Re: [RFC PATCH v4 1/2] sched/fair: Introduce short duration task check

From: Chen Yu
Date: Fri Jan 06 2023 - 09:31:04 EST


On 2023-01-06 at 12:28:26 +0100, Dietmar Eggemann wrote:
> On 06/01/2023 09:34, Chen Yu wrote:
> > Hi Dietmar,
> > thanks for reviewing the patch!
> > On 2023-01-05 at 12:33:16 +0100, Dietmar Eggemann wrote:
> >> On 16/12/2022 07:11, Chen Yu wrote:
> >>
> >> [...]
> >>
> >>> @@ -5995,6 +6005,18 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
> >>>
> >>> static void set_next_buddy(struct sched_entity *se);
> >>>
> >>> +static inline void dur_avg_update(struct task_struct *p, bool task_sleep)
> >>> +{
> >>> + u64 dur;
> >>> +
> >>> + if (!task_sleep)
> >>> + return;
> >>> +
> >>> + dur = p->se.sum_exec_runtime - p->se.prev_sum_exec_runtime_vol;
> >>> + p->se.prev_sum_exec_runtime_vol = p->se.sum_exec_runtime;
> >>
> >> Shouldn't se->prev_sum_exec_runtime_vol be set in enqueue_task_fair()
> >> and not in dequeue_task_fair()->dur_avg_update()? Otherwise `dur` will
> >> contain sleep time.
> >>
> > After the task p is dequeued, p's sum_exec_runtime will not be increased.
>
> True.
>
> > Unless task p is switched in again, p's sum_exec_runtime will continue to
> > increase. So dur should not include the sleep time, because we substract
>
> Not sure I get this sentence? p's se->sum_exec_runtime will only
> increase if p is current, so running?
>
Yes, it was a typo, should be "will not continue to increase".
> > between the sum_exec_runtime rather than rq->clock_task. Not sure if I understand
> > this correctly?
>
> No, you're right. We're not dealing with time snapshots but rather with
> sum_exec_runtime snapshots. So the value will not change between dequeue
> and the next enqueue.
>
> e ... enqueue_task_fair()
> d ... dequeue_task_fair()
> s ... set_next_entity()
> p ... put_prev_entity()
> u ... update_curr_fair()->update_curr()
>
> p1:
>
> ---|---||--|--|---|--|--||---
> d es u p s u pd
>
> ^ ^
> | |
> (A) (B)
>
> Same se->prev_sum_exec_runtime_vol value in (A) and (B).
>
Yes.
> > My original thought was that, record the average run time of every section:
> > Only consider that task voluntarily relinquishes the CPU.
> > For example, suppose on CPU1, task p1 and p2 run alternatively:
> >
> > --------------------> time
> >
> > | p1 runs 1ms | p2 preempt p1 | p1 switch in, runs 0.5ms and blocks |
> > ^ ^ ^
> > |_____________| |_____________________________________|
> > ^
> > |
> > p1 dequeued
> >
> > p1's duration in one section is (1 + 0.5)ms. Because if p2 does not
> > preempt p1, p1 can run 1.5ms. This reflects the nature of a task,
> > how long it wishes to run at most.
> >
> >> Like we do for se->prev_sum_exec_runtime in set_next_entity() but for
> >> one `set_next_entity()-put_prev_entity()` run section.
> >>
> >> AFAICS, you want to measure the exec_runtime sum over all run sections
> >> between enqueue and dequeue.
> > Yes, we tried to record the 'decayed' average exec_runtime for each section.
> > Say, task p runs for a ms , then p is dequeued and blocks for b ms, and then
> > runs for c ms, its average duration is 0.875 * a + 0.125 * c , which is
> > what update_avg() does.
>
> OK.
>
I'll add more descriptions in next version to avoid confusing.

thanks,
Chenyu