LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: mingo@redhat.com, valentin.schneider@arm.com,
	dietmar.eggemann@arm.com, vincent.guittot@linaro.org,
	gaku.inami.xh@renesas.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCHv2 2/7] sched/fair: Add group_misfit_task load-balance type
Date: Wed, 11 Apr 2018 12:45:20 +0200	[thread overview]
Message-ID: <20180411104520.GG4043@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <1521125224-15434-3-git-send-email-morten.rasmussen@arm.com>

On Thu, Mar 15, 2018 at 02:46:59PM +0000, Morten Rasmussen wrote:

> +static inline unsigned long task_util(struct task_struct *p);
> +static inline int task_fits_capacity(struct task_struct *p, long capacity)
> +{
> +	return capacity * 1024 > task_util(p) * capacity_margin;
> +}
> +
> +static inline void update_misfit_status(struct task_struct *p, struct rq *rq)
> +{
> +	if (!static_branch_unlikely(&sched_asym_cpucapacity))
> +		return;
> +
> +	if (!p) {
> +		rq->misfit_task_load = 0;
> +		return;
> +	}
> +
> +	if (task_fits_capacity(p, capacity_of(cpu_of(rq)))) {
> +		rq->misfit_task_load = 0;
> +		return;
> +	}
> +
> +	rq->misfit_task_load = task_h_load(p);
> +}

So RT/IRQ pressure can also cause misfit..


> @@ -7972,6 +8005,10 @@ static inline void update_sg_lb_stats(struct lb_env *env,
>  		 */
>  		if (!nr_running && idle_cpu(i))
>  			sgs->idle_cpus++;
> +
> +		if (env->sd->flags & SD_ASYM_CPUCAPACITY &&
> +		    !sgs->group_misfit_task_load && rq->misfit_task_load)
> +			sgs->group_misfit_task_load = rq->misfit_task_load;
>  	}

Should we not look for the biggest misfit instead of the first?

>  
>  	/* Adjust by relative CPU capacity of the group */

  reply	other threads:[~2018-04-11 10:45 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-15 14:46 [PATCHv2 0/7] sched/fair: Migrate 'misfit' tasks on asymmetric capacity systems Morten Rasmussen
2018-03-15 14:46 ` [PATCHv2 1/7] sched: Add static_key for asymmetric cpu capacity optimizations Morten Rasmussen
2018-04-27 14:04   ` [PATCHv2,1/7] " Jiada Wang
2018-04-29 19:09     ` Valentin Schneider
2018-03-15 14:46 ` [PATCHv2 2/7] sched/fair: Add group_misfit_task load-balance type Morten Rasmussen
2018-04-11 10:45   ` Peter Zijlstra [this message]
2018-03-15 14:47 ` [PATCHv2 3/7] sched/fair: Consider misfit tasks when load-balancing Morten Rasmussen
2018-04-11 10:46   ` Peter Zijlstra
2018-04-11 10:53   ` Peter Zijlstra
2018-03-15 14:47 ` [PATCHv2 4/7] sched/fair: Kick nohz balance if rq->misfit_task Morten Rasmussen
2018-03-15 14:47 ` [PATCHv2 5/7] sched: Change root_domain->overload type to int Morten Rasmussen
2018-03-15 14:47 ` [PATCHv2 6/7] sched: Wrap rq->rd->overload accesses with READ/WRITE_ONCE Morten Rasmussen
2018-03-15 14:47 ` [PATCHv2 7/7] sched/fair: Set sd->overload when misfit Morten Rasmussen
2018-03-20  5:30 ` [PATCHv2 0/7] sched/fair: Migrate 'misfit' tasks on asymmetric capacity systems Gaku Inami
2018-03-20  9:17   ` Morten Rasmussen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180411104520.GG4043@hirez.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=dietmar.eggemann@arm.com \
    --cc=gaku.inami.xh@renesas.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=morten.rasmussen@arm.com \
    --cc=valentin.schneider@arm.com \
    --cc=vincent.guittot@linaro.org \
    --subject='Re: [PATCHv2 2/7] sched/fair: Add group_misfit_task load-balance type' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).