linux-kernel - Re: [PATCH v2 3/3] cpufreq: schedutil: map raw required frequency to driver frequency

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAJZ5v0gxfRBeQTyHACVtRpon50N_t3qytrFPT4ga2zwPhVYYdw@mail.gmail.com>
Date:	Sun, 29 May 2016 02:40:14 +0200
From:	"Rafael J. Wysocki" <rafael@...nel.org>
To:	Viresh Kumar <viresh.kumar@...aro.org>
Cc:	Steve Muckle <steve.muckle@...aro.org>,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...hat.com>,
	"Rafael J. Wysocki" <rafael@...nel.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	"linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
	Vincent Guittot <vincent.guittot@...aro.org>,
	Morten Rasmussen <morten.rasmussen@....com>,
	Dietmar Eggemann <dietmar.eggemann@....com>,
	Juri Lelli <Juri.Lelli@....com>,
	Patrick Bellasi <patrick.bellasi@....com>,
	Michael Turquette <mturquette@...libre.com>
Subject: Re: [PATCH v2 3/3] cpufreq: schedutil: map raw required frequency to
 driver frequency

On Thu, May 26, 2016 at 9:16 AM, Viresh Kumar <viresh.kumar@...aro.org> wrote:
> On 25-05-16, 19:53, Steve Muckle wrote:
>> The slow-path frequency transition path is relatively expensive as it
>> requires waking up a thread to do work. Should support be added for
>> remote CPU cpufreq updates that is also expensive since it requires an
>> IPI. These activities should be avoided if they are not necessary.
>>
>> To that end, calculate the actual driver-supported frequency required by
>> the new utilization value in schedutil by using the recently added
>> cpufreq_driver_resolve_freq callback. If it is the same as the
>> previously requested driver frequency then there is no need to continue
>> with the update assuming the cpu frequency limits have not changed. This
>> will have additional benefits should the semantics of the rate limit be
>> changed to apply solely to frequency transitions rather than to
>> frequency calculations in schedutil.

[cut]

> I also have a doubt (I am quite sure Rafael will have a reason for
> that, which I am failing to understand now), on why we are doing
> next_freq == UINT_MAX in sugov_should_update_freq().
>
> I understand that because the limits might have changed,
> need_freq_update would have been set to true. We should evaluate
> next-freq again without worrying about the load or the time since last
> evaluation.

This is in response to the "limits" event (or to the ->limits call
after my recent patches).  That event basically means "something has
changed, so if you have cached anything, invalidate it" to the
governor.  Accordingly, it invalidates next_freq, because that's a
cached value.

> But what will happen by forcefully calling the cpufreq routines to
> change the frequency, if next_freq hasn't changed even after limits
> updates?

I can't really parse the above question, so I'm not going to try to
answer it. :-)

> Wouldn't that call always return early because the new freq
> and the current freq are going to be same ?
>
> @Rafael: Sorry for asking this so late :(

It is not too late.  If there's a problem somewhere, it needs to be
fixed, but at this point I have no idea what you are asking about.