diff options
author | Paolo Valente <paolo.valente@linaro.org> | 2018-09-14 17:23:09 +0300 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2019-11-20 20:00:00 +0300 |
commit | f927911d4abad751f877a7edc7261276c266db65 (patch) | |
tree | 798e05f0131e65fb518563a5550f266865faafb9 /block | |
parent | e269eb6f1f7b66df81d063ff91a57a0b4abf111e (diff) | |
download | linux-f927911d4abad751f877a7edc7261276c266db65.tar.xz |
blok, bfq: do not plug I/O if all queues are weight-raised
[ Upstream commit c8765de0adfcaaf4ffb2d951e07444f00ffa9453 ]
To reduce latency for interactive and soft real-time applications, bfq
privileges the bfq_queues containing the I/O of these
applications. These privileged queues, referred-to as weight-raised
queues, get a much higher share of the device throughput
w.r.t. non-privileged queues. To preserve this higher share, the I/O
of any non-weight-raised queue must be plugged whenever a sync
weight-raised queue, while being served, remains temporarily empty. To
attain this goal, bfq simply plugs any I/O (from any queue), if a sync
weight-raised queue remains empty while in service.
Unfortunately, this plugging typically lowers throughput with random
I/O, on devices with internal queueing (because it reduces the filling
level of the internal queues of the device).
This commit addresses this issue by restricting the cases where
plugging is performed: if a sync weight-raised queue remains empty
while in service, then I/O plugging is performed only if some of the
active bfq_queues are *not* weight-raised (which is actually the only
circumstance where plugging is needed to preserve the higher share of
the throughput of weight-raised queues). This restriction proved able
to boost throughput in really many use cases needing only maximum
throughput.
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Diffstat (limited to 'block')
-rw-r--r-- | block/bfq-iosched.c | 10 |
1 files changed, 8 insertions, 2 deletions
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index e65b0da1007b..93863c6173e6 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -3314,7 +3314,12 @@ static bool bfq_bfqq_may_idle(struct bfq_queue *bfqq) * whether bfqq is being weight-raised, because * bfq_symmetric_scenario() does not take into account also * weight-raised queues (see comments on - * bfq_weights_tree_add()). + * bfq_weights_tree_add()). In particular, if bfqq is being + * weight-raised, it is important to idle only if there are + * other, non-weight-raised queues that may steal throughput + * to bfqq. Actually, we should be even more precise, and + * differentiate between interactive weight raising and + * soft real-time weight raising. * * As a side note, it is worth considering that the above * device-idling countermeasures may however fail in the @@ -3326,7 +3331,8 @@ static bool bfq_bfqq_may_idle(struct bfq_queue *bfqq) * to let requests be served in the desired order until all * the requests already queued in the device have been served. */ - asymmetric_scenario = bfqq->wr_coeff > 1 || + asymmetric_scenario = (bfqq->wr_coeff > 1 && + bfqd->wr_busy_queues < bfqd->busy_queues) || !bfq_symmetric_scenario(bfqd); /* |