Btrfs: fix ordered extent leak when failing to start a transaction

We cannot just return error before freeing ordered extent and releasing reserved
space when we fail to start a transacion.

Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>

1 files changed, 5 insertions, 2 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 38cda78de5e4..6ba80b902877 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1882,8 +1882,11 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
                                 trans = btrfs_join_transaction_nolock(root);
                         else
                                 trans = btrfs_join_transaction(root);
-                        if (IS_ERR(trans))
-                                return PTR_ERR(trans);
+                        if (IS_ERR(trans)) {
+                                ret = PTR_ERR(trans);
+                                trans = NULL;
+                                goto out;
+                        }
                         trans->block_rsv = &root->fs_info->delalloc_block_rsv;
                         ret = btrfs_update_inode_fallback(trans, root, inode);
                         if (ret) /* -ENOMEM or corruption */