[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <4ab8d0aa-7c03-4160-a2bf-39351c394a6e@bytedance.com>
Date: Tue, 5 Mar 2024 11:15:00 +0800
From: Abel Wu <wuyun.abel@...edance.com>
To: Tianchen Ding <dtcccc@...ux.alibaba.com>, linux-kernel@...r.kernel.org
Cc: Ingo Molnar <mingo@...hat.com>, Peter Zijlstra <peterz@...radead.org>,
Juri Lelli <juri.lelli@...hat.com>,
Vincent Guittot <vincent.guittot@...aro.org>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Steven Rostedt <rostedt@...dmis.org>, Ben Segall <bsegall@...gle.com>,
Mel Gorman <mgorman@...e.de>, Daniel Bristot de Oliveira
<bristot@...hat.com>, Valentin Schneider <vschneid@...hat.com>
Subject: Re: Re: [PATCH 1/2] sched/eevdf: Always update V if se->on_rq when
reweighting
On 3/5/24 10:13 AM, Tianchen Ding Wrote:
> On 2024/3/4 20:06, Abel Wu wrote:
>> On 3/4/24 11:00 AM, Tianchen Ding Wrote:
>>> reweight_eevdf() needs the latest V to do accurate calculation for new
>>> ve and vd. So update V unconditionally when se is runnable.
>>
>> As this should come along with commit eab03c23c2a1, I think it would
>> be better add a Fixes tag?
>>
>
> I'm not sure. In fact avg_vruntime_sub() in reweight_entity() was first introduced by af4cf40470c2, and at that time the algorithm about reweight was not completed...
>
It is because before commit eab03c23c2a1, reweight didn't rely on V.
Powered by blists - more mailing lists