Merge branch 'for-linus-4.11' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs

Pull btrfs updates from Chris Mason:
 "This has a series of fixes and cleanups that Dave Sterba has been
  collecting.

  There is a pretty big variety here, cleaning up internal APIs and
  fixing corner cases"

* 'for-linus-4.11' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (124 commits)
  Btrfs: use the correct type when creating cow dio extent
  Btrfs: fix deadlock between dedup on same file and starting writeback
  btrfs: use btrfs_debug instead of pr_debug in transaction abort
  btrfs: btrfs_truncate_free_space_cache always allocates path
  btrfs: free-space-cache, clean up unnecessary root arguments
  btrfs: convert btrfs_inc_block_group_ro to accept fs_info
  btrfs: flush_space always takes fs_info->fs_root
  btrfs: pass fs_info to (more) routines that are only called with extent_root
  btrfs: qgroup: Move half of the qgroup accounting time out of commit trans
  btrfs: remove unused parameter from adjust_slots_upwards
  btrfs: remove unused parameters from __btrfs_write_out_cache
  btrfs: remove unused parameter from cleanup_write_cache_enospc
  btrfs: remove unused parameter from __add_inode_ref
  btrfs: remove unused parameter from clone_copy_inline_extent
  btrfs: remove unused parameters from btrfs_cmp_data
  btrfs: remove unused parameter from __add_inline_refs
  btrfs: remove unused parameters from scrub_setup_wr_ctx
  btrfs: remove unused parameter from create_snapshot
  btrfs: remove unused parameter from init_first_rw_device
  btrfs: remove unused parameter from __btrfs_alloc_chunk
  ...

fs/btrfs/backref.c

@@ -956,8 +956,7 @@ static int __add_delayed_refs(struct btrfs_delayed_ref_head *head, u64 seq,
 /*
  * add all inline backrefs for bytenr to the list
  */
-static int __add_inline_refs(struct btrfs_fs_info *fs_info,
-			     struct btrfs_path *path, u64 bytenr,
+static int __add_inline_refs(struct btrfs_path *path, u64 bytenr,
 			     int *info_level, struct list_head *prefs,
 			     struct ref_root *ref_tree,
 			     u64 *total_refs, u64 inum)
@@ -1284,7 +1283,7 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans,
 		 */
 		delayed_refs = &trans->transaction->delayed_refs;
 		spin_lock(&delayed_refs->lock);
-		head = btrfs_find_delayed_ref_head(trans, bytenr);
+		head = btrfs_find_delayed_ref_head(delayed_refs, bytenr);
 		if (head) {
 			if (!mutex_trylock(&head->mutex)) {
 				atomic_inc(&head->node.refs);
@@ -1354,7 +1353,7 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans,
 		if (key.objectid == bytenr &&
 		    (key.type == BTRFS_EXTENT_ITEM_KEY ||
 		     key.type == BTRFS_METADATA_ITEM_KEY)) {
-			ret = __add_inline_refs(fs_info, path, bytenr,
+			ret = __add_inline_refs(path, bytenr,
 						&info_level, &prefs,
 						ref_tree, &total_refs,
 						inum);

fs/btrfs/btrfs_inode.h

@@ -224,16 +224,16 @@ static inline void btrfs_insert_inode_hash(struct inode *inode)
 	__insert_inode_hash(inode, h);
 }
 
-static inline u64 btrfs_ino(struct inode *inode)
+static inline u64 btrfs_ino(struct btrfs_inode *inode)
 {
-	u64 ino = BTRFS_I(inode)->location.objectid;
+	u64 ino = inode->location.objectid;
 
 	/*
 	 * !ino: btree_inode
 	 * type == BTRFS_ROOT_ITEM_KEY: subvol dir
 	 */
-	if (!ino || BTRFS_I(inode)->location.type == BTRFS_ROOT_ITEM_KEY)
-		ino = inode->i_ino;
+	if (!ino || inode->location.type == BTRFS_ROOT_ITEM_KEY)
+		ino = inode->vfs_inode.i_ino;
 	return ino;
 }
 
@@ -248,23 +248,21 @@ static inline bool btrfs_is_free_space_inode(struct inode *inode)
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 
 	if (root == root->fs_info->tree_root &&
-	    btrfs_ino(inode) != BTRFS_BTREE_INODE_OBJECTID)
+	    btrfs_ino(BTRFS_I(inode)) != BTRFS_BTREE_INODE_OBJECTID)
 		return true;
 	if (BTRFS_I(inode)->location.objectid == BTRFS_FREE_INO_OBJECTID)
 		return true;
 	return false;
 }
 
-static inline int btrfs_inode_in_log(struct inode *inode, u64 generation)
+static inline int btrfs_inode_in_log(struct btrfs_inode *inode, u64 generation)
 {
 	int ret = 0;
 
-	spin_lock(&BTRFS_I(inode)->lock);
-	if (BTRFS_I(inode)->logged_trans == generation &&
-	    BTRFS_I(inode)->last_sub_trans <=
-	    BTRFS_I(inode)->last_log_commit &&
-	    BTRFS_I(inode)->last_sub_trans <=
-	    BTRFS_I(inode)->root->last_log_commit) {
+	spin_lock(&inode->lock);
+	if (inode->logged_trans == generation &&
+	    inode->last_sub_trans <= inode->last_log_commit &&
+	    inode->last_sub_trans <= inode->root->last_log_commit) {
 		/*
 		 * After a ranged fsync we might have left some extent maps
 		 * (that fall outside the fsync's range). So return false
@@ -272,10 +270,10 @@ static inline int btrfs_inode_in_log(struct inode *inode, u64 generation)
 		 * will be called and process those extent maps.
 		 */
 		smp_mb();
-		if (list_empty(&BTRFS_I(inode)->extent_tree.modified_extents))
+		if (list_empty(&inode->extent_tree.modified_extents))
 			ret = 1;
 	}
-	spin_unlock(&BTRFS_I(inode)->lock);
+	spin_unlock(&inode->lock);
 	return ret;
 }
 
@@ -326,6 +324,24 @@ static inline void btrfs_inode_resume_unlocked_dio(struct inode *inode)
 		  &BTRFS_I(inode)->runtime_flags);
 }
 
+static inline void btrfs_print_data_csum_error(struct inode *inode,
+		u64 logical_start, u32 csum, u32 csum_expected, int mirror_num)
+{
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+
+	/* Output minus objectid, which is more meaningful */
+	if (root->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(BTRFS_I(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(BTRFS_I(inode)),
+			logical_start, csum, csum_expected, mirror_num);
+}
+
 bool btrfs_page_exists_in_range(struct inode *inode, loff_t start, loff_t end);
 
 #endif

fs/btrfs/compression.c

@@ -124,10 +124,8 @@ static int check_compressed_csum(struct inode *inode,
 		kunmap_atomic(kaddr);
 
 		if (csum != *cb_sum) {
-			btrfs_info(BTRFS_I(inode)->root->fs_info,
-			   "csum failed ino %llu extent %llu csum %u wanted %u mirror %d",
-			   btrfs_ino(inode), disk_start, csum, *cb_sum,
-			   cb->mirror_num);
+			btrfs_print_data_csum_error(inode, disk_start, csum,
+						    *cb_sum, cb->mirror_num);
 			ret = -EIO;
 			goto fail;
 		}

fs/btrfs/ctree.c

@@ -28,9 +28,9 @@
 
 static int split_node(struct btrfs_trans_handle *trans, struct btrfs_root
 		      *root, struct btrfs_path *path, int level);
-static int split_leaf(struct btrfs_trans_handle *trans, struct btrfs_root
-		      *root, struct btrfs_key *ins_key,
-		      struct btrfs_path *path, int data_size, int extend);
+static int split_leaf(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+		      const struct btrfs_key *ins_key, struct btrfs_path *path,
+		      int data_size, int extend);
 static int push_node_left(struct btrfs_trans_handle *trans,
 			  struct btrfs_fs_info *fs_info,
 			  struct extent_buffer *dst,
@@ -426,7 +426,7 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 	tm_root = &fs_info->tree_mod_log;
 	for (node = rb_first(tm_root); node; node = next) {
 		next = rb_next(node);
-		tm = container_of(node, struct tree_mod_elem, node);
+		tm = rb_entry(node, struct tree_mod_elem, node);
 		if (tm->seq > min_seq)
 			continue;
 		rb_erase(node, tm_root);
@@ -460,7 +460,7 @@ __tree_mod_log_insert(struct btrfs_fs_info *fs_info, struct tree_mod_elem *tm)
 	tm_root = &fs_info->tree_mod_log;
 	new = &tm_root->rb_node;
 	while (*new) {
-		cur = container_of(*new, struct tree_mod_elem, node);
+		cur = rb_entry(*new, struct tree_mod_elem, node);
 		parent = *new;
 		if (cur->logical < tm->logical)
 			new = &((*new)->rb_left);
@@ -746,7 +746,7 @@ __tree_mod_log_search(struct btrfs_fs_info *fs_info, u64 start, u64 min_seq,
 	tm_root = &fs_info->tree_mod_log;
 	node = tm_root->rb_node;
 	while (node) {
-		cur = container_of(node, struct tree_mod_elem, node);
+		cur = rb_entry(node, struct tree_mod_elem, node);
 		if (cur->logical < start) {
 			node = node->rb_left;
 		} else if (cur->logical > start) {
@@ -1074,7 +1074,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 			ret = btrfs_dec_ref(trans, root, buf, 1);
 			BUG_ON(ret); /* -ENOMEM */
 		}
-		clean_tree_block(trans, fs_info, buf);
+		clean_tree_block(fs_info, buf);
 		*last_ref = 1;
 	}
 	return 0;
@@ -1326,7 +1326,7 @@ __tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct extent_buffer *eb,
 		next = rb_next(&tm->node);
 		if (!next)
 			break;
-		tm = container_of(next, struct tree_mod_elem, node);
+		tm = rb_entry(next, struct tree_mod_elem, node);
 		if (tm->logical != first_tm->logical)
 			break;
 	}
@@ -1580,7 +1580,8 @@ static int close_blocks(u64 blocknr, u64 other, u32 blocksize)
 /*
  * compare two keys in a memcmp fashion
  */
-static int comp_keys(struct btrfs_disk_key *disk, struct btrfs_key *k2)
+static int comp_keys(const struct btrfs_disk_key *disk,
+		     const struct btrfs_key *k2)
 {
 	struct btrfs_key k1;
 
@@ -1592,7 +1593,7 @@ static int comp_keys(struct btrfs_disk_key *disk, struct btrfs_key *k2)
 /*
  * same as comp_keys only with two btrfs_key's
  */
-int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2)
+int btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2)
 {
 	if (k1->objectid > k2->objectid)
 		return 1;
@@ -1732,8 +1733,8 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
  * slot may point to max if the key is bigger than all of the keys
  */
 static noinline int generic_bin_search(struct extent_buffer *eb,
-				       unsigned long p,
-				       int item_size, struct btrfs_key *key,
+				       unsigned long p, int item_size,
+				       const struct btrfs_key *key,
 				       int max, int *slot)
 {
 	int low = 0;
@@ -1802,7 +1803,7 @@ static noinline int generic_bin_search(struct extent_buffer *eb,
  * simple bin_search frontend that does the right thing for
  * leaves vs nodes
  */
-static int bin_search(struct extent_buffer *eb, struct btrfs_key *key,
+static int bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
 		      int level, int *slot)
 {
 	if (level == 0)
@@ -1819,7 +1820,7 @@ static int bin_search(struct extent_buffer *eb, struct btrfs_key *key,
 					  slot);
 }
 
-int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key,
+int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
 		     int level, int *slot)
 {
 	return bin_search(eb, key, level, slot);
@@ -1937,7 +1938,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 
 		path->locks[level] = 0;
 		path->nodes[level] = NULL;
-		clean_tree_block(trans, fs_info, mid);
+		clean_tree_block(fs_info, mid);
 		btrfs_tree_unlock(mid);
 		/* once for the path */
 		free_extent_buffer(mid);
@@ -1998,7 +1999,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		if (wret < 0 && wret != -ENOSPC)
 			ret = wret;
 		if (btrfs_header_nritems(right) == 0) {
-			clean_tree_block(trans, fs_info, right);
+			clean_tree_block(fs_info, right);
 			btrfs_tree_unlock(right);
 			del_ptr(root, path, level + 1, pslot + 1);
 			root_sub_used(root, right->len);
@@ -2042,7 +2043,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		BUG_ON(wret == 1);
 	}
 	if (btrfs_header_nritems(mid) == 0) {
-		clean_tree_block(trans, fs_info, mid);
+		clean_tree_block(fs_info, mid);
 		btrfs_tree_unlock(mid);
 		del_ptr(root, path, level + 1, pslot);
 		root_sub_used(root, mid->len);
@@ -2437,10 +2438,9 @@ noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level)
  * reada.  -EAGAIN is returned and the search must be repeated.
  */
 static int
-read_block_for_search(struct btrfs_trans_handle *trans,
-		       struct btrfs_root *root, struct btrfs_path *p,
-		       struct extent_buffer **eb_ret, int level, int slot,
-		       struct btrfs_key *key, u64 time_seq)
+read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
+		      struct extent_buffer **eb_ret, int level, int slot,
+		      const struct btrfs_key *key)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	u64 blocknr;
@@ -2587,7 +2587,7 @@ setup_nodes_for_search(struct btrfs_trans_handle *trans,
 }
 
 static void key_search_validate(struct extent_buffer *b,
-				struct btrfs_key *key,
+				const struct btrfs_key *key,
 				int level)
 {
 #ifdef CONFIG_BTRFS_ASSERT
@@ -2606,7 +2606,7 @@ static void key_search_validate(struct extent_buffer *b,
 #endif
 }
 
-static int key_search(struct extent_buffer *b, struct btrfs_key *key,
+static int key_search(struct extent_buffer *b, const struct btrfs_key *key,
 		      int level, int *prev_cmp, int *slot)
 {
 	if (*prev_cmp != 0) {
@@ -2668,9 +2668,9 @@ int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
  * tree.  if ins_len < 0, nodes will be merged as we walk down the tree (if
  * possible)
  */
-int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
-		      *root, struct btrfs_key *key, struct btrfs_path *p, int
-		      ins_len, int cow)
+int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+		      const struct btrfs_key *key, struct btrfs_path *p,
+		      int ins_len, int cow)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_buffer *b;
@@ -2870,8 +2870,8 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
 				goto done;
 			}
 
-			err = read_block_for_search(trans, root, p,
-						    &b, level, slot, key, 0);
+			err = read_block_for_search(root, p, &b, level,
+						    slot, key);
 			if (err == -EAGAIN)
 				goto again;
 			if (err) {
@@ -2953,7 +2953,7 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
  * The resulting path and return value will be set up as if we called
  * btrfs_search_slot at that point in time with ins_len and cow both set to 0.
  */
-int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
+int btrfs_search_old_slot(struct btrfs_root *root, const struct btrfs_key *key,
 			  struct btrfs_path *p, u64 time_seq)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -3014,8 +3014,8 @@ int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
 				goto done;
 			}
 
-			err = read_block_for_search(NULL, root, p, &b, level,
-						    slot, key, time_seq);
+			err = read_block_for_search(root, p, &b, level,
+						    slot, key);
 			if (err == -EAGAIN)
 				goto again;
 			if (err) {
@@ -3067,8 +3067,9 @@ int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
  * < 0 on error
  */
 int btrfs_search_slot_for_read(struct btrfs_root *root,
-			       struct btrfs_key *key, struct btrfs_path *p,
-			       int find_higher, int return_any)
+			       const struct btrfs_key *key,
+			       struct btrfs_path *p, int find_higher,
+			       int return_any)
 {
 	int ret;
 	struct extent_buffer *leaf;
@@ -3166,7 +3167,7 @@ static void fixup_low_keys(struct btrfs_fs_info *fs_info,
  */
 void btrfs_set_item_key_safe(struct btrfs_fs_info *fs_info,
 			     struct btrfs_path *path,
-			     struct btrfs_key *new_key)
+			     const struct btrfs_key *new_key)
 {
 	struct btrfs_disk_key disk_key;
 	struct extent_buffer *eb;
@@ -3594,8 +3595,7 @@ noinline int btrfs_leaf_free_space(struct btrfs_fs_info *fs_info,
  * min slot controls the lowest index we're willing to push to the
  * right.  We'll push up to and including min_slot, but no lower
  */
-static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
-				      struct btrfs_fs_info *fs_info,
+static noinline int __push_leaf_right(struct btrfs_fs_info *fs_info,
 				      struct btrfs_path *path,
 				      int data_size, int empty,
 				      struct extent_buffer *right,
@@ -3704,7 +3704,7 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
 	if (left_nritems)
 		btrfs_mark_buffer_dirty(left);
 	else
-		clean_tree_block(trans, fs_info, left);
+		clean_tree_block(fs_info, left);
 
 	btrfs_mark_buffer_dirty(right);
 
@@ -3716,7 +3716,7 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
 	if (path->slots[0] >= left_nritems) {
 		path->slots[0] -= left_nritems;
 		if (btrfs_header_nritems(path->nodes[0]) == 0)
-			clean_tree_block(trans, fs_info, path->nodes[0]);
+			clean_tree_block(fs_info, path->nodes[0]);
 		btrfs_tree_unlock(path->nodes[0]);
 		free_extent_buffer(path->nodes[0]);
 		path->nodes[0] = right;
@@ -3809,7 +3809,7 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
 		return 0;
 	}
 
-	return __push_leaf_right(trans, fs_info, path, min_data_size, empty,
+	return __push_leaf_right(fs_info, path, min_data_size, empty,
 				right, free_space, left_nritems, min_slot);
 out_unlock:
 	btrfs_tree_unlock(right);
@@ -3825,8 +3825,7 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
  * item at 'max_slot' won't be touched.  Use (u32)-1 to make us do all the
  * items
  */
-static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
-				     struct btrfs_fs_info *fs_info,
+static noinline int __push_leaf_left(struct btrfs_fs_info *fs_info,
 				     struct btrfs_path *path, int data_size,
 				     int empty, struct extent_buffer *left,
 				     int free_space, u32 right_nritems,
@@ -3945,7 +3944,7 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
 	if (right_nritems)
 		btrfs_mark_buffer_dirty(right);
 	else
-		clean_tree_block(trans, fs_info, right);
+		clean_tree_block(fs_info, right);
 
 	btrfs_item_key(right, &disk_key, 0);
 	fixup_low_keys(fs_info, path, &disk_key, 1);
@@ -4035,7 +4034,7 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 		goto out;
 	}
 
-	return __push_leaf_left(trans, fs_info, path, min_data_size,
+	return __push_leaf_left(fs_info, path, min_data_size,
 			       empty, left, free_space, right_nritems,
 			       max_slot);
 out:
@@ -4180,7 +4179,7 @@ static noinline int push_for_double_split(struct btrfs_trans_handle *trans,
  */
 static noinline int split_leaf(struct btrfs_trans_handle *trans,
 			       struct btrfs_root *root,
-			       struct btrfs_key *ins_key,
+			       const struct btrfs_key *ins_key,
 			       struct btrfs_path *path, int data_size,
 			       int extend)
 {
@@ -4412,10 +4411,9 @@ static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-static noinline int split_item(struct btrfs_trans_handle *trans,
-			       struct btrfs_fs_info *fs_info,
+static noinline int split_item(struct btrfs_fs_info *fs_info,
 			       struct btrfs_path *path,
-			       struct btrfs_key *new_key,
+			       const struct btrfs_key *new_key,
 			       unsigned long split_offset)
 {
 	struct extent_buffer *leaf;
@@ -4501,7 +4499,7 @@ static noinline int split_item(struct btrfs_trans_handle *trans,
 int btrfs_split_item(struct btrfs_trans_handle *trans,
 		     struct btrfs_root *root,
 		     struct btrfs_path *path,
-		     struct btrfs_key *new_key,
+		     const struct btrfs_key *new_key,
 		     unsigned long split_offset)
 {
 	int ret;
@@ -4510,7 +4508,7 @@ int btrfs_split_item(struct btrfs_trans_handle *trans,
 	if (ret)
 		return ret;
 
-	ret = split_item(trans, root->fs_info, path, new_key, split_offset);
+	ret = split_item(root->fs_info, path, new_key, split_offset);
 	return ret;
 }
 
@@ -4525,7 +4523,7 @@ int btrfs_split_item(struct btrfs_trans_handle *trans,
 int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
 			 struct btrfs_root *root,
 			 struct btrfs_path *path,
-			 struct btrfs_key *new_key)
+			 const struct btrfs_key *new_key)
 {
 	struct extent_buffer *leaf;
 	int ret;
@@ -4726,7 +4724,7 @@ void btrfs_extend_item(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
  * that doesn't call btrfs_search_slot
  */
 void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
-			    struct btrfs_key *cpu_key, u32 *data_size,
+			    const struct btrfs_key *cpu_key, u32 *data_size,
 			    u32 total_data, u32 total_size, int nr)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -4820,7 +4818,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
 			    struct btrfs_root *root,
 			    struct btrfs_path *path,
-			    struct btrfs_key *cpu_key, u32 *data_size,
+			    const struct btrfs_key *cpu_key, u32 *data_size,
 			    int nr)
 {
 	int ret = 0;
@@ -4851,9 +4849,9 @@ int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
  * Given a key and some data, insert an item into the tree.
  * This does all the path init required, making room in the tree if needed.
  */
-int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root
-		      *root, struct btrfs_key *cpu_key, void *data, u32
-		      data_size)
+int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+		      const struct btrfs_key *cpu_key, void *data,
+		      u32 data_size)
 {
 	int ret = 0;
 	struct btrfs_path *path;
@@ -5008,7 +5006,7 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 			btrfs_set_header_level(leaf, 0);
 		} else {
 			btrfs_set_path_blocking(path);
-			clean_tree_block(trans, fs_info, leaf);
+			clean_tree_block(fs_info, leaf);
 			btrfs_del_leaf(trans, root, path, leaf);
 		}
 	} else {
@@ -5243,7 +5241,7 @@ int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key,
 
 static int tree_move_down(struct btrfs_fs_info *fs_info,
 			   struct btrfs_path *path,
-			   int *level, int root_level)
+			   int *level)
 {
 	struct extent_buffer *eb;
 
@@ -5258,8 +5256,7 @@ static int tree_move_down(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
-static int tree_move_next_or_upnext(struct btrfs_fs_info *fs_info,
-				    struct btrfs_path *path,
+static int tree_move_next_or_upnext(struct btrfs_path *path,
 				    int *level, int root_level)
 {
 	int ret = 0;
@@ -5298,10 +5295,9 @@ static int tree_advance(struct btrfs_fs_info *fs_info,
 	int ret;
 
 	if (*level == 0 || !allow_down) {
-		ret = tree_move_next_or_upnext(fs_info, path, level,
-					       root_level);
+		ret = tree_move_next_or_upnext(path, level, root_level);
 	} else {
-		ret = tree_move_down(fs_info, path, level, root_level);
+		ret = tree_move_down(fs_info, path, level);
 	}
 	if (ret >= 0) {
 		if (*level == 0)
@@ -5784,8 +5780,8 @@ int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
 
 		next = c;
 		next_rw_lock = path->locks[level];
-		ret = read_block_for_search(NULL, root, path, &next, level,
-					    slot, &key, 0);
+		ret = read_block_for_search(root, path, &next, level,
+					    slot, &key);
 		if (ret == -EAGAIN)
 			goto again;
 
@@ -5834,8 +5830,8 @@ int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
 		if (!level)
 			break;
 
-		ret = read_block_for_search(NULL, root, path, &next, level,
-					    0, &key, 0);
+		ret = read_block_for_search(root, path, &next, level,
+					    0, &key);
 		if (ret == -EAGAIN)
 			goto again;
 

fs/btrfs/ctree.h

@@ -97,6 +97,14 @@ static const int btrfs_csum_sizes[] = { 4 };
 
 #define BTRFS_MAX_EXTENT_SIZE SZ_128M
 
+/*
+ * Count how many BTRFS_MAX_EXTENT_SIZE cover the @size
+ */
+static inline u32 count_max_extents(u64 size)
+{
+	return div_u64(size + BTRFS_MAX_EXTENT_SIZE - 1, BTRFS_MAX_EXTENT_SIZE);
+}
+
 struct btrfs_mapping_tree {
 	struct extent_map_tree map_tree;
 };
@@ -1953,7 +1961,7 @@ BTRFS_SETGET_STACK_FUNCS(disk_key_offset, struct btrfs_disk_key, offset, 64);
 BTRFS_SETGET_STACK_FUNCS(disk_key_type, struct btrfs_disk_key, type, 8);
 
 static inline void btrfs_disk_key_to_cpu(struct btrfs_key *cpu,
-					 struct btrfs_disk_key *disk)
+					 const struct btrfs_disk_key *disk)
 {
 	cpu->offset = le64_to_cpu(disk->offset);
 	cpu->type = disk->type;
@@ -1961,7 +1969,7 @@ static inline void btrfs_disk_key_to_cpu(struct btrfs_key *cpu,
 }
 
 static inline void btrfs_cpu_key_to_disk(struct btrfs_disk_key *disk,
-					 struct btrfs_key *cpu)
+					 const struct btrfs_key *cpu)
 {
 	disk->offset = cpu_to_le64(cpu->offset);
 	disk->type = cpu->type;
@@ -1993,8 +2001,7 @@ static inline void btrfs_dir_item_key_to_cpu(struct extent_buffer *eb,
 	btrfs_disk_key_to_cpu(key, &disk_key);
 }
 
-
-static inline u8 btrfs_key_type(struct btrfs_key *key)
+static inline u8 btrfs_key_type(const struct btrfs_key *key)
 {
 	return key->type;
 }
@@ -2577,8 +2584,7 @@ int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info,
 				    u64 bytenr, u64 num_bytes);
 int btrfs_exclude_logged_extents(struct btrfs_fs_info *fs_info,
 				 struct extent_buffer *eb);
-int btrfs_cross_ref_exist(struct btrfs_trans_handle *trans,
-			  struct btrfs_root *root,
+int btrfs_cross_ref_exist(struct btrfs_root *root,
 			  u64 objectid, u64 offset, u64 bytenr);
 struct btrfs_block_group_cache *btrfs_lookup_block_group(
 						 struct btrfs_fs_info *info,
@@ -2587,10 +2593,11 @@ void btrfs_get_block_group(struct btrfs_block_group_cache *cache);
 void btrfs_put_block_group(struct btrfs_block_group_cache *cache);
 int get_block_group_index(struct btrfs_block_group_cache *cache);
 struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
-					struct btrfs_root *root, u64 parent,
-					u64 root_objectid,
-					struct btrfs_disk_key *key, int level,
-					u64 hint, u64 empty_size);
+					     struct btrfs_root *root,
+					     u64 parent, u64 root_objectid,
+					     const struct btrfs_disk_key *key,
+					     int level, u64 hint,
+					     u64 empty_size);
 void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 			   struct btrfs_root *root,
 			   struct extent_buffer *buf,
@@ -2623,8 +2630,7 @@ int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
 			       u64 start, u64 len, int delalloc);
 int btrfs_free_and_pin_reserved_extent(struct btrfs_fs_info *fs_info,
 				       u64 start, u64 len);
-void btrfs_prepare_extent_commit(struct btrfs_trans_handle *trans,
-				 struct btrfs_fs_info *fs_info);
+void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info);
 int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans,
 			       struct btrfs_fs_info *fs_info);
 int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
@@ -2696,8 +2702,7 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 				     int nitems,
 				     u64 *qgroup_reserved, bool use_global_rsv);
 void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info,
-				      struct btrfs_block_rsv *rsv,
-				      u64 qgroup_reserved);
+				      struct btrfs_block_rsv *rsv);
 int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes);
 void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes);
 int btrfs_delalloc_reserve_space(struct inode *inode, u64 start, u64 len);
@@ -2724,7 +2729,7 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
 			     struct btrfs_block_rsv *block_rsv,
 			     u64 num_bytes);
-int btrfs_inc_block_group_ro(struct btrfs_root *root,
+int btrfs_inc_block_group_ro(struct btrfs_fs_info *fs_info,
 			     struct btrfs_block_group_cache *cache);
 void btrfs_dec_block_group_ro(struct btrfs_block_group_cache *cache);
 void btrfs_put_block_group_cache(struct btrfs_fs_info *info);
@@ -2750,9 +2755,9 @@ u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
 		       struct btrfs_fs_info *info, u64 start, u64 end);
 
 /* ctree.c */
-int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key,
+int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
 		     int level, int *slot);
-int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2);
+int btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2);
 int btrfs_previous_item(struct btrfs_root *root,
 			struct btrfs_path *path, u64 min_objectid,
 			int type);
@@ -2760,7 +2765,7 @@ int btrfs_previous_extent_item(struct btrfs_root *root,
 			struct btrfs_path *path, u64 min_objectid);
 void btrfs_set_item_key_safe(struct btrfs_fs_info *fs_info,
 			     struct btrfs_path *path,
-			     struct btrfs_key *new_key);
+			     const struct btrfs_key *new_key);
 struct extent_buffer *btrfs_root_node(struct btrfs_root *root);
 struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root);
 int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path,
@@ -2802,22 +2807,23 @@ void btrfs_truncate_item(struct btrfs_fs_info *fs_info,
 int btrfs_split_item(struct btrfs_trans_handle *trans,
 		     struct btrfs_root *root,
 		     struct btrfs_path *path,
-		     struct btrfs_key *new_key,
+		     const struct btrfs_key *new_key,
 		     unsigned long split_offset);
 int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
 			 struct btrfs_root *root,
 			 struct btrfs_path *path,
-			 struct btrfs_key *new_key);
+			 const struct btrfs_key *new_key);
 int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
 		u64 inum, u64 ioff, u8 key_type, struct btrfs_key *found_key);
-int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
-		      *root, struct btrfs_key *key, struct btrfs_path *p, int
-		      ins_len, int cow);
-int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
+int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+		      const struct btrfs_key *key, struct btrfs_path *p,
+		      int ins_len, int cow);
+int btrfs_search_old_slot(struct btrfs_root *root, const struct btrfs_key *key,
 			  struct btrfs_path *p, u64 time_seq);
 int btrfs_search_slot_for_read(struct btrfs_root *root,
-			       struct btrfs_key *key, struct btrfs_path *p,
-			       int find_higher, int return_any);
+			       const struct btrfs_key *key,
+			       struct btrfs_path *p, int find_higher,
+			       int return_any);
 int btrfs_realloc_node(struct btrfs_trans_handle *trans,
 		       struct btrfs_root *root, struct extent_buffer *parent,
 		       int start_slot, u64 *last_ret,
@@ -2840,19 +2846,20 @@ static inline int btrfs_del_item(struct btrfs_trans_handle *trans,
 }
 
 void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
-			    struct btrfs_key *cpu_key, u32 *data_size,
+			    const struct btrfs_key *cpu_key, u32 *data_size,
 			    u32 total_data, u32 total_size, int nr);
-int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root
-		      *root, struct btrfs_key *key, void *data, u32 data_size);
+int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+		      const struct btrfs_key *key, void *data, u32 data_size);
 int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root,
 			     struct btrfs_path *path,
-			     struct btrfs_key *cpu_key, u32 *data_size, int nr);
+			     const struct btrfs_key *cpu_key, u32 *data_size,
+			     int nr);
 
 static inline int btrfs_insert_empty_item(struct btrfs_trans_handle *trans,
 					  struct btrfs_root *root,
 					  struct btrfs_path *path,
-					  struct btrfs_key *key,
+					  const struct btrfs_key *key,
 					  u32 data_size)
 {
 	return btrfs_insert_empty_items(trans, root, path, key, &data_size, 1);
@@ -2941,15 +2948,15 @@ int btrfs_del_root_ref(struct btrfs_trans_handle *trans,
 		       u64 root_id, u64 ref_id, u64 dirid, u64 *sequence,
 		       const char *name, int name_len);
 int btrfs_del_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		   struct btrfs_key *key);
-int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root
-		      *root, struct btrfs_key *key, struct btrfs_root_item
-		      *item);
+		   const struct btrfs_key *key);
+int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+		      const struct btrfs_key *key,
+		      struct btrfs_root_item *item);
 int __must_check btrfs_update_root(struct btrfs_trans_handle *trans,
 				   struct btrfs_root *root,
 				   struct btrfs_key *key,
 				   struct btrfs_root_item *item);
-int btrfs_find_root(struct btrfs_root *root, struct btrfs_key *search_key,
+int btrfs_find_root(struct btrfs_root *root, const struct btrfs_key *search_key,
 		    struct btrfs_path *path, struct btrfs_root_item *root_item,
 		    struct btrfs_key *root_key);
 int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info);
@@ -3119,7 +3126,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry);
 int btrfs_set_inode_index(struct inode *dir, u64 *index);
 int btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 		       struct btrfs_root *root,
-		       struct inode *dir, struct inode *inode,
+		       struct btrfs_inode *dir, struct btrfs_inode *inode,
 		       const char *name, int name_len);
 int btrfs_add_link(struct btrfs_trans_handle *trans,
 		   struct inode *parent_inode, struct inode *inode,
@@ -3447,7 +3454,8 @@ do {								\
 			"BTRFS: Transaction aborted (error %d)\n",	\
 			(errno));					\
 		} else {						\
-			pr_debug("BTRFS: Transaction aborted (error %d)\n", \
+			btrfs_debug((trans)->fs_info,			\
+				    "Transaction aborted (error %d)", \
 				  (errno));			\
 		}						\
 	}							\

fs/btrfs/delayed-inode.c

@@ -72,14 +72,14 @@ static inline int btrfs_is_continuous_delayed_item(
 	return 0;
 }
 
-static struct btrfs_delayed_node *btrfs_get_delayed_node(struct inode *inode)
+static struct btrfs_delayed_node *btrfs_get_delayed_node(
+		struct btrfs_inode *btrfs_inode)
 {
-	struct btrfs_inode *btrfs_inode = BTRFS_I(inode);
 	struct btrfs_root *root = btrfs_inode->root;
-	u64 ino = btrfs_ino(inode);
+	u64 ino = btrfs_ino(btrfs_inode);
 	struct btrfs_delayed_node *node;
 
-	node = ACCESS_ONCE(btrfs_inode->delayed_node);
+	node = READ_ONCE(btrfs_inode->delayed_node);
 	if (node) {
 		atomic_inc(&node->refs);
 		return node;
@@ -107,16 +107,15 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(struct inode *inode)
 
 /* Will return either the node or PTR_ERR(-ENOMEM) */
 static struct btrfs_delayed_node *btrfs_get_or_create_delayed_node(
-							struct inode *inode)
+		struct btrfs_inode *btrfs_inode)
 {
 	struct btrfs_delayed_node *node;
-	struct btrfs_inode *btrfs_inode = BTRFS_I(inode);
 	struct btrfs_root *root = btrfs_inode->root;
-	u64 ino = btrfs_ino(inode);
+	u64 ino = btrfs_ino(btrfs_inode);
 	int ret;
 
 again:
-	node = btrfs_get_delayed_node(inode);
+	node = btrfs_get_delayed_node(btrfs_inode);
 	if (node)
 		return node;
 
@@ -574,7 +573,7 @@ static void btrfs_delayed_item_release_metadata(struct btrfs_fs_info *fs_info,
 static int btrfs_delayed_inode_reserve_metadata(
 					struct btrfs_trans_handle *trans,
 					struct btrfs_root *root,
-					struct inode *inode,
+					struct btrfs_inode *inode,
 					struct btrfs_delayed_node *node)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -603,13 +602,13 @@ static int btrfs_delayed_inode_reserve_metadata(
 	 * worth which is less likely to hurt us.
 	 */
 	if (src_rsv && src_rsv->type == BTRFS_BLOCK_RSV_DELALLOC) {
-		spin_lock(&BTRFS_I(inode)->lock);
+		spin_lock(&inode->lock);
 		if (test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
-				       &BTRFS_I(inode)->runtime_flags))
+				       &inode->runtime_flags))
 			release = true;
 		else
 			src_rsv = NULL;
-		spin_unlock(&BTRFS_I(inode)->lock);
+		spin_unlock(&inode->lock);
 	}
 
 	/*
@@ -1196,7 +1195,7 @@ int btrfs_run_delayed_items_nr(struct btrfs_trans_handle *trans,
 }
 
 int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
-				     struct inode *inode)
+				     struct btrfs_inode *inode)
 {
 	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
 	struct btrfs_path *path;
@@ -1233,9 +1232,9 @@ int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-int btrfs_commit_inode_delayed_inode(struct inode *inode)
+int btrfs_commit_inode_delayed_inode(struct btrfs_inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
 	struct btrfs_trans_handle *trans;
 	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
 	struct btrfs_path *path;
@@ -1288,15 +1287,15 @@ int btrfs_commit_inode_delayed_inode(struct inode *inode)
 	return ret;
 }
 
-void btrfs_remove_delayed_node(struct inode *inode)
+void btrfs_remove_delayed_node(struct btrfs_inode *inode)
 {
 	struct btrfs_delayed_node *delayed_node;
 
-	delayed_node = ACCESS_ONCE(BTRFS_I(inode)->delayed_node);
+	delayed_node = READ_ONCE(inode->delayed_node);
 	if (!delayed_node)
 		return;
 
-	BTRFS_I(inode)->delayed_node = NULL;
+	inode->delayed_node = NULL;
 	btrfs_release_delayed_node(delayed_node);
 }
 
@@ -1434,7 +1433,7 @@ void btrfs_balance_delayed_items(struct btrfs_fs_info *fs_info)
 int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
 				   struct btrfs_fs_info *fs_info,
 				   const char *name, int name_len,
-				   struct inode *dir,
+				   struct btrfs_inode *dir,
 				   struct btrfs_disk_key *disk_key, u8 type,
 				   u64 index)
 {
@@ -1510,7 +1509,7 @@ static int btrfs_delete_delayed_insertion_item(struct btrfs_fs_info *fs_info,
 
 int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
 				   struct btrfs_fs_info *fs_info,
-				   struct inode *dir, u64 index)
+				   struct btrfs_inode *dir, u64 index)
 {
 	struct btrfs_delayed_node *node;
 	struct btrfs_delayed_item *item;
@@ -1558,7 +1557,7 @@ int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-int btrfs_inode_delayed_dir_index_count(struct inode *inode)
+int btrfs_inode_delayed_dir_index_count(struct btrfs_inode *inode)
 {
 	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
 
@@ -1575,7 +1574,7 @@ int btrfs_inode_delayed_dir_index_count(struct inode *inode)
 		return -EINVAL;
 	}
 
-	BTRFS_I(inode)->index_cnt = delayed_node->index_cnt;
+	inode->index_cnt = delayed_node->index_cnt;
 	btrfs_release_delayed_node(delayed_node);
 	return 0;
 }
@@ -1587,7 +1586,7 @@ bool btrfs_readdir_get_delayed_items(struct inode *inode,
 	struct btrfs_delayed_node *delayed_node;
 	struct btrfs_delayed_item *item;
 
-	delayed_node = btrfs_get_delayed_node(inode);
+	delayed_node = btrfs_get_delayed_node(BTRFS_I(inode));
 	if (!delayed_node)
 		return false;
 
@@ -1776,7 +1775,7 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
 	struct btrfs_delayed_node *delayed_node;
 	struct btrfs_inode_item *inode_item;
 
-	delayed_node = btrfs_get_delayed_node(inode);
+	delayed_node = btrfs_get_delayed_node(BTRFS_I(inode));
 	if (!delayed_node)
 		return -ENOENT;
 
@@ -1831,7 +1830,7 @@ int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
 	struct btrfs_delayed_node *delayed_node;
 	int ret = 0;
 
-	delayed_node = btrfs_get_or_create_delayed_node(inode);
+	delayed_node = btrfs_get_or_create_delayed_node(BTRFS_I(inode));
 	if (IS_ERR(delayed_node))
 		return PTR_ERR(delayed_node);
 
@@ -1841,7 +1840,7 @@ int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
 		goto release_node;
 	}
 
-	ret = btrfs_delayed_inode_reserve_metadata(trans, root, inode,
+	ret = btrfs_delayed_inode_reserve_metadata(trans, root, BTRFS_I(inode),
 						   delayed_node);
 	if (ret)
 		goto release_node;
@@ -1856,9 +1855,9 @@ int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-int btrfs_delayed_delete_inode_ref(struct inode *inode)
+int btrfs_delayed_delete_inode_ref(struct btrfs_inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb);
 	struct btrfs_delayed_node *delayed_node;
 
 	/*
@@ -1933,7 +1932,7 @@ static void __btrfs_kill_delayed_node(struct btrfs_delayed_node *delayed_node)
 	mutex_unlock(&delayed_node->mutex);
 }
 
-void btrfs_kill_delayed_inode_items(struct inode *inode)
+void btrfs_kill_delayed_inode_items(struct btrfs_inode *inode)
 {
 	struct btrfs_delayed_node *delayed_node;
 

fs/btrfs/delayed-inode.h

@@ -101,15 +101,15 @@ static inline void btrfs_init_delayed_root(
 int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
 				   struct btrfs_fs_info *fs_info,
 				   const char *name, int name_len,
-				   struct inode *dir,
+				   struct btrfs_inode *dir,
 				   struct btrfs_disk_key *disk_key, u8 type,
 				   u64 index);
 
 int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
 				   struct btrfs_fs_info *fs_info,
-				   struct inode *dir, u64 index);
+				   struct btrfs_inode *dir, u64 index);
 
-int btrfs_inode_delayed_dir_index_count(struct inode *inode);
+int btrfs_inode_delayed_dir_index_count(struct btrfs_inode *inode);
 
 int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
 			    struct btrfs_fs_info *fs_info);
@@ -119,17 +119,17 @@ int btrfs_run_delayed_items_nr(struct btrfs_trans_handle *trans,
 void btrfs_balance_delayed_items(struct btrfs_fs_info *fs_info);
 
 int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
-				     struct inode *inode);
+				     struct btrfs_inode *inode);
 /* Used for evicting the inode. */
-void btrfs_remove_delayed_node(struct inode *inode);
-void btrfs_kill_delayed_inode_items(struct inode *inode);
-int btrfs_commit_inode_delayed_inode(struct inode *inode);
+void btrfs_remove_delayed_node(struct btrfs_inode *inode);
+void btrfs_kill_delayed_inode_items(struct btrfs_inode *inode);
+int btrfs_commit_inode_delayed_inode(struct btrfs_inode *inode);
 
 
 int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
 			       struct btrfs_root *root, struct inode *inode);
 int btrfs_fill_inode(struct inode *inode, u32 *rdev);
-int btrfs_delayed_delete_inode_ref(struct inode *inode);
+int btrfs_delayed_delete_inode_ref(struct btrfs_inode *inode);
 
 /* Used for drop dead root */
 void btrfs_kill_all_delayed_nodes(struct btrfs_root *root);

fs/btrfs/delayed-ref.c

@@ -550,13 +550,14 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 		     struct btrfs_delayed_ref_node *ref,
 		     struct btrfs_qgroup_extent_record *qrecord,
 		     u64 bytenr, u64 num_bytes, u64 ref_root, u64 reserved,
-		     int action, int is_data)
+		     int action, int is_data, int *qrecord_inserted_ret)
 {
 	struct btrfs_delayed_ref_head *existing;
 	struct btrfs_delayed_ref_head *head_ref = NULL;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	int count_mod = 1;
 	int must_insert_reserved = 0;
+	int qrecord_inserted = 0;
 
 	/* If reserved is provided, it must be a data extent. */
 	BUG_ON(!is_data && reserved);
@@ -623,6 +624,8 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 		if(btrfs_qgroup_trace_extent_nolock(fs_info,
 					delayed_refs, qrecord))
 			kfree(qrecord);
+		else
+			qrecord_inserted = 1;
 	}
 
 	spin_lock_init(&head_ref->lock);
@@ -650,6 +653,8 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info,
 		atomic_inc(&delayed_refs->num_entries);
 		trans->delayed_ref_updates++;
 	}
+	if (qrecord_inserted_ret)
+		*qrecord_inserted_ret = qrecord_inserted;
 	return head_ref;
 }
 
@@ -779,6 +784,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 	struct btrfs_delayed_ref_head *head_ref;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	struct btrfs_qgroup_extent_record *record = NULL;
+	int qrecord_inserted;
 
 	BUG_ON(extent_op && extent_op->is_data);
 	ref = kmem_cache_alloc(btrfs_delayed_tree_ref_cachep, GFP_NOFS);
@@ -806,12 +812,15 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
 	 * the spin lock
 	 */
 	head_ref = add_delayed_ref_head(fs_info, trans, &head_ref->node, record,
-					bytenr, num_bytes, 0, 0, action, 0);
+					bytenr, num_bytes, 0, 0, action, 0,
+					&qrecord_inserted);
 
 	add_delayed_tree_ref(fs_info, trans, head_ref, &ref->node, bytenr,
 			     num_bytes, parent, ref_root, level, action);
 	spin_unlock(&delayed_refs->lock);
 
+	if (qrecord_inserted)
+		return btrfs_qgroup_trace_extent_post(fs_info, record);
 	return 0;
 
 free_head_ref:
@@ -829,15 +838,14 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 			       struct btrfs_trans_handle *trans,
 			       u64 bytenr, u64 num_bytes,
 			       u64 parent, u64 ref_root,
-			       u64 owner, u64 offset, u64 reserved, int action,
-			       struct btrfs_delayed_extent_op *extent_op)
+			       u64 owner, u64 offset, u64 reserved, int action)
 {
 	struct btrfs_delayed_data_ref *ref;
 	struct btrfs_delayed_ref_head *head_ref;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	struct btrfs_qgroup_extent_record *record = NULL;
+	int qrecord_inserted;
 
-	BUG_ON(extent_op && !extent_op->is_data);
 	ref = kmem_cache_alloc(btrfs_delayed_data_ref_cachep, GFP_NOFS);
 	if (!ref)
 		return -ENOMEM;
@@ -859,7 +867,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 		}
 	}
 
-	head_ref->extent_op = extent_op;
+	head_ref->extent_op = NULL;
 
 	delayed_refs = &trans->transaction->delayed_refs;
 	spin_lock(&delayed_refs->lock);
@@ -870,13 +878,15 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 	 */
 	head_ref = add_delayed_ref_head(fs_info, trans, &head_ref->node, record,
 					bytenr, num_bytes, ref_root, reserved,
-					action, 1);
+					action, 1, &qrecord_inserted);
 
 	add_delayed_data_ref(fs_info, trans, head_ref, &ref->node, bytenr,
 				   num_bytes, parent, ref_root, owner, offset,
 				   action);
 	spin_unlock(&delayed_refs->lock);
 
+	if (qrecord_inserted)
+		return btrfs_qgroup_trace_extent_post(fs_info, record);
 	return 0;
 }
 
@@ -899,7 +909,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
 
 	add_delayed_ref_head(fs_info, trans, &head_ref->node, NULL, bytenr,
 			     num_bytes, 0, 0, BTRFS_UPDATE_DELAYED_HEAD,
-			     extent_op->is_data);
+			     extent_op->is_data, NULL);
 
 	spin_unlock(&delayed_refs->lock);
 	return 0;
@@ -911,11 +921,8 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
  * the head node if any where found, or NULL if not.
  */
 struct btrfs_delayed_ref_head *
-btrfs_find_delayed_ref_head(struct btrfs_trans_handle *trans, u64 bytenr)
+btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs, u64 bytenr)
 {
-	struct btrfs_delayed_ref_root *delayed_refs;
-
-	delayed_refs = &trans->transaction->delayed_refs;
 	return find_ref_head(&delayed_refs->href_root, bytenr, 0);
 }
 

fs/btrfs/delayed-ref.h

@@ -250,8 +250,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
 			       struct btrfs_trans_handle *trans,
 			       u64 bytenr, u64 num_bytes,
 			       u64 parent, u64 ref_root,
-			       u64 owner, u64 offset, u64 reserved, int action,
-			       struct btrfs_delayed_extent_op *extent_op);
+			       u64 owner, u64 offset, u64 reserved, int action);
 int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
 				struct btrfs_trans_handle *trans,
 				u64 bytenr, u64 num_bytes,
@@ -262,7 +261,8 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
 			      struct btrfs_delayed_ref_head *head);
 
 struct btrfs_delayed_ref_head *
-btrfs_find_delayed_ref_head(struct btrfs_trans_handle *trans, u64 bytenr);
+btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs,
+			    u64 bytenr);
 int btrfs_delayed_ref_lock(struct btrfs_trans_handle *trans,
 			   struct btrfs_delayed_ref_head *head);
 static inline void btrfs_delayed_ref_unlock(struct btrfs_delayed_ref_head *head)

fs/btrfs/dir-item.c

@@ -133,7 +133,7 @@ int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
 	struct btrfs_disk_key disk_key;
 	u32 data_size;
 
-	key.objectid = btrfs_ino(dir);
+	key.objectid = btrfs_ino(BTRFS_I(dir));
 	key.type = BTRFS_DIR_ITEM_KEY;
 	key.offset = btrfs_name_hash(name, name_len);
 
@@ -174,8 +174,7 @@ int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
 	btrfs_release_path(path);
 
 	ret2 = btrfs_insert_delayed_dir_index(trans, root->fs_info, name,
-					      name_len, dir, &disk_key, type,
-					      index);
+			name_len, BTRFS_I(dir), &disk_key, type, index);
 out_free:
 	btrfs_free_path(path);
 	if (ret)

fs/btrfs/disk-io.c

@@ -64,8 +64,7 @@
 static const struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
 static void free_fs_root(struct btrfs_root *root);
-static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
-				    int read_only);
+static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info);
 static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
 static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 				      struct btrfs_fs_info *fs_info);
@@ -1005,7 +1004,7 @@ static int __btree_submit_bio_done(struct inode *inode, struct bio *bio,
 	return ret;
 }
 
-static int check_async_write(struct inode *inode, unsigned long bio_flags)
+static int check_async_write(unsigned long bio_flags)
 {
 	if (bio_flags & EXTENT_BIO_TREE_LOG)
 		return 0;
@@ -1021,7 +1020,7 @@ static int btree_submit_bio_hook(struct inode *inode, struct bio *bio,
 				 u64 bio_offset)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	int async = check_async_write(inode, bio_flags);
+	int async = check_async_write(bio_flags);
 	int ret;
 
 	if (bio_op(bio) != REQ_OP_WRITE) {
@@ -1248,8 +1247,7 @@ struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr,
 
 }
 
-void clean_tree_block(struct btrfs_trans_handle *trans,
-		      struct btrfs_fs_info *fs_info,
+void clean_tree_block(struct btrfs_fs_info *fs_info,
 		      struct extent_buffer *buf)
 {
 	if (btrfs_header_generation(buf) ==
@@ -2802,7 +2800,7 @@ int open_ctree(struct super_block *sb,
 
 	memcpy(fs_info->fsid, fs_info->super_copy->fsid, BTRFS_FSID_SIZE);
 
-	ret = btrfs_check_super_valid(fs_info, sb->s_flags & MS_RDONLY);
+	ret = btrfs_check_super_valid(fs_info);
 	if (ret) {
 		btrfs_err(fs_info, "superblock contains fatal errors");
 		err = -EINVAL;
@@ -3411,7 +3409,7 @@ struct buffer_head *btrfs_read_dev_super(struct block_device *bdev)
  */
 static int write_dev_supers(struct btrfs_device *device,
 			    struct btrfs_super_block *sb,
-			    int do_barriers, int wait, int max_mirrors)
+			    int wait, int max_mirrors)
 {
 	struct buffer_head *bh;
 	int i;
@@ -3696,7 +3694,7 @@ int btrfs_calc_num_tolerated_disk_barrier_failures(
 	return num_tolerated_disk_barrier_failures;
 }
 
-static int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
+int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
 {
 	struct list_head *head;
 	struct btrfs_device *dev;
@@ -3753,7 +3751,7 @@ static int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
 		flags = btrfs_super_flags(sb);
 		btrfs_set_super_flags(sb, flags | BTRFS_HEADER_FLAG_WRITTEN);
 
-		ret = write_dev_supers(dev, sb, do_barriers, 0, max_mirrors);
+		ret = write_dev_supers(dev, sb, 0, max_mirrors);
 		if (ret)
 			total_errors++;
 	}
@@ -3776,7 +3774,7 @@ static int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
 		if (!dev->in_fs_metadata || !dev->writeable)
 			continue;
 
-		ret = write_dev_supers(dev, sb, do_barriers, 1, max_mirrors);
+		ret = write_dev_supers(dev, sb, 1, max_mirrors);
 		if (ret)
 			total_errors++;
 	}
@@ -3790,12 +3788,6 @@ static int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
 	return 0;
 }
 
-int write_ctree_super(struct btrfs_trans_handle *trans,
-		      struct btrfs_fs_info *fs_info, int max_mirrors)
-{
-	return write_all_supers(fs_info, max_mirrors);
-}
-
 /* Drop a fs root from the radix tree and free it. */
 void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 				  struct btrfs_root *root)
@@ -4122,8 +4114,7 @@ int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid)
 	return btree_read_extent_buffer_pages(fs_info, buf, parent_transid);
 }
 
-static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
-			      int read_only)
+static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_super_block *sb = fs_info->super_copy;
 	u64 nodesize = btrfs_super_nodesize(sb);

fs/btrfs/disk-io.h

@@ -52,14 +52,12 @@ int reada_tree_block_flagged(struct btrfs_fs_info *fs_info, u64 bytenr,
 struct extent_buffer *btrfs_find_create_tree_block(
 						struct btrfs_fs_info *fs_info,
 						u64 bytenr);
-void clean_tree_block(struct btrfs_trans_handle *trans,
-		      struct btrfs_fs_info *fs_info, struct extent_buffer *buf);
+void clean_tree_block(struct btrfs_fs_info *fs_info, struct extent_buffer *buf);
 int open_ctree(struct super_block *sb,
 	       struct btrfs_fs_devices *fs_devices,
 	       char *options);
 void close_ctree(struct btrfs_fs_info *fs_info);
-int write_ctree_super(struct btrfs_trans_handle *trans,
-		      struct btrfs_fs_info *fs_info, int max_mirrors);
+int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors);
 struct buffer_head *btrfs_read_dev_super(struct block_device *bdev);
 int btrfs_read_dev_one_super(struct block_device *bdev, int copy_num,
 			struct buffer_head **bh_ret);

fs/btrfs/export.c

@@ -30,7 +30,7 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len,
 	len  = BTRFS_FID_SIZE_NON_CONNECTABLE;
 	type = FILEID_BTRFS_WITHOUT_PARENT;
 
-	fid->objectid = btrfs_ino(inode);
+	fid->objectid = btrfs_ino(BTRFS_I(inode));
 	fid->root_objectid = BTRFS_I(inode)->root->objectid;
 	fid->gen = inode->i_generation;
 
@@ -166,13 +166,13 @@ static struct dentry *btrfs_get_parent(struct dentry *child)
 	if (!path)
 		return ERR_PTR(-ENOMEM);
 
-	if (btrfs_ino(dir) == BTRFS_FIRST_FREE_OBJECTID) {
+	if (btrfs_ino(BTRFS_I(dir)) == BTRFS_FIRST_FREE_OBJECTID) {
 		key.objectid = root->root_key.objectid;
 		key.type = BTRFS_ROOT_BACKREF_KEY;
 		key.offset = (u64)-1;
 		root = fs_info->tree_root;
 	} else {
-		key.objectid = btrfs_ino(dir);
+		key.objectid = btrfs_ino(BTRFS_I(dir));
 		key.type = BTRFS_INODE_REF_KEY;
 		key.offset = (u64)-1;
 	}
@@ -235,13 +235,10 @@ static int btrfs_get_name(struct dentry *parent, char *name,
 	int ret;
 	u64 ino;
 
-	if (!dir || !inode)
-		return -EINVAL;
-
 	if (!S_ISDIR(dir->i_mode))
 		return -EINVAL;
 
-	ino = btrfs_ino(inode);
+	ino = btrfs_ino(BTRFS_I(inode));
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -255,7 +252,7 @@ static int btrfs_get_name(struct dentry *parent, char *name,
 		root = fs_info->tree_root;
 	} else {
 		key.objectid = ino;
-		key.offset = btrfs_ino(dir);
+		key.offset = btrfs_ino(BTRFS_I(dir));
 		key.type = BTRFS_INODE_REF_KEY;
 	}
 

fs/btrfs/extent_io.c

@@ -98,7 +98,7 @@ static inline void __btrfs_debug_check_extent_io_range(const char *caller,
 	if (end >= PAGE_SIZE && (end % 2) == 0 && end != isize - 1) {
 		btrfs_debug_rl(BTRFS_I(inode)->root->fs_info,
 		    "%s: ino %llu isize %llu odd range [%llu,%llu]",
-				caller, btrfs_ino(inode), isize, start, end);
+			caller, btrfs_ino(BTRFS_I(inode)), isize, start, end);
 	}
 }
 #else
@@ -144,7 +144,7 @@ static void add_extent_changeset(struct extent_state *state, unsigned bits,
 	if (!set && (state->state & bits) == 0)
 		return;
 	changeset->bytes_changed += state->end - state->start + 1;
-	ret = ulist_add(changeset->range_changed, state->start, state->end,
+	ret = ulist_add(&changeset->range_changed, state->start, state->end,
 			GFP_ATOMIC);
 	/* ENOMEM */
 	BUG_ON(ret < 0);
@@ -226,6 +226,11 @@ static struct extent_state *alloc_extent_state(gfp_t mask)
 {
 	struct extent_state *state;
 
+	/*
+	 * The given mask might be not appropriate for the slab allocator,
+	 * drop the unsupported bits
+	 */
+	mask &= ~(__GFP_DMA32|__GFP_HIGHMEM);
 	state = kmem_cache_alloc(extent_state_cache, mask);
 	if (!state)
 		return state;
@@ -1549,33 +1554,24 @@ static noinline u64 find_delalloc_range(struct extent_io_tree *tree,
 	return found;
 }
 
+static int __process_pages_contig(struct address_space *mapping,
+				  struct page *locked_page,
+				  pgoff_t start_index, pgoff_t end_index,
+				  unsigned long page_ops, pgoff_t *index_ret);
+
 static noinline void __unlock_for_delalloc(struct inode *inode,
 					   struct page *locked_page,
 					   u64 start, u64 end)
 {
-	int ret;
-	struct page *pages[16];
 	unsigned long index = start >> PAGE_SHIFT;
 	unsigned long end_index = end >> PAGE_SHIFT;
-	unsigned long nr_pages = end_index - index + 1;
-	int i;
 
+	ASSERT(locked_page);
 	if (index == locked_page->index && end_index == index)
 		return;
 
-	while (nr_pages > 0) {
-		ret = find_get_pages_contig(inode->i_mapping, index,
-				     min_t(unsigned long, nr_pages,
-				     ARRAY_SIZE(pages)), pages);
-		for (i = 0; i < ret; i++) {
-			if (pages[i] != locked_page)
-				unlock_page(pages[i]);
-			put_page(pages[i]);
-		}
-		nr_pages -= ret;
-		index += ret;
-		cond_resched();
-	}
+	__process_pages_contig(inode->i_mapping, locked_page, index, end_index,
+			       PAGE_UNLOCK, NULL);
 }
 
 static noinline int lock_delalloc_pages(struct inode *inode,
@@ -1584,59 +1580,19 @@ static noinline int lock_delalloc_pages(struct inode *inode,
 					u64 delalloc_end)
 {
 	unsigned long index = delalloc_start >> PAGE_SHIFT;
-	unsigned long start_index = index;
+	unsigned long index_ret = index;
 	unsigned long end_index = delalloc_end >> PAGE_SHIFT;
-	unsigned long pages_locked = 0;
-	struct page *pages[16];
-	unsigned long nrpages;
 	int ret;
-	int i;
 
-	/* the caller is responsible for locking the start index */
+	ASSERT(locked_page);
 	if (index == locked_page->index && index == end_index)
 		return 0;
 
-	/* skip the page at the start index */
-	nrpages = end_index - index + 1;
-	while (nrpages > 0) {
-		ret = find_get_pages_contig(inode->i_mapping, index,
-				     min_t(unsigned long,
-				     nrpages, ARRAY_SIZE(pages)), pages);
-		if (ret == 0) {
-			ret = -EAGAIN;
-			goto done;
-		}
-		/* now we have an array of pages, lock them all */
-		for (i = 0; i < ret; i++) {
-			/*
-			 * the caller is taking responsibility for
-			 * locked_page
-			 */
-			if (pages[i] != locked_page) {
-				lock_page(pages[i]);
-				if (!PageDirty(pages[i]) ||
-				    pages[i]->mapping != inode->i_mapping) {
-					ret = -EAGAIN;
-					unlock_page(pages[i]);
-					put_page(pages[i]);
-					goto done;
-				}
-			}
-			put_page(pages[i]);
-			pages_locked++;
-		}
-		nrpages -= ret;
-		index += ret;
-		cond_resched();
-	}
-	ret = 0;
-done:
-	if (ret && pages_locked) {
-		__unlock_for_delalloc(inode, locked_page,
-			      delalloc_start,
-			      ((u64)(start_index + pages_locked - 1)) <<
-			      PAGE_SHIFT);
-	}
+	ret = __process_pages_contig(inode->i_mapping, locked_page, index,
+				     end_index, PAGE_LOCK, &index_ret);
+	if (ret == -EAGAIN)
+		__unlock_for_delalloc(inode, locked_page, delalloc_start,
+				      (u64)index_ret << PAGE_SHIFT);
 	return ret;
 }
 
@@ -1726,37 +1682,47 @@ STATIC u64 find_lock_delalloc_range(struct inode *inode,
 	return found;
 }
 
-void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
-				 u64 delalloc_end, struct page *locked_page,
-				 unsigned clear_bits,
-				 unsigned long page_ops)
+static int __process_pages_contig(struct address_space *mapping,
+				  struct page *locked_page,
+				  pgoff_t start_index, pgoff_t end_index,
+				  unsigned long page_ops, pgoff_t *index_ret)
 {
-	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
-	int ret;
+	unsigned long nr_pages = end_index - start_index + 1;
+	unsigned long pages_locked = 0;
+	pgoff_t index = start_index;
 	struct page *pages[16];
-	unsigned long index = start >> PAGE_SHIFT;
-	unsigned long end_index = end >> PAGE_SHIFT;
-	unsigned long nr_pages = end_index - index + 1;
+	unsigned ret;
+	int err = 0;
 	int i;
 
-	clear_extent_bit(tree, start, end, clear_bits, 1, 0, NULL, GFP_NOFS);
-	if (page_ops == 0)
-		return;
+	if (page_ops & PAGE_LOCK) {
+		ASSERT(page_ops == PAGE_LOCK);
+		ASSERT(index_ret && *index_ret == start_index);
+	}
 
 	if ((page_ops & PAGE_SET_ERROR) && nr_pages > 0)
-		mapping_set_error(inode->i_mapping, -EIO);
+		mapping_set_error(mapping, -EIO);
 
 	while (nr_pages > 0) {
-		ret = find_get_pages_contig(inode->i_mapping, index,
+		ret = find_get_pages_contig(mapping, index,
 				     min_t(unsigned long,
 				     nr_pages, ARRAY_SIZE(pages)), pages);
-		for (i = 0; i < ret; i++) {
+		if (ret == 0) {
+			/*
+			 * Only if we're going to lock these pages,
+			 * can we find nothing at @index.
+			 */
+			ASSERT(page_ops & PAGE_LOCK);
+			return ret;
+		}
 
+		for (i = 0; i < ret; i++) {
 			if (page_ops & PAGE_SET_PRIVATE2)
 				SetPagePrivate2(pages[i]);
 
 			if (pages[i] == locked_page) {
 				put_page(pages[i]);
+				pages_locked++;
 				continue;
 			}
 			if (page_ops & PAGE_CLEAR_DIRTY)
@@ -1769,12 +1735,40 @@ void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
 				end_page_writeback(pages[i]);
 			if (page_ops & PAGE_UNLOCK)
 				unlock_page(pages[i]);
+			if (page_ops & PAGE_LOCK) {
+				lock_page(pages[i]);
+				if (!PageDirty(pages[i]) ||
+				    pages[i]->mapping != mapping) {
+					unlock_page(pages[i]);
+					put_page(pages[i]);
+					err = -EAGAIN;
+					goto out;
+				}
+			}
 			put_page(pages[i]);
+			pages_locked++;
 		}
 		nr_pages -= ret;
 		index += ret;
 		cond_resched();
 	}
+out:
+	if (err && index_ret)
+		*index_ret = start_index + pages_locked - 1;
+	return err;
+}
+
+void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
+				 u64 delalloc_end, struct page *locked_page,
+				 unsigned clear_bits,
+				 unsigned long page_ops)
+{
+	clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, clear_bits, 1, 0,
+			 NULL, GFP_NOFS);
+
+	__process_pages_contig(inode->i_mapping, locked_page,
+			       start >> PAGE_SHIFT, end >> PAGE_SHIFT,
+			       page_ops, NULL);
 }
 
 /*
@@ -2060,7 +2054,7 @@ int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
 
 	btrfs_info_rl_in_rcu(fs_info,
 		"read error corrected: ino %llu off %llu (dev %s sector %llu)",
-				  btrfs_ino(inode), start,
+				  btrfs_ino(BTRFS_I(inode)), start,
 				  rcu_str_deref(dev->name), sector);
 	btrfs_bio_counter_dec(fs_info);
 	bio_put(bio);
@@ -2765,7 +2759,6 @@ static int submit_extent_page(int op, int op_flags, struct extent_io_tree *tree,
 			      size_t size, unsigned long offset,
 			      struct block_device *bdev,
 			      struct bio **bio_ret,
-			      unsigned long max_pages,
 			      bio_end_io_t end_io_func,
 			      int mirror_num,
 			      unsigned long prev_bio_flags,
@@ -2931,7 +2924,6 @@ static int __do_readpage(struct extent_io_tree *tree,
 		}
 	}
 	while (cur <= end) {
-		unsigned long pnr = (last_byte >> PAGE_SHIFT) + 1;
 		bool force_bio_submit = false;
 
 		if (cur >= last_byte) {
@@ -3066,10 +3058,9 @@ static int __do_readpage(struct extent_io_tree *tree,
 			continue;
 		}
 
-		pnr -= page->index;
 		ret = submit_extent_page(REQ_OP_READ, read_flags, tree, NULL,
 					 page, sector, disk_io_size, pg_offset,
-					 bdev, bio, pnr,
+					 bdev, bio,
 					 end_bio_extent_readpage, mirror_num,
 					 *bio_flags,
 					 this_bio_flag,
@@ -3210,7 +3201,7 @@ int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
 	return ret;
 }
 
-static void update_nr_written(struct page *page, struct writeback_control *wbc,
+static void update_nr_written(struct writeback_control *wbc,
 			      unsigned long nr_written)
 {
 	wbc->nr_to_write -= nr_written;
@@ -3330,7 +3321,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 	u64 block_start;
 	u64 iosize;
 	sector_t sector;
-	struct extent_state *cached_state = NULL;
 	struct extent_map *em;
 	struct block_device *bdev;
 	size_t pg_offset = 0;
@@ -3349,10 +3339,9 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 			else
 				redirty_page_for_writepage(wbc, page);
 
-			update_nr_written(page, wbc, nr_written);
+			update_nr_written(wbc, nr_written);
 			unlock_page(page);
-			ret = 1;
-			goto done_unlocked;
+			return 1;
 		}
 	}
 
@@ -3360,7 +3349,7 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 	 * we don't want to touch the inode after unlocking the page,
 	 * so we update the mapping writeback index now
 	 */
-	update_nr_written(page, wbc, nr_written + 1);
+	update_nr_written(wbc, nr_written + 1);
 
 	end = page_end;
 	if (i_size <= start) {
@@ -3374,7 +3363,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 
 	while (cur <= end) {
 		u64 em_end;
-		unsigned long max_nr;
 
 		if (cur >= i_size) {
 			if (tree->ops && tree->ops->writepage_end_io_hook)
@@ -3431,8 +3419,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 			continue;
 		}
 
-		max_nr = (i_size >> PAGE_SHIFT) + 1;
-
 		set_range_writeback(tree, cur, cur + iosize - 1);
 		if (!PageWriteback(page)) {
 			btrfs_err(BTRFS_I(inode)->root->fs_info,
@@ -3442,11 +3428,14 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 
 		ret = submit_extent_page(REQ_OP_WRITE, write_flags, tree, wbc,
 					 page, sector, iosize, pg_offset,
-					 bdev, &epd->bio, max_nr,
+					 bdev, &epd->bio,
 					 end_bio_extent_writepage,
 					 0, 0, 0, false);
-		if (ret)
+		if (ret) {
 			SetPageError(page);
+			if (PageWriteback(page))
+				end_page_writeback(page);
+		}
 
 		cur = cur + iosize;
 		pg_offset += iosize;
@@ -3454,11 +3443,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 	}
 done:
 	*nr_ret = nr;
-
-done_unlocked:
-
-	/* drop our reference on any cached states */
-	free_extent_state(cached_state);
 	return ret;
 }
 
@@ -3761,20 +3745,21 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
 		set_page_writeback(p);
 		ret = submit_extent_page(REQ_OP_WRITE, write_flags, tree, wbc,
 					 p, offset >> 9, PAGE_SIZE, 0, bdev,
-					 &epd->bio, -1,
+					 &epd->bio,
 					 end_bio_extent_buffer_writepage,
 					 0, epd->bio_flags, bio_flags, false);
 		epd->bio_flags = bio_flags;
 		if (ret) {
 			set_btree_ioerr(p);
-			end_page_writeback(p);
+			if (PageWriteback(p))
+				end_page_writeback(p);
 			if (atomic_sub_and_test(num_pages - i, &eb->io_pages))
 				end_extent_buffer_writeback(eb);
 			ret = -EIO;
 			break;
 		}
 		offset += PAGE_SIZE;
-		update_nr_written(p, wbc, 1);
+		update_nr_written(wbc, 1);
 		unlock_page(p);
 	}
 
@@ -3926,8 +3911,7 @@ int btree_write_cache_pages(struct address_space *mapping,
  * WB_SYNC_ALL then we were called for data integrity and we must wait for
  * existing IO to complete.
  */
-static int extent_write_cache_pages(struct extent_io_tree *tree,
-			     struct address_space *mapping,
+static int extent_write_cache_pages(struct address_space *mapping,
 			     struct writeback_control *wbc,
 			     writepage_t writepage, void *data,
 			     void (*flush_fn)(void *))
@@ -4168,8 +4152,7 @@ int extent_writepages(struct extent_io_tree *tree,
 		.bio_flags = 0,
 	};
 
-	ret = extent_write_cache_pages(tree, mapping, wbc,
-				       __extent_writepage, &epd,
+	ret = extent_write_cache_pages(mapping, wbc, __extent_writepage, &epd,
 				       flush_write_bio);
 	flush_epd_write_bio(&epd);
 	return ret;
@@ -4264,8 +4247,6 @@ static int try_release_extent_state(struct extent_map_tree *map,
 			   EXTENT_IOBITS, 0, NULL))
 		ret = 0;
 	else {
-		if ((mask & GFP_NOFS) == GFP_NOFS)
-			mask = GFP_NOFS;
 		/*
 		 * at this point we can safely clear everything except the
 		 * locked bit and the nodatasum bit
@@ -4410,8 +4391,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	 * lookup the last file extent.  We're not using i_size here
 	 * because there might be preallocation past i_size
 	 */
-	ret = btrfs_lookup_file_extent(NULL, root, path, btrfs_ino(inode), -1,
-				       0);
+	ret = btrfs_lookup_file_extent(NULL, root, path,
+			btrfs_ino(BTRFS_I(inode)), -1, 0);
 	if (ret < 0) {
 		btrfs_free_path(path);
 		return ret;
@@ -4426,7 +4407,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	found_type = found_key.type;
 
 	/* No extents, but there might be delalloc bits */
-	if (found_key.objectid != btrfs_ino(inode) ||
+	if (found_key.objectid != btrfs_ino(BTRFS_I(inode)) ||
 	    found_type != BTRFS_EXTENT_DATA_KEY) {
 		/* have to trust i_size as the end */
 		last = (u64)-1;
@@ -4535,8 +4516,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			 * lookup stuff.
 			 */
 			ret = btrfs_check_shared(trans, root->fs_info,
-						 root->objectid,
-						 btrfs_ino(inode), bytenr);
+					root->objectid,
+					btrfs_ino(BTRFS_I(inode)), bytenr);
 			if (trans)
 				btrfs_end_transaction(trans);
 			if (ret < 0)

fs/btrfs/extent_io.h

@@ -45,13 +45,14 @@
 #define EXTENT_BUFFER_IN_TREE 10
 #define EXTENT_BUFFER_WRITE_ERR 11    /* write IO error */
 
-/* these are flags for extent_clear_unlock_delalloc */
+/* these are flags for __process_pages_contig */
 #define PAGE_UNLOCK		(1 << 0)
 #define PAGE_CLEAR_DIRTY	(1 << 1)
 #define PAGE_SET_WRITEBACK	(1 << 2)
 #define PAGE_END_WRITEBACK	(1 << 3)
 #define PAGE_SET_PRIVATE2	(1 << 4)
 #define PAGE_SET_ERROR		(1 << 5)
+#define PAGE_LOCK		(1 << 6)
 
 /*
  * page->private values.  Every page that is controlled by the extent
@@ -192,7 +193,7 @@ struct extent_changeset {
 	u64 bytes_changed;
 
 	/* Changed ranges */
-	struct ulist *range_changed;
+	struct ulist range_changed;
 };
 
 static inline void extent_set_compress_type(unsigned long *bio_flags,

fs/btrfs/file-item.c

@@ -255,7 +255,7 @@ static int __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
 				} else {
 					btrfs_info_rl(fs_info,
 						   "no csum found for inode %llu start %llu",
-					       btrfs_ino(inode), offset);
+					       btrfs_ino(BTRFS_I(inode)), offset);
 				}
 				item = NULL;
 				btrfs_release_path(path);
@@ -856,8 +856,8 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
 		tmp = min(tmp, (next_offset - file_key.offset) >>
 					 fs_info->sb->s_blocksize_bits);
 
-		tmp = max((u64)1, tmp);
-		tmp = min(tmp, (u64)MAX_CSUM_ITEMS(fs_info, csum_size));
+		tmp = max_t(u64, 1, tmp);
+		tmp = min_t(u64, tmp, MAX_CSUM_ITEMS(fs_info, csum_size));
 		ins_size = csum_size * tmp;
 	} else {
 		ins_size = csum_size;
@@ -977,7 +977,7 @@ void btrfs_extent_item_to_extent_map(struct inode *inode,
 	} else {
 		btrfs_err(fs_info,
 			  "unknown file extent item type %d, inode %llu, offset %llu, root %llu",
-			  type, btrfs_ino(inode), extent_start,
+			  type, btrfs_ino(BTRFS_I(inode)), extent_start,
 			  root->root_key.objectid);
 	}
 }

fs/btrfs/file.c

@@ -168,7 +168,7 @@ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
 	if (!defrag)
 		return -ENOMEM;
 
-	defrag->ino = btrfs_ino(inode);
+	defrag->ino = btrfs_ino(BTRFS_I(inode));
 	defrag->transid = transid;
 	defrag->root = root->root_key.objectid;
 
@@ -702,7 +702,7 @@ int __btrfs_drop_extents(struct btrfs_trans_handle *trans,
 	struct btrfs_file_extent_item *fi;
 	struct btrfs_key key;
 	struct btrfs_key new_key;
-	u64 ino = btrfs_ino(inode);
+	u64 ino = btrfs_ino(BTRFS_I(inode));
 	u64 search_start = start;
 	u64 disk_bytenr = 0;
 	u64 num_bytes = 0;
@@ -1102,7 +1102,7 @@ int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
 	int del_slot = 0;
 	int recow;
 	int ret;
-	u64 ino = btrfs_ino(inode);
+	u64 ino = btrfs_ino(BTRFS_I(inode));
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -2062,7 +2062,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 * commit does not start nor waits for ordered extents to complete.
 	 */
 	smp_mb();
-	if (btrfs_inode_in_log(inode, fs_info->generation) ||
+	if (btrfs_inode_in_log(BTRFS_I(inode), fs_info->generation) ||
 	    (full_sync && BTRFS_I(inode)->last_trans <=
 	     fs_info->last_trans_committed) ||
 	    (!btrfs_have_ordered_extents_in_range(inode, start, len) &&
@@ -2203,7 +2203,7 @@ static int hole_mergeable(struct inode *inode, struct extent_buffer *leaf,
 		return 0;
 
 	btrfs_item_key_to_cpu(leaf, &key, slot);
-	if (key.objectid != btrfs_ino(inode) ||
+	if (key.objectid != btrfs_ino(BTRFS_I(inode)) ||
 	    key.type != BTRFS_EXTENT_DATA_KEY)
 		return 0;
 
@@ -2237,7 +2237,7 @@ static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode,
 	if (btrfs_fs_incompat(fs_info, NO_HOLES))
 		goto out;
 
-	key.objectid = btrfs_ino(inode);
+	key.objectid = btrfs_ino(BTRFS_I(inode));
 	key.type = BTRFS_EXTENT_DATA_KEY;
 	key.offset = offset;
 
@@ -2285,9 +2285,8 @@ static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode,
 	}
 	btrfs_release_path(path);
 
-	ret = btrfs_insert_file_extent(trans, root, btrfs_ino(inode), offset,
-				       0, 0, end - offset, 0, end - offset,
-				       0, 0, 0);
+	ret = btrfs_insert_file_extent(trans, root, btrfs_ino(BTRFS_I(inode)),
+			offset, 0, 0, end - offset, 0, end - offset, 0, 0, 0);
 	if (ret)
 		return ret;
 

fs/btrfs/free-space-cache.c

@@ -94,12 +94,11 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
 	return inode;
 }
 
-struct inode *lookup_free_space_inode(struct btrfs_root *root,
+struct inode *lookup_free_space_inode(struct btrfs_fs_info *fs_info,
 				      struct btrfs_block_group_cache
 				      *block_group, struct btrfs_path *path)
 {
 	struct inode *inode = NULL;
-	struct btrfs_fs_info *fs_info = root->fs_info;
 	u32 flags = BTRFS_INODE_NODATASUM | BTRFS_INODE_NODATACOW;
 
 	spin_lock(&block_group->lock);
@@ -109,7 +108,7 @@ struct inode *lookup_free_space_inode(struct btrfs_root *root,
 	if (inode)
 		return inode;
 
-	inode = __lookup_free_space_inode(root, path,
+	inode = __lookup_free_space_inode(fs_info->tree_root, path,
 					  block_group->key.objectid);
 	if (IS_ERR(inode))
 		return inode;
@@ -192,7 +191,7 @@ static int __create_free_space_inode(struct btrfs_root *root,
 	return 0;
 }
 
-int create_free_space_inode(struct btrfs_root *root,
+int create_free_space_inode(struct btrfs_fs_info *fs_info,
 			    struct btrfs_trans_handle *trans,
 			    struct btrfs_block_group_cache *block_group,
 			    struct btrfs_path *path)
@@ -200,11 +199,11 @@ int create_free_space_inode(struct btrfs_root *root,
 	int ret;
 	u64 ino;
 
-	ret = btrfs_find_free_objectid(root, &ino);
+	ret = btrfs_find_free_objectid(fs_info->tree_root, &ino);
 	if (ret < 0)
 		return ret;
 
-	return __create_free_space_inode(root, trans, path, ino,
+	return __create_free_space_inode(fs_info->tree_root, trans, path, ino,
 					 block_group->key.objectid);
 }
 
@@ -227,21 +226,21 @@ int btrfs_check_trunc_cache_free_space(struct btrfs_fs_info *fs_info,
 	return ret;
 }
 
-int btrfs_truncate_free_space_cache(struct btrfs_root *root,
-				    struct btrfs_trans_handle *trans,
+int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans,
 				    struct btrfs_block_group_cache *block_group,
 				    struct inode *inode)
 {
+	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int ret = 0;
-	struct btrfs_path *path = btrfs_alloc_path();
 	bool locked = false;
 
-	if (!path) {
-		ret = -ENOMEM;
-		goto fail;
-	}
-
 	if (block_group) {
+		struct btrfs_path *path = btrfs_alloc_path();
+
+		if (!path) {
+			ret = -ENOMEM;
+			goto fail;
+		}
 		locked = true;
 		mutex_lock(&trans->transaction->cache_write_mutex);
 		if (!list_empty(&block_group->io_list)) {
@@ -258,8 +257,8 @@ int btrfs_truncate_free_space_cache(struct btrfs_root *root,
 		spin_lock(&block_group->lock);
 		block_group->disk_cache_state = BTRFS_DC_CLEAR;
 		spin_unlock(&block_group->lock);
+		btrfs_free_path(path);
 	}
-	btrfs_free_path(path);
 
 	btrfs_i_size_write(inode, 0);
 	truncate_pagecache(inode, 0);
@@ -286,14 +285,14 @@ int btrfs_truncate_free_space_cache(struct btrfs_root *root,
 	return ret;
 }
 
-static int readahead_cache(struct inode *inode)
+static void readahead_cache(struct inode *inode)
 {
 	struct file_ra_state *ra;
 	unsigned long last_index;
 
 	ra = kzalloc(sizeof(*ra), GFP_NOFS);
 	if (!ra)
-		return -ENOMEM;
+		return;
 
 	file_ra_state_init(ra, inode->i_mapping);
 	last_index = (i_size_read(inode) - 1) >> PAGE_SHIFT;
@@ -301,8 +300,6 @@ static int readahead_cache(struct inode *inode)
 	page_cache_sync_readahead(inode->i_mapping, ra, NULL, 0, last_index);
 
 	kfree(ra);
-
-	return 0;
 }
 
 static int io_ctl_init(struct btrfs_io_ctl *io_ctl, struct inode *inode,
@@ -313,7 +310,7 @@ static int io_ctl_init(struct btrfs_io_ctl *io_ctl, struct inode *inode,
 
 	num_pages = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE);
 
-	if (btrfs_ino(inode) != BTRFS_FREE_INO_OBJECTID)
+	if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FREE_INO_OBJECTID)
 		check_crcs = 1;
 
 	/* Make sure we can fit our crcs into the first page */
@@ -730,9 +727,7 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
 	if (ret)
 		return ret;
 
-	ret = readahead_cache(inode);
-	if (ret)
-		goto out;
+	readahead_cache(inode);
 
 	ret = io_ctl_prepare_pages(&io_ctl, inode, 1);
 	if (ret)
@@ -828,7 +823,6 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
 			  struct btrfs_block_group_cache *block_group)
 {
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
-	struct btrfs_root *root = fs_info->tree_root;
 	struct inode *inode;
 	struct btrfs_path *path;
 	int ret = 0;
@@ -852,7 +846,7 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
 	path->search_commit_root = 1;
 	path->skip_locking = 1;
 
-	inode = lookup_free_space_inode(root, block_group, path);
+	inode = lookup_free_space_inode(fs_info, block_group, path);
 	if (IS_ERR(inode)) {
 		btrfs_free_path(path);
 		return 0;
@@ -1128,8 +1122,7 @@ cleanup_bitmap_list(struct list_head *bitmap_list)
 static void noinline_for_stack
 cleanup_write_cache_enospc(struct inode *inode,
 			   struct btrfs_io_ctl *io_ctl,
-			   struct extent_state **cached_state,
-			   struct list_head *bitmap_list)
+			   struct extent_state **cached_state)
 {
 	io_ctl_drop_pages(io_ctl);
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0,
@@ -1225,8 +1218,6 @@ int btrfs_wait_cache_io(struct btrfs_trans_handle *trans,
  * @ctl - the free space cache we are going to write out
  * @block_group - the block_group for this cache if it belongs to a block_group
  * @trans - the trans handle
- * @path - the path to use
- * @offset - the offset for the key we'll insert
  *
  * This function writes out a free space cache struct to disk for quick recovery
  * on mount.  This will return 0 if it was successful in writing the cache out,
@@ -1236,8 +1227,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 				   struct btrfs_free_space_ctl *ctl,
 				   struct btrfs_block_group_cache *block_group,
 				   struct btrfs_io_ctl *io_ctl,
-				   struct btrfs_trans_handle *trans,
-				   struct btrfs_path *path, u64 offset)
+				   struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_state *cached_state = NULL;
@@ -1365,7 +1355,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 	mutex_unlock(&ctl->cache_writeout_mutex);
 
 out_nospc:
-	cleanup_write_cache_enospc(inode, io_ctl, &cached_state, &bitmap_list);
+	cleanup_write_cache_enospc(inode, io_ctl, &cached_state);
 
 	if (block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA))
 		up_write(&block_group->data_rwsem);
@@ -1378,7 +1368,6 @@ int btrfs_write_out_cache(struct btrfs_fs_info *fs_info,
 			  struct btrfs_block_group_cache *block_group,
 			  struct btrfs_path *path)
 {
-	struct btrfs_root *root = fs_info->tree_root;
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	struct inode *inode;
 	int ret = 0;
@@ -1390,13 +1379,12 @@ int btrfs_write_out_cache(struct btrfs_fs_info *fs_info,
 	}
 	spin_unlock(&block_group->lock);
 
-	inode = lookup_free_space_inode(root, block_group, path);
+	inode = lookup_free_space_inode(fs_info, block_group, path);
 	if (IS_ERR(inode))
 		return 0;
 
-	ret = __btrfs_write_out_cache(root, inode, ctl, block_group,
-				      &block_group->io_ctl, trans,
-				      path, block_group->key.objectid);
+	ret = __btrfs_write_out_cache(fs_info->tree_root, inode, ctl,
+				block_group, &block_group->io_ctl, trans);
 	if (ret) {
 #ifdef DEBUG
 		btrfs_err(fs_info,
@@ -3543,8 +3531,7 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
 		return 0;
 
 	memset(&io_ctl, 0, sizeof(io_ctl));
-	ret = __btrfs_write_out_cache(root, inode, ctl, NULL, &io_ctl,
-				      trans, path, 0);
+	ret = __btrfs_write_out_cache(root, inode, ctl, NULL, &io_ctl, trans);
 	if (!ret) {
 		/*
 		 * At this point writepages() didn't error out, so our metadata

fs/btrfs/free-space-cache.h

@@ -51,18 +51,17 @@ struct btrfs_free_space_op {
 
 struct btrfs_io_ctl;
 
-struct inode *lookup_free_space_inode(struct btrfs_root *root,
+struct inode *lookup_free_space_inode(struct btrfs_fs_info *fs_info,
 				      struct btrfs_block_group_cache
 				      *block_group, struct btrfs_path *path);
-int create_free_space_inode(struct btrfs_root *root,
+int create_free_space_inode(struct btrfs_fs_info *fs_info,
 			    struct btrfs_trans_handle *trans,
 			    struct btrfs_block_group_cache *block_group,
 			    struct btrfs_path *path);
 
 int btrfs_check_trunc_cache_free_space(struct btrfs_fs_info *fs_info,
 				       struct btrfs_block_rsv *rsv);
-int btrfs_truncate_free_space_cache(struct btrfs_root *root,
-				    struct btrfs_trans_handle *trans,
+int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans,
 				    struct btrfs_block_group_cache *block_group,
 				    struct inode *inode);
 int load_free_space_cache(struct btrfs_fs_info *fs_info,

fs/btrfs/free-space-tree.c

@@ -1269,7 +1269,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
 	list_del(&free_space_root->dirty_list);
 
 	btrfs_tree_lock(free_space_root->node);
-	clean_tree_block(trans, fs_info, free_space_root->node);
+	clean_tree_block(fs_info, free_space_root->node);
 	btrfs_tree_unlock(free_space_root->node);
 	btrfs_free_tree_block(trans, free_space_root, free_space_root->node,
 			      0, 1);

fs/btrfs/inode-map.c

@@ -467,7 +467,7 @@ int btrfs_save_ino_cache(struct btrfs_root *root,
 	}
 
 	if (i_size_read(inode) > 0) {
-		ret = btrfs_truncate_free_space_cache(root, trans, NULL, inode);
+		ret = btrfs_truncate_free_space_cache(trans, NULL, inode);
 		if (ret) {
 			if (ret != -ENOSPC)
 				btrfs_abort_transaction(trans, ret);

fs/btrfs/ioctl.c

@@ -395,7 +395,7 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
 		q = bdev_get_queue(device->bdev);
 		if (blk_queue_discard(q)) {
 			num_devices++;
-			minlen = min((u64)q->limits.discard_granularity,
+			minlen = min_t(u64, q->limits.discard_granularity,
 				     minlen);
 		}
 	}
@@ -487,8 +487,7 @@ static noinline int create_subvol(struct inode *dir,
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
-		btrfs_subvolume_release_metadata(fs_info, &block_rsv,
-						 qgroup_reserved);
+		btrfs_subvolume_release_metadata(fs_info, &block_rsv);
 		goto fail_free;
 	}
 	trans->block_rsv = &block_rsv;
@@ -601,7 +600,7 @@ static noinline int create_subvol(struct inode *dir,
 
 	ret = btrfs_add_root_ref(trans, fs_info,
 				 objectid, root->root_key.objectid,
-				 btrfs_ino(dir), index, name, namelen);
+				 btrfs_ino(BTRFS_I(dir)), index, name, namelen);
 	BUG_ON(ret);
 
 	ret = btrfs_uuid_tree_add(trans, fs_info, root_item->uuid,
@@ -613,7 +612,7 @@ static noinline int create_subvol(struct inode *dir,
 	kfree(root_item);
 	trans->block_rsv = NULL;
 	trans->bytes_reserved = 0;
-	btrfs_subvolume_release_metadata(fs_info, &block_rsv, qgroup_reserved);
+	btrfs_subvolume_release_metadata(fs_info, &block_rsv);
 
 	if (async_transid) {
 		*async_transid = trans->transid;
@@ -657,7 +656,7 @@ static void btrfs_wait_for_no_snapshoting_writes(struct btrfs_root *root)
 }
 
 static int create_snapshot(struct btrfs_root *root, struct inode *dir,
-			   struct dentry *dentry, char *name, int namelen,
+			   struct dentry *dentry,
 			   u64 *async_transid, bool readonly,
 			   struct btrfs_qgroup_inherit *inherit)
 {
@@ -670,12 +669,12 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
 		return -EINVAL;
 
-	pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
+	pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_KERNEL);
 	if (!pending_snapshot)
 		return -ENOMEM;
 
 	pending_snapshot->root_item = kzalloc(sizeof(struct btrfs_root_item),
-			GFP_NOFS);
+			GFP_KERNEL);
 	pending_snapshot->path = btrfs_alloc_path();
 	if (!pending_snapshot->root_item || !pending_snapshot->path) {
 		ret = -ENOMEM;
@@ -753,9 +752,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	d_instantiate(dentry, inode);
 	ret = 0;
 fail:
-	btrfs_subvolume_release_metadata(fs_info,
-					 &pending_snapshot->block_rsv,
-					 pending_snapshot->qgroup_reserved);
+	btrfs_subvolume_release_metadata(fs_info, &pending_snapshot->block_rsv);
 dec_and_free:
 	if (atomic_dec_and_test(&root->will_be_snapshoted))
 		wake_up_atomic_t(&root->will_be_snapshoted);
@@ -874,7 +871,7 @@ static noinline int btrfs_mksubvol(const struct path *parent,
 		goto out_up_read;
 
 	if (snap_src) {
-		error = create_snapshot(snap_src, dir, dentry, name, namelen,
+		error = create_snapshot(snap_src, dir, dentry,
 					async_transid, readonly, inherit);
 	} else {
 		error = create_subvol(dir, dentry, name, namelen,
@@ -941,7 +938,7 @@ static int find_new_extents(struct btrfs_root *root,
 	struct btrfs_file_extent_item *extent;
 	int type;
 	int ret;
-	u64 ino = btrfs_ino(inode);
+	u64 ino = btrfs_ino(BTRFS_I(inode));
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -1780,7 +1777,7 @@ static noinline int btrfs_ioctl_subvol_getflags(struct file *file,
 	int ret = 0;
 	u64 flags = 0;
 
-	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID)
+	if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID)
 		return -EINVAL;
 
 	down_read(&fs_info->subvol_sem);
@@ -1812,7 +1809,7 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file,
 	if (ret)
 		goto out;
 
-	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
+	if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) {
 		ret = -EINVAL;
 		goto out_drop_write;
 	}
@@ -2446,7 +2443,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 	if (err)
 		goto out_dput;
 
-	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
+	if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) {
 		err = -EINVAL;
 		goto out_dput;
 	}
@@ -2497,7 +2494,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 	trans->block_rsv = &block_rsv;
 	trans->bytes_reserved = block_rsv.size;
 
-	btrfs_record_snapshot_destroy(trans, dir);
+	btrfs_record_snapshot_destroy(trans, BTRFS_I(dir));
 
 	ret = btrfs_unlink_subvol(trans, root, dir,
 				dest->root_key.objectid,
@@ -2555,7 +2552,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 		err = ret;
 	inode->i_flags |= S_DEAD;
 out_release:
-	btrfs_subvolume_release_metadata(fs_info, &block_rsv, qgroup_reserved);
+	btrfs_subvolume_release_metadata(fs_info, &block_rsv);
 out_up_write:
 	up_write(&fs_info->subvol_sem);
 	if (err) {
@@ -2613,9 +2610,6 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
 			goto out;
 		}
 		ret = btrfs_defrag_root(root);
-		if (ret)
-			goto out;
-		ret = btrfs_defrag_root(root->fs_info->extent_root);
 		break;
 	case S_IFREG:
 		if (!(file->f_mode & FMODE_WRITE)) {
@@ -3047,11 +3041,21 @@ static int btrfs_cmp_data_prepare(struct inode *src, u64 loff,
 	cmp->src_pages = src_pgarr;
 	cmp->dst_pages = dst_pgarr;
 
-	ret = gather_extent_pages(src, cmp->src_pages, cmp->num_pages, loff);
+	/*
+	 * If deduping ranges in the same inode, locking rules make it mandatory
+	 * to always lock pages in ascending order to avoid deadlocks with
+	 * concurrent tasks (such as starting writeback/delalloc).
+	 */
+	if (src == dst && dst_loff < loff) {
+		swap(src_pgarr, dst_pgarr);
+		swap(loff, dst_loff);
+	}
+
+	ret = gather_extent_pages(src, src_pgarr, cmp->num_pages, loff);
 	if (ret)
 		goto out;
 
-	ret = gather_extent_pages(dst, cmp->dst_pages, cmp->num_pages, dst_loff);
+	ret = gather_extent_pages(dst, dst_pgarr, cmp->num_pages, dst_loff);
 
 out:
 	if (ret)
@@ -3059,8 +3063,7 @@ static int btrfs_cmp_data_prepare(struct inode *src, u64 loff,
 	return 0;
 }
 
-static int btrfs_cmp_data(struct inode *src, u64 loff, struct inode *dst,
-			  u64 dst_loff, u64 len, struct cmp_pages *cmp)
+static int btrfs_cmp_data(u64 len, struct cmp_pages *cmp)
 {
 	int ret = 0;
 	int i;
@@ -3128,26 +3131,27 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
 	int ret;
 	u64 len = olen;
 	struct cmp_pages cmp;
-	int same_inode = 0;
+	bool same_inode = (src == dst);
 	u64 same_lock_start = 0;
 	u64 same_lock_len = 0;
 
-	if (src == dst)
-		same_inode = 1;
-
 	if (len == 0)
 		return 0;
 
-	if (same_inode) {
+	if (same_inode)
 		inode_lock(src);
+	else
+		btrfs_double_inode_lock(src, dst);
 
-		ret = extent_same_check_offsets(src, loff, &len, olen);
-		if (ret)
-			goto out_unlock;
-		ret = extent_same_check_offsets(src, dst_loff, &len, olen);
-		if (ret)
-			goto out_unlock;
+	ret = extent_same_check_offsets(src, loff, &len, olen);
+	if (ret)
+		goto out_unlock;
+
+	ret = extent_same_check_offsets(dst, dst_loff, &len, olen);
+	if (ret)
+		goto out_unlock;
 
+	if (same_inode) {
 		/*
 		 * Single inode case wants the same checks, except we
 		 * don't want our length pushed out past i_size as
@@ -3175,16 +3179,6 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
 
 		same_lock_start = min_t(u64, loff, dst_loff);
 		same_lock_len = max_t(u64, loff, dst_loff) + len - same_lock_start;
-	} else {
-		btrfs_double_inode_lock(src, dst);
-
-		ret = extent_same_check_offsets(src, loff, &len, olen);
-		if (ret)
-			goto out_unlock;
-
-		ret = extent_same_check_offsets(dst, dst_loff, &len, olen);
-		if (ret)
-			goto out_unlock;
 	}
 
 	/* don't make the dst file partly checksummed */
@@ -3236,7 +3230,7 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
 	}
 
 	/* pass original length for comparison so we stay within i_size */
-	ret = btrfs_cmp_data(src, loff, dst, dst_loff, olen, &cmp);
+	ret = btrfs_cmp_data(olen, &cmp);
 	if (ret == 0)
 		ret = btrfs_clone(src, dst, loff, olen, len, dst_loff, 1);
 
@@ -3399,8 +3393,7 @@ static void clone_update_extent_map(struct inode *inode,
  * data into the destination inode's inline extent if the later is greater then
  * the former.
  */
-static int clone_copy_inline_extent(struct inode *src,
-				    struct inode *dst,
+static int clone_copy_inline_extent(struct inode *dst,
 				    struct btrfs_trans_handle *trans,
 				    struct btrfs_path *path,
 				    struct btrfs_key *new_key,
@@ -3420,7 +3413,7 @@ static int clone_copy_inline_extent(struct inode *src,
 	if (new_key->offset > 0)
 		return -EOPNOTSUPP;
 
-	key.objectid = btrfs_ino(dst);
+	key.objectid = btrfs_ino(BTRFS_I(dst));
 	key.type = BTRFS_EXTENT_DATA_KEY;
 	key.offset = 0;
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
@@ -3435,7 +3428,7 @@ static int clone_copy_inline_extent(struct inode *src,
 				goto copy_inline_extent;
 		}
 		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
-		if (key.objectid == btrfs_ino(dst) &&
+		if (key.objectid == btrfs_ino(BTRFS_I(dst)) &&
 		    key.type == BTRFS_EXTENT_DATA_KEY) {
 			ASSERT(key.offset > 0);
 			return -EOPNOTSUPP;
@@ -3469,7 +3462,7 @@ static int clone_copy_inline_extent(struct inode *src,
 		} else if (ret == 0) {
 			btrfs_item_key_to_cpu(path->nodes[0], &key,
 					      path->slots[0]);
-			if (key.objectid == btrfs_ino(dst) &&
+			if (key.objectid == btrfs_ino(BTRFS_I(dst)) &&
 			    key.type == BTRFS_EXTENT_DATA_KEY)
 				return -EOPNOTSUPP;
 		}
@@ -3563,7 +3556,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 
 	path->reada = READA_FORWARD;
 	/* clone data */
-	key.objectid = btrfs_ino(src);
+	key.objectid = btrfs_ino(BTRFS_I(src));
 	key.type = BTRFS_EXTENT_DATA_KEY;
 	key.offset = off;
 
@@ -3606,7 +3599,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 
 		btrfs_item_key_to_cpu(leaf, &key, slot);
 		if (key.type > BTRFS_EXTENT_DATA_KEY ||
-		    key.objectid != btrfs_ino(src))
+		    key.objectid != btrfs_ino(BTRFS_I(src)))
 			break;
 
 		if (key.type == BTRFS_EXTENT_DATA_KEY) {
@@ -3659,7 +3652,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 			path->leave_spinning = 0;
 
 			memcpy(&new_key, &key, sizeof(new_key));
-			new_key.objectid = btrfs_ino(inode);
+			new_key.objectid = btrfs_ino(BTRFS_I(inode));
 			if (off <= key.offset)
 				new_key.offset = key.offset + destoff - off;
 			else
@@ -3749,7 +3742,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 							fs_info,
 							disko, diskl, 0,
 							root->root_key.objectid,
-							btrfs_ino(inode),
+							btrfs_ino(BTRFS_I(inode)),
 							new_key.offset - datao);
 					if (ret) {
 						btrfs_abort_transaction(trans,
@@ -3779,7 +3772,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 				size -= skip + trim;
 				datal -= skip + trim;
 
-				ret = clone_copy_inline_extent(src, inode,
+				ret = clone_copy_inline_extent(inode,
 							       trans, path,
 							       &new_key,
 							       drop_start,
@@ -5129,7 +5122,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file,
 
 	down_write(&fs_info->subvol_sem);
 
-	if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
+	if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) {
 		ret = -EINVAL;
 		goto out;
 	}

fs/btrfs/ordered-data.c

@@ -432,7 +432,7 @@ int btrfs_dec_test_ordered_pending(struct inode *inode,
 }
 
 /* Needs to either be called under a log transaction or the log_mutex */
-void btrfs_get_logged_extents(struct inode *inode,
+void btrfs_get_logged_extents(struct btrfs_inode *inode,
 			      struct list_head *logged_list,
 			      const loff_t start,
 			      const loff_t end)
@@ -442,7 +442,7 @@ void btrfs_get_logged_extents(struct inode *inode,
 	struct rb_node *n;
 	struct rb_node *prev;
 
-	tree = &BTRFS_I(inode)->ordered_tree;
+	tree = &inode->ordered_tree;
 	spin_lock_irq(&tree->lock);
 	n = __tree_search(&tree->tree, end, &prev);
 	if (!n)
@@ -984,8 +984,18 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 	}
 	disk_i_size = BTRFS_I(inode)->disk_i_size;
 
-	/* truncate file */
-	if (disk_i_size > i_size) {
+	/*
+	 * truncate file.
+	 * If ordered is not NULL, then this is called from endio and
+	 * disk_i_size will be updated by either truncate itself or any
+	 * in-flight IOs which are inside the disk_i_size.
+	 *
+	 * Because btrfs_setsize() may set i_size with disk_i_size if truncate
+	 * fails somehow, we need to make sure we have a precise disk_i_size by
+	 * updating it as usual.
+	 *
+	 */
+	if (!ordered && disk_i_size > i_size) {
 		BTRFS_I(inode)->disk_i_size = orig_offset;
 		ret = 0;
 		goto out;
@@ -1032,25 +1042,22 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 		/* We treat this entry as if it doesn't exist */
 		if (test_bit(BTRFS_ORDERED_UPDATED_ISIZE, &test->flags))
 			continue;
-		if (test->file_offset + test->len <= disk_i_size)
+
+		if (entry_end(test) <= disk_i_size)
 			break;
 		if (test->file_offset >= i_size)
 			break;
-		if (entry_end(test) > disk_i_size) {
-			/*
-			 * we don't update disk_i_size now, so record this
-			 * undealt i_size. Or we will not know the real
-			 * i_size.
-			 */
-			if (test->outstanding_isize < offset)
-				test->outstanding_isize = offset;
-			if (ordered &&
-			    ordered->outstanding_isize >
-			    test->outstanding_isize)
-				test->outstanding_isize =
-						ordered->outstanding_isize;
-			goto out;
-		}
+
+		/*
+		 * We don't update disk_i_size now, so record this undealt
+		 * i_size. Or we will not know the real i_size.
+		 */
+		if (test->outstanding_isize < offset)
+			test->outstanding_isize = offset;
+		if (ordered &&
+		    ordered->outstanding_isize > test->outstanding_isize)
+			test->outstanding_isize = ordered->outstanding_isize;
+		goto out;
 	}
 	new_i_size = min_t(u64, offset, i_size);
 

fs/btrfs/ordered-data.h

@@ -75,6 +75,8 @@ struct btrfs_ordered_sum {
 				 * in the logging code. */
 #define BTRFS_ORDERED_PENDING 11 /* We are waiting for this ordered extent to
 				  * complete in the current transaction. */
+#define BTRFS_ORDERED_REGULAR 12 /* Regular IO for COW */
+
 struct btrfs_ordered_extent {
 	/* logical offset in the file */
 	u64 file_offset;
@@ -201,7 +203,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
 			       const u64 range_start, const u64 range_len);
 int btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
 			      const u64 range_start, const u64 range_len);
-void btrfs_get_logged_extents(struct inode *inode,
+void btrfs_get_logged_extents(struct btrfs_inode *inode,
 			      struct list_head *logged_list,
 			      const loff_t start,
 			      const loff_t end);

fs/btrfs/props.c

@@ -279,7 +279,7 @@ static void inode_prop_iterator(void *ctx,
 	if (unlikely(ret))
 		btrfs_warn(root->fs_info,
 			   "error applying prop %s to ino %llu (root %llu): %d",
-			   handler->xattr_name, btrfs_ino(inode),
+			   handler->xattr_name, btrfs_ino(BTRFS_I(inode)),
 			   root->root_key.objectid, ret);
 	else
 		set_bit(BTRFS_INODE_HAS_PROPS, &BTRFS_I(inode)->runtime_flags);
@@ -288,7 +288,7 @@ static void inode_prop_iterator(void *ctx,
 int btrfs_load_inode_props(struct inode *inode, struct btrfs_path *path)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	u64 ino = btrfs_ino(inode);
+	u64 ino = btrfs_ino(BTRFS_I(inode));
 	int ret;
 
 	ret = iterate_object_props(root, path, ino, inode_prop_iterator, inode);

fs/btrfs/qgroup.c

@@ -319,7 +319,7 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
 		return 0;
 
-	fs_info->qgroup_ulist = ulist_alloc(GFP_NOFS);
+	fs_info->qgroup_ulist = ulist_alloc(GFP_KERNEL);
 	if (!fs_info->qgroup_ulist) {
 		ret = -ENOMEM;
 		goto out;
@@ -876,7 +876,7 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	fs_info->qgroup_ulist = ulist_alloc(GFP_NOFS);
+	fs_info->qgroup_ulist = ulist_alloc(GFP_KERNEL);
 	if (!fs_info->qgroup_ulist) {
 		ret = -ENOMEM;
 		goto out;
@@ -1019,7 +1019,7 @@ int btrfs_quota_disable(struct btrfs_trans_handle *trans,
 	list_del(&quota_root->dirty_list);
 
 	btrfs_tree_lock(quota_root->node);
-	clean_tree_block(trans, fs_info, quota_root->node);
+	clean_tree_block(fs_info, quota_root->node);
 	btrfs_tree_unlock(quota_root->node);
 	btrfs_free_tree_block(trans, quota_root, quota_root->node, 0, 1);
 
@@ -1038,6 +1038,15 @@ static void qgroup_dirty(struct btrfs_fs_info *fs_info,
 		list_add(&qgroup->dirty, &fs_info->dirty_qgroups);
 }
 
+static void report_reserved_underflow(struct btrfs_fs_info *fs_info,
+				      struct btrfs_qgroup *qgroup,
+				      u64 num_bytes)
+{
+	btrfs_warn(fs_info,
+		"qgroup %llu reserved space underflow, have: %llu, to free: %llu",
+		qgroup->qgroupid, qgroup->reserved, num_bytes);
+	qgroup->reserved = 0;
+}
 /*
  * The easy accounting, if we are adding/removing the only ref for an extent
  * then this qgroup and all of the parent qgroups get their reference and
@@ -1065,8 +1074,12 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info,
 	WARN_ON(sign < 0 && qgroup->excl < num_bytes);
 	qgroup->excl += sign * num_bytes;
 	qgroup->excl_cmpr += sign * num_bytes;
-	if (sign > 0)
-		qgroup->reserved -= num_bytes;
+	if (sign > 0) {
+		if (WARN_ON(qgroup->reserved < num_bytes))
+			report_reserved_underflow(fs_info, qgroup, num_bytes);
+		else
+			qgroup->reserved -= num_bytes;
+	}
 
 	qgroup_dirty(fs_info, qgroup);
 
@@ -1086,8 +1099,13 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info,
 		qgroup->rfer_cmpr += sign * num_bytes;
 		WARN_ON(sign < 0 && qgroup->excl < num_bytes);
 		qgroup->excl += sign * num_bytes;
-		if (sign > 0)
-			qgroup->reserved -= num_bytes;
+		if (sign > 0) {
+			if (WARN_ON(qgroup->reserved < num_bytes))
+				report_reserved_underflow(fs_info, qgroup,
+							  num_bytes);
+			else
+				qgroup->reserved -= num_bytes;
+		}
 		qgroup->excl_cmpr += sign * num_bytes;
 		qgroup_dirty(fs_info, qgroup);
 
@@ -1156,7 +1174,7 @@ int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
 	if (btrfs_qgroup_level(src) >= btrfs_qgroup_level(dst))
 		return -EINVAL;
 
-	tmp = ulist_alloc(GFP_NOFS);
+	tmp = ulist_alloc(GFP_KERNEL);
 	if (!tmp)
 		return -ENOMEM;
 
@@ -1205,7 +1223,7 @@ int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-int __del_qgroup_relation(struct btrfs_trans_handle *trans,
+static int __del_qgroup_relation(struct btrfs_trans_handle *trans,
 			      struct btrfs_fs_info *fs_info, u64 src, u64 dst)
 {
 	struct btrfs_root *quota_root;
@@ -1216,7 +1234,7 @@ int __del_qgroup_relation(struct btrfs_trans_handle *trans,
 	int ret = 0;
 	int err;
 
-	tmp = ulist_alloc(GFP_NOFS);
+	tmp = ulist_alloc(GFP_KERNEL);
 	if (!tmp)
 		return -ENOMEM;
 
@@ -1446,8 +1464,9 @@ int btrfs_qgroup_prepare_account_extents(struct btrfs_trans_handle *trans,
 	while (node) {
 		record = rb_entry(node, struct btrfs_qgroup_extent_record,
 				  node);
-		ret = btrfs_find_all_roots(NULL, fs_info, record->bytenr, 0,
-					   &record->old_roots);
+		if (WARN_ON(!record->old_roots))
+			ret = btrfs_find_all_roots(NULL, fs_info,
+					record->bytenr, 0, &record->old_roots);
 		if (ret < 0)
 			break;
 		if (qgroup_to_skip)
@@ -1486,6 +1505,28 @@ int btrfs_qgroup_trace_extent_nolock(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
+int btrfs_qgroup_trace_extent_post(struct btrfs_fs_info *fs_info,
+				   struct btrfs_qgroup_extent_record *qrecord)
+{
+	struct ulist *old_root;
+	u64 bytenr = qrecord->bytenr;
+	int ret;
+
+	ret = btrfs_find_all_roots(NULL, fs_info, bytenr, 0, &old_root);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Here we don't need to get the lock of
+	 * trans->transaction->delayed_refs, since inserted qrecord won't
+	 * be deleted, only qrecord->node may be modified (new qrecord insert)
+	 *
+	 * So modifying qrecord->old_roots is safe here
+	 */
+	qrecord->old_roots = old_root;
+	return 0;
+}
+
 int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
 		struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes,
 		gfp_t gfp_flag)
@@ -1511,9 +1552,11 @@ int btrfs_qgroup_trace_extent(struct btrfs_trans_handle *trans,
 	spin_lock(&delayed_refs->lock);
 	ret = btrfs_qgroup_trace_extent_nolock(fs_info, delayed_refs, record);
 	spin_unlock(&delayed_refs->lock);
-	if (ret > 0)
+	if (ret > 0) {
 		kfree(record);
-	return 0;
+		return 0;
+	}
+	return btrfs_qgroup_trace_extent_post(fs_info, record);
 }
 
 int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
@@ -1571,8 +1614,7 @@ int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
  * If we increment the root nodes slot counter past the number of
  * elements, 1 is returned to signal completion of the search.
  */
-static int adjust_slots_upwards(struct btrfs_root *root,
-				struct btrfs_path *path, int root_level)
+static int adjust_slots_upwards(struct btrfs_path *path, int root_level)
 {
 	int level = 0;
 	int nr, slot;
@@ -1713,7 +1755,7 @@ int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 				goto out;
 
 			/* Nonzero return here means we completed our search */
-			ret = adjust_slots_upwards(root, path, root_level);
+			ret = adjust_slots_upwards(path, root_level);
 			if (ret)
 				break;
 
@@ -1927,13 +1969,14 @@ btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans,
 	u64 nr_old_roots = 0;
 	int ret = 0;
 
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
+		return 0;
+
 	if (new_roots)
 		nr_new_roots = new_roots->nnodes;
 	if (old_roots)
 		nr_old_roots = old_roots->nnodes;
 
-	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
-		goto out_free;
 	BUG_ON(!fs_info->quota_root);
 
 	trace_btrfs_qgroup_account_extent(fs_info, bytenr, num_bytes,
@@ -2170,9 +2213,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 			goto out;
 		}
 
-		rcu_read_lock();
 		level_size = fs_info->nodesize;
-		rcu_read_unlock();
 	}
 
 	/*
@@ -2306,7 +2347,20 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
+static bool qgroup_check_limits(const struct btrfs_qgroup *qg, u64 num_bytes)
+{
+	if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_RFER) &&
+	    qg->reserved + (s64)qg->rfer + num_bytes > qg->max_rfer)
+		return false;
+
+	if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_EXCL) &&
+	    qg->reserved + (s64)qg->excl + num_bytes > qg->max_excl)
+		return false;
+
+	return true;
+}
+
+static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes, bool enforce)
 {
 	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *qgroup;
@@ -2347,16 +2401,7 @@ static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
 
 		qg = unode_aux_to_qgroup(unode);
 
-		if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_RFER) &&
-		    qg->reserved + (s64)qg->rfer + num_bytes >
-		    qg->max_rfer) {
-			ret = -EDQUOT;
-			goto out;
-		}
-
-		if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_EXCL) &&
-		    qg->reserved + (s64)qg->excl + num_bytes >
-		    qg->max_excl) {
+		if (enforce && !qgroup_check_limits(qg, num_bytes)) {
 			ret = -EDQUOT;
 			goto out;
 		}
@@ -2424,7 +2469,10 @@ void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
 
 		qg = unode_aux_to_qgroup(unode);
 
-		qg->reserved -= num_bytes;
+		if (WARN_ON(qg->reserved < num_bytes))
+			report_reserved_underflow(fs_info, qg, num_bytes);
+		else
+			qg->reserved -= num_bytes;
 
 		list_for_each_entry(glist, &qg->groups, next_group) {
 			ret = ulist_add(fs_info->qgroup_ulist,
@@ -2439,11 +2487,6 @@ void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
 	spin_unlock(&fs_info->qgroup_lock);
 }
 
-static inline void qgroup_free(struct btrfs_root *root, u64 num_bytes)
-{
-	return btrfs_qgroup_free_refroot(root->fs_info, root->objectid,
-					 num_bytes);
-}
 void assert_qgroups_uptodate(struct btrfs_trans_handle *trans)
 {
 	if (list_empty(&trans->qgroup_ref_list) && !trans->delayed_ref_elem.seq)
@@ -2803,7 +2846,7 @@ int btrfs_qgroup_reserve_data(struct inode *inode, u64 start, u64 len)
 		return 0;
 
 	changeset.bytes_changed = 0;
-	changeset.range_changed = ulist_alloc(GFP_NOFS);
+	ulist_init(&changeset.range_changed);
 	ret = set_record_extent_bits(&BTRFS_I(inode)->io_tree, start,
 			start + len -1, EXTENT_QGROUP_RESERVED, &changeset);
 	trace_btrfs_qgroup_reserve_data(inode, start, len,
@@ -2811,21 +2854,21 @@ int btrfs_qgroup_reserve_data(struct inode *inode, u64 start, u64 len)
 					QGROUP_RESERVE);
 	if (ret < 0)
 		goto cleanup;
-	ret = qgroup_reserve(root, changeset.bytes_changed);
+	ret = qgroup_reserve(root, changeset.bytes_changed, true);
 	if (ret < 0)
 		goto cleanup;
 
-	ulist_free(changeset.range_changed);
+	ulist_release(&changeset.range_changed);
 	return ret;
 
 cleanup:
 	/* cleanup already reserved ranges */
 	ULIST_ITER_INIT(&uiter);
-	while ((unode = ulist_next(changeset.range_changed, &uiter)))
+	while ((unode = ulist_next(&changeset.range_changed, &uiter)))
 		clear_extent_bit(&BTRFS_I(inode)->io_tree, unode->val,
 				 unode->aux, EXTENT_QGROUP_RESERVED, 0, 0, NULL,
 				 GFP_NOFS);
-	ulist_free(changeset.range_changed);
+	ulist_release(&changeset.range_changed);
 	return ret;
 }
 
@@ -2837,23 +2880,22 @@ static int __btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len,
 	int ret;
 
 	changeset.bytes_changed = 0;
-	changeset.range_changed = ulist_alloc(GFP_NOFS);
-	if (!changeset.range_changed)
-		return -ENOMEM;
-
+	ulist_init(&changeset.range_changed);
 	ret = clear_record_extent_bits(&BTRFS_I(inode)->io_tree, start, 
 			start + len -1, EXTENT_QGROUP_RESERVED, &changeset);
 	if (ret < 0)
 		goto out;
 
 	if (free) {
-		qgroup_free(BTRFS_I(inode)->root, changeset.bytes_changed);
+		btrfs_qgroup_free_refroot(BTRFS_I(inode)->root->fs_info,
+				BTRFS_I(inode)->root->objectid,
+				changeset.bytes_changed);
 		trace_op = QGROUP_FREE;
 	}
 	trace_btrfs_qgroup_release_data(inode, start, len,
 					changeset.bytes_changed, trace_op);
 out:
-	ulist_free(changeset.range_changed);
+	ulist_release(&changeset.range_changed);
 	return ret;
 }
 
@@ -2892,7 +2934,8 @@ int btrfs_qgroup_release_data(struct inode *inode, u64 start, u64 len)
 	return __btrfs_qgroup_release_data(inode, start, len, 0);
 }
 
-int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes)
+int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes,
+			      bool enforce)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret;
@@ -2902,7 +2945,7 @@ int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes)
 		return 0;
 
 	BUG_ON(num_bytes != round_down(num_bytes, fs_info->nodesize));
-	ret = qgroup_reserve(root, num_bytes);
+	ret = qgroup_reserve(root, num_bytes, enforce);
 	if (ret < 0)
 		return ret;
 	atomic_add(num_bytes, &root->qgroup_meta_rsv);
@@ -2921,7 +2964,7 @@ void btrfs_qgroup_free_meta_all(struct btrfs_root *root)
 	reserved = atomic_xchg(&root->qgroup_meta_rsv, 0);
 	if (reserved == 0)
 		return;
-	qgroup_free(root, reserved);
+	btrfs_qgroup_free_refroot(fs_info, root->objectid, reserved);
 }
 
 void btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes)
@@ -2935,7 +2978,7 @@ void btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes)
 	BUG_ON(num_bytes != round_down(num_bytes, fs_info->nodesize));
 	WARN_ON(atomic_read(&root->qgroup_meta_rsv) < num_bytes);
 	atomic_sub(num_bytes, &root->qgroup_meta_rsv);
-	qgroup_free(root, num_bytes);
+	btrfs_qgroup_free_refroot(fs_info, root->objectid, num_bytes);
 }
 
 /*
@@ -2950,22 +2993,22 @@ void btrfs_qgroup_check_reserved_leak(struct inode *inode)
 	int ret;
 
 	changeset.bytes_changed = 0;
-	changeset.range_changed = ulist_alloc(GFP_NOFS);
-	if (WARN_ON(!changeset.range_changed))
-		return;
-
+	ulist_init(&changeset.range_changed);
 	ret = clear_record_extent_bits(&BTRFS_I(inode)->io_tree, 0, (u64)-1,
 			EXTENT_QGROUP_RESERVED, &changeset);
 
 	WARN_ON(ret < 0);
 	if (WARN_ON(changeset.bytes_changed)) {
 		ULIST_ITER_INIT(&iter);
-		while ((unode = ulist_next(changeset.range_changed, &iter))) {
+		while ((unode = ulist_next(&changeset.range_changed, &iter))) {
 			btrfs_warn(BTRFS_I(inode)->root->fs_info,
 				"leaking qgroup reserved space, ino: %lu, start: %llu, end: %llu",
 				inode->i_ino, unode->val, unode->aux);
 		}
-		qgroup_free(BTRFS_I(inode)->root, changeset.bytes_changed);
+		btrfs_qgroup_free_refroot(BTRFS_I(inode)->root->fs_info,
+				BTRFS_I(inode)->root->objectid,
+				changeset.bytes_changed);
+
 	}
-	ulist_free(changeset.range_changed);
+	ulist_release(&changeset.range_changed);
 }

fs/btrfs/raid56.c

@@ -677,11 +677,9 @@ static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
 	struct btrfs_raid_bio *freeit = NULL;
 	struct btrfs_raid_bio *cache_drop = NULL;
 	int ret = 0;
-	int walk = 0;
 
 	spin_lock_irqsave(&h->lock, flags);
 	list_for_each_entry(cur, &h->hash_list, hash_list) {
-		walk++;
 		if (cur->bbio->raid_map[0] == rbio->bbio->raid_map[0]) {
 			spin_lock(&cur->bio_list_lock);
 

fs/btrfs/root-tree.c

@@ -74,7 +74,7 @@ static void btrfs_read_root_item(struct extent_buffer *eb, int slot,
  *
  * If we find something return 0, otherwise > 0, < 0 on error.
  */
-int btrfs_find_root(struct btrfs_root *root, struct btrfs_key *search_key,
+int btrfs_find_root(struct btrfs_root *root, const struct btrfs_key *search_key,
 		    struct btrfs_path *path, struct btrfs_root_item *root_item,
 		    struct btrfs_key *root_key)
 {
@@ -207,7 +207,7 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
 }
 
 int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		      struct btrfs_key *key, struct btrfs_root_item *item)
+		      const struct btrfs_key *key, struct btrfs_root_item *item)
 {
 	/*
 	 * Make sure generation v1 and v2 match. See update_root for details.
@@ -337,7 +337,7 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 
 /* drop the root item for 'key' from 'root' */
 int btrfs_del_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		   struct btrfs_key *key)
+		   const struct btrfs_key *key)
 {
 	struct btrfs_path *path;
 	int ret;

include/trace/events/btrfs.h

@@ -184,7 +184,7 @@ DEFINE_EVENT(btrfs__inode, btrfs_inode_evict,
 
 TRACE_EVENT_CONDITION(btrfs_get_extent,
 
-	TP_PROTO(struct btrfs_root *root, struct inode *inode,
+	TP_PROTO(struct btrfs_root *root, struct btrfs_inode *inode,
 		 struct extent_map *map),
 
 	TP_ARGS(root, inode, map),