From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yan Zheng Subject: [PATCH] Fix race in btrfs_mark_extent_written Date: Wed, 12 Nov 2008 17:04:46 +0800 Message-ID: <491A9C2E.1070603@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 To: linux-btrfs@vger.kernel.org, chris Mason Return-path: List-ID: Hello, When extent needs to be split, btrfs_mark_extent_written truncates the extent at first, then inserts a new extent and increases the reference count. The race happens if someone else deletes the old extent before the new extent is inserted. The fix here is increase the reference count in advance. This race is similar to the race in btrfs_drop_extents that fixed a few days ago. Thank you, Signed-off-by: Yan Zheng --- diff -urp 1/fs/btrfs/file.c 4/fs/btrfs/file.c --- 1/fs/btrfs/file.c 2008-11-12 08:06:09.000000000 +0800 +++ 4/fs/btrfs/file.c 2008-11-12 13:13:03.000000000 +0800 @@ -746,6 +746,7 @@ int btrfs_mark_extent_written(struct btr u64 other_end; u64 split = start; u64 locked_end = end; + u64 orig_parent; int extent_type; int split_end = 1; int ret; @@ -890,6 +891,12 @@ again: } btrfs_mark_buffer_dirty(leaf); + + orig_parent = leaf->start; + ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, + orig_parent, root->root_key.objectid, + trans->transid, inode->i_ino); + BUG_ON(ret); btrfs_release_path(root, path); key.offset = start; @@ -910,10 +917,13 @@ again: btrfs_set_file_extent_encryption(leaf, fi, 0); btrfs_set_file_extent_other_encoding(leaf, fi, 0); - ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, - leaf->start, root->root_key.objectid, - trans->transid, inode->i_ino); - BUG_ON(ret); + if (orig_parent != leaf->start) { + ret = btrfs_update_extent_ref(trans, root, bytenr, + orig_parent, leaf->start, + root->root_key.objectid, + trans->transid, inode->i_ino); + BUG_ON(ret); + } done: btrfs_mark_buffer_dirty(leaf); btrfs_release_path(root, path);