Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 27 Jul 2021 15:29:44 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Mel Gorman <mgorman@techsingularity.net>
Cc:     Christian Borntraeger <borntraeger@de.ibm.com>, bristot@redhat.com,
        bsegall@google.com, dietmar.eggemann@arm.com, joshdon@google.com,
        juri.lelli@redhat.com, kvm@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-s390@vger.kernel.org,
        linux@rasmusvillemoes.dk, mgorman@suse.de, mingo@kernel.org,
        rostedt@goodmis.org, valentin.schneider@arm.com,
        vincent.guittot@linaro.org
Subject: Re: [PATCH 1/1] sched/fair: improve yield_to vs fairness
Message-ID: <YQAKSO759lvZurgw@hirez.programming.kicks-ass.net>
References: <YIlXQ43b6+7sUl+f@hirez.programming.kicks-ass.net>
 <20210707123402.13999-1-borntraeger@de.ibm.com>
 <20210707123402.13999-2-borntraeger@de.ibm.com>
 <20210723093523.GX3809@techsingularity.net>
 <ddb81bc9-1429-c392-adac-736e23977c84@de.ibm.com>
 <20210723162137.GY3809@techsingularity.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210723162137.GY3809@techsingularity.net>
Precedence: bulk

On Fri, Jul 23, 2021 at 05:21:37PM +0100, Mel Gorman wrote:

> I'm still not a fan because vruntime gets distorted. From the docs
> 
>    Small detail: on "ideal" hardware, at any time all tasks would have the same
>    p->se.vruntime value --- i.e., tasks would execute simultaneously and no task
>    would ever get "out of balance" from the "ideal" share of CPU time
> 
> If yield_to impacts this "ideal share" then it could have other
> consequences.

Note that we already violate this ideal both subtly and explicitly.

For an example of the latter consider pretty much the entirety of
place_entity() with GENTLE_FAIR_SLEEPERS being the most egregious
example.

That said; adding to vruntime will only penalize the task itself, while
subtracting from vruntime will penalize everyone else. And in that sense
addition to vruntime is a safe option.

I've not fully considered the case at hand; just wanted to give some
context.