EINVAL);
+	disk_super = btrfs_read_disk_super(bdev, bytenr);
+	if (IS_ERR(disk_super)) {
+		device = ERR_CAST(disk_super);
 		goto error_bdev_put;
 	}
 
@@ -2663,8 +2661,18 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 		ret = btrfs_commit_transaction(trans);
 	}
 
-	/* Update ctime/mtime for libblkid */
+	/*
+	 * Now that we have written a new super block to this device, check all
+	 * other fs_devices list if device_path alienates any other scanned
+	 * device.
+	 * We can ignore the return value as it typically returns -EINVAL and
+	 * only succeeds if the device was an alien.
+	 */
+	btrfs_forget_devices(device_path);
+
+	/* Update ctime/mtime for blkid or udev */
 	update_dev_time(device_path);
+
 	return ret;
 
 error_sysfs: