Merge tag 'for-4.20-part1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "This is the first batch with fixes and some nice performance
  improvements.

  Preliminary results show eg. more files/sec in fsmark, better perf on
  multi-threaded workloads (filebench, dbench), fewer context switches
  and overall better memory allocation characteristics (multiple
  benchmarks).

  Apart from general performance, there's an improvement for qgroups +
  balance workload that's been troubling our users.

  Note for stable: there are 20+ patches tagged for stable, out of 90.
  Not all of them apply cleanly on all stable versions but the conflicts
  are mostly due to simple cleanups and resolving should be obvious. The
  fixes are otherwise independent.

  Performance improvements:

   - transition between blocking and spinning modes of path is gone,
     which originally resulted to more unnecessary wakeups and updates
     to the path locks, the effects are measurable and improve latency
     and scalability

   - qgroups: first batch of changes that should speedup balancing with
     qgroups on, skip quota accounting on unchanged subtrees, overall
     gain is about 30+% in runtime

   - use rb-tree with cached first node for several structures, small
     improvement to avoid pointer chasing

  Fixes:

   - trim
      - fix: some blockgroups could have been missed if their logical
        address was past the total filesystem size (ie. after a lot of
        balancing)
      - better error reporting, after processing blockgroups and whole
        device
      - fix: continue trimming block groups after an error is
        encountered
      - check for trim support of the device earlier and avoid some
        unnecessary work
      - less interaction with transaction commit that improves latency
        on slower storage (eg. image files over NFS)

   - fsync
      - fix warning when replaying log after fsync of a O_TMPFILE
      - fix wrong dentries after fsync of file that got its parent
        replaced

   - qgroups: fix rescan that might misc some dirty groups

   - don't clean dirty pages during buffered writes, this could lead to
     lost updates in some corner cases

   - some block groups could have been delayed in creation, if the
     allocation triggered another one

   - error handling improvements

  Cleanups:

   - removed unused struct members and variables

   - function return type cleanups

   - delayed refs code refactoring

   - protect against deadlock that could be caused by crafted image that
     tries to allocate from a tree that's locked already"

* tag 'for-4.20-part1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (93 commits)
  btrfs: switch return_bigger to bool in find_ref_head
  btrfs: remove fs_info from btrfs_should_throttle_delayed_refs
  btrfs: remove fs_info from btrfs_check_space_for_delayed_refs
  btrfs: delayed-ref: pass delayed_refs directly to btrfs_delayed_ref_lock
  btrfs: delayed-ref: pass delayed_refs directly to btrfs_select_ref_head
  btrfs: qgroup: move the qgroup->members check out from (!qgroup)'s else branch
  btrfs: relocation: Remove redundant tree level check
  btrfs: relocation: Cleanup while loop using rbtree_postorder_for_each_entry_safe
  btrfs: qgroup: Avoid calling qgroup functions if qgroup is not enabled
  Btrfs: fix wrong dentries after fsync of file that got its parent replaced
  Btrfs: fix warning when replaying log after fsync of a tmpfile
  btrfs: drop min_size from evict_refill_and_join
  btrfs: assert on non-empty delayed iputs
  btrfs: make sure we create all new block groups
  btrfs: reset max_extent_size on clear in a bitmap
  btrfs: protect space cache inode alloc with GFP_NOFS
  btrfs: release metadata before running delayed refs
  Btrfs: kill btrfs_clear_path_blocking
  btrfs: dev-replace: remove pointless assert in write unlock
  btrfs: dev-replace: move replace members out of fs_info
  ...

fs/btrfs/backref.c

@@ -112,11 +112,11 @@ static int find_extent_in_eb(const struct extent_buffer *eb,
 }
 
 struct preftree {
-	struct rb_root root;
+	struct rb_root_cached root;
 	unsigned int count;
 };
 
-#define PREFTREE_INIT	{ .root = RB_ROOT, .count = 0 }
+#define PREFTREE_INIT	{ .root = RB_ROOT_CACHED, .count = 0 }
 
 struct preftrees {
 	struct preftree direct;    /* BTRFS_SHARED_[DATA|BLOCK]_REF_KEY */
@@ -225,14 +225,15 @@ static void prelim_ref_insert(const struct btrfs_fs_info *fs_info,
 			      struct prelim_ref *newref,
 			      struct share_check *sc)
 {
-	struct rb_root *root;
+	struct rb_root_cached *root;
 	struct rb_node **p;
 	struct rb_node *parent = NULL;
 	struct prelim_ref *ref;
 	int result;
+	bool leftmost = true;
 
 	root = &preftree->root;
-	p = &root->rb_node;
+	p = &root->rb_root.rb_node;
 
 	while (*p) {
 		parent = *p;
@@ -242,6 +243,7 @@ static void prelim_ref_insert(const struct btrfs_fs_info *fs_info,
 			p = &(*p)->rb_left;
 		} else if (result > 0) {
 			p = &(*p)->rb_right;
+			leftmost = false;
 		} else {
 			/* Identical refs, merge them and free @newref */
 			struct extent_inode_elem *eie = ref->inode_list;
@@ -272,7 +274,7 @@ static void prelim_ref_insert(const struct btrfs_fs_info *fs_info,
 	preftree->count++;
 	trace_btrfs_prelim_ref_insert(fs_info, newref, NULL, preftree->count);
 	rb_link_node(&newref->rbnode, parent, p);
-	rb_insert_color(&newref->rbnode, root);
+	rb_insert_color_cached(&newref->rbnode, root, leftmost);
 }
 
 /*
@@ -283,11 +285,11 @@ static void prelim_release(struct preftree *preftree)
 {
 	struct prelim_ref *ref, *next_ref;
 
-	rbtree_postorder_for_each_entry_safe(ref, next_ref, &preftree->root,
-					     rbnode)
+	rbtree_postorder_for_each_entry_safe(ref, next_ref,
+					     &preftree->root.rb_root, rbnode)
 		free_pref(ref);
 
-	preftree->root = RB_ROOT;
+	preftree->root = RB_ROOT_CACHED;
 	preftree->count = 0;
 }
 
@@ -627,7 +629,7 @@ static int resolve_indirect_refs(struct btrfs_fs_info *fs_info,
 	 * freeing the entire indirect tree when we're done.  In some test
 	 * cases, the tree can grow quite large (~200k objects).
 	 */
-	while ((rnode = rb_first(&preftrees->indirect.root))) {
+	while ((rnode = rb_first_cached(&preftrees->indirect.root))) {
 		struct prelim_ref *ref;
 
 		ref = rb_entry(rnode, struct prelim_ref, rbnode);
@@ -637,7 +639,7 @@ static int resolve_indirect_refs(struct btrfs_fs_info *fs_info,
 			goto out;
 		}
 
-		rb_erase(&ref->rbnode, &preftrees->indirect.root);
+		rb_erase_cached(&ref->rbnode, &preftrees->indirect.root);
 		preftrees->indirect.count--;
 
 		if (ref->count == 0) {
@@ -717,9 +719,9 @@ static int add_missing_keys(struct btrfs_fs_info *fs_info,
 	struct preftree *tree = &preftrees->indirect_missing_keys;
 	struct rb_node *node;
 
-	while ((node = rb_first(&tree->root))) {
+	while ((node = rb_first_cached(&tree->root))) {
 		ref = rb_entry(node, struct prelim_ref, rbnode);
-		rb_erase(node, &tree->root);
+		rb_erase_cached(node, &tree->root);
 
 		BUG_ON(ref->parent);	/* should not be a direct ref */
 		BUG_ON(ref->key_for_search.type);
@@ -769,7 +771,7 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info,
 		btrfs_disk_key_to_cpu(&tmp_op_key, &extent_op->key);
 
 	spin_lock(&head->lock);
-	for (n = rb_first(&head->ref_tree); n; n = rb_next(n)) {
+	for (n = rb_first_cached(&head->ref_tree); n; n = rb_next(n)) {
 		node = rb_entry(n, struct btrfs_delayed_ref_node,
 				ref_node);
 		if (node->seq > seq)
@@ -1229,14 +1231,14 @@ again:
 	if (ret)
 		goto out;
 
-	WARN_ON(!RB_EMPTY_ROOT(&preftrees.indirect_missing_keys.root));
+	WARN_ON(!RB_EMPTY_ROOT(&preftrees.indirect_missing_keys.root.rb_root));
 
 	ret = resolve_indirect_refs(fs_info, path, time_seq, &preftrees,
 				    extent_item_pos, total_refs, sc, ignore_offset);
 	if (ret)
 		goto out;
 
-	WARN_ON(!RB_EMPTY_ROOT(&preftrees.indirect.root));
+	WARN_ON(!RB_EMPTY_ROOT(&preftrees.indirect.root.rb_root));
 
 	/*
 	 * This walks the tree of merged and resolved refs. Tree blocks are
@@ -1245,7 +1247,7 @@ again:
 	 *
 	 * We release the entire tree in one go before returning.
 	 */
-	node = rb_first(&preftrees.direct.root);
+	node = rb_first_cached(&preftrees.direct.root);
 	while (node) {
 		ref = rb_entry(node, struct prelim_ref, rbnode);
 		node = rb_next(&ref->rbnode);
@@ -1468,7 +1470,7 @@ int btrfs_check_shared(struct btrfs_root *root, u64 inum, u64 bytenr)
 	struct seq_list elem = SEQ_LIST_INIT(elem);
 	int ret = 0;
 	struct share_check shared = {
-		.root_objectid = root->objectid,
+		.root_objectid = root->root_key.objectid,
 		.inum = inum,
 		.share_count = 0,
 	};
@@ -2031,7 +2033,8 @@ static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root,
 			/* path must be released before calling iterate()! */
 			btrfs_debug(fs_root->fs_info,
 				"following ref at offset %u for inode %llu in tree %llu",
-				cur, found_key.objectid, fs_root->objectid);
+				cur, found_key.objectid,
+				fs_root->root_key.objectid);
 			ret = iterate(parent, name_len,
 				      (unsigned long)(iref + 1), eb, ctx);
 			if (ret)

fs/btrfs/btrfs_inode.h

@@ -206,7 +206,7 @@ static inline struct btrfs_inode *BTRFS_I(const struct inode *inode)
 static inline unsigned long btrfs_inode_hash(u64 objectid,
 					     const struct btrfs_root *root)
 {
-	u64 h = objectid ^ (root->objectid * GOLDEN_RATIO_PRIME);
+	u64 h = objectid ^ (root->root_key.objectid * GOLDEN_RATIO_PRIME);
 
 #if BITS_PER_LONG == 32
 	h = (h >> 32) ^ (h & 0xffffffff);
@@ -339,15 +339,15 @@ static inline void btrfs_print_data_csum_error(struct btrfs_inode *inode,
 	struct btrfs_root *root = inode->root;
 
 	/* Output minus objectid, which is more meaningful */
-	if (root->objectid >= BTRFS_LAST_FREE_OBJECTID)
+	if (root->root_key.objectid >= BTRFS_LAST_FREE_OBJECTID)
 		btrfs_warn_rl(root->fs_info,
 	"csum failed root %lld ino %lld off %llu csum 0x%08x expected csum 0x%08x mirror %d",
-			root->objectid, btrfs_ino(inode),
+			root->root_key.objectid, btrfs_ino(inode),
 			logical_start, csum, csum_expected, mirror_num);
 	else
 		btrfs_warn_rl(root->fs_info,
 	"csum failed root %llu ino %llu off %llu csum 0x%08x expected csum 0x%08x mirror %d",
-			root->objectid, btrfs_ino(inode),
+			root->root_key.objectid, btrfs_ino(inode),
 			logical_start, csum, csum_expected, mirror_num);
 }
 

fs/btrfs/check-integrity.c

@@ -1594,6 +1594,7 @@ static int btrfsic_read_block(struct btrfsic_state *state,
 {
 	unsigned int num_pages;
 	unsigned int i;
+	size_t size;
 	u64 dev_bytenr;
 	int ret;
 
@@ -1608,9 +1609,8 @@ static int btrfsic_read_block(struct btrfsic_state *state,
 
 	num_pages = (block_ctx->len + (u64)PAGE_SIZE - 1) >>
 		    PAGE_SHIFT;
-	block_ctx->mem_to_free = kcalloc(sizeof(*block_ctx->datav) +
-						sizeof(*block_ctx->pagev),
-					 num_pages, GFP_NOFS);
+	size = sizeof(*block_ctx->datav) + sizeof(*block_ctx->pagev);
+	block_ctx->mem_to_free = kcalloc(num_pages, size, GFP_NOFS);
 	if (!block_ctx->mem_to_free)
 		return -ENOMEM;
 	block_ctx->datav = block_ctx->mem_to_free;

fs/btrfs/compression.c

@@ -528,7 +528,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 				 int mirror_num, unsigned long bio_flags)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct extent_io_tree *tree;
 	struct extent_map_tree *em_tree;
 	struct compressed_bio *cb;
 	unsigned long compressed_len;
@@ -545,7 +544,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	int faili = 0;
 	u32 *sums;
 
-	tree = &BTRFS_I(inode)->io_tree;
 	em_tree = &BTRFS_I(inode)->extent_tree;
 
 	/* we need the actual starting offset of this extent in the file */

fs/btrfs/ctree.c

@@ -52,42 +52,6 @@ noinline void btrfs_set_path_blocking(struct btrfs_path *p)
 	}
 }
 
-/*
- * reset all the locked nodes in the patch to spinning locks.
- *
- * held is used to keep lockdep happy, when lockdep is enabled
- * we set held to a blocking lock before we go around and
- * retake all the spinlocks in the path.  You can safely use NULL
- * for held
- */
-noinline void btrfs_clear_path_blocking(struct btrfs_path *p,
-					struct extent_buffer *held, int held_rw)
-{
-	int i;
-
-	if (held) {
-		btrfs_set_lock_blocking_rw(held, held_rw);
-		if (held_rw == BTRFS_WRITE_LOCK)
-			held_rw = BTRFS_WRITE_LOCK_BLOCKING;
-		else if (held_rw == BTRFS_READ_LOCK)
-			held_rw = BTRFS_READ_LOCK_BLOCKING;
-	}
-	btrfs_set_path_blocking(p);
-
-	for (i = BTRFS_MAX_LEVEL - 1; i >= 0; i--) {
-		if (p->nodes[i] && p->locks[i]) {
-			btrfs_clear_lock_blocking_rw(p->nodes[i], p->locks[i]);
-			if (p->locks[i] == BTRFS_WRITE_LOCK_BLOCKING)
-				p->locks[i] = BTRFS_WRITE_LOCK;
-			else if (p->locks[i] == BTRFS_READ_LOCK_BLOCKING)
-				p->locks[i] = BTRFS_READ_LOCK;
-		}
-	}
-
-	if (held)
-		btrfs_clear_lock_blocking_rw(held, held_rw);
-}
-
 /* this also releases the path */
 void btrfs_free_path(struct btrfs_path *p)
 {
@@ -207,7 +171,7 @@ static void add_root_to_dirty_list(struct btrfs_root *root)
 	spin_lock(&fs_info->trans_lock);
 	if (!test_and_set_bit(BTRFS_ROOT_DIRTY, &root->state)) {
 		/* Want the extent tree to be the last on the list */
-		if (root->objectid == BTRFS_EXTENT_TREE_OBJECTID)
+		if (root->root_key.objectid == BTRFS_EXTENT_TREE_OBJECTID)
 			list_move_tail(&root->dirty_list,
 				       &fs_info->dirty_cowonly_roots);
 		else
@@ -1306,7 +1270,6 @@ tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
 		}
 	}
 
-	btrfs_clear_path_blocking(path, NULL, BTRFS_READ_LOCK);
 	btrfs_tree_read_unlock_blocking(eb);
 	free_extent_buffer(eb);
 
@@ -1815,8 +1778,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 	int orig_slot = path->slots[level];
 	u64 orig_ptr;
 
-	if (level == 0)
-		return 0;
+	ASSERT(level > 0);
 
 	mid = path->nodes[level];
 
@@ -2483,7 +2445,6 @@ setup_nodes_for_search(struct btrfs_trans_handle *trans,
 		btrfs_set_path_blocking(p);
 		reada_for_balance(fs_info, p, level);
 		sret = split_node(trans, root, p, level);
-		btrfs_clear_path_blocking(p, NULL, 0);
 
 		BUG_ON(sret > 0);
 		if (sret) {
@@ -2504,7 +2465,6 @@ setup_nodes_for_search(struct btrfs_trans_handle *trans,
 		btrfs_set_path_blocking(p);
 		reada_for_balance(fs_info, p, level);
 		sret = balance_level(trans, root, p, level);
-		btrfs_clear_path_blocking(p, NULL, 0);
 
 		if (sret) {
 			ret = sret;
@@ -2789,7 +2749,10 @@ again:
 		}
 cow_done:
 		p->nodes[level] = b;
-		btrfs_clear_path_blocking(p, NULL, 0);
+		/*
+		 * Leave path with blocking locks to avoid massive
+		 * lock context switch, this is made on purpose.
+		 */
 
 		/*
 		 * we have a lock on b and as long as we aren't changing
@@ -2871,8 +2834,6 @@ cow_done:
 					if (!err) {
 						btrfs_set_path_blocking(p);
 						btrfs_tree_lock(b);
-						btrfs_clear_path_blocking(p, b,
-								  BTRFS_WRITE_LOCK);
 					}
 					p->locks[level] = BTRFS_WRITE_LOCK;
 				} else {
@@ -2880,8 +2841,6 @@ cow_done:
 					if (!err) {
 						btrfs_set_path_blocking(p);
 						btrfs_tree_read_lock(b);
-						btrfs_clear_path_blocking(p, b,
-								  BTRFS_READ_LOCK);
 					}
 					p->locks[level] = BTRFS_READ_LOCK;
 				}
@@ -2900,7 +2859,6 @@ cow_done:
 				btrfs_set_path_blocking(p);
 				err = split_leaf(trans, root, key,
 						 p, ins_len, ret == 0);
-				btrfs_clear_path_blocking(p, NULL, 0);
 
 				BUG_ON(err > 0);
 				if (err) {
@@ -2910,7 +2868,7 @@ cow_done:
 			}
 			if (!p->search_for_split)
 				unlock_up(p, level, lowest_unlock,
-					  min_write_lock_level, &write_lock_level);
+					  min_write_lock_level, NULL);
 			goto done;
 		}
 	}
@@ -2961,13 +2919,16 @@ int btrfs_search_old_slot(struct btrfs_root *root, const struct btrfs_key *key,
 
 again:
 	b = get_old_root(root, time_seq);
+	if (!b) {
+		ret = -EIO;
+		goto done;
+	}
 	level = btrfs_header_level(b);
 	p->locks[level] = BTRFS_READ_LOCK;
 
 	while (b) {
 		level = btrfs_header_level(b);
 		p->nodes[level] = b;
-		btrfs_clear_path_blocking(p, NULL, 0);
 
 		/*
 		 * we have a lock on b and as long as we aren't changing
@@ -3013,8 +2974,6 @@ again:
 			if (!err) {
 				btrfs_set_path_blocking(p);
 				btrfs_tree_read_lock(b);
-				btrfs_clear_path_blocking(p, b,
-							  BTRFS_READ_LOCK);
 			}
 			b = tree_mod_log_rewind(fs_info, p, b, time_seq);
 			if (!b) {
@@ -5198,7 +5157,6 @@ find_next_key:
 		path->locks[level - 1] = BTRFS_READ_LOCK;
 		path->nodes[level - 1] = cur;
 		unlock_up(path, level, 1, 0, NULL);
-		btrfs_clear_path_blocking(path, NULL, 0);
 	}
 out:
 	path->keep_locks = keep_locks;
@@ -5783,8 +5741,6 @@ again:
 			if (!ret) {
 				btrfs_set_path_blocking(path);
 				btrfs_tree_read_lock(next);
-				btrfs_clear_path_blocking(path, next,
-							  BTRFS_READ_LOCK);
 			}
 			next_rw_lock = BTRFS_READ_LOCK;
 		}
@@ -5820,8 +5776,6 @@ again:
 			if (!ret) {
 				btrfs_set_path_blocking(path);
 				btrfs_tree_read_lock(next);
-				btrfs_clear_path_blocking(path, next,
-							  BTRFS_READ_LOCK);
 			}
 			next_rw_lock = BTRFS_READ_LOCK;
 		}

fs/btrfs/ctree.h

@@ -41,12 +41,6 @@ extern struct kmem_cache *btrfs_path_cachep;
 extern struct kmem_cache *btrfs_free_space_cachep;
 struct btrfs_ordered_sum;
 
-#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-#define STATIC noinline
-#else
-#define STATIC static noinline
-#endif
-
 #define BTRFS_MAGIC 0x4D5F53665248425FULL /* ascii _BHRfS_M, no null */
 
 #define BTRFS_MAX_MIRRORS 3
@@ -367,11 +361,13 @@ struct btrfs_dev_replace {
 
 	struct mutex lock_finishing_cancel_unmount;
 	rwlock_t lock;
-	atomic_t read_locks;
 	atomic_t blocking_readers;
 	wait_queue_head_t read_lock_wq;
 
 	struct btrfs_scrub_progress scrub_progress;
+
+	struct percpu_counter bio_counter;
+	wait_queue_head_t replace_wait;
 };
 
 /* For raid type sysfs entries */
@@ -1094,9 +1090,6 @@ struct btrfs_fs_info {
 	/* device replace state */
 	struct btrfs_dev_replace dev_replace;
 
-	struct percpu_counter bio_counter;
-	wait_queue_head_t replace_wait;
-
 	struct semaphore uuid_tree_rescan_sem;
 
 	/* Used to reclaim the metadata space in the background. */
@@ -1202,18 +1195,12 @@ struct btrfs_root {
 	int last_log_commit;
 	pid_t log_start_pid;
 
-	u64 objectid;
 	u64 last_trans;
 
 	u32 type;
 
 	u64 highest_objectid;
 
-#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-	/* only used with CONFIG_BTRFS_FS_RUN_SANITY_TESTS is enabled */
-	u64 alloc_bytenr;
-#endif
-
 	u64 defrag_trans_start;
 	struct btrfs_key defrag_progress;
 	struct btrfs_key defrag_max;
@@ -1286,6 +1273,10 @@ struct btrfs_root {
 	spinlock_t qgroup_meta_rsv_lock;
 	u64 qgroup_meta_rsv_pertrans;
 	u64 qgroup_meta_rsv_prealloc;
+
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+	u64 alloc_bytenr;
+#endif
 };
 
 struct btrfs_file_private {
@@ -2607,10 +2598,8 @@ static inline u64 btrfs_calc_trunc_metadata_size(struct btrfs_fs_info *fs_info,
 	return (u64)fs_info->nodesize * BTRFS_MAX_LEVEL * num_items;
 }
 
-int btrfs_should_throttle_delayed_refs(struct btrfs_trans_handle *trans,
-				       struct btrfs_fs_info *fs_info);
-int btrfs_check_space_for_delayed_refs(struct btrfs_trans_handle *trans,
-				       struct btrfs_fs_info *fs_info);
+int btrfs_should_throttle_delayed_refs(struct btrfs_trans_handle *trans);
+int btrfs_check_space_for_delayed_refs(struct btrfs_trans_handle *trans);
 void btrfs_dec_block_group_reservations(struct btrfs_fs_info *fs_info,
 					 const u64 start);
 void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg);
@@ -2771,7 +2760,7 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
 			   enum btrfs_reserve_flush_enum flush);
 int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv,
 			    struct btrfs_block_rsv *dst_rsv, u64 num_bytes,
-			    int update_size);
+			    bool update_size);
 int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 			     struct btrfs_block_rsv *dest, u64 num_bytes,
 			     int min_factor);
@@ -2877,8 +2866,6 @@ void btrfs_release_path(struct btrfs_path *p);
 struct btrfs_path *btrfs_alloc_path(void);
 void btrfs_free_path(struct btrfs_path *p);
 void btrfs_set_path_blocking(struct btrfs_path *p);
-void btrfs_clear_path_blocking(struct btrfs_path *p,
-			       struct extent_buffer *held, int held_rw);
 void btrfs_unlock_up_safe(struct btrfs_path *p, int level);
 
 int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
@@ -3021,8 +3008,7 @@ int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info,
 /* dir-item.c */
 int btrfs_check_dir_item_collision(struct btrfs_root *root, u64 dir,
 			  const char *name, int name_len);
-int btrfs_insert_dir_item(struct btrfs_trans_handle *trans,
-			  struct btrfs_root *root, const char *name,
+int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, const char *name,
 			  int name_len, struct btrfs_inode *dir,
 			  struct btrfs_key *location, u8 type, u64 index);
 struct btrfs_dir_item *btrfs_lookup_dir_item(struct btrfs_trans_handle *trans,
@@ -3180,8 +3166,8 @@ void __cold btrfs_destroy_cachep(void);
 struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
 			 struct btrfs_root *root, int *was_new);
 struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
-		struct page *page, size_t pg_offset,
-		u64 start, u64 end, int create);
+				    struct page *page, size_t pg_offset,
+				    u64 start, u64 end, int create);
 int btrfs_update_inode(struct btrfs_trans_handle *trans,
 			      struct btrfs_root *root,
 			      struct inode *inode);
@@ -3201,9 +3187,6 @@ int btrfs_prealloc_file_range_trans(struct inode *inode,
 				    u64 start, u64 num_bytes, u64 min_size,
 				    loff_t actual_len, u64 *alloc_hint);
 extern const struct dentry_operations btrfs_dentry_operations;
-#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-void btrfs_test_inode_set_ops(struct inode *inode);
-#endif
 
 /* ioctl.c */
 long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
@@ -3716,18 +3699,19 @@ static inline int btrfs_defrag_cancelled(struct btrfs_fs_info *fs_info)
 
 /* Sanity test specific functions */
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+void btrfs_test_inode_set_ops(struct inode *inode);
 void btrfs_test_destroy_inode(struct inode *inode);
-#endif
 
 static inline int btrfs_is_testing(struct btrfs_fs_info *fs_info)
 {
-#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-	if (unlikely(test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO,
-			      &fs_info->fs_state)))
-		return 1;
-#endif
+	return test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state);
+}
+#else
+static inline int btrfs_is_testing(struct btrfs_fs_info *fs_info)
+{
 	return 0;
 }
+#endif
 
 static inline void cond_wake_up(struct wait_queue_head *wq)
 {

fs/btrfs/delayed-inode.c

@@ -42,8 +42,8 @@ static inline void btrfs_init_delayed_node(
 	delayed_node->root = root;
 	delayed_node->inode_id = inode_id;
 	refcount_set(&delayed_node->refs, 0);
-	delayed_node->ins_root = RB_ROOT;
-	delayed_node->del_root = RB_ROOT;
+	delayed_node->ins_root = RB_ROOT_CACHED;
+	delayed_node->del_root = RB_ROOT_CACHED;
 	mutex_init(&delayed_node->mutex);
 	INIT_LIST_HEAD(&delayed_node->n_list);
 	INIT_LIST_HEAD(&delayed_node->p_list);
@@ -390,7 +390,7 @@ static struct btrfs_delayed_item *__btrfs_lookup_delayed_insertion_item(
 					struct btrfs_delayed_node *delayed_node,
 					struct btrfs_key *key)
 {
-	return __btrfs_lookup_delayed_item(&delayed_node->ins_root, key,
+	return __btrfs_lookup_delayed_item(&delayed_node->ins_root.rb_root, key,
 					   NULL, NULL);
 }
 
@@ -400,9 +400,10 @@ static int __btrfs_add_delayed_item(struct btrfs_delayed_node *delayed_node,
 {
 	struct rb_node **p, *node;
 	struct rb_node *parent_node = NULL;
-	struct rb_root *root;
+	struct rb_root_cached *root;
 	struct btrfs_delayed_item *item;
 	int cmp;
+	bool leftmost = true;
 
 	if (action == BTRFS_DELAYED_INSERTION_ITEM)
 		root = &delayed_node->ins_root;
@@ -410,7 +411,7 @@ static int __btrfs_add_delayed_item(struct btrfs_delayed_node *delayed_node,
 		root = &delayed_node->del_root;
 	else
 		BUG();
-	p = &root->rb_node;
+	p = &root->rb_root.rb_node;
 	node = &ins->rb_node;
 
 	while (*p) {
@@ -419,16 +420,18 @@ static int __btrfs_add_delayed_item(struct btrfs_delayed_node *delayed_node,
 				 rb_node);
 
 		cmp = btrfs_comp_cpu_keys(&item->key, &ins->key);
-		if (cmp < 0)
+		if (cmp < 0) {
 			p = &(*p)->rb_right;
-		else if (cmp > 0)
+			leftmost = false;
+		} else if (cmp > 0) {
 			p = &(*p)->rb_left;
-		else
+		} else {
 			return -EEXIST;
+		}
 	}
 
 	rb_link_node(node, parent_node, p);
-	rb_insert_color(node, root);
+	rb_insert_color_cached(node, root, leftmost);
 	ins->delayed_node = delayed_node;
 	ins->ins_or_del = action;
 
@@ -468,7 +471,7 @@ static void finish_one_item(struct btrfs_delayed_root *delayed_root)
 
 static void __btrfs_remove_delayed_item(struct btrfs_delayed_item *delayed_item)
 {
-	struct rb_root *root;
+	struct rb_root_cached *root;
 	struct btrfs_delayed_root *delayed_root;
 
 	delayed_root = delayed_item->delayed_node->root->fs_info->delayed_root;
@@ -482,7 +485,7 @@ static void __btrfs_remove_delayed_item(struct btrfs_delayed_item *delayed_item)
 	else
 		root = &delayed_item->delayed_node->del_root;
 
-	rb_erase(&delayed_item->rb_node, root);
+	rb_erase_cached(&delayed_item->rb_node, root);
 	delayed_item->delayed_node->count--;
 
 	finish_one_item(delayed_root);
@@ -503,7 +506,7 @@ static struct btrfs_delayed_item *__btrfs_first_delayed_insertion_item(
 	struct rb_node *p;
 	struct btrfs_delayed_item *item = NULL;
 
-	p = rb_first(&delayed_node->ins_root);
+	p = rb_first_cached(&delayed_node->ins_root);
 	if (p)
 		item = rb_entry(p, struct btrfs_delayed_item, rb_node);
 
@@ -516,7 +519,7 @@ static struct btrfs_delayed_item *__btrfs_first_delayed_deletion_item(
 	struct rb_node *p;
 	struct btrfs_delayed_item *item = NULL;
 
-	p = rb_first(&delayed_node->del_root);
+	p = rb_first_cached(&delayed_node->del_root);
 	if (p)
 		item = rb_entry(p, struct btrfs_delayed_item, rb_node);
 
@@ -559,7 +562,7 @@ static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans,
 	 * reserved space when starting a transaction.  So no need to reserve
 	 * qgroup space here.
 	 */
-	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
+	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, true);
 	if (!ret) {
 		trace_btrfs_space_reservation(fs_info, "delayed_item",
 					      item->key.objectid,
@@ -647,7 +650,7 @@ static int btrfs_delayed_inode_reserve_metadata(
 		return ret;
 	}
 
-	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
+	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, true);
 	if (!ret) {
 		trace_btrfs_space_reservation(fs_info, "delayed_inode",
 					      btrfs_ino(inode), num_bytes, 1);
@@ -762,9 +765,6 @@ static int btrfs_batch_insert_items(struct btrfs_root *root,
 		i++;
 	}
 
-	/* reset all the locked nodes in the patch to spinning locks. */
-	btrfs_clear_path_blocking(path, NULL, 0);
-
 	/* insert the keys of the items */
 	setup_items_for_insert(root, path, keys, data_size,
 			       total_data_size, total_size, nitems);
@@ -1462,7 +1462,7 @@ int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
 	if (unlikely(ret)) {
 		btrfs_err(trans->fs_info,
 			  "err add delayed dir index item(name: %.*s) into the insertion tree of the delayed node(root id: %llu, inode id: %llu, errno: %d)",
-			  name_len, name, delayed_node->root->objectid,
+			  name_len, name, delayed_node->root->root_key.objectid,
 			  delayed_node->inode_id, ret);
 		BUG();
 	}
@@ -1533,7 +1533,8 @@ int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
 	if (unlikely(ret)) {
 		btrfs_err(trans->fs_info,
 			  "err add delayed dir index item(index: %llu) into the deletion tree of the delayed node(root id: %llu, inode id: %llu, errno: %d)",
-			  index, node->root->objectid, node->inode_id, ret);
+			  index, node->root->root_key.objectid,
+			  node->inode_id, ret);
 		BUG();
 	}
 	mutex_unlock(&node->mutex);

fs/btrfs/delayed-inode.h

@@ -50,8 +50,8 @@ struct btrfs_delayed_node {
 	 * is waiting to be dealt with by the async worker.
 	 */
 	struct list_head p_list;
-	struct rb_root ins_root;
-	struct rb_root del_root;
+	struct rb_root_cached ins_root;
+	struct rb_root_cached del_root;
 	struct mutex mutex;
 	struct btrfs_inode_item inode_item;
 	refcount_t refs;

fs/btrfs/delayed-ref.c

@@ -101,14 +101,15 @@ static int comp_refs(struct btrfs_delayed_ref_node *ref1,
 }
 
 /* insert a new ref to head ref rbtree */
-static struct btrfs_delayed_ref_head *htree_insert(struct rb_root *root,
+static struct btrfs_delayed_ref_head *htree_insert(struct rb_root_cached *root,
 						   struct rb_node *node)
 {
-	struct rb_node **p = &root->rb_node;
+	struct rb_node **p = &root->rb_root.rb_node;
 	struct rb_node *parent_node = NULL;
 	struct btrfs_delayed_ref_head *entry;
 	struct btrfs_delayed_ref_head *ins;
 	u64 bytenr;
+	bool leftmost = true;
 
 	ins = rb_entry(node, struct btrfs_delayed_ref_head, href_node);
 	bytenr = ins->bytenr;
@@ -117,26 +118,29 @@ static struct btrfs_delayed_ref_head *htree_insert(struct rb_root *root,
 		entry = rb_entry(parent_node, struct btrfs_delayed_ref_head,
 				 href_node);
 
-		if (bytenr < entry->bytenr)
+		if (bytenr < entry->bytenr) {
 			p = &(*p)->rb_left;
-		else if (bytenr > entry->bytenr)
+		} else if (bytenr > entry->bytenr) {
 			p = &(*p)->rb_right;
-		else
+			leftmost = false;
+		} else {
 			return entry;
+		}
 	}
 
 	rb_link_node(node, parent_node, p);
-	rb_insert_color(node, root);
+	rb_insert_color_cached(node, root, leftmost);
 	return NULL;
 }
 
-static struct btrfs_delayed_ref_node* tree_insert(struct rb_root *root,
+static struct btrfs_delayed_ref_node* tree_insert(struct rb_root_cached *root,
 		struct btrfs_delayed_ref_node *ins)
 {
-	struct rb_node **p = &root->rb_node;
+	struct rb_node **p = &root->rb_root.rb_node;
 	struct rb_node *node = &ins->ref_node;
 	struct rb_node *parent_node = NULL;
 	struct btrfs_delayed_ref_node *entry;
+	bool leftmost = true;
 
 	while (*p) {
 		int comp;
@@ -145,16 +149,18 @@ static struct btrfs_delayed_ref_node* tree_insert(struct rb_root *root,
 		entry = rb_entry(parent_node, struct btrfs_delayed_ref_node,
 				 ref_node);
 		comp = comp_refs(ins, entry, true);
-		if (comp < 0)
+		if (comp < 0) {
 			p = &(*p)->rb_left;
-		else if (comp > 0)
+		} else if (comp > 0) {
 			p = &(*p)->rb_right;
-		else
+			leftmost = false;
+		} else {
 			return entry;
+		}
 	}
 
 	rb_link_node(node, parent_node, p);
-	rb_insert_color(node, root);
+	rb_insert_color_cached(node, root, leftmost);
 	return NULL;
 }
 
@@ -162,12 +168,14 @@ static struct btrfs_delayed_ref_node* tree_insert(struct rb_root *root,
  * find an head entry based on bytenr. This returns the delayed ref
  * head if it was able to find one, or NULL if nothing was in that spot.
  * If return_bigger is given, the next bigger entry is returned if no exact
- * match is found.
+ * match is found. But if no bigger one is found then the first node of the
+ * ref head tree will be returned.
  */
-static struct btrfs_delayed_ref_head *
-find_ref_head(struct rb_root *root, u64 bytenr,
-	      int return_bigger)
+static struct btrfs_delayed_ref_head* find_ref_head(
+		struct btrfs_delayed_ref_root *dr, u64 bytenr,
+		bool return_bigger)
 {
+	struct rb_root *root = &dr->href_root.rb_root;
 	struct rb_node *n;
 	struct btrfs_delayed_ref_head *entry;
 
@@ -187,7 +195,7 @@ find_ref_head(struct rb_root *root, u64 bytenr,
 		if (bytenr > entry->bytenr) {
 			n = rb_next(&entry->href_node);
 			if (!n)
-				n = rb_first(root);
+				n = rb_first_cached(&dr->href_root);
 			entry = rb_entry(n, struct btrfs_delayed_ref_head,
 					 href_node);
 			return entry;
@@ -197,12 +205,9 @@ find_ref_head(struct rb_root *root, u64 bytenr,
 	return NULL;
 }
 
-int btrfs_delayed_ref_lock(struct btrfs_trans_handle *trans,
+int btrfs_delayed_ref_lock(struct btrfs_delayed_ref_root *delayed_refs,
 			   struct btrfs_delayed_ref_head *head)
 {
-	struct btrfs_delayed_ref_root *delayed_refs;
-
-	delayed_refs = &trans->transaction->delayed_refs;
 	lockdep_assert_held(&delayed_refs->lock);
 	if (mutex_trylock(&head->mutex))
 		return 0;
@@ -227,7 +232,7 @@ static inline void drop_delayed_ref(struct btrfs_trans_handle *trans,
 				    struct btrfs_delayed_ref_node *ref)
 {
 	lockdep_assert_held(&head->lock);
-	rb_erase(&ref->ref_node, &head->ref_tree);
+	rb_erase_cached(&ref->ref_node, &head->ref_tree);
 	RB_CLEAR_NODE(&ref->ref_node);
 	if (!list_empty(&ref->add_list))
 		list_del(&ref->add_list);
@@ -296,7 +301,7 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
 
 	lockdep_assert_held(&head->lock);
 
-	if (RB_EMPTY_ROOT(&head->ref_tree))
+	if (RB_EMPTY_ROOT(&head->ref_tree.rb_root))
 		return;
 
 	/* We don't have too many refs to merge for data. */
@@ -314,7 +319,8 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
 	spin_unlock(&fs_info->tree_mod_seq_lock);
 
 again:
-	for (node = rb_first(&head->ref_tree); node; node = rb_next(node)) {
+	for (node = rb_first_cached(&head->ref_tree); node;
+	     node = rb_next(node)) {
 		ref = rb_entry(node, struct btrfs_delayed_ref_node, ref_node);
 		if (seq && ref->seq >= seq)
 			continue;
@@ -345,24 +351,21 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq)
 	return ret;
 }
 
-struct btrfs_delayed_ref_head *
-btrfs_select_ref_head(struct btrfs_trans_handle *trans)
+struct btrfs_delayed_ref_head *btrfs_select_ref_head(
+		struct btrfs_delayed_ref_root *delayed_refs)
 {
-	struct btrfs_delayed_ref_root *delayed_refs;
 	struct btrfs_delayed_ref_head *head;
 	u64 start;
 	bool loop = false;
 
-	delayed_refs = &trans->transaction->delayed_refs;
-
 again:
 	start = delayed_refs->run_delayed_start;
-	head = find_ref_head(&delayed_refs->href_root, start, 1);
+	head = find_ref_head(delayed_refs, start, true);
 	if (!head && !loop) {
 		delayed_refs->run_delayed_start = 0;
 		start = 0;
 		loop = true;
-		head = find_ref_head(&delayed_refs->href_root, start, 1);
+		head = find_ref_head(delayed_refs, start, true);
 		if (!head)
 			return NULL;
 	} else if (!head && loop) {
@@ -569,7 +572,7 @@ static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref,
 	head_ref->must_insert_reserved = must_insert_reserved;
 	head_ref->is_data = is_data;
 	head_ref->is_system = is_system;
-	head_ref->ref_tree = RB_ROOT;
+	head_ref->ref_tree = RB_ROOT_CACHED;
 	INIT_LIST_HEAD(&head_ref->ref_add_list);
 	RB_CLEAR_NODE(&head_ref->href_node);
 	head_ref->processing = 0;
@@ -903,7 +906,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
 struct btrfs_delayed_ref_head *
 btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs, u64 bytenr)
 {
-	return find_ref_head(&delayed_refs->href_root, bytenr, 0);
+	return find_ref_head(delayed_refs, bytenr, false);
 }
 
 void __cold btrfs_delayed_ref_exit(void)

fs/btrfs/delayed-ref.h

@@ -79,7 +79,7 @@ struct btrfs_delayed_ref_head {
 	struct mutex mutex;
 
 	spinlock_t lock;
-	struct rb_root ref_tree;
+	struct rb_root_cached ref_tree;
 	/* accumulate add BTRFS_ADD_DELAYED_REF nodes to this ref_add_list. */
 	struct list_head ref_add_list;
 
@@ -148,7 +148,7 @@ struct btrfs_delayed_data_ref {
 
 struct btrfs_delayed_ref_root {
 	/* head ref rbtree */
-	struct rb_root href_root;
+	struct rb_root_cached href_root;
 
 	/* dirty extent records */
 	struct rb_root dirty_extent_root;
@@ -255,7 +255,7 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
 struct btrfs_delayed_ref_head *
 btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs,
 			    u64 bytenr);
-int btrfs_delayed_ref_lock(struct btrfs_trans_handle *trans,
+int btrfs_delayed_ref_lock(struct btrfs_delayed_ref_root *delayed_refs,
 			   struct btrfs_delayed_ref_head *head);
 static inline void btrfs_delayed_ref_unlock(struct btrfs_delayed_ref_head *head)
 {
@@ -263,8 +263,8 @@ static inline void btrfs_delayed_ref_unlock(struct btrfs_delayed_ref_head *head)
 }
 
 
-struct btrfs_delayed_ref_head *
-btrfs_select_ref_head(struct btrfs_trans_handle *trans);
+struct btrfs_delayed_ref_head *btrfs_select_ref_head(
+		struct btrfs_delayed_ref_root *delayed_refs);
 
 int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq);
 

fs/btrfs/dev-replace.c

@@ -382,14 +382,6 @@ out:
 	return ret;
 }
 
-void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info)
-{
-	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
-
-	dev_replace->committed_cursor_left =
-		dev_replace->cursor_left_last_write_of_item;
-}
-
 static char* btrfs_dev_name(struct btrfs_device *device)
 {
 	if (!device || test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
@@ -408,11 +400,12 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	int ret;
 	struct btrfs_device *tgt_device = NULL;
 	struct btrfs_device *src_device = NULL;
+	bool need_unlock;
 
-	ret = btrfs_find_device_by_devspec(fs_info, srcdevid,
-					    srcdev_name, &src_device);
-	if (ret)
-		return ret;
+	src_device = btrfs_find_device_by_devspec(fs_info, srcdevid,
+						  srcdev_name);
+	if (IS_ERR(src_device))
+		return PTR_ERR(src_device);
 
 	ret = btrfs_init_dev_replace_tgtdev(fs_info, tgtdev_name,
 					    src_device, &tgt_device);
@@ -432,6 +425,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 		return PTR_ERR(trans);
 	}
 
+	need_unlock = true;
 	btrfs_dev_replace_write_lock(dev_replace);
 	switch (dev_replace->replace_state) {
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
@@ -440,6 +434,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 		break;
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
+		ASSERT(0);
 		ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
 		goto leave;
 	}
@@ -470,6 +465,7 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	atomic64_set(&dev_replace->num_write_errors, 0);
 	atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
 	btrfs_dev_replace_write_unlock(dev_replace);
+	need_unlock = false;
 
 	ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device);
 	if (ret)
@@ -481,7 +477,12 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
+		need_unlock = true;
 		btrfs_dev_replace_write_lock(dev_replace);
+		dev_replace->replace_state =
+			BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED;
+		dev_replace->srcdev = NULL;
+		dev_replace->tgtdev = NULL;
 		goto leave;
 	}
 
@@ -503,9 +504,8 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	return ret;
 
 leave:
-	dev_replace->srcdev = NULL;
-	dev_replace->tgtdev = NULL;
-	btrfs_dev_replace_write_unlock(dev_replace);
+	if (need_unlock)
+		btrfs_dev_replace_write_unlock(dev_replace);
 	btrfs_destroy_dev_replace_tgtdev(tgt_device);
 	return ret;
 }
@@ -545,8 +545,8 @@ int btrfs_dev_replace_by_ioctl(struct btrfs_fs_info *fs_info,
 static void btrfs_rm_dev_replace_blocked(struct btrfs_fs_info *fs_info)
 {
 	set_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state);
-	wait_event(fs_info->replace_wait, !percpu_counter_sum(
-		   &fs_info->bio_counter));
+	wait_event(fs_info->dev_replace.replace_wait, !percpu_counter_sum(
+		   &fs_info->dev_replace.bio_counter));
 }
 
 /*
@@ -555,7 +555,7 @@ static void btrfs_rm_dev_replace_blocked(struct btrfs_fs_info *fs_info)
 static void btrfs_rm_dev_replace_unblocked(struct btrfs_fs_info *fs_info)
 {
 	clear_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state);
-	wake_up(&fs_info->replace_wait);
+	wake_up(&fs_info->dev_replace.replace_wait);
 }
 
 static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
@@ -961,13 +961,10 @@ int btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace)
 void btrfs_dev_replace_read_lock(struct btrfs_dev_replace *dev_replace)
 {
 	read_lock(&dev_replace->lock);
-	atomic_inc(&dev_replace->read_locks);
 }
 
 void btrfs_dev_replace_read_unlock(struct btrfs_dev_replace *dev_replace)
 {
-	ASSERT(atomic_read(&dev_replace->read_locks) > 0);
-	atomic_dec(&dev_replace->read_locks);
 	read_unlock(&dev_replace->lock);
 }
 
@@ -985,7 +982,6 @@ again:
 
 void btrfs_dev_replace_write_unlock(struct btrfs_dev_replace *dev_replace)
 {
-	ASSERT(atomic_read(&dev_replace->blocking_readers) == 0);
 	write_unlock(&dev_replace->lock);
 }
 
@@ -994,45 +990,31 @@ void btrfs_dev_replace_set_lock_blocking(
 					struct btrfs_dev_replace *dev_replace)
 {
 	/* only set blocking for read lock */
-	ASSERT(atomic_read(&dev_replace->read_locks) > 0);
 	atomic_inc(&dev_replace->blocking_readers);
 	read_unlock(&dev_replace->lock);
 }
 
-/* acquire read lock and dec blocking cnt */
-void btrfs_dev_replace_clear_lock_blocking(
-					struct btrfs_dev_replace *dev_replace)
-{
-	/* only set blocking for read lock */
-	ASSERT(atomic_read(&dev_replace->read_locks) > 0);
-	ASSERT(atomic_read(&dev_replace->blocking_readers) > 0);
-	read_lock(&dev_replace->lock);
-	/* Barrier implied by atomic_dec_and_test */
-	if (atomic_dec_and_test(&dev_replace->blocking_readers))
-		cond_wake_up_nomb(&dev_replace->read_lock_wq);
-}
-
 void btrfs_bio_counter_inc_noblocked(struct btrfs_fs_info *fs_info)
 {
-	percpu_counter_inc(&fs_info->bio_counter);
+	percpu_counter_inc(&fs_info->dev_replace.bio_counter);
 }
 
 void btrfs_bio_counter_sub(struct btrfs_fs_info *fs_info, s64 amount)
 {
-	percpu_counter_sub(&fs_info->bio_counter, amount);
-	cond_wake_up_nomb(&fs_info->replace_wait);
+	percpu_counter_sub(&fs_info->dev_replace.bio_counter, amount);
+	cond_wake_up_nomb(&fs_info->dev_replace.replace_wait);
 }
 
 void btrfs_bio_counter_inc_blocked(struct btrfs_fs_info *fs_info)
 {
 	while (1) {
-		percpu_counter_inc(&fs_info->bio_counter);
+		percpu_counter_inc(&fs_info->dev_replace.bio_counter);
 		if (likely(!test_bit(BTRFS_FS_STATE_DEV_REPLACING,
 				     &fs_info->fs_state)))
 			break;
 
 		btrfs_bio_counter_dec(fs_info);
-		wait_event(fs_info->replace_wait,
+		wait_event(fs_info->dev_replace.replace_wait,
 			   !test_bit(BTRFS_FS_STATE_DEV_REPLACING,
 				     &fs_info->fs_state));
 	}

fs/btrfs/dev-replace.h

@@ -11,7 +11,6 @@ struct btrfs_ioctl_dev_replace_args;
 int btrfs_init_dev_replace(struct btrfs_fs_info *fs_info);
 int btrfs_run_dev_replace(struct btrfs_trans_handle *trans,
 			  struct btrfs_fs_info *fs_info);
-void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info);
 int btrfs_dev_replace_by_ioctl(struct btrfs_fs_info *fs_info,
 			    struct btrfs_ioctl_dev_replace_args *args);
 int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
@@ -28,12 +27,5 @@ void btrfs_dev_replace_read_unlock(struct btrfs_dev_replace *dev_replace);
 void btrfs_dev_replace_write_lock(struct btrfs_dev_replace *dev_replace);
 void btrfs_dev_replace_write_unlock(struct btrfs_dev_replace *dev_replace);
 void btrfs_dev_replace_set_lock_blocking(struct btrfs_dev_replace *dev_replace);
-void btrfs_dev_replace_clear_lock_blocking(
-					struct btrfs_dev_replace *dev_replace);
-
-static inline void btrfs_dev_replace_stats_inc(atomic64_t *stat_value)
-{
-	atomic64_inc(stat_value);
-}
 
 #endif

fs/btrfs/dir-item.c

@@ -105,13 +105,13 @@ int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
  * to use for the second index (if one is created).
  * Will return 0 or -ENOMEM
  */
-int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
-			  *root, const char *name, int name_len,
-			  struct btrfs_inode *dir, struct btrfs_key *location,
-			  u8 type, u64 index)
+int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, const char *name,
+			  int name_len, struct btrfs_inode *dir,
+			  struct btrfs_key *location, u8 type, u64 index)
 {
 	int ret = 0;
 	int ret2 = 0;
+	struct btrfs_root *root = dir->root;
 	struct btrfs_path *path;
 	struct btrfs_dir_item *dir_item;
 	struct extent_buffer *leaf;

fs/btrfs/disk-io.c

@@ -125,8 +125,8 @@ struct async_submit_bio {
  * Different roots are used for different purposes and may nest inside each
  * other and they require separate keysets.  As lockdep keys should be
  * static, assign keysets according to the purpose of the root as indicated
- * by btrfs_root->objectid.  This ensures that all special purpose roots
- * have separate keysets.
+ * by btrfs_root->root_key.objectid.  This ensures that all special purpose
+ * roots have separate keysets.
  *
  * Lock-nesting across peer nodes is always done with the immediate parent
  * node locked thus preventing deadlock.  As lockdep doesn't know this, use
@@ -1148,7 +1148,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	root->state = 0;
 	root->orphan_cleanup_state = 0;
 
-	root->objectid = objectid;
 	root->last_trans = 0;
 	root->highest_objectid = 0;
 	root->nr_delalloc_inodes = 0;
@@ -2156,9 +2155,8 @@ static void btrfs_init_dev_replace_locks(struct btrfs_fs_info *fs_info)
 {
 	mutex_init(&fs_info->dev_replace.lock_finishing_cancel_unmount);
 	rwlock_init(&fs_info->dev_replace.lock);
-	atomic_set(&fs_info->dev_replace.read_locks, 0);
 	atomic_set(&fs_info->dev_replace.blocking_readers, 0);
-	init_waitqueue_head(&fs_info->replace_wait);
+	init_waitqueue_head(&fs_info->dev_replace.replace_wait);
 	init_waitqueue_head(&fs_info->dev_replace.read_lock_wq);
 }
 
@@ -2648,7 +2646,8 @@ int open_ctree(struct super_block *sb,
 		goto fail_dirty_metadata_bytes;
 	}
 
-	ret = percpu_counter_init(&fs_info->bio_counter, 0, GFP_KERNEL);
+	ret = percpu_counter_init(&fs_info->dev_replace.bio_counter, 0,
+			GFP_KERNEL);
 	if (ret) {
 		err = ret;
 		goto fail_delalloc_bytes;
@@ -3309,7 +3308,7 @@ fail_iput:
 
 	iput(fs_info->btree_inode);
 fail_bio_counter:
-	percpu_counter_destroy(&fs_info->bio_counter);
+	percpu_counter_destroy(&fs_info->dev_replace.bio_counter);
 fail_delalloc_bytes:
 	percpu_counter_destroy(&fs_info->delalloc_bytes);
 fail_dirty_metadata_bytes:
@@ -3977,6 +3976,7 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 	kthread_stop(fs_info->transaction_kthread);
 	kthread_stop(fs_info->cleaner_kthread);
 
+	ASSERT(list_empty(&fs_info->delayed_iputs));
 	set_bit(BTRFS_FS_CLOSING_DONE, &fs_info->flags);
 
 	btrfs_free_qgroup_config(fs_info);
@@ -4018,7 +4018,7 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 
 	percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
 	percpu_counter_destroy(&fs_info->delalloc_bytes);
-	percpu_counter_destroy(&fs_info->bio_counter);
+	percpu_counter_destroy(&fs_info->dev_replace.bio_counter);
 	cleanup_srcu_struct(&fs_info->subvol_srcu);
 
 	btrfs_free_stripe_hash_table(fs_info);
@@ -4204,7 +4204,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 		return ret;
 	}
 
-	while ((node = rb_first(&delayed_refs->href_root)) != NULL) {
+	while ((node = rb_first_cached(&delayed_refs->href_root)) != NULL) {
 		struct btrfs_delayed_ref_head *head;
 		struct rb_node *n;
 		bool pin_bytes = false;
@@ -4222,11 +4222,11 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 			continue;
 		}
 		spin_lock(&head->lock);
-		while ((n = rb_first(&head->ref_tree)) != NULL) {
+		while ((n = rb_first_cached(&head->ref_tree)) != NULL) {
 			ref = rb_entry(n, struct btrfs_delayed_ref_node,
 				       ref_node);
 			ref->in_tree = 0;
-			rb_erase(&ref->ref_node, &head->ref_tree);
+			rb_erase_cached(&ref->ref_node, &head->ref_tree);
 			RB_CLEAR_NODE(&ref->ref_node);
 			if (!list_empty(&ref->add_list))
 				list_del(&ref->add_list);
@@ -4240,7 +4240,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 		if (head->processing == 0)
 			delayed_refs->num_heads_ready--;
 		atomic_dec(&delayed_refs->num_entries);
-		rb_erase(&head->href_node, &delayed_refs->href_root);
+		rb_erase_cached(&head->href_node, &delayed_refs->href_root);
 		RB_CLEAR_NODE(&head->href_node);
 		spin_unlock(&head->lock);
 		spin_unlock(&delayed_refs->lock);

fs/btrfs/export.c

@@ -33,7 +33,7 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len,
 	type = FILEID_BTRFS_WITHOUT_PARENT;
 
 	fid->objectid = btrfs_ino(BTRFS_I(inode));
-	fid->root_objectid = BTRFS_I(inode)->root->objectid;
+	fid->root_objectid = BTRFS_I(inode)->root->root_key.objectid;
 	fid->gen = inode->i_generation;
 
 	if (parent) {
@@ -41,7 +41,7 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len,
 
 		fid->parent_objectid = BTRFS_I(parent)->location.objectid;
 		fid->parent_gen = parent->i_generation;
-		parent_root_id = BTRFS_I(parent)->root->objectid;
+		parent_root_id = BTRFS_I(parent)->root->root_key.objectid;
 
 		if (parent_root_id != fid->root_objectid) {
 			fid->parent_root_objectid = parent_root_id;

fs/btrfs/extent-tree.c

@@ -2374,7 +2374,7 @@ select_delayed_ref(struct btrfs_delayed_ref_head *head)
 {
 	struct btrfs_delayed_ref_node *ref;
 
-	if (RB_EMPTY_ROOT(&head->ref_tree))
+	if (RB_EMPTY_ROOT(&head->ref_tree.rb_root))
 		return NULL;
 
 	/*
@@ -2387,7 +2387,7 @@ select_delayed_ref(struct btrfs_delayed_ref_head *head)
 		return list_first_entry(&head->ref_add_list,
 				struct btrfs_delayed_ref_node, add_list);
 
-	ref = rb_entry(rb_first(&head->ref_tree),
+	ref = rb_entry(rb_first_cached(&head->ref_tree),
 		       struct btrfs_delayed_ref_node, ref_node);
 	ASSERT(list_empty(&ref->add_list));
 	return ref;
@@ -2448,13 +2448,13 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans,
 	spin_unlock(&head->lock);
 	spin_lock(&delayed_refs->lock);
 	spin_lock(&head->lock);
-	if (!RB_EMPTY_ROOT(&head->ref_tree) || head->extent_op) {
+	if (!RB_EMPTY_ROOT(&head->ref_tree.rb_root) || head->extent_op) {
 		spin_unlock(&head->lock);
 		spin_unlock(&delayed_refs->lock);
 		return 1;
 	}
 	delayed_refs->num_heads--;
-	rb_erase(&head->href_node, &delayed_refs->href_root);
+	rb_erase_cached(&head->href_node, &delayed_refs->href_root);
 	RB_CLEAR_NODE(&head->href_node);
 	spin_unlock(&head->lock);
 	spin_unlock(&delayed_refs->lock);
@@ -2502,102 +2502,66 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
-/*
- * Returns 0 on success or if called with an already aborted transaction.
- * Returns -ENOMEM or -EIO on failure and will abort the transaction.
- */
-static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
-					     unsigned long nr)
+static struct btrfs_delayed_ref_head *btrfs_obtain_ref_head(
+					struct btrfs_trans_handle *trans)
+{
+	struct btrfs_delayed_ref_root *delayed_refs =
+		&trans->transaction->delayed_refs;
+	struct btrfs_delayed_ref_head *head = NULL;
+	int ret;
+
+	spin_lock(&delayed_refs->lock);
+	head = btrfs_select_ref_head(delayed_refs);
+	if (!head) {
+		spin_unlock(&delayed_refs->lock);
+		return head;
+	}
+
+	/*
+	 * Grab the lock that says we are going to process all the refs for
+	 * this head
+	 */
+	ret = btrfs_delayed_ref_lock(delayed_refs, head);
+	spin_unlock(&delayed_refs->lock);
+
+	/*
+	 * We may have dropped the spin lock to get the head mutex lock, and
+	 * that might have given someone else time to free the head.  If that's
+	 * true, it has been removed from our list and we can move on.
+	 */
+	if (ret == -EAGAIN)
+		head = ERR_PTR(-EAGAIN);
+
+	return head;
+}
+
+static int btrfs_run_delayed_refs_for_head(struct btrfs_trans_handle *trans,
+				    struct btrfs_delayed_ref_head *locked_ref,
+				    unsigned long *run_refs)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_delayed_ref_root *delayed_refs;
-	struct btrfs_delayed_ref_node *ref;
-	struct btrfs_delayed_ref_head *locked_ref = NULL;
 	struct btrfs_delayed_extent_op *extent_op;
-	ktime_t start = ktime_get();
-	int ret;
-	unsigned long count = 0;
-	unsigned long actual_count = 0;
+	struct btrfs_delayed_ref_node *ref;
 	int must_insert_reserved = 0;
+	int ret;
 
 	delayed_refs = &trans->transaction->delayed_refs;
-	while (1) {
-		if (!locked_ref) {
-			if (count >= nr)
-				break;
 
-			spin_lock(&delayed_refs->lock);
-			locked_ref = btrfs_select_ref_head(trans);
-			if (!locked_ref) {
-				spin_unlock(&delayed_refs->lock);
-				break;
-			}
+	lockdep_assert_held(&locked_ref->mutex);
+	lockdep_assert_held(&locked_ref->lock);
 
-			/* grab the lock that says we are going to process
-			 * all the refs for this head */
-			ret = btrfs_delayed_ref_lock(trans, locked_ref);
-			spin_unlock(&delayed_refs->lock);
-			/*
-			 * we may have dropped the spin lock to get the head
-			 * mutex lock, and that might have given someone else
-			 * time to free the head.  If that's true, it has been
-			 * removed from our list and we can move on.
-			 */
-			if (ret == -EAGAIN) {
-				locked_ref = NULL;
-				count++;
-				continue;
-			}
-		}
-
-		/*
-		 * We need to try and merge add/drops of the same ref since we
-		 * can run into issues with relocate dropping the implicit ref
-		 * and then it being added back again before the drop can
-		 * finish.  If we merged anything we need to re-loop so we can
-		 * get a good ref.
-		 * Or we can get node references of the same type that weren't
-		 * merged when created due to bumps in the tree mod seq, and
-		 * we need to merge them to prevent adding an inline extent
-		 * backref before dropping it (triggering a BUG_ON at
-		 * insert_inline_extent_backref()).
-		 */
-		spin_lock(&locked_ref->lock);
-		btrfs_merge_delayed_refs(trans, delayed_refs, locked_ref);
-
-		ref = select_delayed_ref(locked_ref);
-
-		if (ref && ref->seq &&
+	while ((ref = select_delayed_ref(locked_ref))) {
+		if (ref->seq &&
 		    btrfs_check_delayed_seq(fs_info, ref->seq)) {
 			spin_unlock(&locked_ref->lock);
 			unselect_delayed_ref_head(delayed_refs, locked_ref);
-			locked_ref = NULL;
-			cond_resched();
-			count++;
-			continue;
-		}
-
-		/*
-		 * We're done processing refs in this ref_head, clean everything
-		 * up and move on to the next ref_head.
-		 */
-		if (!ref) {
-			ret = cleanup_ref_head(trans, locked_ref);
-			if (ret > 0 ) {
-				/* We dropped our lock, we need to loop. */
-				ret = 0;
-				continue;
-			} else if (ret) {
-				return ret;
-			}
-			locked_ref = NULL;
-			count++;
-			continue;
+			return -EAGAIN;
 		}
 
-		actual_count++;
+		(*run_refs)++;
 		ref->in_tree = 0;
-		rb_erase(&ref->ref_node, &locked_ref->ref_tree);
+		rb_erase_cached(&ref->ref_node, &locked_ref->ref_tree);
 		RB_CLEAR_NODE(&ref->ref_node);
 		if (!list_empty(&ref->add_list))
 			list_del(&ref->add_list);
@@ -2619,8 +2583,8 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 		atomic_dec(&delayed_refs->num_entries);
 
 		/*
-		 * Record the must-insert_reserved flag before we drop the spin
-		 * lock.
+		 * Record the must_insert_reserved flag before we drop the
+		 * spin lock.
 		 */
 		must_insert_reserved = locked_ref->must_insert_reserved;
 		locked_ref->must_insert_reserved = 0;
@@ -2642,10 +2606,90 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 		}
 
 		btrfs_put_delayed_ref(ref);
-		count++;
 		cond_resched();
+
+		spin_lock(&locked_ref->lock);
+		btrfs_merge_delayed_refs(trans, delayed_refs, locked_ref);
 	}
 
+	return 0;
+}
+
+/*
+ * Returns 0 on success or if called with an already aborted transaction.
+ * Returns -ENOMEM or -EIO on failure and will abort the transaction.
+ */
+static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
+					     unsigned long nr)
+{
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_delayed_ref_root *delayed_refs;
+	struct btrfs_delayed_ref_head *locked_ref = NULL;
+	ktime_t start = ktime_get();
+	int ret;
+	unsigned long count = 0;
+	unsigned long actual_count = 0;
+
+	delayed_refs = &trans->transaction->delayed_refs;
+	do {
+		if (!locked_ref) {
+			locked_ref = btrfs_obtain_ref_head(trans);
+			if (IS_ERR_OR_NULL(locked_ref)) {
+				if (PTR_ERR(locked_ref) == -EAGAIN) {
+					continue;
+				} else {
+					break;
+				}
+			}
+			count++;
+		}
+		/*
+		 * We need to try and merge add/drops of the same ref since we
+		 * can run into issues with relocate dropping the implicit ref
+		 * and then it being added back again before the drop can
+		 * finish.  If we merged anything we need to re-loop so we can
+		 * get a good ref.
+		 * Or we can get node references of the same type that weren't
+		 * merged when created due to bumps in the tree mod seq, and
+		 * we need to merge them to prevent adding an inline extent
+		 * backref before dropping it (triggering a BUG_ON at
+		 * insert_inline_extent_backref()).
+		 */
+		spin_lock(&locked_ref->lock);
+		btrfs_merge_delayed_refs(trans, delayed_refs, locked_ref);
+
+		ret = btrfs_run_delayed_refs_for_head(trans, locked_ref,
+						      &actual_count);
+		if (ret < 0 && ret != -EAGAIN) {
+			/*
+			 * Error, btrfs_run_delayed_refs_for_head already
+			 * unlocked everything so just bail out
+			 */
+			return ret;
+		} else if (!ret) {
+			/*
+			 * Success, perform the usual cleanup of a processed
+			 * head
+			 */
+			ret = cleanup_ref_head(trans, locked_ref);
+			if (ret > 0 ) {
+				/* We dropped our lock, we need to loop. */
+				ret = 0;
+				continue;
+			} else if (ret) {
+				return ret;
+			}
+		}
+
+		/*
+		 * Either success case or btrfs_run_delayed_refs_for_head
+		 * returned -EAGAIN, meaning we need to select another head
+		 */
+
+		locked_ref = NULL;
+		cond_resched();
+	} while ((nr != -1 && count < nr) || locked_ref);
+
 	/*
 	 * We don't want to include ref heads since we can have empty ref heads
 	 * and those will drastically skew our runtime down since we just do
@@ -2745,9 +2789,9 @@ u64 btrfs_csum_bytes_to_leaves(struct btrfs_fs_info *fs_info, u64 csum_bytes)
 	return num_csums;
 }
 
-int btrfs_check_space_for_delayed_refs(struct btrfs_trans_handle *trans,
-				       struct btrfs_fs_info *fs_info)
+int btrfs_check_space_for_delayed_refs(struct btrfs_trans_handle *trans)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_block_rsv *global_rsv;
 	u64 num_heads = trans->transaction->delayed_refs.num_heads_ready;
 	u64 csum_bytes = trans->transaction->delayed_refs.pending_csums;
@@ -2782,8 +2826,7 @@ int btrfs_check_space_for_delayed_refs(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-int btrfs_should_throttle_delayed_refs(struct btrfs_trans_handle *trans,
-				       struct btrfs_fs_info *fs_info)
+int btrfs_should_throttle_delayed_refs(struct btrfs_trans_handle *trans)
 {
 	u64 num_entries =
 		atomic_read(&trans->transaction->delayed_refs.num_entries);
@@ -2791,14 +2834,14 @@ int btrfs_should_throttle_delayed_refs(struct btrfs_trans_handle *trans,
 	u64 val;
 
 	smp_mb();
-	avg_runtime = fs_info->avg_delayed_ref_runtime;
+	avg_runtime = trans->fs_info->avg_delayed_ref_runtime;
 	val = num_entries * avg_runtime;
 	if (val >= NSEC_PER_SEC)
 		return 1;
 	if (val >= NSEC_PER_SEC / 2)
 		return 2;
 
-	return btrfs_check_space_for_delayed_refs(trans, fs_info);
+	return btrfs_check_space_for_delayed_refs(trans);
 }
 
 struct async_delayed_refs {
@@ -2940,7 +2983,7 @@ again:
 			btrfs_create_pending_block_groups(trans);
 
 		spin_lock(&delayed_refs->lock);
-		node = rb_first(&delayed_refs->href_root);
+		node = rb_first_cached(&delayed_refs->href_root);
 		if (!node) {
 			spin_unlock(&delayed_refs->lock);
 			goto out;
@@ -3040,7 +3083,8 @@ static noinline int check_delayed_ref(struct btrfs_root *root,
 	 * XXX: We should replace this with a proper search function in the
 	 * future.
 	 */
-	for (node = rb_first(&head->ref_tree); node; node = rb_next(node)) {
+	for (node = rb_first_cached(&head->ref_tree); node;
+	     node = rb_next(node)) {
 		ref = rb_entry(node, struct btrfs_delayed_ref_node, ref_node);
 		/* If it's a shared ref we know a cross reference exists */
 		if (ref->type != BTRFS_EXTENT_DATA_REF_KEY) {
@@ -3139,7 +3183,6 @@ int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset,
 {
 	struct btrfs_path *path;
 	int ret;
-	int ret2;
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -3151,17 +3194,9 @@ int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset,
 		if (ret && ret != -ENOENT)
 			goto out;
 
-		ret2 = check_delayed_ref(root, path, objectid,
-					 offset, bytenr);
-	} while (ret2 == -EAGAIN);
-
-	if (ret2 && ret2 != -ENOENT) {
-		ret = ret2;
-		goto out;
-	}
+		ret = check_delayed_ref(root, path, objectid, offset, bytenr);
+	} while (ret == -EAGAIN);
 
-	if (ret != -ENOENT || ret2 != -ENOENT)
-		ret = 0;
 out:
 	btrfs_free_path(path);
 	if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID)
@@ -5284,7 +5319,7 @@ static int block_rsv_use_bytes(struct btrfs_block_rsv *block_rsv,
 }
 
 static void block_rsv_add_bytes(struct btrfs_block_rsv *block_rsv,
-				u64 num_bytes, int update_size)
+				u64 num_bytes, bool update_size)
 {
 	spin_lock(&block_rsv->lock);
 	block_rsv->reserved += num_bytes;
@@ -5316,7 +5351,7 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 		global_rsv->full = 0;
 	spin_unlock(&global_rsv->lock);
 
-	block_rsv_add_bytes(dest, num_bytes, 1);
+	block_rsv_add_bytes(dest, num_bytes, true);
 	return 0;
 }
 
@@ -5479,7 +5514,7 @@ static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info,
 
 int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src,
 			    struct btrfs_block_rsv *dst, u64 num_bytes,
-			    int update_size)
+			    bool update_size)
 {
 	int ret;
 
@@ -5539,10 +5574,8 @@ int btrfs_block_rsv_add(struct btrfs_root *root,
 		return 0;
 
 	ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush);
-	if (!ret) {
-		block_rsv_add_bytes(block_rsv, num_bytes, 1);
-		return 0;
-	}
+	if (!ret)
+		block_rsv_add_bytes(block_rsv, num_bytes, true);
 
 	return ret;
 }
@@ -5587,7 +5620,7 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
 
 	ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush);
 	if (!ret) {
-		block_rsv_add_bytes(block_rsv, num_bytes, 0);
+		block_rsv_add_bytes(block_rsv, num_bytes, false);
 		return 0;
 	}
 
@@ -5629,7 +5662,7 @@ static int btrfs_inode_rsv_refill(struct btrfs_inode *inode,
 		return ret;
 	ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush);
 	if (!ret) {
-		block_rsv_add_bytes(block_rsv, num_bytes, 0);
+		block_rsv_add_bytes(block_rsv, num_bytes, false);
 		trace_btrfs_space_reservation(root->fs_info, "delalloc",
 					      btrfs_ino(inode), num_bytes, 1);
 
@@ -5835,7 +5868,7 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 				  BTRFS_RESERVE_FLUSH_ALL);
 
 	if (ret == -ENOSPC && use_global_rsv)
-		ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes, 1);
+		ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes, true);
 
 	if (ret && qgroup_num_bytes)
 		btrfs_qgroup_free_meta_prealloc(root, qgroup_num_bytes);
@@ -6399,10 +6432,6 @@ static int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache,
 	} else {
 		cache->reserved += num_bytes;
 		space_info->bytes_reserved += num_bytes;
-
-		trace_btrfs_space_reservation(cache->fs_info,
-				"space_info", space_info->flags,
-				ram_bytes, 0);
 		space_info->bytes_may_use -= ram_bytes;
 		if (delalloc)
 			cache->delalloc_bytes += num_bytes;
@@ -6424,11 +6453,10 @@ static int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache,
  * reserve set to 0 in order to clear the reservation.
  */
 
-static int btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
-				     u64 num_bytes, int delalloc)
+static void btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
+				      u64 num_bytes, int delalloc)
 {
 	struct btrfs_space_info *space_info = cache->space_info;
-	int ret = 0;
 
 	spin_lock(&space_info->lock);
 	spin_lock(&cache->lock);
@@ -6441,7 +6469,6 @@ static int btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
 		cache->delalloc_bytes -= num_bytes;
 	spin_unlock(&cache->lock);
 	spin_unlock(&space_info->lock);
-	return ret;
 }
 void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info)
 {
@@ -6925,7 +6952,7 @@ static noinline int check_ref_cleanup(struct btrfs_trans_handle *trans,
 		goto out_delayed_unlock;
 
 	spin_lock(&head->lock);
-	if (!RB_EMPTY_ROOT(&head->ref_tree))
+	if (!RB_EMPTY_ROOT(&head->ref_tree.rb_root))
 		goto out;
 
 	if (head->extent_op) {
@@ -6946,7 +6973,7 @@ static noinline int check_ref_cleanup(struct btrfs_trans_handle *trans,
 	 * at this point we have a head with no other entries.  Go
 	 * ahead and process it.
 	 */
-	rb_erase(&head->href_node, &delayed_refs->href_root);
+	rb_erase_cached(&head->href_node, &delayed_refs->href_root);
 	RB_CLEAR_NODE(&head->href_node);
 	atomic_dec(&delayed_refs->num_entries);
 
@@ -8119,6 +8146,19 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	if (IS_ERR(buf))
 		return buf;
 
+	/*
+	 * Extra safety check in case the extent tree is corrupted and extent
+	 * allocator chooses to use a tree block which is already used and
+	 * locked.
+	 */
+	if (buf->lock_owner == current->pid) {
+		btrfs_err_rl(fs_info,
+"tree block %llu owner %llu already locked by pid=%d, extent tree corruption detected",
+			buf->start, btrfs_header_owner(buf), current->pid);
+		free_extent_buffer(buf);
+		return ERR_PTR(-EUCLEAN);
+	}
+
 	btrfs_set_buffer_lockdep_class(root->root_key.objectid, buf, level);
 	btrfs_tree_lock(buf);
 	clean_tree_block(fs_info, buf);
@@ -8215,7 +8255,7 @@ try_reserve:
 static void unuse_block_rsv(struct btrfs_fs_info *fs_info,
 			    struct btrfs_block_rsv *block_rsv, u32 blocksize)
 {
-	block_rsv_add_bytes(block_rsv, blocksize, 0);
+	block_rsv_add_bytes(block_rsv, blocksize, false);
 	block_rsv_release_bytes(fs_info, block_rsv, NULL, 0, NULL);
 }
 
@@ -8642,7 +8682,13 @@ skip:
 			parent = 0;
 		}
 
-		if (need_account) {
+		/*
+		 * Reloc tree doesn't contribute to qgroup numbers, and we have
+		 * already accounted them at merge time (replace_path),
+		 * thus we could skip expensive subtree trace here.
+		 */
+		if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID &&
+		    need_account) {
 			ret = btrfs_qgroup_trace_subtree(trans, next,
 							 generation, level - 1);
 			if (ret) {
@@ -8763,15 +8809,14 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
 	if (eb == root->node) {
 		if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
 			parent = eb->start;
-		else
-			BUG_ON(root->root_key.objectid !=
-			       btrfs_header_owner(eb));
+		else if (root->root_key.objectid != btrfs_header_owner(eb))
+			goto owner_mismatch;
 	} else {
 		if (wc->flags[level + 1] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
 			parent = path->nodes[level + 1]->start;
-		else
-			BUG_ON(root->root_key.objectid !=
-			       btrfs_header_owner(path->nodes[level + 1]));
+		else if (root->root_key.objectid !=
+			 btrfs_header_owner(path->nodes[level + 1]))
+			goto owner_mismatch;
 	}
 
 	btrfs_free_tree_block(trans, root, eb, parent, wc->refs[level] == 1);
@@ -8779,6 +8824,11 @@ out:
 	wc->refs[level] = 0;
 	wc->flags[level] = 0;
 	return 0;
+
+owner_mismatch:
+	btrfs_err_rl(fs_info, "unexpected tree owner, have %llu expect %llu",
+		     btrfs_header_owner(eb), root->root_key.objectid);
+	return -EUCLEAN;
 }
 
 static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
@@ -8832,6 +8882,8 @@ static noinline int walk_up_tree(struct btrfs_trans_handle *trans,
 			ret = walk_up_proc(trans, root, path, wc);
 			if (ret > 0)
 				return 0;
+			if (ret < 0)
+				return ret;
 
 			if (path->locks[level]) {
 				btrfs_tree_unlock_rw(path->nodes[level],
@@ -8875,7 +8927,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	int level;
 	bool root_dropped = false;
 
-	btrfs_debug(fs_info, "Drop subvolume %llu", root->objectid);
+	btrfs_debug(fs_info, "Drop subvolume %llu", root->root_key.objectid);
 
 	path = btrfs_alloc_path();
 	if (!path) {
@@ -9613,6 +9665,7 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info)
 
 		block_group = btrfs_lookup_first_block_group(info, last);
 		while (block_group) {
+			wait_block_group_cache_done(block_group);
 			spin_lock(&block_group->lock);
 			if (block_group->iref)
 				break;
@@ -10074,7 +10127,7 @@ error:
 void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_block_group_cache *block_group, *tmp;
+	struct btrfs_block_group_cache *block_group;
 	struct btrfs_root *extent_root = fs_info->extent_root;
 	struct btrfs_block_group_item item;
 	struct btrfs_key key;
@@ -10082,7 +10135,10 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 	bool can_flush_pending_bgs = trans->can_flush_pending_bgs;
 
 	trans->can_flush_pending_bgs = false;
-	list_for_each_entry_safe(block_group, tmp, &trans->new_bgs, bg_list) {
+	while (!list_empty(&trans->new_bgs)) {
+		block_group = list_first_entry(&trans->new_bgs,
+					       struct btrfs_block_group_cache,
+					       bg_list);
 		if (ret)
 			goto next;
 
@@ -10753,14 +10809,16 @@ int btrfs_error_unpin_extent_range(struct btrfs_fs_info *fs_info,
  * We don't want a transaction for this since the discard may take a
  * substantial amount of time.  We don't require that a transaction be
  * running, but we do need to take a running transaction into account
- * to ensure that we're not discarding chunks that were released in
- * the current transaction.
+ * to ensure that we're not discarding chunks that were released or
+ * allocated in the current transaction.
  *
  * Holding the chunks lock will prevent other threads from allocating
  * or releasing chunks, but it won't prevent a running transaction
  * from committing and releasing the memory that the pending chunks
  * list head uses.  For that, we need to take a reference to the
- * transaction.
+ * transaction and hold the commit root sem.  We only need to hold
+ * it while performing the free space search since we have already
+ * held back allocations.
  */
 static int btrfs_trim_free_extents(struct btrfs_device *device,
 				   u64 minlen, u64 *trimmed)
@@ -10770,6 +10828,10 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
 
 	*trimmed = 0;
 
+	/* Discard not supported = nothing to do. */
+	if (!blk_queue_discard(bdev_get_queue(device->bdev)))
+		return 0;
+
 	/* Not writeable = nothing to do. */
 	if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state))
 		return 0;
@@ -10787,9 +10849,13 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
 
 		ret = mutex_lock_interruptible(&fs_info->chunk_mutex);
 		if (ret)
-			return ret;
+			break;
 
-		down_read(&fs_info->commit_root_sem);
+		ret = down_read_killable(&fs_info->commit_root_sem);
+		if (ret) {
+			mutex_unlock(&fs_info->chunk_mutex);
+			break;
+		}
 
 		spin_lock(&fs_info->trans_lock);
 		trans = fs_info->running_transaction;
@@ -10797,13 +10863,17 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
 			refcount_inc(&trans->use_count);
 		spin_unlock(&fs_info->trans_lock);
 
+		if (!trans)
+			up_read(&fs_info->commit_root_sem);
+
 		ret = find_free_dev_extent_start(trans, device, minlen, start,
 						 &start, &len);
-		if (trans)
+		if (trans) {
+			up_read(&fs_info->commit_root_sem);
 			btrfs_put_transaction(trans);
+		}
 
 		if (ret) {
-			up_read(&fs_info->commit_root_sem);
 			mutex_unlock(&fs_info->chunk_mutex);
 			if (ret == -ENOSPC)
 				ret = 0;
@@ -10811,7 +10881,6 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
 		}
 
 		ret = btrfs_issue_discard(device->bdev, start, len, &bytes);
-		up_read(&fs_info->commit_root_sem);
 		mutex_unlock(&fs_info->chunk_mutex);
 
 		if (ret)
@@ -10831,6 +10900,15 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
 	return ret;
 }
 
+/*
+ * Trim the whole filesystem by:
+ * 1) trimming the free space in each block group
+ * 2) trimming the unallocated space on each device
+ *
+ * This will also continue trimming even if a block group or device encounters
+ * an error.  The return value will be the last error, or 0 if nothing bad
+ * happens.
+ */
 int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
 {
 	struct btrfs_block_group_cache *cache = NULL;
@@ -10840,18 +10918,14 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
 	u64 start;
 	u64 end;
 	u64 trimmed = 0;
-	u64 total_bytes = btrfs_super_total_bytes(fs_info->super_copy);
+	u64 bg_failed = 0;
+	u64 dev_failed = 0;
+	int bg_ret = 0;
+	int dev_ret = 0;
 	int ret = 0;
 
-	/*
-	 * try to trim all FS space, our block group may start from non-zero.
-	 */
-	if (range->len == total_bytes)
-		cache = btrfs_lookup_first_block_group(fs_info, range->start);
-	else
-		cache = btrfs_lookup_block_group(fs_info, range->start);
-
-	while (cache) {
+	cache = btrfs_lookup_first_block_group(fs_info, range->start);
+	for (; cache; cache = next_block_group(fs_info, cache)) {
 		if (cache->key.objectid >= (range->start + range->len)) {
 			btrfs_put_block_group(cache);
 			break;
@@ -10865,13 +10939,15 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
 			if (!block_group_cache_done(cache)) {
 				ret = cache_block_group(cache, 0);
 				if (ret) {
-					btrfs_put_block_group(cache);
-					break;
+					bg_failed++;
+					bg_ret = ret;
+					continue;
 				}
 				ret = wait_block_group_cache_done(cache);
 				if (ret) {
-					btrfs_put_block_group(cache);
-					break;
+					bg_failed++;
+					bg_ret = ret;
+					continue;
 				}
 			}
 			ret = btrfs_trim_block_group(cache,
@@ -10882,28 +10958,40 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
 
 			trimmed += group_trimmed;
 			if (ret) {
-				btrfs_put_block_group(cache);
-				break;
+				bg_failed++;
+				bg_ret = ret;
+				continue;
 			}
 		}
-
-		cache = next_block_group(fs_info, cache);
 	}
 
+	if (bg_failed)
+		btrfs_warn(fs_info,
+			"failed to trim %llu block group(s), last error %d",
+			bg_failed, bg_ret);
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
-	devices = &fs_info->fs_devices->alloc_list;
-	list_for_each_entry(device, devices, dev_alloc_list) {
+	devices = &fs_info->fs_devices->devices;
+	list_for_each_entry(device, devices, dev_list) {
 		ret = btrfs_trim_free_extents(device, range->minlen,
 					      &group_trimmed);
-		if (ret)
+		if (ret) {
+			dev_failed++;
+			dev_ret = ret;
 			break;
+		}
 
 		trimmed += group_trimmed;
 	}
 	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
+	if (dev_failed)
+		btrfs_warn(fs_info,
+			"failed to trim %llu device(s), last error %d",
+			dev_failed, dev_ret);
 	range->len = trimmed;
-	return ret;
+	if (bg_ret)
+		return bg_ret;
+	return dev_ret;
 }
 
 /*

fs/btrfs/extent_io.c

@@ -1424,20 +1424,15 @@ int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
 			  struct extent_state **cached_state)
 {
 	struct extent_state *state;
-	struct rb_node *n;
 	int ret = 1;
 
 	spin_lock(&tree->lock);
 	if (cached_state && *cached_state) {
 		state = *cached_state;
 		if (state->end == start - 1 && extent_state_in_tree(state)) {
-			n = rb_next(&state->rb_node);
-			while (n) {
-				state = rb_entry(n, struct extent_state,
-						 rb_node);
+			while ((state = next_state(state)) != NULL) {
 				if (state->state & bits)
 					goto got_it;
-				n = rb_next(n);
 			}
 			free_extent_state(*cached_state);
 			*cached_state = NULL;
@@ -1568,7 +1563,7 @@ static noinline int lock_delalloc_pages(struct inode *inode,
  *
  * 1 is returned if we find something, 0 if nothing was in the tree
  */
-STATIC u64 find_lock_delalloc_range(struct inode *inode,
+static noinline_for_stack u64 find_lock_delalloc_range(struct inode *inode,
 				    struct extent_io_tree *tree,
 				    struct page *locked_page, u64 *start,
 				    u64 *end, u64 max_bytes)
@@ -1648,6 +1643,17 @@ out_failed:
 	return found;
 }
 
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+u64 btrfs_find_lock_delalloc_range(struct inode *inode,
+				    struct extent_io_tree *tree,
+				    struct page *locked_page, u64 *start,
+				    u64 *end, u64 max_bytes)
+{
+	return find_lock_delalloc_range(inode, tree, locked_page, start, end,
+			max_bytes);
+}
+#endif
+
 static int __process_pages_contig(struct address_space *mapping,
 				  struct page *locked_page,
 				  pgoff_t start_index, pgoff_t end_index,
@@ -5165,11 +5171,11 @@ void clear_extent_buffer_dirty(struct extent_buffer *eb)
 	WARN_ON(atomic_read(&eb->refs) == 0);
 }
 
-int set_extent_buffer_dirty(struct extent_buffer *eb)
+bool set_extent_buffer_dirty(struct extent_buffer *eb)
 {
 	int i;
 	int num_pages;
-	int was_dirty = 0;
+	bool was_dirty;
 
 	check_buffer_tree_ref(eb);
 
@@ -5179,8 +5185,15 @@ int set_extent_buffer_dirty(struct extent_buffer *eb)
 	WARN_ON(atomic_read(&eb->refs) == 0);
 	WARN_ON(!test_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags));
 
+	if (!was_dirty)
+		for (i = 0; i < num_pages; i++)
+			set_page_dirty(eb->pages[i]);
+
+#ifdef CONFIG_BTRFS_DEBUG
 	for (i = 0; i < num_pages; i++)
-		set_page_dirty(eb->pages[i]);
+		ASSERT(PageDirty(eb->pages[i]));
+#endif
+
 	return was_dirty;
 }
 

fs/btrfs/extent_io.h

@@ -479,7 +479,7 @@ void extent_buffer_bitmap_set(struct extent_buffer *eb, unsigned long start,
 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);
-int set_extent_buffer_dirty(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);
@@ -546,7 +546,7 @@ int free_io_failure(struct extent_io_tree *failure_tree,
 		    struct extent_io_tree *io_tree,
 		    struct io_failure_record *rec);
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-noinline u64 find_lock_delalloc_range(struct inode *inode,
+u64 btrfs_find_lock_delalloc_range(struct inode *inode,
 				      struct extent_io_tree *tree,
 				      struct page *locked_page, u64 *start,
 				      u64 *end, u64 max_bytes);

fs/btrfs/extent_map.c

@@ -34,7 +34,7 @@ void __cold extent_map_exit(void)
  */
 void extent_map_tree_init(struct extent_map_tree *tree)
 {
-	tree->map = RB_ROOT;
+	tree->map = RB_ROOT_CACHED;
 	INIT_LIST_HEAD(&tree->modified_extents);
 	rwlock_init(&tree->lock);
 }
@@ -90,24 +90,27 @@ static u64 range_end(u64 start, u64 len)
 	return start + len;
 }
 
-static int tree_insert(struct rb_root *root, struct extent_map *em)
+static int tree_insert(struct rb_root_cached *root, struct extent_map *em)
 {
-	struct rb_node **p = &root->rb_node;
+	struct rb_node **p = &root->rb_root.rb_node;
 	struct rb_node *parent = NULL;
 	struct extent_map *entry = NULL;
 	struct rb_node *orig_parent = NULL;
 	u64 end = range_end(em->start, em->len);
+	bool leftmost = true;
 
 	while (*p) {
 		parent = *p;
 		entry = rb_entry(parent, struct extent_map, rb_node);
 
-		if (em->start < entry->start)
+		if (em->start < entry->start) {
 			p = &(*p)->rb_left;
-		else if (em->start >= extent_map_end(entry))
+		} else if (em->start >= extent_map_end(entry)) {
 			p = &(*p)->rb_right;
-		else
+			leftmost = false;
+		} else {
 			return -EEXIST;
+		}
 	}
 
 	orig_parent = parent;
@@ -130,7 +133,7 @@ static int tree_insert(struct rb_root *root, struct extent_map *em)
 			return -EEXIST;
 
 	rb_link_node(&em->rb_node, orig_parent, p);
-	rb_insert_color(&em->rb_node, root);
+	rb_insert_color_cached(&em->rb_node, root, leftmost);
 	return 0;
 }
 
@@ -242,7 +245,7 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em)
 			em->mod_start = merge->mod_start;
 			em->generation = max(em->generation, merge->generation);
 
-			rb_erase(&merge->rb_node, &tree->map);
+			rb_erase_cached(&merge->rb_node, &tree->map);
 			RB_CLEAR_NODE(&merge->rb_node);
 			free_extent_map(merge);
 		}
@@ -254,7 +257,7 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em)
 	if (rb && mergable_maps(em, merge)) {
 		em->len += merge->len;
 		em->block_len += merge->block_len;
-		rb_erase(&merge->rb_node, &tree->map);
+		rb_erase_cached(&merge->rb_node, &tree->map);
 		RB_CLEAR_NODE(&merge->rb_node);
 		em->mod_len = (merge->mod_start + merge->mod_len) - em->mod_start;
 		em->generation = max(em->generation, merge->generation);
@@ -367,7 +370,7 @@ __lookup_extent_mapping(struct extent_map_tree *tree,
 	struct rb_node *next = NULL;
 	u64 end = range_end(start, len);
 
-	rb_node = __tree_search(&tree->map, start, &prev, &next);
+	rb_node = __tree_search(&tree->map.rb_root, start, &prev, &next);
 	if (!rb_node) {
 		if (prev)
 			rb_node = prev;
@@ -428,16 +431,13 @@ struct extent_map *search_extent_mapping(struct extent_map_tree *tree,
  * Removes @em from @tree.  No reference counts are dropped, and no checks
  * are done to see if the range is in use
  */
-int remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em)
+void remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em)
 {
-	int ret = 0;
-
 	WARN_ON(test_bit(EXTENT_FLAG_PINNED, &em->flags));
-	rb_erase(&em->rb_node, &tree->map);
+	rb_erase_cached(&em->rb_node, &tree->map);
 	if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags))
 		list_del_init(&em->list);
 	RB_CLEAR_NODE(&em->rb_node);
-	return ret;
 }
 
 void replace_extent_mapping(struct extent_map_tree *tree,
@@ -449,7 +449,7 @@ void replace_extent_mapping(struct extent_map_tree *tree,
 	ASSERT(extent_map_in_tree(cur));
 	if (!test_bit(EXTENT_FLAG_LOGGING, &cur->flags))
 		list_del_init(&cur->list);
-	rb_replace_node(&cur->rb_node, &new->rb_node, &tree->map);
+	rb_replace_node_cached(&cur->rb_node, &new->rb_node, &tree->map);
 	RB_CLEAR_NODE(&cur->rb_node);
 
 	setup_extent_mapping(tree, new, modified);

fs/btrfs/extent_map.h

@@ -49,7 +49,7 @@ struct extent_map {
 };
 
 struct extent_map_tree {
-	struct rb_root map;
+	struct rb_root_cached map;
 	struct list_head modified_extents;
 	rwlock_t lock;
 };
@@ -78,7 +78,7 @@ struct extent_map *lookup_extent_mapping(struct extent_map_tree *tree,
 					 u64 start, u64 len);
 int add_extent_mapping(struct extent_map_tree *tree,
 		       struct extent_map *em, int modified);
-int remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em);
+void remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em);
 void replace_extent_mapping(struct extent_map_tree *tree,
 			    struct extent_map *cur,
 			    struct extent_map *new,

fs/btrfs/file.c

@@ -531,6 +531,14 @@ int btrfs_dirty_pages(struct inode *inode, struct page **pages,
 
 	end_of_last_block = start_pos + num_bytes - 1;
 
+	/*
+	 * The pages may have already been dirty, clear out old accounting so
+	 * we can set things up properly
+	 */
+	clear_extent_bit(&BTRFS_I(inode)->io_tree, start_pos, end_of_last_block,
+			 EXTENT_DIRTY | EXTENT_DELALLOC |
+			 EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0, cached);
+
 	if (!btrfs_is_free_space_inode(BTRFS_I(inode))) {
 		if (start_pos >= isize &&
 		    !(BTRFS_I(inode)->flags & BTRFS_INODE_PREALLOC)) {
@@ -1500,18 +1508,27 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 		}
 		if (ordered)
 			btrfs_put_ordered_extent(ordered);
-		clear_extent_bit(&inode->io_tree, start_pos, last_pos,
-				 EXTENT_DIRTY | EXTENT_DELALLOC |
-				 EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
-				 0, 0, cached_state);
+
 		*lockstart = start_pos;
 		*lockend = last_pos;
 		ret = 1;
 	}
 
+	/*
+	 * It's possible the pages are dirty right now, but we don't want
+	 * to clean them yet because copy_from_user may catch a page fault
+	 * and we might have to fall back to one page at a time.  If that
+	 * happens, we'll unlock these pages and we'd have a window where
+	 * reclaim could sneak in and drop the once-dirty page on the floor
+	 * without writing it.
+	 *
+	 * We have the pages locked and the extent range locked, so there's
+	 * no way someone can start IO on any dirty pages in this range.
+	 *
+	 * We'll call btrfs_dirty_pages() later on, and that will flip around
+	 * delalloc bits and dirty the pages as required.
+	 */
 	for (i = 0; i < num_pages; i++) {
-		if (clear_page_dirty_for_io(pages[i]))
-			account_page_redirty(pages[i]);
 		set_page_extent_mapped(pages[i]);
 		WARN_ON(!PageLocked(pages[i]));
 	}
@@ -2544,7 +2561,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 	}
 
 	ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv,
-				      min_size, 0);
+				      min_size, false);
 	BUG_ON(ret);
 	trans->block_rsv = rsv;
 
@@ -2594,7 +2611,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 		}
 
 		ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv,
-					      rsv, min_size, 0);
+					      rsv, min_size, false);
 		BUG_ON(ret);	/* shouldn't happen */
 		trans->block_rsv = rsv;
 

fs/btrfs/free-space-cache.c

@@ -10,6 +10,7 @@
 #include <linux/math64.h>
 #include <linux/ratelimit.h>
 #include <linux/error-injection.h>
+#include <linux/sched/mm.h>
 #include "ctree.h"
 #include "free-space-cache.h"
 #include "transaction.h"
@@ -47,6 +48,7 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
 	struct btrfs_free_space_header *header;
 	struct extent_buffer *leaf;
 	struct inode *inode = NULL;
+	unsigned nofs_flag;
 	int ret;
 
 	key.objectid = BTRFS_FREE_SPACE_OBJECTID;
@@ -68,7 +70,13 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
 	btrfs_disk_key_to_cpu(&location, &disk_key);
 	btrfs_release_path(path);
 
+	/*
+	 * We are often under a trans handle at this point, so we need to make
+	 * sure NOFS is set to keep us from deadlocking.
+	 */
+	nofs_flag = memalloc_nofs_save();
 	inode = btrfs_iget(fs_info->sb, &location, root, NULL);
+	memalloc_nofs_restore(nofs_flag);
 	if (IS_ERR(inode))
 		return inode;
 
@@ -1679,6 +1687,8 @@ static inline void __bitmap_clear_bits(struct btrfs_free_space_ctl *ctl,
 	bitmap_clear(info->bitmap, start, count);
 
 	info->bytes -= bytes;
+	if (info->max_extent_size > ctl->unit)
+		info->max_extent_size = 0;
 }
 
 static void bitmap_clear_bits(struct btrfs_free_space_ctl *ctl,
@@ -2110,8 +2120,7 @@ new_bitmap:
 
 out:
 	if (info) {
-		if (info->bitmap)
-			kfree(info->bitmap);
+		kfree(info->bitmap);
 		kmem_cache_free(btrfs_free_space_cachep, info);
 	}
 
@@ -3601,8 +3610,7 @@ again:
 
 	if (info)
 		kmem_cache_free(btrfs_free_space_cachep, info);
-	if (map)
-		kfree(map);
+	kfree(map);
 	return 0;
 }
 

fs/btrfs/inode.c

@@ -64,7 +64,6 @@ static const struct inode_operations btrfs_dir_ro_inode_operations;
 static const struct inode_operations btrfs_special_inode_operations;
 static const struct inode_operations btrfs_file_inode_operations;
 static const struct address_space_operations btrfs_aops;
-static const struct address_space_operations btrfs_symlink_aops;
 static const struct file_operations btrfs_dir_file_operations;
 static const struct extent_io_ops btrfs_extent_io_ops;
 
@@ -2750,12 +2749,9 @@ static void relink_file_extents(struct new_sa_defrag_extent *new)
 	struct btrfs_path *path;
 	struct sa_defrag_extent_backref *backref;
 	struct sa_defrag_extent_backref *prev = NULL;
-	struct inode *inode;
 	struct rb_node *node;
 	int ret;
 
-	inode = new->inode;
-
 	path = btrfs_alloc_path();
 	if (!path)
 		return;
@@ -3471,8 +3467,6 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 
 		/* this will do delete_inode and everything for us */
 		iput(inode);
-		if (ret)
-			goto out;
 	}
 	/* release the path since we're done with it */
 	btrfs_release_path(path);
@@ -3738,7 +3732,7 @@ cache_acl:
 	case S_IFLNK:
 		inode->i_op = &btrfs_symlink_inode_operations;
 		inode_nohighmem(inode);
-		inode->i_mapping->a_ops = &btrfs_symlink_aops;
+		inode->i_mapping->a_ops = &btrfs_aops;
 		break;
 	default:
 		inode->i_op = &btrfs_special_inode_operations;
@@ -3910,12 +3904,8 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 	path->leave_spinning = 1;
 	di = btrfs_lookup_dir_item(trans, root, path, dir_ino,
 				    name, name_len, -1);
-	if (IS_ERR(di)) {
-		ret = PTR_ERR(di);
-		goto err;
-	}
-	if (!di) {
-		ret = -ENOENT;
+	if (IS_ERR_OR_NULL(di)) {
+		ret = di ? PTR_ERR(di) : -ENOENT;
 		goto err;
 	}
 	leaf = path->nodes[0];
@@ -4075,10 +4065,7 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 	di = btrfs_lookup_dir_item(trans, root, path, dir_ino,
 				   name, name_len, -1);
 	if (IS_ERR_OR_NULL(di)) {
-		if (!di)
-			ret = -ENOENT;
-		else
-			ret = PTR_ERR(di);
+		ret = di ? PTR_ERR(di) : -ENOENT;
 		goto out;
 	}
 
@@ -4270,18 +4257,17 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry)
 	 * again is not run concurrently.
 	 */
 	spin_lock(&dest->root_item_lock);
-	root_flags = btrfs_root_flags(&dest->root_item);
-	if (dest->send_in_progress == 0) {
-		btrfs_set_root_flags(&dest->root_item,
-				root_flags | BTRFS_ROOT_SUBVOL_DEAD);
-		spin_unlock(&dest->root_item_lock);
-	} else {
+	if (dest->send_in_progress) {
 		spin_unlock(&dest->root_item_lock);
 		btrfs_warn(fs_info,
 			   "attempt to delete subvolume %llu during send",
 			   dest->root_key.objectid);
 		return -EPERM;
 	}
+	root_flags = btrfs_root_flags(&dest->root_item);
+	btrfs_set_root_flags(&dest->root_item,
+			     root_flags | BTRFS_ROOT_SUBVOL_DEAD);
+	spin_unlock(&dest->root_item_lock);
 
 	down_write(&fs_info->subvol_sem);
 
@@ -4727,7 +4713,7 @@ delete:
 				btrfs_abort_transaction(trans, ret);
 				break;
 			}
-			if (btrfs_should_throttle_delayed_refs(trans, fs_info))
+			if (btrfs_should_throttle_delayed_refs(trans))
 				btrfs_async_run_delayed_refs(fs_info,
 					trans->delayed_ref_updates * 2,
 					trans->transid, 0);
@@ -4736,8 +4722,7 @@ delete:
 							 extent_num_bytes)) {
 					should_end = true;
 				}
-				if (btrfs_should_throttle_delayed_refs(trans,
-								       fs_info))
+				if (btrfs_should_throttle_delayed_refs(trans))
 					should_throttle = true;
 			}
 		}
@@ -5235,10 +5220,10 @@ static void evict_inode_truncate_pages(struct inode *inode)
 	truncate_inode_pages_final(&inode->i_data);
 
 	write_lock(&map_tree->lock);
-	while (!RB_EMPTY_ROOT(&map_tree->map)) {
+	while (!RB_EMPTY_ROOT(&map_tree->map.rb_root)) {
 		struct extent_map *em;
 
-		node = rb_first(&map_tree->map);
+		node = rb_first_cached(&map_tree->map);
 		em = rb_entry(node, struct extent_map, rb_node);
 		clear_bit(EXTENT_FLAG_PINNED, &em->flags);
 		clear_bit(EXTENT_FLAG_LOGGING, &em->flags);
@@ -5306,8 +5291,7 @@ static void evict_inode_truncate_pages(struct inode *inode)
 }
 
 static struct btrfs_trans_handle *evict_refill_and_join(struct btrfs_root *root,
-							struct btrfs_block_rsv *rsv,
-							u64 min_size)
+							struct btrfs_block_rsv *rsv)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
@@ -5317,7 +5301,7 @@ static struct btrfs_trans_handle *evict_refill_and_join(struct btrfs_root *root,
 		struct btrfs_trans_handle *trans;
 		int ret;
 
-		ret = btrfs_block_rsv_refill(root, rsv, min_size,
+		ret = btrfs_block_rsv_refill(root, rsv, rsv->size,
 					     BTRFS_RESERVE_FLUSH_LIMIT);
 
 		if (ret && ++failures > 2) {
@@ -5334,8 +5318,8 @@ static struct btrfs_trans_handle *evict_refill_and_join(struct btrfs_root *root,
 		 * Try to steal from the global reserve if there is space for
 		 * it.
 		 */
-		if (!btrfs_check_space_for_delayed_refs(trans, fs_info) &&
-		    !btrfs_block_rsv_migrate(global_rsv, rsv, min_size, 0))
+		if (!btrfs_check_space_for_delayed_refs(trans) &&
+		    !btrfs_block_rsv_migrate(global_rsv, rsv, rsv->size, false))
 			return trans;
 
 		/* If not, commit and try again. */
@@ -5351,7 +5335,6 @@ void btrfs_evict_inode(struct inode *inode)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_block_rsv *rsv;
-	u64 min_size;
 	int ret;
 
 	trace_btrfs_inode_evict(inode);
@@ -5361,8 +5344,6 @@ void btrfs_evict_inode(struct inode *inode)
 		return;
 	}
 
-	min_size = btrfs_calc_trunc_metadata_size(fs_info, 1);
-
 	evict_inode_truncate_pages(inode);
 
 	if (inode->i_nlink &&
@@ -5373,9 +5354,6 @@ void btrfs_evict_inode(struct inode *inode)
 
 	if (is_bad_inode(inode))
 		goto no_delete;
-	/* do we really want it for ->i_nlink > 0 and zero btrfs_root_refs? */
-	if (!special_file(inode->i_mode))
-		btrfs_wait_ordered_range(inode, 0, (u64)-1);
 
 	btrfs_free_io_failure_record(BTRFS_I(inode), 0, (u64)-1);
 
@@ -5395,13 +5373,13 @@ void btrfs_evict_inode(struct inode *inode)
 	rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP);
 	if (!rsv)
 		goto no_delete;
-	rsv->size = min_size;
+	rsv->size = btrfs_calc_trunc_metadata_size(fs_info, 1);
 	rsv->failfast = 1;
 
 	btrfs_i_size_write(BTRFS_I(inode), 0);
 
 	while (1) {
-		trans = evict_refill_and_join(root, rsv, min_size);
+		trans = evict_refill_and_join(root, rsv);
 		if (IS_ERR(trans))
 			goto free_rsv;
 
@@ -5426,7 +5404,7 @@ void btrfs_evict_inode(struct inode *inode)
 	 * If it turns out that we are dropping too many of these, we might want
 	 * to add a mechanism for retrying these after a commit.
 	 */
-	trans = evict_refill_and_join(root, rsv, min_size);
+	trans = evict_refill_and_join(root, rsv);
 	if (!IS_ERR(trans)) {
 		trans->block_rsv = rsv;
 		btrfs_orphan_del(trans, BTRFS_I(inode));
@@ -5471,12 +5449,8 @@ static int btrfs_inode_by_name(struct inode *dir, struct dentry *dentry,
 
 	di = btrfs_lookup_dir_item(NULL, root, path, btrfs_ino(BTRFS_I(dir)),
 			name, namelen, 0);
-	if (!di) {
-		ret = -ENOENT;
-		goto out;
-	}
-	if (IS_ERR(di)) {
-		ret = PTR_ERR(di);
+	if (IS_ERR_OR_NULL(di)) {
+		ret = di ? PTR_ERR(di) : -ENOENT;
 		goto out;
 	}
 
@@ -6390,8 +6364,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
 	if (ret)
 		return ret;
 
-	ret = btrfs_insert_dir_item(trans, root, name, name_len,
-				    parent_inode, &key,
+	ret = btrfs_insert_dir_item(trans, name, name_len, parent_inode, &key,
 				    btrfs_inode_type(&inode->vfs_inode), index);
 	if (ret == -EEXIST || ret == -EOVERFLOW)
 		goto fail_dir_item;
@@ -6584,7 +6557,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	int drop_inode = 0;
 
 	/* do not allow sys_link's with other subvols of the same device */
-	if (root->objectid != BTRFS_I(inode)->root->objectid)
+	if (root->root_key.objectid != BTRFS_I(inode)->root->root_key.objectid)
 		return -EXDEV;
 
 	if (inode->i_nlink >= BTRFS_LINK_MAX)
@@ -6777,9 +6750,9 @@ static noinline int uncompress_inline(struct btrfs_path *path,
  * This also copies inline extents directly into the page.
  */
 struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
-		struct page *page,
-	    size_t pg_offset, u64 start, u64 len,
-		int create)
+				    struct page *page,
+				    size_t pg_offset, u64 start, u64 len,
+				    int create)
 {
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	int ret;
@@ -6823,19 +6796,21 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	em->len = (u64)-1;
 	em->block_len = (u64)-1;
 
+	path = btrfs_alloc_path();
 	if (!path) {
-		path = btrfs_alloc_path();
-		if (!path) {
-			err = -ENOMEM;
-			goto out;
-		}
-		/*
-		 * Chances are we'll be called again, so go ahead and do
-		 * readahead
-		 */
-		path->reada = READA_FORWARD;
+		err = -ENOMEM;
+		goto out;
 	}
 
+	/* Chances are we'll be called again, so go ahead and do readahead */
+	path->reada = READA_FORWARD;
+
+	/*
+	 * Unless we're going to uncompress the inline extent, no sleep would
+	 * happen.
+	 */
+	path->leave_spinning = 1;
+
 	ret = btrfs_lookup_file_extent(NULL, root, path, objectid, start, 0);
 	if (ret < 0) {
 		err = ret;
@@ -6938,6 +6913,8 @@ next:
 		em->orig_block_len = em->len;
 		em->orig_start = em->start;
 		ptr = btrfs_file_extent_inline_start(item) + extent_offset;
+
+		btrfs_set_path_blocking(path);
 		if (!PageUptodate(page)) {
 			if (btrfs_file_extent_compression(leaf, item) !=
 			    BTRFS_COMPRESS_NONE) {
@@ -6985,10 +6962,10 @@ insert:
 	err = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len);
 	write_unlock(&em_tree->lock);
 out:
+	btrfs_free_path(path);
 
 	trace_btrfs_get_extent(root, inode, em);
 
-	btrfs_free_path(path);
 	if (err) {
 		free_extent_map(em);
 		return ERR_PTR(err);
@@ -9021,7 +8998,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 
 	/* Migrate the slack space for the truncate to our reserve */
 	ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv,
-				      min_size, 0);
+				      min_size, false);
 	BUG_ON(ret);
 
 	/*
@@ -9058,7 +9035,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 
 		btrfs_block_rsv_release(fs_info, rsv, -1);
 		ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv,
-					      rsv, min_size, 0);
+					      rsv, min_size, false);
 		BUG_ON(ret);	/* shouldn't happen */
 		trans->block_rsv = rsv;
 	}
@@ -10191,7 +10168,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 
 	inode->i_op = &btrfs_symlink_inode_operations;
 	inode_nohighmem(inode);
-	inode->i_mapping->a_ops = &btrfs_symlink_aops;
+	inode->i_mapping->a_ops = &btrfs_aops;
 	inode_set_bytes(inode, name_len);
 	btrfs_i_size_write(BTRFS_I(inode), name_len);
 	err = btrfs_update_inode(trans, root, inode);
@@ -10567,13 +10544,6 @@ static const struct address_space_operations btrfs_aops = {
 	.error_remove_page = generic_error_remove_page,
 };
 
-static const struct address_space_operations btrfs_symlink_aops = {
-	.readpage	= btrfs_readpage,
-	.writepage	= btrfs_writepage,
-	.invalidatepage = btrfs_invalidatepage,
-	.releasepage	= btrfs_releasepage,
-};
-
 static const struct inode_operations btrfs_file_inode_operations = {
 	.getattr	= btrfs_getattr,
 	.setattr	= btrfs_setattr,

fs/btrfs/ioctl.c

@@ -491,7 +491,6 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
 	struct fstrim_range range;
 	u64 minlen = ULLONG_MAX;
 	u64 num_devices = 0;
-	u64 total_bytes = btrfs_super_total_bytes(fs_info->super_copy);
 	int ret;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -515,11 +514,15 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
 		return -EOPNOTSUPP;
 	if (copy_from_user(&range, arg, sizeof(range)))
 		return -EFAULT;
-	if (range.start > total_bytes ||
-	    range.len < fs_info->sb->s_blocksize)
+
+	/*
+	 * NOTE: Don't truncate the range using super->total_bytes.  Bytenr of
+	 * block group is in the logical address space, which can be any
+	 * sectorsize aligned bytenr in  the range [0, U64_MAX].
+	 */
+	if (range.len < fs_info->sb->s_blocksize)
 		return -EINVAL;
 
-	range.len = min(range.len, total_bytes - range.start);
 	range.minlen = max(range.minlen, minlen);
 	ret = btrfs_trim_fs(fs_info, &range);
 	if (ret < 0)
@@ -686,8 +689,7 @@ static noinline int create_subvol(struct inode *dir,
 		goto fail;
 	}
 
-	ret = btrfs_insert_dir_item(trans, root,
-				    name, namelen, BTRFS_I(dir), &key,
+	ret = btrfs_insert_dir_item(trans, name, namelen, BTRFS_I(dir), &key,
 				    BTRFS_FT_DIR, index);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
@@ -1324,7 +1326,7 @@ again:
 
 	if (i_done != page_cnt) {
 		spin_lock(&BTRFS_I(inode)->lock);
-		BTRFS_I(inode)->outstanding_extents++;
+		btrfs_mod_outstanding_extents(BTRFS_I(inode), 1);
 		spin_unlock(&BTRFS_I(inode)->lock);
 		btrfs_delalloc_release_space(inode, data_reserved,
 				start_index << PAGE_SHIFT,
@@ -4393,7 +4395,7 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
 		ret = PTR_ERR(new_root);
 		goto out;
 	}
-	if (!is_fstree(new_root->objectid)) {
+	if (!is_fstree(new_root->root_key.objectid)) {
 		ret = -ENOENT;
 		goto out;
 	}

fs/btrfs/qgroup.c

@@ -1416,13 +1416,14 @@ int btrfs_remove_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid)
 	if (!qgroup) {
 		ret = -ENOENT;
 		goto out;
-	} else {
-		/* check if there are no children of this qgroup */
-		if (!list_empty(&qgroup->members)) {
-			ret = -EBUSY;
-			goto out;
-		}
 	}
+
+	/* Check if there are no children of this qgroup */
+	if (!list_empty(&qgroup->members)) {
+		ret = -EBUSY;
+		goto out;
+	}
+
 	ret = del_qgroup_item(trans, qgroupid);
 	if (ret && ret != -ENOENT)
 		goto out;
@@ -1712,6 +1713,416 @@ static int adjust_slots_upwards(struct btrfs_path *path, int root_level)
 	return 0;
 }
 
+/*
+ * Helper function to trace a subtree tree block swap.
+ *
+ * The swap will happen in highest tree block, but there may be a lot of
+ * tree blocks involved.
+ *
+ * For example:
+ *  OO = Old tree blocks
+ *  NN = New tree blocks allocated during balance
+ *
+ *           File tree (257)                  Reloc tree for 257
+ * L2              OO                                NN
+ *               /    \                            /    \
+ * L1          OO      OO (a)                    OO      NN (a)
+ *            / \     / \                       / \     / \
+ * L0       OO   OO OO   OO                   OO   OO NN   NN
+ *                  (b)  (c)                          (b)  (c)
+ *
+ * When calling qgroup_trace_extent_swap(), we will pass:
+ * @src_eb = OO(a)
+ * @dst_path = [ nodes[1] = NN(a), nodes[0] = NN(c) ]
+ * @dst_level = 0
+ * @root_level = 1
+ *
+ * In that case, qgroup_trace_extent_swap() will search from OO(a) to
+ * reach OO(c), then mark both OO(c) and NN(c) as qgroup dirty.
+ *
+ * The main work of qgroup_trace_extent_swap() can be split into 3 parts:
+ *
+ * 1) Tree search from @src_eb
+ *    It should acts as a simplified btrfs_search_slot().
+ *    The key for search can be extracted from @dst_path->nodes[dst_level]
+ *    (first key).
+ *
+ * 2) Mark the final tree blocks in @src_path and @dst_path qgroup dirty
+ *    NOTE: In above case, OO(a) and NN(a) won't be marked qgroup dirty.
+ *    They should be marked during preivous (@dst_level = 1) iteration.
+ *
+ * 3) Mark file extents in leaves dirty
+ *    We don't have good way to pick out new file extents only.
+ *    So we still follow the old method by scanning all file extents in
+ *    the leave.
+ *
+ * This function can free us from keeping two pathes, thus later we only need
+ * to care about how to iterate all new tree blocks in reloc tree.
+ */
+static int qgroup_trace_extent_swap(struct btrfs_trans_handle* trans,
+				    struct extent_buffer *src_eb,
+				    struct btrfs_path *dst_path,
+				    int dst_level, int root_level,
+				    bool trace_leaf)
+{
+	struct btrfs_key key;
+	struct btrfs_path *src_path;
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	u32 nodesize = fs_info->nodesize;
+	int cur_level = root_level;
+	int ret;
+
+	BUG_ON(dst_level > root_level);
+	/* Level mismatch */
+	if (btrfs_header_level(src_eb) != root_level)
+		return -EINVAL;
+
+	src_path = btrfs_alloc_path();
+	if (!src_path) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	if (dst_level)
+		btrfs_node_key_to_cpu(dst_path->nodes[dst_level], &key, 0);
+	else
+		btrfs_item_key_to_cpu(dst_path->nodes[dst_level], &key, 0);
+
+	/* For src_path */
+	extent_buffer_get(src_eb);
+	src_path->nodes[root_level] = src_eb;
+	src_path->slots[root_level] = dst_path->slots[root_level];
+	src_path->locks[root_level] = 0;
+
+	/* A simplified version of btrfs_search_slot() */
+	while (cur_level >= dst_level) {
+		struct btrfs_key src_key;
+		struct btrfs_key dst_key;
+
+		if (src_path->nodes[cur_level] == NULL) {
+			struct btrfs_key first_key;
+			struct extent_buffer *eb;
+			int parent_slot;
+			u64 child_gen;
+			u64 child_bytenr;
+
+			eb = src_path->nodes[cur_level + 1];
+			parent_slot = src_path->slots[cur_level + 1];
+			child_bytenr = btrfs_node_blockptr(eb, parent_slot);
+			child_gen = btrfs_node_ptr_generation(eb, parent_slot);
+			btrfs_node_key_to_cpu(eb, &first_key, parent_slot);
+
+			eb = read_tree_block(fs_info, child_bytenr, child_gen,
+					     cur_level, &first_key);
+			if (IS_ERR(eb)) {
+				ret = PTR_ERR(eb);
+				goto out;
+			} else if (!extent_buffer_uptodate(eb)) {
+				free_extent_buffer(eb);
+				ret = -EIO;
+				goto out;
+			}
+
+			src_path->nodes[cur_level] = eb;
+
+			btrfs_tree_read_lock(eb);
+			btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
+			src_path->locks[cur_level] = BTRFS_READ_LOCK_BLOCKING;
+		}
+
+		src_path->slots[cur_level] = dst_path->slots[cur_level];
+		if (cur_level) {
+			btrfs_node_key_to_cpu(dst_path->nodes[cur_level],
+					&dst_key, dst_path->slots[cur_level]);
+			btrfs_node_key_to_cpu(src_path->nodes[cur_level],
+					&src_key, src_path->slots[cur_level]);
+		} else {
+			btrfs_item_key_to_cpu(dst_path->nodes[cur_level],
+					&dst_key, dst_path->slots[cur_level]);
+			btrfs_item_key_to_cpu(src_path->nodes[cur_level],
+					&src_key, src_path->slots[cur_level]);
+		}
+		/* Content mismatch, something went wrong */
+		if (btrfs_comp_cpu_keys(&dst_key, &src_key)) {
+			ret = -ENOENT;
+			goto out;
+		}
+		cur_level--;
+	}
+
+	/*
+	 * Now both @dst_path and @src_path have been populated, record the tree
+	 * blocks for qgroup accounting.
+	 */
+	ret = btrfs_qgroup_trace_extent(trans, src_path->nodes[dst_level]->start,
+			nodesize, GFP_NOFS);
+	if (ret < 0)
+		goto out;
+	ret = btrfs_qgroup_trace_extent(trans,
+			dst_path->nodes[dst_level]->start,
+			nodesize, GFP_NOFS);
+	if (ret < 0)
+		goto out;
+
+	/* Record leaf file extents */
+	if (dst_level == 0 && trace_leaf) {
+		ret = btrfs_qgroup_trace_leaf_items(trans, src_path->nodes[0]);
+		if (ret < 0)
+			goto out;
+		ret = btrfs_qgroup_trace_leaf_items(trans, dst_path->nodes[0]);
+	}
+out:
+	btrfs_free_path(src_path);
+	return ret;
+}
+
+/*
+ * Helper function to do recursive generation-aware depth-first search, to
+ * locate all new tree blocks in a subtree of reloc tree.
+ *
+ * E.g. (OO = Old tree blocks, NN = New tree blocks, whose gen == last_snapshot)
+ *         reloc tree
+ * L2         NN (a)
+ *          /    \
+ * L1    OO        NN (b)
+ *      /  \      /  \
+ * L0  OO  OO    OO  NN
+ *               (c) (d)
+ * If we pass:
+ * @dst_path = [ nodes[1] = NN(b), nodes[0] = NULL ],
+ * @cur_level = 1
+ * @root_level = 1
+ *
+ * We will iterate through tree blocks NN(b), NN(d) and info qgroup to trace
+ * above tree blocks along with their counter parts in file tree.
+ * While during search, old tree blocsk OO(c) will be skiped as tree block swap
+ * won't affect OO(c).
+ */
+static int qgroup_trace_new_subtree_blocks(struct btrfs_trans_handle* trans,
+					   struct extent_buffer *src_eb,
+					   struct btrfs_path *dst_path,
+					   int cur_level, int root_level,
+					   u64 last_snapshot, bool trace_leaf)
+{
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct extent_buffer *eb;
+	bool need_cleanup = false;
+	int ret = 0;
+	int i;
+
+	/* Level sanity check */
+	if (cur_level < 0 || cur_level >= BTRFS_MAX_LEVEL ||
+	    root_level < 0 || root_level >= BTRFS_MAX_LEVEL ||
+	    root_level < cur_level) {
+		btrfs_err_rl(fs_info,
+			"%s: bad levels, cur_level=%d root_level=%d",
+			__func__, cur_level, root_level);
+		return -EUCLEAN;
+	}
+
+	/* Read the tree block if needed */
+	if (dst_path->nodes[cur_level] == NULL) {
+		struct btrfs_key first_key;
+		int parent_slot;
+		u64 child_gen;
+		u64 child_bytenr;
+
+		/*
+		 * dst_path->nodes[root_level] must be initialized before
+		 * calling this function.
+		 */
+		if (cur_level == root_level) {
+			btrfs_err_rl(fs_info,
+	"%s: dst_path->nodes[%d] not initialized, root_level=%d cur_level=%d",
+				__func__, root_level, root_level, cur_level);
+			return -EUCLEAN;
+		}
+
+		/*
+		 * We need to get child blockptr/gen from parent before we can
+		 * read it.
+		  */
+		eb = dst_path->nodes[cur_level + 1];
+		parent_slot = dst_path->slots[cur_level + 1];
+		child_bytenr = btrfs_node_blockptr(eb, parent_slot);
+		child_gen = btrfs_node_ptr_generation(eb, parent_slot);
+		btrfs_node_key_to_cpu(eb, &first_key, parent_slot);
+
+		/* This node is old, no need to trace */
+		if (child_gen < last_snapshot)
+			goto out;
+
+		eb = read_tree_block(fs_info, child_bytenr, child_gen,
+				     cur_level, &first_key);
+		if (IS_ERR(eb)) {
+			ret = PTR_ERR(eb);
+			goto out;
+		} else if (!extent_buffer_uptodate(eb)) {
+			free_extent_buffer(eb);
+			ret = -EIO;
+			goto out;
+		}
+
+		dst_path->nodes[cur_level] = eb;
+		dst_path->slots[cur_level] = 0;
+
+		btrfs_tree_read_lock(eb);
+		btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
+		dst_path->locks[cur_level] = BTRFS_READ_LOCK_BLOCKING;
+		need_cleanup = true;
+	}
+
+	/* Now record this tree block and its counter part for qgroups */
+	ret = qgroup_trace_extent_swap(trans, src_eb, dst_path, cur_level,
+				       root_level, trace_leaf);
+	if (ret < 0)
+		goto cleanup;
+
+	eb = dst_path->nodes[cur_level];
+
+	if (cur_level > 0) {
+		/* Iterate all child tree blocks */
+		for (i = 0; i < btrfs_header_nritems(eb); i++) {
+			/* Skip old tree blocks as they won't be swapped */
+			if (btrfs_node_ptr_generation(eb, i) < last_snapshot)
+				continue;
+			dst_path->slots[cur_level] = i;
+
+			/* Recursive call (at most 7 times) */
+			ret = qgroup_trace_new_subtree_blocks(trans, src_eb,
+					dst_path, cur_level - 1, root_level,
+					last_snapshot, trace_leaf);
+			if (ret < 0)
+				goto cleanup;
+		}
+	}
+
+cleanup:
+	if (need_cleanup) {
+		/* Clean up */
+		btrfs_tree_unlock_rw(dst_path->nodes[cur_level],
+				     dst_path->locks[cur_level]);
+		free_extent_buffer(dst_path->nodes[cur_level]);
+		dst_path->nodes[cur_level] = NULL;
+		dst_path->slots[cur_level] = 0;
+		dst_path->locks[cur_level] = 0;
+	}
+out:
+	return ret;
+}
+
+/*
+ * Inform qgroup to trace subtree swap used in balance.
+ *
+ * Unlike btrfs_qgroup_trace_subtree(), this function will only trace
+ * new tree blocks whose generation is equal to (or larger than) @last_snapshot.
+ *
+ * Will go down the tree block pointed by @dst_eb (pointed by @dst_parent and
+ * @dst_slot), and find any tree blocks whose generation is at @last_snapshot,
+ * and then go down @src_eb (pointed by @src_parent and @src_slot) to find
+ * the conterpart of the tree block, then mark both tree blocks as qgroup dirty,
+ * and skip all tree blocks whose generation is smaller than last_snapshot.
+ *
+ * This would skip tons of tree blocks of original btrfs_qgroup_trace_subtree(),
+ * which could be the cause of very slow balance if the file tree is large.
+ *
+ * @src_parent, @src_slot: pointer to src (file tree) eb.
+ * @dst_parent, @dst_slot: pointer to dst (reloc tree) eb.
+ */
+int btrfs_qgroup_trace_subtree_swap(struct btrfs_trans_handle *trans,
+				struct btrfs_block_group_cache *bg_cache,
+				struct extent_buffer *src_parent, int src_slot,
+				struct extent_buffer *dst_parent, int dst_slot,
+				u64 last_snapshot)
+{
+	struct btrfs_fs_info *fs_info = trans->fs_info;
+	struct btrfs_path *dst_path = NULL;
+	struct btrfs_key first_key;
+	struct extent_buffer *src_eb = NULL;
+	struct extent_buffer *dst_eb = NULL;
+	bool trace_leaf = false;
+	u64 child_gen;
+	u64 child_bytenr;
+	int level;
+	int ret;
+
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
+		return 0;
+
+	/* Check parameter order */
+	if (btrfs_node_ptr_generation(src_parent, src_slot) >
+	    btrfs_node_ptr_generation(dst_parent, dst_slot)) {
+		btrfs_err_rl(fs_info,
+		"%s: bad parameter order, src_gen=%llu dst_gen=%llu", __func__,
+			btrfs_node_ptr_generation(src_parent, src_slot),
+			btrfs_node_ptr_generation(dst_parent, dst_slot));
+		return -EUCLEAN;
+	}
+
+	/*
+	 * Only trace leaf if we're relocating data block groups, this could
+	 * reduce tons of data extents tracing for meta/sys bg relocation.
+	 */
+	if (bg_cache->flags & BTRFS_BLOCK_GROUP_DATA)
+		trace_leaf = true;
+	/* Read out real @src_eb, pointed by @src_parent and @src_slot */
+	child_bytenr = btrfs_node_blockptr(src_parent, src_slot);
+	child_gen = btrfs_node_ptr_generation(src_parent, src_slot);
+	btrfs_node_key_to_cpu(src_parent, &first_key, src_slot);
+
+	src_eb = read_tree_block(fs_info, child_bytenr, child_gen,
+			btrfs_header_level(src_parent) - 1, &first_key);
+	if (IS_ERR(src_eb)) {
+		ret = PTR_ERR(src_eb);
+		goto out;
+	}
+
+	/* Read out real @dst_eb, pointed by @src_parent and @src_slot */
+	child_bytenr = btrfs_node_blockptr(dst_parent, dst_slot);
+	child_gen = btrfs_node_ptr_generation(dst_parent, dst_slot);
+	btrfs_node_key_to_cpu(dst_parent, &first_key, dst_slot);
+
+	dst_eb = read_tree_block(fs_info, child_bytenr, child_gen,
+			btrfs_header_level(dst_parent) - 1, &first_key);
+	if (IS_ERR(dst_eb)) {
+		ret = PTR_ERR(dst_eb);
+		goto out;
+	}
+
+	if (!extent_buffer_uptodate(src_eb) || !extent_buffer_uptodate(dst_eb)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	level = btrfs_header_level(dst_eb);
+	dst_path = btrfs_alloc_path();
+	if (!dst_path) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	/* For dst_path */
+	extent_buffer_get(dst_eb);
+	dst_path->nodes[level] = dst_eb;
+	dst_path->slots[level] = 0;
+	dst_path->locks[level] = 0;
+
+	/* Do the generation-aware breadth-first search */
+	ret = qgroup_trace_new_subtree_blocks(trans, src_eb, dst_path, level,
+					      level, last_snapshot, trace_leaf);
+	if (ret < 0)
+		goto out;
+	ret = 0;
+
+out:
+	free_extent_buffer(src_eb);
+	free_extent_buffer(dst_eb);
+	btrfs_free_path(dst_path);
+	if (ret < 0)
+		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+	return ret;
+}
+
 int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 			       struct extent_buffer *root_eb,
 			       u64 root_gen, int root_level)
@@ -2132,6 +2543,7 @@ int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans)
 	struct btrfs_delayed_ref_root *delayed_refs;
 	struct ulist *new_roots = NULL;
 	struct rb_node *node;
+	u64 num_dirty_extents = 0;
 	u64 qgroup_to_skip;
 	int ret = 0;
 
@@ -2141,6 +2553,7 @@ int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans)
 		record = rb_entry(node, struct btrfs_qgroup_extent_record,
 				  node);
 
+		num_dirty_extents++;
 		trace_btrfs_qgroup_account_extents(fs_info, record);
 
 		if (!ret) {
@@ -2186,6 +2599,8 @@ cleanup:
 		kfree(record);
 
 	}
+	trace_qgroup_num_dirty_extents(fs_info, trans->transid,
+				       num_dirty_extents);
 	return ret;
 }
 
@@ -2897,6 +3312,7 @@ qgroup_rescan_zero_tracking(struct btrfs_fs_info *fs_info)
 		qgroup->rfer_cmpr = 0;
 		qgroup->excl = 0;
 		qgroup->excl_cmpr = 0;
+		qgroup_dirty(fs_info, qgroup);
 	}
 	spin_unlock(&fs_info->qgroup_lock);
 }
@@ -3004,7 +3420,7 @@ int btrfs_qgroup_reserve_data(struct inode *inode,
 	int ret;
 
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags) ||
-	    !is_fstree(root->objectid) || len == 0)
+	    !is_fstree(root->root_key.objectid) || len == 0)
 		return 0;
 
 	/* @reserved parameter is mandatory for qgroup */
@@ -3090,7 +3506,7 @@ static int qgroup_free_reserved_data(struct inode *inode,
 			goto out;
 		freed += changeset.bytes_changed;
 	}
-	btrfs_qgroup_free_refroot(root->fs_info, root->objectid, freed,
+	btrfs_qgroup_free_refroot(root->fs_info, root->root_key.objectid, freed,
 				  BTRFS_QGROUP_RSV_DATA);
 	ret = freed;
 out:
@@ -3106,6 +3522,10 @@ static int __btrfs_qgroup_release_data(struct inode *inode,
 	int trace_op = QGROUP_RELEASE;
 	int ret;
 
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED,
+		      &BTRFS_I(inode)->root->fs_info->flags))
+		return 0;
+
 	/* In release case, we shouldn't have @reserved */
 	WARN_ON(!free && reserved);
 	if (free && reserved)
@@ -3122,7 +3542,7 @@ static int __btrfs_qgroup_release_data(struct inode *inode,
 					changeset.bytes_changed, trace_op);
 	if (free)
 		btrfs_qgroup_free_refroot(BTRFS_I(inode)->root->fs_info,
-				BTRFS_I(inode)->root->objectid,
+				BTRFS_I(inode)->root->root_key.objectid,
 				changeset.bytes_changed, BTRFS_QGROUP_RSV_DATA);
 	ret = changeset.bytes_changed;
 out:
@@ -3215,7 +3635,7 @@ int __btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes,
 	int ret;
 
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) ||
-	    !is_fstree(root->objectid) || num_bytes == 0)
+	    !is_fstree(root->root_key.objectid) || num_bytes == 0)
 		return 0;
 
 	BUG_ON(num_bytes != round_down(num_bytes, fs_info->nodesize));
@@ -3240,13 +3660,13 @@ void btrfs_qgroup_free_meta_all_pertrans(struct btrfs_root *root)
 	struct btrfs_fs_info *fs_info = root->fs_info;
 
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) ||
-	    !is_fstree(root->objectid))
+	    !is_fstree(root->root_key.objectid))
 		return;
 
 	/* TODO: Update trace point to handle such free */
 	trace_qgroup_meta_free_all_pertrans(root);
 	/* Special value -1 means to free all reserved space */
-	btrfs_qgroup_free_refroot(fs_info, root->objectid, (u64)-1,
+	btrfs_qgroup_free_refroot(fs_info, root->root_key.objectid, (u64)-1,
 				  BTRFS_QGROUP_RSV_META_PERTRANS);
 }
 
@@ -3256,7 +3676,7 @@ void __btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes,
 	struct btrfs_fs_info *fs_info = root->fs_info;
 
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) ||
-	    !is_fstree(root->objectid))
+	    !is_fstree(root->root_key.objectid))
 		return;
 
 	/*
@@ -3267,7 +3687,8 @@ void __btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes,
 	num_bytes = sub_root_meta_rsv(root, num_bytes, type);
 	BUG_ON(num_bytes != round_down(num_bytes, fs_info->nodesize));
 	trace_qgroup_meta_reserve(root, type, -(s64)num_bytes);
-	btrfs_qgroup_free_refroot(fs_info, root->objectid, num_bytes, type);
+	btrfs_qgroup_free_refroot(fs_info, root->root_key.objectid,
+				  num_bytes, type);
 }
 
 static void qgroup_convert_meta(struct btrfs_fs_info *fs_info, u64 ref_root,
@@ -3321,13 +3742,13 @@ void btrfs_qgroup_convert_reserved_meta(struct btrfs_root *root, int num_bytes)
 	struct btrfs_fs_info *fs_info = root->fs_info;
 
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) ||
-	    !is_fstree(root->objectid))
+	    !is_fstree(root->root_key.objectid))
 		return;
 	/* Same as btrfs_qgroup_free_meta_prealloc() */
 	num_bytes = sub_root_meta_rsv(root, num_bytes,
 				      BTRFS_QGROUP_RSV_META_PREALLOC);
 	trace_qgroup_meta_convert(root, num_bytes);
-	qgroup_convert_meta(fs_info, root->objectid, num_bytes);
+	qgroup_convert_meta(fs_info, root->root_key.objectid, num_bytes);
 }
 
 /*
@@ -3354,7 +3775,7 @@ void btrfs_qgroup_check_reserved_leak(struct inode *inode)
 				inode->i_ino, unode->val, unode->aux);
 		}
 		btrfs_qgroup_free_refroot(BTRFS_I(inode)->root->fs_info,
-				BTRFS_I(inode)->root->objectid,
+				BTRFS_I(inode)->root->root_key.objectid,
 				changeset.bytes_changed, BTRFS_QGROUP_RSV_DATA);
 
 	}

fs/btrfs/qgroup.h

@@ -236,6 +236,12 @@ int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
 int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 			       struct extent_buffer *root_eb,
 			       u64 root_gen, int root_level);
+
+int btrfs_qgroup_trace_subtree_swap(struct btrfs_trans_handle *trans,
+				struct btrfs_block_group_cache *bg_cache,
+				struct extent_buffer *src_parent, int src_slot,
+				struct extent_buffer *dst_parent, int dst_slot,
+				u64 last_snapshot);
 int btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans, u64 bytenr,
 				u64 num_bytes, struct ulist *old_roots,
 				struct ulist *new_roots);
@@ -249,6 +255,8 @@ void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
 static inline void btrfs_qgroup_free_delayed_ref(struct btrfs_fs_info *fs_info,
 						 u64 ref_root, u64 num_bytes)
 {
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
+		return;
 	trace_btrfs_qgroup_free_delayed_ref(fs_info, ref_root, num_bytes);
 	btrfs_qgroup_free_refroot(fs_info, ref_root, num_bytes,
 				  BTRFS_QGROUP_RSV_DATA);

fs/btrfs/ref-verify.c

@@ -732,7 +732,7 @@ int btrfs_ref_tree_mod(struct btrfs_root *root, u64 bytenr, u64 num_bytes,
 
 	INIT_LIST_HEAD(&ra->list);
 	ra->action = action;
-	ra->root = root->objectid;
+	ra->root = root->root_key.objectid;
 
 	/*
 	 * This is an allocation, preallocate the block_entry in case we haven't
@@ -787,8 +787,8 @@ int btrfs_ref_tree_mod(struct btrfs_root *root, u64 bytenr, u64 num_bytes,
 			 * one we want to lookup below when we modify the
 			 * re->num_refs.
 			 */
-			ref_root = root->objectid;
-			re->root_objectid = root->objectid;
+			ref_root = root->root_key.objectid;
+			re->root_objectid = root->root_key.objectid;
 			re->num_refs = 0;
 		}
 
@@ -862,7 +862,7 @@ int btrfs_ref_tree_mod(struct btrfs_root *root, u64 bytenr, u64 num_bytes,
 			 * didn't thik of some other corner case.
 			 */
 			btrfs_err(fs_info, "failed to find root %llu for %llu",
-				  root->objectid, be->bytenr);
+				  root->root_key.objectid, be->bytenr);
 			dump_block_entry(fs_info, be);
 			dump_ref_action(fs_info, ra);
 			kfree(ra);

fs/btrfs/relocation.c

@@ -648,8 +648,8 @@ struct backref_node *build_backref_tree(struct reloc_control *rc,
 					int level, u64 bytenr)
 {
 	struct backref_cache *cache = &rc->backref_cache;
-	struct btrfs_path *path1;
-	struct btrfs_path *path2;
+	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;
@@ -662,7 +662,7 @@ struct backref_node *build_backref_tree(struct reloc_control *rc,
 	struct btrfs_key key;
 	unsigned long end;
 	unsigned long ptr;
-	LIST_HEAD(list);
+	LIST_HEAD(list); /* Pending edge list, upper node needs to be checked */
 	LIST_HEAD(useless);
 	int cowonly;
 	int ret;
@@ -778,6 +778,10 @@ again:
 				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) ||
@@ -787,11 +791,12 @@ again:
 			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 of this type.
+				 * Only root blocks of reloc trees use backref
+				 * pointing to itself.
 				 */
 				root = find_reloc_root(rc, cur->bytenr);
 				ASSERT(root);
@@ -840,7 +845,11 @@ again:
 			goto next;
 		}
 
-		/* key.type == BTRFS_TREE_BLOCK_REF_KEY */
+		/*
+		 * 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);
@@ -863,10 +872,7 @@ again:
 
 		level = cur->level + 1;
 
-		/*
-		 * searching the tree to find upper level blocks
-		 * reference the block.
-		 */
+		/* Search the tree to find parent blocks referring the block. */
 		path2->search_commit_root = 1;
 		path2->skip_locking = 1;
 		path2->lowest_level = level;
@@ -884,7 +890,8 @@ again:
 		    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->objectid,
+				  cur->bytenr, level - 1,
+				  root->root_key.objectid,
 				  node_key->objectid, node_key->type,
 				  node_key->offset);
 			err = -ENOENT;
@@ -892,6 +899,8 @@ again:
 		}
 		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) ==
@@ -1281,7 +1290,7 @@ static void __del_reloc_root(struct btrfs_root *root)
 	struct mapping_node *node = NULL;
 	struct reloc_control *rc = fs_info->reloc_ctl;
 
-	if (rc) {
+	if (rc && root->node) {
 		spin_lock(&rc->reloc_root_tree.lock);
 		rb_node = tree_search(&rc->reloc_root_tree.rb_root,
 				      root->node->start);
@@ -1735,7 +1744,7 @@ int memcmp_node_keys(struct extent_buffer *eb, int slot,
  * errors, a negative error number is returned.
  */
 static noinline_for_stack
-int replace_path(struct btrfs_trans_handle *trans,
+int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc,
 		 struct btrfs_root *dest, struct btrfs_root *src,
 		 struct btrfs_path *path, struct btrfs_key *next_key,
 		 int lowest_level, int max_level)
@@ -1879,14 +1888,9 @@ again:
 		 *    and tree block numbers, if current trans doesn't free
 		 *    data reloc tree inode.
 		 */
-		ret = btrfs_qgroup_trace_subtree(trans, parent,
-				btrfs_header_generation(parent),
-				btrfs_header_level(parent));
-		if (ret < 0)
-			break;
-		ret = btrfs_qgroup_trace_subtree(trans, path->nodes[level],
-				btrfs_header_generation(path->nodes[level]),
-				btrfs_header_level(path->nodes[level]));
+		ret = btrfs_qgroup_trace_subtree_swap(trans, rc->block_group,
+				parent, slot, path->nodes[level],
+				path->slots[level], last_snapshot);
 		if (ret < 0)
 			break;
 
@@ -2205,7 +2209,7 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
 		    btrfs_comp_cpu_keys(&next_key, &key) >= 0) {
 			ret = 0;
 		} else {
-			ret = replace_path(trans, root, reloc_root, path,
+			ret = replace_path(trans, rc, root, reloc_root, path,
 					   &next_key, level, max_level);
 		}
 		if (ret < 0) {
@@ -2911,7 +2915,6 @@ static int get_tree_block_key(struct btrfs_fs_info *fs_info,
 		free_extent_buffer(eb);
 		return -EIO;
 	}
-	WARN_ON(btrfs_header_level(eb) != block->level);
 	if (block->level == 0)
 		btrfs_item_key_to_cpu(eb, &block->key, 0);
 	else
@@ -2987,7 +2990,7 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
 	struct backref_node *node;
 	struct btrfs_path *path;
 	struct tree_block *block;
-	struct rb_node *rb_node;
+	struct tree_block *next;
 	int ret;
 	int err = 0;
 
@@ -2997,29 +3000,23 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
 		goto out_free_blocks;
 	}
 
-	rb_node = rb_first(blocks);
-	while (rb_node) {
-		block = rb_entry(rb_node, struct tree_block, rb_node);
+	/* Kick in readahead for tree blocks with missing keys */
+	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
 		if (!block->key_ready)
 			readahead_tree_block(fs_info, block->bytenr);
-		rb_node = rb_next(rb_node);
 	}
 
-	rb_node = rb_first(blocks);
-	while (rb_node) {
-		block = rb_entry(rb_node, struct tree_block, rb_node);
+	/* Get first keys */
+	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
 		if (!block->key_ready) {
 			err = get_tree_block_key(fs_info, block);
 			if (err)
 				goto out_free_path;
 		}
-		rb_node = rb_next(rb_node);
 	}
 
-	rb_node = rb_first(blocks);
-	while (rb_node) {
-		block = rb_entry(rb_node, struct tree_block, rb_node);
-
+	/* Do tree relocation */
+	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
 		node = build_backref_tree(rc, &block->key,
 					  block->level, block->bytenr);
 		if (IS_ERR(node)) {
@@ -3030,11 +3027,10 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
 		ret = relocate_tree_block(trans, rc, node, &block->key,
 					  path);
 		if (ret < 0) {
-			if (ret != -EAGAIN || rb_node == rb_first(blocks))
+			if (ret != -EAGAIN || &block->rb_node == rb_first(blocks))
 				err = ret;
 			goto out;
 		}
-		rb_node = rb_next(rb_node);
 	}
 out:
 	err = finish_pending_nodes(trans, rc, path, err);
@@ -4669,7 +4665,7 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 	if (rc->merge_reloc_tree) {
 		ret = btrfs_block_rsv_migrate(&pending->block_rsv,
 					      rc->block_rsv,
-					      rc->nodes_relocated, 1);
+					      rc->nodes_relocated, true);
 		if (ret)
 			return ret;
 	}

fs/btrfs/scrub.c

@@ -1124,7 +1124,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 
 			if (scrub_write_page_to_dev_replace(sblock_other,
 							    page_num) != 0) {
-				btrfs_dev_replace_stats_inc(
+				atomic64_inc(
 					&fs_info->dev_replace.num_write_errors);
 				success = 0;
 			}
@@ -1564,8 +1564,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
 		if (btrfsic_submit_bio_wait(bio)) {
 			btrfs_dev_stat_inc_and_print(page_bad->dev,
 				BTRFS_DEV_STAT_WRITE_ERRS);
-			btrfs_dev_replace_stats_inc(
-				&fs_info->dev_replace.num_write_errors);
+			atomic64_inc(&fs_info->dev_replace.num_write_errors);
 			bio_put(bio);
 			return -EIO;
 		}
@@ -1592,8 +1591,7 @@ static void scrub_write_block_to_dev_replace(struct scrub_block *sblock)
 
 		ret = scrub_write_page_to_dev_replace(sblock, page_num);
 		if (ret)
-			btrfs_dev_replace_stats_inc(
-				&fs_info->dev_replace.num_write_errors);
+			atomic64_inc(&fs_info->dev_replace.num_write_errors);
 	}
 }
 
@@ -1726,8 +1724,7 @@ static void scrub_wr_bio_end_io_worker(struct btrfs_work *work)
 			struct scrub_page *spage = sbio->pagev[i];
 
 			spage->io_error = 1;
-			btrfs_dev_replace_stats_inc(&dev_replace->
-						    num_write_errors);
+			atomic64_inc(&dev_replace->num_write_errors);
 		}
 	}
 
@@ -3022,8 +3019,7 @@ 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 is_dev_replace)
+					   int num, u64 base, u64 length)
 {
 	struct btrfs_path *path, *ppath;
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
@@ -3299,7 +3295,7 @@ again:
 			extent_physical = extent_logical - logical + physical;
 			extent_dev = scrub_dev;
 			extent_mirror_num = mirror_num;
-			if (is_dev_replace)
+			if (sctx->is_dev_replace)
 				scrub_remap_extent(fs_info, extent_logical,
 						   extent_len, &extent_physical,
 						   &extent_dev,
@@ -3397,8 +3393,7 @@ static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx,
 					  struct btrfs_device *scrub_dev,
 					  u64 chunk_offset, u64 length,
 					  u64 dev_offset,
-					  struct btrfs_block_group_cache *cache,
-					  int is_dev_replace)
+					  struct btrfs_block_group_cache *cache)
 {
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
 	struct btrfs_mapping_tree *map_tree = &fs_info->mapping_tree;
@@ -3435,8 +3430,7 @@ 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,
-					   is_dev_replace);
+					   chunk_offset, length);
 			if (ret)
 				goto out;
 		}
@@ -3449,8 +3443,7 @@ out:
 
 static noinline_for_stack
 int scrub_enumerate_chunks(struct scrub_ctx *sctx,
-			   struct btrfs_device *scrub_dev, u64 start, u64 end,
-			   int is_dev_replace)
+			   struct btrfs_device *scrub_dev, u64 start, u64 end)
 {
 	struct btrfs_dev_extent *dev_extent = NULL;
 	struct btrfs_path *path;
@@ -3544,7 +3537,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		 */
 		scrub_pause_on(fs_info);
 		ret = btrfs_inc_block_group_ro(cache);
-		if (!ret && is_dev_replace) {
+		if (!ret && sctx->is_dev_replace) {
 			/*
 			 * If we are doing a device replace wait for any tasks
 			 * that started dellaloc right before we set the block
@@ -3609,7 +3602,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		dev_replace->item_needs_writeback = 1;
 		btrfs_dev_replace_write_unlock(&fs_info->dev_replace);
 		ret = scrub_chunk(sctx, scrub_dev, chunk_offset, length,
-				  found_key.offset, cache, is_dev_replace);
+				  found_key.offset, cache);
 
 		/*
 		 * flush, submit all pending read and write bios, afterwards
@@ -3670,7 +3663,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		btrfs_put_block_group(cache);
 		if (ret)
 			break;
-		if (is_dev_replace &&
+		if (sctx->is_dev_replace &&
 		    atomic64_read(&dev_replace->num_write_errors) > 0) {
 			ret = -EIO;
 			break;
@@ -3893,8 +3886,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	}
 
 	if (!ret)
-		ret = scrub_enumerate_chunks(sctx, dev, start, end,
-					     is_dev_replace);
+		ret = scrub_enumerate_chunks(sctx, dev, start, end);
 
 	wait_event(sctx->list_wait, atomic_read(&sctx->bios_in_flight) == 0);
 	atomic_dec(&fs_info->scrubs_running);

fs/btrfs/send.c

@@ -1186,9 +1186,9 @@ static int __clone_root_cmp_bsearch(const void *key, const void *elt)
 	u64 root = (u64)(uintptr_t)key;
 	struct clone_root *cr = (struct clone_root *)elt;
 
-	if (root < cr->root->objectid)
+	if (root < cr->root->root_key.objectid)
 		return -1;
-	if (root > cr->root->objectid)
+	if (root > cr->root->root_key.objectid)
 		return 1;
 	return 0;
 }
@@ -1198,9 +1198,9 @@ static int __clone_root_cmp_sort(const void *e1, const void *e2)
 	struct clone_root *cr1 = (struct clone_root *)e1;
 	struct clone_root *cr2 = (struct clone_root *)e2;
 
-	if (cr1->root->objectid < cr2->root->objectid)
+	if (cr1->root->root_key.objectid < cr2->root->root_key.objectid)
 		return -1;
-	if (cr1->root->objectid > cr2->root->objectid)
+	if (cr1->root->root_key.objectid > cr2->root->root_key.objectid)
 		return 1;
 	return 0;
 }
@@ -1693,12 +1693,8 @@ static int lookup_dir_item_inode(struct btrfs_root *root,
 
 	di = btrfs_lookup_dir_item(NULL, root, path,
 			dir, name, name_len, 0);
-	if (!di) {
-		ret = -ENOENT;
-		goto out;
-	}
-	if (IS_ERR(di)) {
-		ret = PTR_ERR(di);
+	if (IS_ERR_OR_NULL(di)) {
+		ret = di ? PTR_ERR(di) : -ENOENT;
 		goto out;
 	}
 	btrfs_dir_item_key_to_cpu(path->nodes[0], di, &key);
@@ -2346,7 +2342,7 @@ static int send_subvol_begin(struct send_ctx *sctx)
 		return -ENOMEM;
 	}
 
-	key.objectid = send_root->objectid;
+	key.objectid = send_root->root_key.objectid;
 	key.type = BTRFS_ROOT_BACKREF_KEY;
 	key.offset = 0;
 
@@ -2362,7 +2358,7 @@ static int send_subvol_begin(struct send_ctx *sctx)
 	leaf = path->nodes[0];
 	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 	if (key.type != BTRFS_ROOT_BACKREF_KEY ||
-	    key.objectid != send_root->objectid) {
+	    key.objectid != send_root->root_key.objectid) {
 		ret = -ENOENT;
 		goto out;
 	}
@@ -4907,8 +4903,8 @@ static int send_clone(struct send_ctx *sctx,
 
 	btrfs_debug(sctx->send_root->fs_info,
 		    "send_clone offset=%llu, len=%d, clone_root=%llu, clone_inode=%llu, clone_offset=%llu",
-		    offset, len, clone_root->root->objectid, clone_root->ino,
-		    clone_root->offset);
+		    offset, len, clone_root->root->root_key.objectid,
+		    clone_root->ino, clone_root->offset);
 
 	p = fs_path_alloc();
 	if (!p)

fs/btrfs/super.c

@@ -2177,8 +2177,10 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_fsid.val[0] = be32_to_cpu(fsid[0]) ^ be32_to_cpu(fsid[2]);
 	buf->f_fsid.val[1] = be32_to_cpu(fsid[1]) ^ be32_to_cpu(fsid[3]);
 	/* Mask in the root object ID too, to disambiguate subvols */
-	buf->f_fsid.val[0] ^= BTRFS_I(d_inode(dentry))->root->objectid >> 32;
-	buf->f_fsid.val[1] ^= BTRFS_I(d_inode(dentry))->root->objectid;
+	buf->f_fsid.val[0] ^=
+		BTRFS_I(d_inode(dentry))->root->root_key.objectid >> 32;
+	buf->f_fsid.val[1] ^=
+		BTRFS_I(d_inode(dentry))->root->root_key.objectid;
 
 	return 0;
 }

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

@@ -106,7 +106,7 @@ static int test_find_delalloc(u32 sectorsize)
 	set_extent_delalloc(&tmp, 0, sectorsize - 1, 0, NULL);
 	start = 0;
 	end = 0;
-	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+	found = btrfs_find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (!found) {
 		test_err("should have found at least one delalloc");
@@ -137,7 +137,7 @@ static int test_find_delalloc(u32 sectorsize)
 	set_extent_delalloc(&tmp, sectorsize, max_bytes - 1, 0, NULL);
 	start = test_start;
 	end = 0;
-	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+	found = btrfs_find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (!found) {
 		test_err("couldn't find delalloc in our range");
@@ -171,7 +171,7 @@ static int test_find_delalloc(u32 sectorsize)
 	}
 	start = test_start;
 	end = 0;
-	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+	found = btrfs_find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (found) {
 		test_err("found range when we shouldn't have");
@@ -192,7 +192,7 @@ static int test_find_delalloc(u32 sectorsize)
 	set_extent_delalloc(&tmp, max_bytes, total_dirty - 1, 0, NULL);
 	start = test_start;
 	end = 0;
-	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+	found = btrfs_find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (!found) {
 		test_err("didn't find our range");
@@ -233,7 +233,7 @@ static int test_find_delalloc(u32 sectorsize)
 	 * this changes at any point in the future we will need to fix this
 	 * tests expected behavior.
 	 */
-	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+	found = btrfs_find_lock_delalloc_range(inode, &tmp, locked_page, &start,
 					 &end, max_bytes);
 	if (!found) {
 		test_err("didn't find our range");

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

@@ -12,8 +12,8 @@ static void free_extent_map_tree(struct extent_map_tree *em_tree)
 	struct extent_map *em;
 	struct rb_node *node;
 
-	while (!RB_EMPTY_ROOT(&em_tree->map)) {
-		node = rb_first(&em_tree->map);
+	while (!RB_EMPTY_ROOT(&em_tree->map.rb_root)) {
+		node = rb_first_cached(&em_tree->map);
 		em = rb_entry(node, struct extent_map, rb_node);
 		remove_extent_mapping(em_tree, em);
 

fs/btrfs/transaction.c

@@ -44,7 +44,8 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction)
 	WARN_ON(refcount_read(&transaction->use_count) == 0);
 	if (refcount_dec_and_test(&transaction->use_count)) {
 		BUG_ON(!list_empty(&transaction->list));
-		WARN_ON(!RB_EMPTY_ROOT(&transaction->delayed_refs.href_root));
+		WARN_ON(!RB_EMPTY_ROOT(
+				&transaction->delayed_refs.href_root.rb_root));
 		if (transaction->delayed_refs.pending_csums)
 			btrfs_err(transaction->fs_info,
 				  "pending csums is %llu",
@@ -118,7 +119,7 @@ static noinline void switch_commit_roots(struct btrfs_transaction *trans)
 		list_del_init(&root->dirty_list);
 		free_extent_buffer(root->commit_root);
 		root->commit_root = btrfs_root_node(root);
-		if (is_fstree(root->objectid))
+		if (is_fstree(root->root_key.objectid))
 			btrfs_unpin_free_ino(root);
 		clear_btree_io_tree(&root->dirty_log_pages);
 	}
@@ -245,7 +246,7 @@ loop:
 
 	memset(&cur_trans->delayed_refs, 0, sizeof(cur_trans->delayed_refs));
 
-	cur_trans->delayed_refs.href_root = RB_ROOT;
+	cur_trans->delayed_refs.href_root = RB_ROOT_CACHED;
 	cur_trans->delayed_refs.dirty_extent_root = RB_ROOT;
 	atomic_set(&cur_trans->delayed_refs.num_entries, 0);
 
@@ -759,7 +760,7 @@ static int should_end_transaction(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 
-	if (btrfs_check_space_for_delayed_refs(trans, fs_info))
+	if (btrfs_check_space_for_delayed_refs(trans))
 		return 1;
 
 	return !!btrfs_block_rsv_check(&fs_info->global_block_rsv, 5);
@@ -834,7 +835,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 	trans->delayed_ref_updates = 0;
 	if (!trans->sync) {
 		must_run_delayed_refs =
-			btrfs_should_throttle_delayed_refs(trans, info);
+			btrfs_should_throttle_delayed_refs(trans);
 		cur = max_t(unsigned long, cur, 32);
 
 		/*
@@ -1197,7 +1198,10 @@ again:
 
 	list_add_tail(&fs_info->extent_root->dirty_list,
 		      &trans->transaction->switch_commits);
-	btrfs_after_dev_replace_commit(fs_info);
+
+	/* Update dev-replace pointer once everything is committed */
+	fs_info->dev_replace.committed_cursor_left =
+		fs_info->dev_replace.cursor_left_last_write_of_item;
 
 	return 0;
 }
@@ -1613,10 +1617,9 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	if (ret < 0)
 		goto fail;
 
-	ret = btrfs_insert_dir_item(trans, parent_root,
-				    dentry->d_name.name, dentry->d_name.len,
-				    BTRFS_I(parent_inode), &key,
-				    BTRFS_FT_DIR, index);
+	ret = btrfs_insert_dir_item(trans, dentry->d_name.name,
+				    dentry->d_name.len, BTRFS_I(parent_inode),
+				    &key, BTRFS_FT_DIR, index);
 	/* We have check then name at the beginning, so it is impossible. */
 	BUG_ON(ret == -EEXIST || ret == -EOVERFLOW);
 	if (ret) {
@@ -1929,6 +1932,9 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 		return ret;
 	}
 
+	btrfs_trans_release_metadata(trans);
+	trans->block_rsv = NULL;
+
 	/* make a pass through all the delayed refs we have so far
 	 * any runnings procs may add more while we are here
 	 */
@@ -1938,9 +1944,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 		return ret;
 	}
 
-	btrfs_trans_release_metadata(trans);
-	trans->block_rsv = NULL;
-
 	cur_trans = trans->transaction;
 
 	/*
@@ -2330,7 +2333,7 @@ int btrfs_clean_one_deleted_snapshot(struct btrfs_root *root)
 	list_del_init(&root->root_list);
 	spin_unlock(&fs_info->trans_lock);
 
-	btrfs_debug(fs_info, "cleaner removing %llu", root->objectid);
+	btrfs_debug(fs_info, "cleaner removing %llu", root->root_key.objectid);
 
 	btrfs_kill_all_delayed_nodes(root);
 

fs/btrfs/tree-checker.c

@@ -487,6 +487,13 @@ static int check_leaf(struct btrfs_fs_info *fs_info, struct extent_buffer *leaf,
 	u32 nritems = btrfs_header_nritems(leaf);
 	int slot;
 
+	if (btrfs_header_level(leaf) != 0) {
+		generic_err(fs_info, leaf, 0,
+			"invalid level for leaf, have %d expect 0",
+			btrfs_header_level(leaf));
+		return -EUCLEAN;
+	}
+
 	/*
 	 * Extent buffers from a relocation tree have a owner field that
 	 * corresponds to the subvolume tree they are based on. So just from an
@@ -645,9 +652,16 @@ int btrfs_check_node(struct btrfs_fs_info *fs_info, struct extent_buffer *node)
 	unsigned long nr = btrfs_header_nritems(node);
 	struct btrfs_key key, next_key;
 	int slot;
+	int level = btrfs_header_level(node);
 	u64 bytenr;
 	int ret = 0;
 
+	if (level <= 0 || level >= BTRFS_MAX_LEVEL) {
+		generic_err(fs_info, node, 0,
+			"invalid level for node, have %d expect [1, %d]",
+			level, BTRFS_MAX_LEVEL - 1);
+		return -EUCLEAN;
+	}
 	if (nr == 0 || nr > BTRFS_NODEPTRS_PER_BLOCK(fs_info)) {
 		btrfs_crit(fs_info,
 "corrupt node: root=%llu block=%llu, nritems too %s, have %lu expect range [1,%u]",

fs/btrfs/tree-log.c

@@ -205,14 +205,11 @@ static int join_running_log_trans(struct btrfs_root *root)
  * until you call btrfs_end_log_trans() or it makes any future
  * log transactions wait until you call btrfs_end_log_trans()
  */
-int btrfs_pin_log_trans(struct btrfs_root *root)
+void btrfs_pin_log_trans(struct btrfs_root *root)
 {
-	int ret = -ENOENT;
-
 	mutex_lock(&root->log_mutex);
 	atomic_inc(&root->log_writers);
 	mutex_unlock(&root->log_mutex);
-	return ret;
 }
 
 /*
@@ -258,6 +255,13 @@ struct walk_control {
 	/* what stage of the replay code we're currently in */
 	int stage;
 
+	/*
+	 * Ignore any items from the inode currently being processed. Needs
+	 * to be set every time we find a BTRFS_INODE_ITEM_KEY and we are in
+	 * the LOG_WALK_REPLAY_INODES stage.
+	 */
+	bool ignore_cur_inode;
+
 	/* the root we are currently replaying */
 	struct btrfs_root *replay_dest;
 
@@ -2487,6 +2491,20 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 
 			inode_item = btrfs_item_ptr(eb, i,
 					    struct btrfs_inode_item);
+			/*
+			 * If we have a tmpfile (O_TMPFILE) that got fsync'ed
+			 * and never got linked before the fsync, skip it, as
+			 * replaying it is pointless since it would be deleted
+			 * later. We skip logging tmpfiles, but it's always
+			 * possible we are replaying a log created with a kernel
+			 * that used to log tmpfiles.
+			 */
+			if (btrfs_inode_nlink(eb, inode_item) == 0) {
+				wc->ignore_cur_inode = true;
+				continue;
+			} else {
+				wc->ignore_cur_inode = false;
+			}
 			ret = replay_xattr_deletes(wc->trans, root, log,
 						   path, key.objectid);
 			if (ret)
@@ -2524,16 +2542,8 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 					     root->fs_info->sectorsize);
 				ret = btrfs_drop_extents(wc->trans, root, inode,
 							 from, (u64)-1, 1);
-				/*
-				 * If the nlink count is zero here, the iput
-				 * will free the inode.  We bump it to make
-				 * sure it doesn't get freed until the link
-				 * count fixup is done.
-				 */
 				if (!ret) {
-					if (inode->i_nlink == 0)
-						inc_nlink(inode);
-					/* Update link count and nbytes. */
+					/* Update the inode's nbytes. */
 					ret = btrfs_update_inode(wc->trans,
 								 root, inode);
 				}
@@ -2548,6 +2558,9 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 				break;
 		}
 
+		if (wc->ignore_cur_inode)
+			continue;
+
 		if (key.type == BTRFS_DIR_INDEX_KEY &&
 		    wc->stage == LOG_WALK_REPLAY_DIR_INDEX) {
 			ret = replay_one_dir_item(wc->trans, root, path,
@@ -3196,9 +3209,12 @@ static void free_log_tree(struct btrfs_trans_handle *trans,
 	};
 
 	ret = walk_log_tree(trans, log, &wc);
-	/* I don't think this can happen but just in case */
-	if (ret)
-		btrfs_abort_transaction(trans, ret);
+	if (ret) {
+		if (trans)
+			btrfs_abort_transaction(trans, ret);
+		else
+			btrfs_handle_fs_error(log->fs_info, ret, NULL);
+	}
 
 	while (1) {
 		ret = find_first_extent_bit(&log->dirty_log_pages,
@@ -5564,9 +5580,33 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
 
 			dir_inode = btrfs_iget(fs_info->sb, &inode_key,
 					       root, NULL);
-			/* If parent inode was deleted, skip it. */
-			if (IS_ERR(dir_inode))
-				continue;
+			/*
+			 * If the parent inode was deleted, return an error to
+			 * fallback to a transaction commit. This is to prevent
+			 * getting an inode that was moved from one parent A to
+			 * a parent B, got its former parent A deleted and then
+			 * it got fsync'ed, from existing at both parents after
+			 * a log replay (and the old parent still existing).
+			 * Example:
+			 *
+			 * mkdir /mnt/A
+			 * mkdir /mnt/B
+			 * touch /mnt/B/bar
+			 * sync
+			 * mv /mnt/B/bar /mnt/A/bar
+			 * mv -T /mnt/A /mnt/B
+			 * fsync /mnt/B/bar
+			 * <power fail>
+			 *
+			 * If we ignore the old parent B which got deleted,
+			 * after a log replay we would have file bar linked
+			 * at both parents and the old parent B would still
+			 * exist.
+			 */
+			if (IS_ERR(dir_inode)) {
+				ret = PTR_ERR(dir_inode);
+				goto out;
+			}
 
 			if (ctx)
 				ctx->log_new_dentries = false;
@@ -5640,7 +5680,13 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 	if (ret)
 		goto end_no_trans;
 
-	if (btrfs_inode_in_log(inode, trans->transid)) {
+	/*
+	 * Skip already logged inodes or inodes corresponding to tmpfiles
+	 * (since logging them is pointless, a link count of 0 means they
+	 * will never be accessible).
+	 */
+	if (btrfs_inode_in_log(inode, trans->transid) ||
+	    inode->vfs_inode.i_nlink == 0) {
 		ret = BTRFS_NO_LOG_SYNC;
 		goto end_no_trans;
 	}

fs/btrfs/tree-log.h

@@ -65,7 +65,7 @@ int btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans,
 			       const char *name, int name_len,
 			       struct btrfs_inode *inode, u64 dirid);
 void btrfs_end_log_trans(struct btrfs_root *root);
-int btrfs_pin_log_trans(struct btrfs_root *root);
+void btrfs_pin_log_trans(struct btrfs_root *root);
 void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
 			     struct btrfs_inode *dir, struct btrfs_inode *inode,
 			     int for_rename);

fs/btrfs/volumes.c

@@ -1613,7 +1613,7 @@ static u64 find_next_chunk(struct btrfs_fs_info *fs_info)
 
 	em_tree = &fs_info->mapping_tree.map_tree;
 	read_lock(&em_tree->lock);
-	n = rb_last(&em_tree->map);
+	n = rb_last(&em_tree->map.rb_root);
 	if (n) {
 		em = rb_entry(n, struct extent_map, rb_node);
 		ret = em->start + em->len;
@@ -1854,6 +1854,24 @@ void btrfs_assign_next_active_device(struct btrfs_device *device,
 		fs_info->fs_devices->latest_bdev = next_device->bdev;
 }
 
+/*
+ * Return btrfs_fs_devices::num_devices excluding the device that's being
+ * currently replaced.
+ */
+static u64 btrfs_num_devices(struct btrfs_fs_info *fs_info)
+{
+	u64 num_devices = fs_info->fs_devices->num_devices;
+
+	btrfs_dev_replace_read_lock(&fs_info->dev_replace);
+	if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace)) {
+		ASSERT(num_devices > 1);
+		num_devices--;
+	}
+	btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
+
+	return num_devices;
+}
+
 int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 		u64 devid)
 {
@@ -1865,22 +1883,22 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 
 	mutex_lock(&uuid_mutex);
 
-	num_devices = fs_devices->num_devices;
-	btrfs_dev_replace_read_lock(&fs_info->dev_replace);
-	if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace)) {
-		WARN_ON(num_devices < 1);
-		num_devices--;
-	}
-	btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
+	num_devices = btrfs_num_devices(fs_info);
 
 	ret = btrfs_check_raid_min_devices(fs_info, num_devices - 1);
 	if (ret)
 		goto out;
 
-	ret = btrfs_find_device_by_devspec(fs_info, devid, device_path,
-					   &device);
-	if (ret)
+	device = btrfs_find_device_by_devspec(fs_info, devid, device_path);
+
+	if (IS_ERR(device)) {
+		if (PTR_ERR(device) == -ENOENT &&
+		    strcmp(device_path, "missing") == 0)
+			ret = BTRFS_ERROR_DEV_MISSING_NOT_FOUND;
+		else
+			ret = PTR_ERR(device);
 		goto out;
+	}
 
 	if (test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) {
 		ret = BTRFS_ERROR_DEV_TGT_REPLACE;
@@ -2096,9 +2114,8 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev)
 	call_rcu(&tgtdev->rcu, free_device_rcu);
 }
 
-static int btrfs_find_device_by_path(struct btrfs_fs_info *fs_info,
-				     const char *device_path,
-				     struct btrfs_device **device)
+static struct btrfs_device *btrfs_find_device_by_path(
+		struct btrfs_fs_info *fs_info, const char *device_path)
 {
 	int ret = 0;
 	struct btrfs_super_block *disk_super;
@@ -2106,28 +2123,27 @@ static int btrfs_find_device_by_path(struct btrfs_fs_info *fs_info,
 	u8 *dev_uuid;
 	struct block_device *bdev;
 	struct buffer_head *bh;
+	struct btrfs_device *device;
 
-	*device = NULL;
 	ret = btrfs_get_bdev_and_sb(device_path, FMODE_READ,
 				    fs_info->bdev_holder, 0, &bdev, &bh);
 	if (ret)
-		return ret;
+		return ERR_PTR(ret);
 	disk_super = (struct btrfs_super_block *)bh->b_data;
 	devid = btrfs_stack_device_id(&disk_super->dev_item);
 	dev_uuid = disk_super->dev_item.uuid;
-	*device = btrfs_find_device(fs_info, devid, dev_uuid, disk_super->fsid);
+	device = btrfs_find_device(fs_info, devid, dev_uuid, disk_super->fsid);
 	brelse(bh);
-	if (!*device)
-		ret = -ENOENT;
+	if (!device)
+		device = ERR_PTR(-ENOENT);
 	blkdev_put(bdev, FMODE_READ);
-	return ret;
+	return device;
 }
 
-int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
-					 const char *device_path,
-					 struct btrfs_device **device)
+static struct btrfs_device *btrfs_find_device_missing_or_by_path(
+		struct btrfs_fs_info *fs_info, const char *device_path)
 {
-	*device = NULL;
+	struct btrfs_device *device = NULL;
 	if (strcmp(device_path, "missing") == 0) {
 		struct list_head *devices;
 		struct btrfs_device *tmp;
@@ -2136,42 +2152,38 @@ int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
 		list_for_each_entry(tmp, devices, dev_list) {
 			if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
 					&tmp->dev_state) && !tmp->bdev) {
-				*device = tmp;
+				device = tmp;
 				break;
 			}
 		}
 
-		if (!*device)
-			return BTRFS_ERROR_DEV_MISSING_NOT_FOUND;
-
-		return 0;
+		if (!device)
+			return ERR_PTR(-ENOENT);
 	} else {
-		return btrfs_find_device_by_path(fs_info, device_path, device);
+		device = btrfs_find_device_by_path(fs_info, device_path);
 	}
+
+	return device;
 }
 
 /*
  * Lookup a device given by device id, or the path if the id is 0.
  */
-int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
-				 const char *devpath,
-				 struct btrfs_device **device)
+struct btrfs_device *btrfs_find_device_by_devspec(
+		struct btrfs_fs_info *fs_info, u64 devid, const char *devpath)
 {
-	int ret;
+	struct btrfs_device *device;
 
 	if (devid) {
-		ret = 0;
-		*device = btrfs_find_device(fs_info, devid, NULL, NULL);
-		if (!*device)
-			ret = -ENOENT;
+		device = btrfs_find_device(fs_info, devid, NULL, NULL);
+		if (!device)
+			return ERR_PTR(-ENOENT);
 	} else {
 		if (!devpath || !devpath[0])
-			return -EINVAL;
-
-		ret = btrfs_find_device_missing_or_by_path(fs_info, devpath,
-							   device);
+			return ERR_PTR(-EINVAL);
+		device = btrfs_find_device_missing_or_by_path(fs_info, devpath);
 	}
-	return ret;
+	return device;
 }
 
 /*
@@ -3679,7 +3691,7 @@ static int alloc_profile_is_valid(u64 flags, int extended)
 		return !extended; /* "0" is valid for usual profiles */
 
 	/* true if exactly one bit set */
-	return (flags & (flags - 1)) == 0;
+	return is_power_of_2(flags);
 }
 
 static inline int balance_need_close(struct btrfs_fs_info *fs_info)
@@ -3740,13 +3752,8 @@ int btrfs_balance(struct btrfs_fs_info *fs_info,
 		}
 	}
 
-	num_devices = fs_info->fs_devices->num_devices;
-	btrfs_dev_replace_read_lock(&fs_info->dev_replace);
-	if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace)) {
-		BUG_ON(num_devices < 1);
-		num_devices--;
-	}
-	btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
+	num_devices = btrfs_num_devices(fs_info);
+
 	allowed = BTRFS_AVAIL_ALLOC_BIT_SINGLE | BTRFS_BLOCK_GROUP_DUP;
 	if (num_devices > 1)
 		allowed |= (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID1);
@@ -5897,7 +5904,11 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
 	}
 out:
 	if (dev_replace_is_ongoing) {
-		btrfs_dev_replace_clear_lock_blocking(dev_replace);
+		ASSERT(atomic_read(&dev_replace->blocking_readers) > 0);
+		btrfs_dev_replace_read_lock(dev_replace);
+		/* Barrier implied by atomic_dec_and_test */
+		if (atomic_dec_and_test(&dev_replace->blocking_readers))
+			cond_wake_up_nomb(&dev_replace->read_lock_wq);
 		btrfs_dev_replace_read_unlock(dev_replace);
 	}
 	free_extent_map(em);
@@ -7438,7 +7449,7 @@ static int verify_chunk_dev_extent_mapping(struct btrfs_fs_info *fs_info)
 	int ret = 0;
 
 	read_lock(&em_tree->lock);
-	for (node = rb_first(&em_tree->map); node; node = rb_next(node)) {
+	for (node = rb_first_cached(&em_tree->map); node; node = rb_next(node)) {
 		em = rb_entry(node, struct extent_map, rb_node);
 		if (em->map_lookup->num_stripes !=
 		    em->map_lookup->verified_stripes) {

fs/btrfs/volumes.h

@@ -410,12 +410,9 @@ int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
 void btrfs_free_extra_devids(struct btrfs_fs_devices *fs_devices, int step);
 void btrfs_assign_next_active_device(struct btrfs_device *device,
 				     struct btrfs_device *this_dev);
-int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
-					 const char *device_path,
-					 struct btrfs_device **device);
-int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
-					 const char *devpath,
-					 struct btrfs_device **device);
+struct btrfs_device *btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info,
+						  u64 devid,
+						  const char *devpath);
 struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 					const u64 *devid,
 					const u8 *uuid);

include/trace/events/btrfs.h

@@ -316,7 +316,7 @@ DECLARE_EVENT_CLASS(btrfs__file_extent_item_regular,
 	),
 
 	TP_fast_assign_btrfs(bi->root->fs_info,
-		__entry->root_obj	= bi->root->objectid;
+		__entry->root_obj	= bi->root->root_key.objectid;
 		__entry->ino		= btrfs_ino(bi);
 		__entry->isize		= bi->vfs_inode.i_size;
 		__entry->disk_isize	= bi->disk_i_size;
@@ -367,7 +367,7 @@ DECLARE_EVENT_CLASS(
 
 	TP_fast_assign_btrfs(
 		bi->root->fs_info,
-		__entry->root_obj	= bi->root->objectid;
+		__entry->root_obj	= bi->root->root_key.objectid;
 		__entry->ino		= btrfs_ino(bi);
 		__entry->isize		= bi->vfs_inode.i_size;
 		__entry->disk_isize	= bi->disk_i_size;
@@ -1477,7 +1477,8 @@ DECLARE_EVENT_CLASS(btrfs__qgroup_rsv_data,
 	),
 
 	TP_fast_assign_btrfs(btrfs_sb(inode->i_sb),
-		__entry->rootid		= BTRFS_I(inode)->root->objectid;
+		__entry->rootid		=
+			BTRFS_I(inode)->root->root_key.objectid;
 		__entry->ino		= btrfs_ino(BTRFS_I(inode));
 		__entry->start		= start;
 		__entry->len		= len;
@@ -1575,6 +1576,27 @@ DEFINE_EVENT(btrfs_qgroup_extent, btrfs_qgroup_trace_extent,
 	TP_ARGS(fs_info, rec)
 );
 
+TRACE_EVENT(qgroup_num_dirty_extents,
+
+	TP_PROTO(const struct btrfs_fs_info *fs_info, u64 transid,
+		 u64 num_dirty_extents),
+
+	TP_ARGS(fs_info, transid, num_dirty_extents),
+
+	TP_STRUCT__entry_btrfs(
+		__field(	u64, transid			)
+		__field(	u64, num_dirty_extents		)
+	),
+
+	TP_fast_assign_btrfs(fs_info,
+		__entry->transid	   = transid;
+		__entry->num_dirty_extents = num_dirty_extents;
+	),
+
+	TP_printk_btrfs("transid=%llu num_dirty_extents=%llu",
+		__entry->transid, __entry->num_dirty_extents)
+);
+
 TRACE_EVENT(btrfs_qgroup_account_extent,
 
 	TP_PROTO(const struct btrfs_fs_info *fs_info, u64 transid, u64 bytenr,
@@ -1675,7 +1697,7 @@ TRACE_EVENT(qgroup_meta_reserve,
 	),
 
 	TP_fast_assign_btrfs(root->fs_info,
-		__entry->refroot	= root->objectid;
+		__entry->refroot	= root->root_key.objectid;
 		__entry->diff		= diff;
 	),
 
@@ -1697,7 +1719,7 @@ TRACE_EVENT(qgroup_meta_convert,
 	),
 
 	TP_fast_assign_btrfs(root->fs_info,
-		__entry->refroot	= root->objectid;
+		__entry->refroot	= root->root_key.objectid;
 		__entry->diff		= diff;
 	),
 
@@ -1721,7 +1743,7 @@ TRACE_EVENT(qgroup_meta_free_all_pertrans,
 	),
 
 	TP_fast_assign_btrfs(root->fs_info,
-		__entry->refroot	= root->objectid;
+		__entry->refroot	= root->root_key.objectid;
 		spin_lock(&root->qgroup_meta_rsv_lock);
 		__entry->diff		= -(s64)root->qgroup_meta_rsv_pertrans;
 		spin_unlock(&root->qgroup_meta_rsv_lock);
@@ -1802,7 +1824,7 @@ TRACE_EVENT(btrfs_inode_mod_outstanding_extents,
 	),
 
 	TP_fast_assign_btrfs(root->fs_info,
-		__entry->root_objectid	= root->objectid;
+		__entry->root_objectid	= root->root_key.objectid;
 		__entry->ino		= ino;
 		__entry->mod		= mod;
 	),