block: Add flag for telling the IO schedulers NOT to anticipate more IO

By default, CFQ will anticipate more IO from a given io context if the
previously completed IO was sync. This used to be fine, since the only
sync IO was reads and O_DIRECT writes. But with more "normal" sync writes
being used now, we don't want to anticipate for those.

Add a bio/request flag that informs the IO scheduler that this is a sync
request that we should not idle for. Introduce WRITE_ODIRECT specifically
for O_DIRECT writes, and make sure that the other sync writes set this
flag.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 3 insertions, 1 deletions

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 664ebfd092ec..9e809345f71a 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -1992,8 +1992,10 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq)
                 }
                 if (cfq_slice_used(cfqq) || cfq_class_idle(cfqq))
                         cfq_slice_expired(cfqd, 1);
-                else if (sync && RB_EMPTY_ROOT(&cfqq->sort_list))
+                else if (sync && !rq_noidle(rq) &&
+                         RB_EMPTY_ROOT(&cfqq->sort_list)) {
                         cfq_arm_slice_timer(cfqd);
+                }
         }
 
         if (!cfqd->rq_in_driver)