[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <0880d81b-9c80-423c-95e6-2fe973e7d739@arm.com>
Date: Mon, 15 Jan 2024 21:05:22 +0100
From: Dietmar Eggemann <dietmar.eggemann@....com>
To: Vincent Guittot <vincent.guittot@...aro.org>
Cc: Qais Yousef <qyousef@...alina.io>, Wyes Karny <wkarny@...il.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Ingo Molnar <mingo@...nel.org>, linux-kernel@...r.kernel.org,
Peter Zijlstra <peterz@...radead.org>, Thomas Gleixner <tglx@...utronix.de>,
Juri Lelli <juri.lelli@...hat.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: [GIT PULL] Scheduler changes for v6.8
On 15/01/2024 16:26, Vincent Guittot wrote:
> On Mon, 15 Jan 2024 at 15:03, Dietmar Eggemann <dietmar.eggemann@....com> wrote:
>>
>> On 15/01/2024 14:26, Vincent Guittot wrote:
>>> On Mon, 15 Jan 2024 at 13:09, Qais Yousef <qyousef@...alina.io> wrote:
>>>>
>>>> On 01/15/24 09:21, Vincent Guittot wrote:
[...]
> Now, util can't be higher than max to handle clamping use cases
> whereas it could be the case before. The jump to next OPP was
> previously done with util being higher than max and it's now done with
> freq being higher than policy->cur
Ah, OK:
util = map_util_perf(util) <- (1) util *= 1.25
in the old code was doing this. And you do this now via frequency for
the !freq_inv case. I just saw that you already mention this in the
patch header of:
https://lkml.kernel.org/r/20240114183600.135316-1-vincent.guittot@linaro.org
Powered by blists - more mailing lists