From: Mikulas Patocka When processing barriers, skip the second flush if processing the bio failed with -EOPNOTSUPP. This can happen with discard+barrier requests. If the device doesn't support discard, there would be two useless SYNCHRONIZE CACHE commands. The first dm_flush cannot be so easily optimized out, so we leave it there. Previously, -EOPNOTSUPP could be received in dec_pending only with empty barriers and we ignored that error, assuming the device not supporting cache flushes has cache always consistent. With the addition of discard barriers, this -EOPNOTSUPP can also be generated by discards and we must record it in md->barrier_error for process_barrier. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) Index: linux-2.6.35/drivers/md/dm.c =================================================================== --- linux-2.6.35.orig/drivers/md/dm.c +++ linux-2.6.35/drivers/md/dm.c @@ -639,8 +639,14 @@ static void dec_pending(struct dm_io *io * There can be just one barrier request so we use * a per-device variable for error reporting. * Note that you can't touch the bio after end_io_acct + * + * We ignore -EOPNOTSUPP for empty flush reported by + * underlying devices. We assume that if the device + * doesn't support empty barriers, it doesn't need + * cache flushing commands. */ - if (!md->barrier_error && io_error != -EOPNOTSUPP) + if (!md->barrier_error && + !(bio_empty_barrier(bio) && io_error == -EOPNOTSUPP)) md->barrier_error = io_error; end_io_acct(io); free_io(md, io); @@ -2284,7 +2290,12 @@ static void process_barrier(struct mappe if (!bio_empty_barrier(bio)) { __split_and_process_bio(md, bio); - dm_flush(md); + /* + * If the request isn't supported, don't waste time with + * the second flush. + */ + if (md->barrier_error != -EOPNOTSUPP) + dm_flush(md); } if (md->barrier_error != DM_ENDIO_REQUEUE)