lisa/android_kernel_xiaomi_sm8350
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

block, bfq: push up injection only after setting service time

Browse Source 
If equal to 0, the injection limit for a bfq_queue is pushed to 1
after a first sample of the total service time of the I/O requests of
the queue is computed (to allow injection to start). Yet, because of a
mistake in the branch that performs this action, the push may happen
also in some other case. This commit fixes this issue.

Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Paolo Valente 2019-08-22 17:20:37 +02:00 committed by Jens Axboe
parent 17c3d26602
commit 58494c980f
1 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
block/bfq-iosched.c
View File

@ -5823,12 +5823,14 @@ static void bfq_update_inject_limit(struct bfq_data *bfqd,
*/
if ((bfqq->last_serv_time_ns == 0 && bfqd->rq_in_driver == 1) ||
tot_time_ns < bfqq->last_serv_time_ns) {
if (bfqq->last_serv_time_ns == 0) {
/*
* Now we certainly have a base value: make sure we
* start trying injection.
*/
bfqq->inject_limit = max_t(unsigned int, 1, old_limit);
}
bfqq->last_serv_time_ns = tot_time_ns;
/*
* Now we certainly have a base value: make sure we
* start trying injection.
*/
bfqq->inject_limit = max_t(unsigned int, 1, old_limit);
} else if (!bfqd->rqs_injected && bfqd->rq_in_driver == 1)
/*
* No I/O injected and no request still in service in