Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:8 as permitted sender) client-ip=2620:137:e000::3:8;
Message-ID: <377e0f71-80f9-4d83-a09d-e86bd61ce9ae@bytedance.com>
Date:   Thu, 16 Nov 2023 19:16:29 +0800
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH 1/1] sched/fair: Update min_vruntime in more relaxed way
Content-Language: en-US
To:     Yiwei Lin <s921975628@gmail.com>, mingo@redhat.com,
        peterz@infradead.org
Cc:     vincent.guittot@linaro.org, dietmar.eggemann@arm.com,
        linux-kernel@vger.kernel.org
References: <20231116105425.84773-1-s921975628@gmail.com>
 <20231116105425.84773-2-s921975628@gmail.com>
From:   Abel Wu <wuyun.abel@bytedance.com>
In-Reply-To: <20231116105425.84773-2-s921975628@gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 11/16/23 6:54 PM, Yiwei Lin Wrote:
> As EEVDF adopts lag-based solution which is irrespective of
> min_vruntime like CFS before, min_vruntime is only used as
> an offset to avoid overflow on evaluation of avg_vruntime now.
> Rely on the fact we will always update_curr() before change
> to cfs_rq, it seems to make sense if we just
> update_min_vruntime() with update_curr() to reduce the cost.
> 
> Signed-off-by: Yiwei Lin <s921975628@gmail.com>
> ---
>   kernel/sched/fair.c | 20 +-------------------
>   1 file changed, 1 insertion(+), 19 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 07f555857..5c40adfae 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -3815,17 +3815,8 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se,
>   	enqueue_load_avg(cfs_rq, se);
>   	if (se->on_rq) {
>   		update_load_add(&cfs_rq->load, se->load.weight);
> -		if (!curr) {
> -			/*
> -			 * The entity's vruntime has been adjusted, so let's check
> -			 * whether the rq-wide min_vruntime needs updated too. Since
> -			 * the calculations above require stable min_vruntime rather
> -			 * than up-to-date one, we do the update at the end of the
> -			 * reweight process.
> -			 */
> +		if (!curr)
>   			__enqueue_entity(cfs_rq, se);
> -			update_min_vruntime(cfs_rq);
> -		}
>   	}
>   }
>   
> @@ -5347,15 +5338,6 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
>   
>   	update_cfs_group(se);
>   
> -	/*
> -	 * Now advance min_vruntime if @se was the entity holding it back,
> -	 * except when: DEQUEUE_SAVE && !DEQUEUE_MOVE, in this case we'll be
> -	 * put back on, and if we advance min_vruntime, we'll be placed back
> -	 * further than we started -- ie. we'll be penalized.
> -	 */
> -	if ((flags & (DEQUEUE_SAVE | DEQUEUE_MOVE)) != DEQUEUE_SAVE)
> -		update_min_vruntime(cfs_rq);
> -
>   	if (cfs_rq->nr_running == 0)
>   		update_idle_cfs_rq_clock_pelt(cfs_rq);
>   }

For now, core pick of core scheduling relies on min_vruntime to be fresh,
so please just fix commit eab03c23c2a1 to preserve its original behavior
by moving update_min_vruntime() to proper position. And behavior change
can be posted separated.

BTW it seems unnecessary to include a cover-letter for a single patch.

Thanks,
	Abel