[PATCH 5.5 098/399] block, bfq: do not plug I/O for bfq_queues with no proc refs

From: Greg Kroah-Hartman
Date: Fri Feb 21 2020 - 02:47:43 EST


From: Paolo Valente <paolo.valente@xxxxxxxxxx>

[ Upstream commit f718b093277df582fbf8775548a4f163e664d282 ]

Commit 478de3380c1c ("block, bfq: deschedule empty bfq_queues not
referred by any process") fixed commit 3726112ec731 ("block, bfq:
re-schedule empty queues if they deserve I/O plugging") by
descheduling an empty bfq_queue when it remains with not process
reference. Yet, this still left a case uncovered: an empty bfq_queue
with not process reference that remains in service. This happens for
an in-service sync bfq_queue that is deemed to deserve I/O-dispatch
plugging when it remains empty. Yet no new requests will arrive for
such a bfq_queue if no process sends requests to it any longer. Even
worse, the bfq_queue may happen to be prematurely freed while still in
service (because there may remain no reference to it any longer).

This commit solves this problem by preventing I/O dispatch from being
plugged for the in-service bfq_queue, if the latter has no process
reference (the bfq_queue is then prevented from remaining in service).

Fixes: 3726112ec731 ("block, bfq: re-schedule empty queues if they deserve I/O plugging")
Tested-by: Oleksandr Natalenko <oleksandr@xxxxxxxxxxxxxx>
Reported-by: Patrick Dung <patdung100@xxxxxxxxx>
Tested-by: Patrick Dung <patdung100@xxxxxxxxx>
Signed-off-by: Paolo Valente <paolo.valente@xxxxxxxxxx>
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
block/bfq-iosched.c | 12 ++++++++++++
1 file changed, 12 insertions(+)

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index ad4af4aaf2ced..5c239c540c47a 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -3444,6 +3444,10 @@ static void bfq_dispatch_remove(struct request_queue *q, struct request *rq)
static bool idling_needed_for_service_guarantees(struct bfq_data *bfqd,
struct bfq_queue *bfqq)
{
+ /* No point in idling for bfqq if it won't get requests any longer */
+ if (unlikely(!bfqq_process_refs(bfqq)))
+ return false;
+
return (bfqq->wr_coeff > 1 &&
(bfqd->wr_busy_queues <
bfq_tot_busy_queues(bfqd) ||
@@ -4077,6 +4081,10 @@ static bool idling_boosts_thr_without_issues(struct bfq_data *bfqd,
bfqq_sequential_and_IO_bound,
idling_boosts_thr;

+ /* No point in idling for bfqq if it won't get requests any longer */
+ if (unlikely(!bfqq_process_refs(bfqq)))
+ return false;
+
bfqq_sequential_and_IO_bound = !BFQQ_SEEKY(bfqq) &&
bfq_bfqq_IO_bound(bfqq) && bfq_bfqq_has_short_ttime(bfqq);

@@ -4170,6 +4178,10 @@ static bool bfq_better_to_idle(struct bfq_queue *bfqq)
struct bfq_data *bfqd = bfqq->bfqd;
bool idling_boosts_thr_with_no_issue, idling_needed_for_service_guar;

+ /* No point in idling for bfqq if it won't get requests any longer */
+ if (unlikely(!bfqq_process_refs(bfqq)))
+ return false;
+
if (unlikely(bfqd->strict_guarantees))
return true;

--
2.20.1