Jan Kara jack@suse.cz writes:
Commit d279c80e0bac ("iomap: inline iomap_dio_bio_opflags()") has broken the logic in iomap_dio_bio_iter() in a way that when the device does support FUA (or has no writeback cache) and the direct IO happens to freshly allocated or unwritten extents, we will *not* issue fsync after completing direct IO O_SYNC / O_DSYNC write because the IOMAP_DIO_WRITE_THROUGH flag stays mistakenly set. Fix the problem by clearing IOMAP_DIO_WRITE_THROUGH whenever we do not perform FUA write as it was originally intended.
CC: John Garry john.g.garry@oracle.com CC: "Ritesh Harjani (IBM)" ritesh.list@gmail.com Fixes: d279c80e0bac ("iomap: inline iomap_dio_bio_opflags()") CC: stable@vger.kernel.org Signed-off-by: Jan Kara jack@suse.cz
Nice catch. It's been there since v6.15 I guess. Looks like it didn't get caught in xfstests either then.
We have generic/737, but looks like it only covers O_SYNC dio writes. Let me enhance that to cover O_DYSNC dio writes too.
Looks good to me. Please feel free to add:
Reviewed-by: Ritesh Harjani (IBM) ritesh.list@gmail.com
-ritesh