Re: [PATCH v2] sched/fair: Sync task util before slow-path wakeup

From: Brendan Jackman
Date: Wed Sep 20 2017 - 13:17:18 EST


Hi Peter,

Ping.

Log of previous discussion: https://patchwork.kernel.org/patch/9876769/

Cheers,
Brendan

On Tue, Aug 08 2017 at 09:55, Brendan Jackman wrote:
> We use task_util in find_idlest_group via capacity_spare_wake. This
> task_util is updated in wake_cap. However wake_cap is not the only
> reason for ending up in find_idlest_group - we could have been sent
> there by wake_wide. So explicitly sync the task util with prev_cpu
> when we are about to head to find_idlest_group.
>
> We could simply do this at the beginning of
> select_task_rq_fair (i.e. irrespective of whether we're heading to
> select_idle_sibling or find_idlest_group & co), but I didn't want to
> slow down the select_idle_sibling path more than necessary.
>
> Don't do this during fork balancing, we won't need the task_util and
> we'd just clobber the last_update_time, which is supposed to be 0.
>
> Signed-off-by: Brendan Jackman <brendan.jackman@xxxxxxx>
> Cc: Dietmar Eggemann <dietmar.eggemann@xxxxxxx>
> Cc: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> Cc: Josef Bacik <josef@xxxxxxxxxxxxxx>
> Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> Cc: Morten Rasmussen <morten.rasmussen@xxxxxxx>
> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> ---
>
> changes v1 -> v2: Just cosmetic
>
> kernel/sched/fair.c | 11 ++++++++++-
> 1 file changed, 10 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index c95880e216f6..20275178604d 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -5913,8 +5913,17 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f
> new_cpu = cpu;
> }
>
> + if (sd && !(sd_flag & SD_BALANCE_FORK)) {
> + /*
> + * We're going to need the task's util for capacity_spare_wake
> + * in find_idlest_group. Sync it up to prev_cpu's
> + * last_update_time.
> + */
> + sync_entity_load_avg(&p->se);
> + }
> +
> if (!sd) {
> - pick_cpu:
> +pick_cpu:
> if (sd_flag & SD_BALANCE_WAKE) /* XXX always ? */
> new_cpu = select_idle_sibling(p, prev_cpu, new_cpu);