Date:   Mon, 20 Feb 2023 11:13:18 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     Vincent Guittot <vincent.guittot@linaro.org>
Cc:     Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Qais Yousef <qyousef@layalina.io>,
        Kajetan Puchalski <kajetan.puchalski@arm.com>,
        Jian-Min Liu <jian-min.liu@mediatek.com>,
        Ingo Molnar <mingo@kernel.org>,
        Morten Rasmussen <morten.rasmussen@arm.com>,
        Vincent Donnefort <vdonnefort@google.com>,
        Quentin Perret <qperret@google.com>,
        Patrick Bellasi <patrick.bellasi@matbug.net>,
        Abhijeet Dharmapurikar <adharmap@quicinc.com>,
        Qais Yousef <qais.yousef@arm.com>,
        linux-kernel@vger.kernel.org,
        Jonathan JMChen <jonathan.jmchen@mediatek.com>
Subject: Re: [RFC PATCH 0/1] sched/pelt: Change PELT halflife at runtime
Message-ID: <Y/NHvvcgOov6DoLc@hirez.programming.kicks-ass.net>
References: <0f82011994be68502fd9833e499749866539c3df.camel@mediatek.com>
 <YzVpqweg21yIn30A@hirez.programming.kicks-ass.net>
 <YzV9Gejo/+DL3UjK@e126311.manchester.arm.com>
 <YzV/yT6OYMgaq0kD@hirez.programming.kicks-ass.net>
 <YzWuq5ShtJC6KWqe@e126311.manchester.arm.com>
 <Y2kLA8x40IiBEPYg@hirez.programming.kicks-ass.net>
 <20221108194843.i4qckcu7zwqstyis@airbuntu>
 <Y2vMBWpPlIArwnI7@hirez.programming.kicks-ass.net>
 <424e2c81-987d-f10e-106d-8b4c611768bc@arm.com>
 <CAKfTPtD0ZOndFef3-JxBn3G9tcX=cZEObjHZ0iqiVTJz7+QrmQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAKfTPtD0ZOndFef3-JxBn3G9tcX=cZEObjHZ0iqiVTJz7+QrmQ@mail.gmail.com>
Precedence: bulk

On Thu, Feb 09, 2023 at 05:16:46PM +0100, Vincent Guittot wrote:

> > The results is very similar to PELT halflife reduction. The advantage is
> > that 'util_est_faster' is only activated selectively when the runtime of
> > the current task in its current activation is long enough to create this
> > CPU util boost.
> 
> IIUC how util_est_faster works, it removes the waiting time when
> sharing cpu time with other tasks. So as long as there is no (runnable
> but not running time), the result is the same as current util_est.

Uh.. it's double the speed, no? Even if there is no contention, the
fake/in-situ pelt sum runs at double time and thus will ramp up faster
than normal.

> util_est_faster makes a difference only when the task alternates
> between runnable and running slices.

UTIL_EST was supposed to help mitigate some of that, but yes. Also note
that _FASTER sorta sucks here because it starts from 0 every time, if it
were to start from the state saved by util_est_dequeue(), it would ramp
up faster still.

Patch has a comment along those lines I think.

> Have you considered using runnable_avg metrics in the increase of cpu
> freq ? This takes into the runnable slice and not only the running
> time and increase faster than util_avg when tasks compete for the same
> CPU

Interesting! Indeed, that's boosting the DVFS for contention. And as
deggeman's reply shows, it seems to work well.

I wonder if that one place where it regresses is exactly the case
without contention.