ext4: BUG_ON assertion repeated for inode1, not done for inode2
authorDavid Moore <dmoorefo@gmail.com>
Mon, 8 Jun 2015 15:59:12 +0000 (11:59 -0400)
committerTheodore Ts'o <tytso@mit.edu>
Mon, 8 Jun 2015 15:59:12 +0000 (11:59 -0400)
During a source code review of fs/ext4/extents.c I noted identical
consecutive lines. An assertion is repeated for inode1 and never done
for inode2. This is not in keeping with the rest of the code in the
ext4_swap_extents function and appears to be a bug.

Assert that the inode2 mutex is not locked.

Signed-off-by: David Moore <dmoorefo@gmail.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
fs/ext4/extents.c

index e003a1e81dc351c76465908bec21702ecc6d4b91..f38a6d6c4a9341747771dde97a5dac99a2f0e556 100644 (file)
@@ -5542,7 +5542,7 @@ ext4_swap_extents(handle_t *handle, struct inode *inode1,
        BUG_ON(!rwsem_is_locked(&EXT4_I(inode1)->i_data_sem));
        BUG_ON(!rwsem_is_locked(&EXT4_I(inode2)->i_data_sem));
        BUG_ON(!mutex_is_locked(&inode1->i_mutex));
-       BUG_ON(!mutex_is_locked(&inode1->i_mutex));
+       BUG_ON(!mutex_is_locked(&inode2->i_mutex));
 
        *erp = ext4_es_remove_extent(inode1, lblk1, count);
        if (unlikely(*erp))