diff options
| author |   Jan Kara <jack@suse.cz> | 2014-10-30 10:53:17 -0400 |
|---|---|---|
| committer |   Theodore Ts'o <tytso@mit.edu> | 2014-10-30 10:53:17 -0400 |
| commit | ae9e9c6aeea6f91ccb4fb369d7dd8f1a8b5f6a58 (patch) | |
| tree | 7eba907e57a324fb69f42336fc4083421a6761a9 /fs | |
| parent | ext4: bail early when clearing inode journal flag fails (diff) | |
| download | linux-dev-ae9e9c6aeea6f91ccb4fb369d7dd8f1a8b5f6a58.tar.xz linux-dev-ae9e9c6aeea6f91ccb4fb369d7dd8f1a8b5f6a58.zip | |
ext4: make ext4_ext_convert_to_initialized() return proper number of blocks
ext4_ext_convert_to_initialized() can return more blocks than are
actually allocated from map->m_lblk in case where initial part of the
on-disk extent is zeroed out. Luckily this doesn't have serious
consequences because the caller currently uses the return value
only to unmap metadata buffers. Anyway this is a data
corruption/exposure problem waiting to happen so fix it.
Coverity-id: 1226848
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Diffstat (limited to 'fs')
| -rw-r--r-- | fs/ext4/extents.c | 9 |
1 files changed, 4 insertions, 5 deletions
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 37043d0b2be8..0b16fb4c06d3 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3603,11 +3603,10 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, } } - allocated = ext4_split_extent(handle, inode, ppath, - &split_map, split_flag, flags); - if (allocated < 0) - err = allocated; - + err = ext4_split_extent(handle, inode, ppath, &split_map, split_flag, + flags); + if (err > 0) + err = 0; out: /* If we have gotten a failure, don't zero out status tree */ if (!err) |