[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20210219194509.00005884.zbestahu@gmail.com>
Date: Fri, 19 Feb 2021 19:45:09 +0800
From: Yue Hu <zbestahu@...il.com>
To: Viresh Kumar <viresh.kumar@...aro.org>
Cc: rjw@...ysocki.net, mingo@...hat.com, peterz@...radead.org,
juri.lelli@...hat.com, vincent.guittot@...aro.org,
dietmar.eggemann@....com, rostedt@...dmis.org, bsegall@...gle.com,
linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org,
huyue2@...ong.com, zbestahu@....com
Subject: Re: [PATCH] cpufreq: schedutil: Don't consider freq reduction to
busy CPU if need_freq_update is set
On Fri, 19 Feb 2021 15:05:51 +0530
Viresh Kumar <viresh.kumar@...aro.org> wrote:
> On 19-02-21, 16:20, Yue Hu wrote:
> > However, we will skip the update if need_freq_update is not set.
>
> Not really, we will update freq periodically nevertheless, around
> every 10ms or something..
>
> > And do the update if need_freq_update is set.
>
> Yeah, that breaks the periodic cycle to attend to some urgent request.
>
> > Note that there are unnecessary fast switch check and spin
> > lock/unlock operations in freq skip path.
>
> Maybe, I am not sure. We are all up for optimizations if there are
> any.
We will set next_f to next_freq(previous freq) if next_f is
reduced for busy CPU. Then the next sugov_update_next_freq() will check
if next_freq matches next_f if need_freq_update is not set.
Obviously, we will do nothing for the case. And The related check to
fast_switch_enabled and raw_spin_{lock,unlock} operations are
unnecessary.
>
> > If we consider unnecessary behaviors above, then we should return
> > right away rather than continue to execute following code.
>
> As I said earlier, we may end up updating the frequency even if
> need_freq_update is unset.
>
Powered by blists - more mailing lists