mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Dietmar Eggemann <dietmar.eggemann@arm.com>
To: peterz@infradead.org, mingo@redhat.com,
	vincent.guittot@linaro.org, vdonnefort@gmail.com
Cc: linux-kernel@vger.kernel.org, morten.rasmussen@arm.com,
	chris.redpath@arm.com, qperret@google.com, tao.zhou@linux.dev
Subject: Re: [PATCH v8 6/7] sched/fair: Remove task_util from effective utilization in feec()
Date: Thu, 5 May 2022 20:41:45 +0200	[thread overview]
Message-ID: <cd29f36a-9063-dae6-4526-85494c221bd2@arm.com> (raw)
In-Reply-To: <20220429141148.181816-7-vincent.donnefort@arm.com>

+ vdonnefort@gmail.com
- vincent.donnefort@arm.com

On 29/04/2022 16:11, Vincent Donnefort wrote:

[...]

> Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>

Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>

[...]

> -static long
> -compute_energy(struct task_struct *p, int dst_cpu, struct cpumask *cpus,
> -	       struct perf_domain *pd)
> +static inline void eenv_task_busy_time(struct energy_env *eenv,
> +				       struct task_struct *p, int prev_cpu)
>  {
> -	unsigned long max_util = 0, sum_util = 0, cpu_cap;
> +	unsigned long max_cap = arch_scale_cpu_capacity(prev_cpu);
> +	unsigned long irq = cpu_util_irq(cpu_rq(prev_cpu));
> +
> +	if (unlikely(irq >= max_cap)) {
> +		eenv->task_busy_time = max_cap;
> +		return;
> +	}
> +
> +	eenv->task_busy_time =
> +		scale_irq_capacity(task_util_est(p), irq, max_cap);

Nit-pick:

Maybe a little bit lighter:

 static inline void eenv_task_busy_time(struct energy_env *eenv,
                                       struct task_struct *p, int prev_cpu)
 {
-       unsigned long max_cap = arch_scale_cpu_capacity(prev_cpu);
+       unsigned long busy_time, max_cap = arch_scale_cpu_capacity(prev_cpu);
        unsigned long irq = cpu_util_irq(cpu_rq(prev_cpu));
 
-       if (unlikely(irq >= max_cap)) {
-               eenv->task_busy_time = max_cap;
-               return;
-       }
+       if (unlikely(irq >= max_cap))
+               busy_time = max_cap;
+       else
+               busy_time = scale_irq_capacity(task_util_est(p), irq, max_cap);
 
-       eenv->task_busy_time =
-               scale_irq_capacity(task_util_est(p), irq, max_cap);
+       eenv->task_busy_time = busy_time;
 }

[...]

> +static inline unsigned long
> +compute_energy(struct energy_env *eenv, struct perf_domain *pd,
> +	       struct cpumask *pd_cpus, struct task_struct *p, int dst_cpu)
> +{
> +	unsigned long max_util = eenv_pd_max_util(eenv, pd_cpus, p, dst_cpu);
> +	unsigned long busy_time = eenv->pd_busy_time;
> +
> +	if (dst_cpu >= 0)
> +		busy_time = min(eenv->pd_cap,
> +				eenv->pd_busy_time + eenv->task_busy_time);

Nit-pick:

        if (dst_cpu >= 0)
-               busy_time = min(eenv->pd_cap,
-                               eenv->pd_busy_time + eenv->task_busy_time);
+               busy_time = min(eenv->pd_cap, busy_time + eenv->task_busy_time);

[..]

> @@ -6905,12 +6976,14 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
>  			continue;
>  
>  		/* Compute the 'base' energy of the pd, without @p */
> -		base_energy_pd = compute_energy(p, -1, cpus, pd);
> +		eenv_pd_busy_time(&eenv, cpus, p);

Nit-pick:

Move comment /* Compute the 'base' energy of the pd, without @p */ here
since eenv->pd_busy_time is also used to calculate prev_delta and cur_delta. 


> +		base_energy_pd = compute_energy(&eenv, pd, cpus, p, -1);>  		base_energy += base_energy_pd;

[...]

  reply	other threads:[~2022-05-05 18:46 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-29 14:11 [PATCH v8 0/7] feec() energy margin removal Vincent Donnefort
2022-04-29 14:11 ` [PATCH v8 1/7] sched/fair: Provide u64 read for 32-bits arch helper Vincent Donnefort
2022-05-05 18:25   ` Dietmar Eggemann
2022-05-05 18:44     ` Dietmar Eggemann
2022-04-29 14:11 ` [PATCH v8 2/7] sched/fair: Decay task PELT values during wakeup migration Vincent Donnefort
2022-04-29 17:09   ` Tao Zhou
2022-05-06 13:58     ` Vincent Guittot
2022-05-05 18:41   ` Dietmar Eggemann
2022-04-29 14:11 ` [PATCH v8 3/7] sched, drivers: Remove max param from effective_cpu_util()/sched_cpu_util() Vincent Donnefort
2022-04-29 14:11 ` [PATCH v8 4/7] sched/fair: Rename select_idle_mask to select_rq_mask Vincent Donnefort
2022-04-29 14:11 ` [PATCH v8 5/7] sched/fair: Use the same cpumask per-PD throughout find_energy_efficient_cpu() Vincent Donnefort
2022-04-29 14:11 ` [PATCH v8 6/7] sched/fair: Remove task_util from effective utilization in feec() Vincent Donnefort
2022-05-05 18:41   ` Dietmar Eggemann [this message]
2022-04-29 14:11 ` [PATCH v8 7/7] sched/fair: Remove the energy margin " Vincent Donnefort
2022-05-05 18:42   ` Dietmar Eggemann

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=cd29f36a-9063-dae6-4526-85494c221bd2@arm.com \
    --to=dietmar.eggemann@arm.com \
    --cc=chris.redpath@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=morten.rasmussen@arm.com \
    --cc=peterz@infradead.org \
    --cc=qperret@google.com \
    --cc=tao.zhou@linux.dev \
    --cc=vdonnefort@gmail.com \
    --cc=vincent.guittot@linaro.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®