ext4: fix race between unwritten extent conversion and truncate
authorJeff Moyer <jmoyer@redhat.com>
Mon, 20 Feb 2012 22:59:24 +0000 (17:59 -0500)
committerTheodore Ts'o <tytso@mit.edu>
Mon, 20 Feb 2012 22:59:24 +0000 (17:59 -0500)
commit266991b13890049ee1a6bb95b9817f06339ee3d7
tree55c4f0e3c0b978e563ced079002bcf7ea60d9767
parentd4dc462f556afe510d58d3b12b3d66c2cabff539
ext4: fix race between unwritten extent conversion and truncate

The following comment in ext4_end_io_dio caught my attention:

/* XXX: probably should move into the real I/O completion handler */
        inode_dio_done(inode);

The truncate code takes i_mutex, then calls inode_dio_wait.  Because the
ext4 code path above will end up dropping the mutex before it is
reacquired by the worker thread that does the extent conversion, it
seems to me that the truncate can happen out of order.  Jan Kara
mentioned that this might result in error messages in the system logs,
but that should be the extent of the "damage."

The fix is pretty straight-forward: don't call inode_dio_done until the
extent conversion is complete.

Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@vger.kernel.org
fs/ext4/ext4.h
fs/ext4/inode.c
fs/ext4/page-io.c