Merge tag 'for-5.2-rc1-tag' of git://git.kernel.org/pub/scm/linux/ker…

…nel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "Notable highlights:

   - fixes for some long-standing bugs in fsync that were quite hard to
     catch but now finaly fixed

   - some fixups to error handling paths that did not properly clean up
     (locking, memory)

   - fix to space reservation for inheriting properties"

* tag 'for-5.2-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  Btrfs: tree-checker: detect file extent items with overlapping ranges
  Btrfs: fix race between ranged fsync and writeback of adjacent ranges
  Btrfs: avoid fallback to transaction commit during fsync of files with holes
  btrfs: extent-tree: Fix a bug that btrfs is unable to add pinned bytes
  btrfs: sysfs: don't leak memory when failing add fsid
  btrfs: sysfs: Fix error path kobject memory leak
  Btrfs: do not abort transaction at btrfs_update_root() after failure to COW path
  btrfs: use the existing reserved items for our first prop for inheritance
  btrfs: don't double unlock on error in btrfs_punch_hole
  btrfs: Check the compression level before getting a workspace

fs/btrfs/compression.c

@@ -1008,6 +1008,7 @@ int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping,
 	struct list_head *workspace;
 	int ret;
 
+	level = btrfs_compress_op[type]->set_level(level);
 	workspace = get_workspace(type, level);
 	ret = btrfs_compress_op[type]->compress_pages(workspace, mapping,
 						      start, pages,

fs/btrfs/extent-tree.c

@@ -757,12 +757,14 @@ static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
 }
 
 static void add_pinned_bytes(struct btrfs_fs_info *fs_info,
-			     struct btrfs_ref *ref)
+			     struct btrfs_ref *ref, int sign)
 {
 	struct btrfs_space_info *space_info;
-	s64 num_bytes = -ref->len;
+	s64 num_bytes;
 	u64 flags;
 
+	ASSERT(sign == 1 || sign == -1);
+	num_bytes = sign * ref->len;
 	if (ref->type == BTRFS_REF_METADATA) {
 		if (ref->tree_ref.root == BTRFS_CHUNK_TREE_OBJECTID)
 			flags = BTRFS_BLOCK_GROUP_SYSTEM;
@@ -2063,7 +2065,7 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 	btrfs_ref_tree_mod(fs_info, generic_ref);
 
 	if (ret == 0 && old_ref_mod < 0 && new_ref_mod >= 0)
-		add_pinned_bytes(fs_info, generic_ref);
+		add_pinned_bytes(fs_info, generic_ref, -1);
 
 	return ret;
 }
@@ -3882,8 +3884,7 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags)
 				    info->space_info_kobj, "%s",
 				    alloc_name(space_info->flags));
 	if (ret) {
-		percpu_counter_destroy(&space_info->total_bytes_pinned);
-		kfree(space_info);
+		kobject_put(&space_info->kobj);
 		return ret;
 	}
 
@@ -7190,7 +7191,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 	}
 out:
 	if (pin)
-		add_pinned_bytes(fs_info, &generic_ref);
+		add_pinned_bytes(fs_info, &generic_ref, 1);
 
 	if (last_ref) {
 		/*
@@ -7238,7 +7239,7 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref)
 		btrfs_ref_tree_mod(fs_info, ref);
 
 	if (ret == 0 && old_ref_mod >= 0 && new_ref_mod < 0)
-		add_pinned_bytes(fs_info, ref);
+		add_pinned_bytes(fs_info, ref, 1);
 
 	return ret;
 }

fs/btrfs/file.c

@@ -2067,6 +2067,18 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	int ret = 0, err;
 	u64 len;
 
+	/*
+	 * If the inode needs a full sync, make sure we use a full range to
+	 * avoid log tree corruption, due to hole detection racing with ordered
+	 * extent completion for adjacent ranges, and assertion failures during
+	 * hole detection.
+	 */
+	if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+		     &BTRFS_I(inode)->runtime_flags)) {
+		start = 0;
+		end = LLONG_MAX;
+	}
+
 	/*
 	 * The range length can be represented by u64, we have to do the typecasts
 	 * to avoid signed overflow if it's [0, LLONG_MAX] eg. from fsync()
@@ -2554,10 +2566,8 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 
 	ret = btrfs_punch_hole_lock_range(inode, lockstart, lockend,
 					  &cached_state);
-	if (ret) {
-		inode_unlock(inode);
+	if (ret)
 		goto out_only_mutex;
-	}
 
 	path = btrfs_alloc_path();
 	if (!path) {

fs/btrfs/props.c

@@ -332,6 +332,7 @@ static int inherit_props(struct btrfs_trans_handle *trans,
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret;
 	int i;
+	bool need_reserve = false;
 
 	if (!test_bit(BTRFS_INODE_HAS_PROPS,
 		      &BTRFS_I(parent)->runtime_flags))
@@ -357,11 +358,20 @@ static int inherit_props(struct btrfs_trans_handle *trans,
 		if (ret)
 			continue;
 
-		num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
-		ret = btrfs_block_rsv_add(root, trans->block_rsv,
-					  num_bytes, BTRFS_RESERVE_NO_FLUSH);
-		if (ret)
-			return ret;
+		/*
+		 * Currently callers should be reserving 1 item for properties,
+		 * since we only have 1 property that we currently support.  If
+		 * we add more in the future we need to try and reserve more
+		 * space for them.  But we should also revisit how we do space
+		 * reservations if we do add more properties in the future.
+		 */
+		if (need_reserve) {
+			num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
+			ret = btrfs_block_rsv_add(root, trans->block_rsv,
+					num_bytes, BTRFS_RESERVE_NO_FLUSH);
+			if (ret)
+				return ret;
+		}
 
 		ret = btrfs_setxattr(trans, inode, h->xattr_name, value,
 				     strlen(value), 0);
@@ -375,9 +385,13 @@ static int inherit_props(struct btrfs_trans_handle *trans,
 					&BTRFS_I(inode)->runtime_flags);
 		}
 
-		btrfs_block_rsv_release(fs_info, trans->block_rsv, num_bytes);
-		if (ret)
-			return ret;
+		if (need_reserve) {
+			btrfs_block_rsv_release(fs_info, trans->block_rsv,
+					num_bytes);
+			if (ret)
+				return ret;
+		}
+		need_reserve = true;
 	}
 
 	return 0;

fs/btrfs/root-tree.c

@@ -132,10 +132,8 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
 		return -ENOMEM;
 
 	ret = btrfs_search_slot(trans, root, key, path, 0, 1);
-	if (ret < 0) {
-		btrfs_abort_transaction(trans, ret);
+	if (ret < 0)
 		goto out;
-	}
 
 	if (ret > 0) {
 		btrfs_crit(fs_info,

fs/btrfs/sysfs.c

@@ -825,7 +825,12 @@ int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs,
 	fs_devs->fsid_kobj.kset = btrfs_kset;
 	error = kobject_init_and_add(&fs_devs->fsid_kobj,
 				&btrfs_ktype, parent, "%pU", fs_devs->fsid);
-	return error;
+	if (error) {
+		kobject_put(&fs_devs->fsid_kobj);
+		return error;
+	}
+
+	return 0;
 }
 
 int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info)

fs/btrfs/tree-checker.c

@@ -107,8 +107,26 @@ static void file_extent_err(const struct extent_buffer *eb, int slot,
 	(!IS_ALIGNED(btrfs_file_extent_##name((leaf), (fi)), (alignment)));   \
 })
 
+static u64 file_extent_end(struct extent_buffer *leaf,
+			   struct btrfs_key *key,
+			   struct btrfs_file_extent_item *extent)
+{
+	u64 end;
+	u64 len;
+
+	if (btrfs_file_extent_type(leaf, extent) == BTRFS_FILE_EXTENT_INLINE) {
+		len = btrfs_file_extent_ram_bytes(leaf, extent);
+		end = ALIGN(key->offset + len, leaf->fs_info->sectorsize);
+	} else {
+		len = btrfs_file_extent_num_bytes(leaf, extent);
+		end = key->offset + len;
+	}
+	return end;
+}
+
 static int check_extent_data_item(struct extent_buffer *leaf,
-				  struct btrfs_key *key, int slot)
+				  struct btrfs_key *key, int slot,
+				  struct btrfs_key *prev_key)
 {
 	struct btrfs_fs_info *fs_info = leaf->fs_info;
 	struct btrfs_file_extent_item *fi;
@@ -188,6 +206,28 @@ static int check_extent_data_item(struct extent_buffer *leaf,
 	    CHECK_FE_ALIGNED(leaf, slot, fi, offset, sectorsize) ||
 	    CHECK_FE_ALIGNED(leaf, slot, fi, num_bytes, sectorsize))
 		return -EUCLEAN;
+
+	/*
+	 * Check that no two consecutive file extent items, in the same leaf,
+	 * present ranges that overlap each other.
+	 */
+	if (slot > 0 &&
+	    prev_key->objectid == key->objectid &&
+	    prev_key->type == BTRFS_EXTENT_DATA_KEY) {
+		struct btrfs_file_extent_item *prev_fi;
+		u64 prev_end;
+
+		prev_fi = btrfs_item_ptr(leaf, slot - 1,
+					 struct btrfs_file_extent_item);
+		prev_end = file_extent_end(leaf, prev_key, prev_fi);
+		if (prev_end > key->offset) {
+			file_extent_err(leaf, slot - 1,
+"file extent end range (%llu) goes beyond start offset (%llu) of the next file extent",
+					prev_end, key->offset);
+			return -EUCLEAN;
+		}
+	}
+
 	return 0;
 }
 
@@ -774,14 +814,15 @@ static int check_inode_item(struct extent_buffer *leaf,
  * Common point to switch the item-specific validation.
  */
 static int check_leaf_item(struct extent_buffer *leaf,
-			   struct btrfs_key *key, int slot)
+			   struct btrfs_key *key, int slot,
+			   struct btrfs_key *prev_key)
 {
 	int ret = 0;
 	struct btrfs_chunk *chunk;
 
 	switch (key->type) {
 	case BTRFS_EXTENT_DATA_KEY:
-		ret = check_extent_data_item(leaf, key, slot);
+		ret = check_extent_data_item(leaf, key, slot, prev_key);
 		break;
 	case BTRFS_EXTENT_CSUM_KEY:
 		ret = check_csum_item(leaf, key, slot);
@@ -928,7 +969,7 @@ static int check_leaf(struct extent_buffer *leaf, bool check_item_data)
 			 * Check if the item size and content meet other
 			 * criteria
 			 */
-			ret = check_leaf_item(leaf, &key, slot);
+			ret = check_leaf_item(leaf, &key, slot, &prev_key);
 			if (ret < 0)
 				return ret;
 		}

fs/btrfs/tree-log.c

@@ -4182,6 +4182,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 							       *last_extent, 0,
 							       0, len, 0, len,
 							       0, 0, 0);
+				*last_extent += len;
 			}
 		}
 	}