Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 8 May 2018 14:17:35 +0200
From:   Juri Lelli <juri.lelli@redhat.com>
To:     Viresh Kumar <viresh.kumar@linaro.org>
Cc:     Dietmar Eggemann <dietmar.eggemann@arm.com>,
        linux-kernel@vger.kernel.org,
        Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>, linux-pm@vger.kernel.org,
        Pavan Kondeti <pkondeti@codeaurora.org>,
        "Rafael J . Wysocki" <rafael.j.wysocki@intel.com>,
        Joel Fernandes <joelaf@google.com>,
        Patrick Bellasi <patrick.bellasi@arm.com>,
        Quentin Perret <quentin.perret@arm.com>
Subject: Re: [PATCH] Revert "cpufreq: schedutil: Don't restrict kthread to
 related_cpus unnecessarily"
Message-ID: <20180508121735.GD19168@localhost.localdomain>
References: <20180508073340.13114-1-dietmar.eggemann@arm.com>
 <20180508082242.bre6sjfvefhz6xc3@vireshk-i7>
 <8cf21b1a-ca6e-fed7-43c5-94c66ff5986b@arm.com>
 <20180508094526.ajyjrwytguhv4xpe@vireshk-i7>
 <7922d081-0bfb-0e55-7caf-ec9fb5d7bab0@arm.com>
 <20180508105332.rxphnwo4byr42gmy@vireshk-i7>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180508105332.rxphnwo4byr42gmy@vireshk-i7>
User-Agent: Mutt/1.9.2 (2017-12-15)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 08/05/18 16:23, Viresh Kumar wrote:
> On 08-05-18, 12:36, Dietmar Eggemann wrote:
> > That's true but where is the benefit by doing so? (Multiple) per-cluster or
> > per-cpu frequency domains, why should the sugov kthread run on a foreign
> > cpu?
> 
> I am not sure I know the answer, but I have a question which you can
> answer :)
> 
> Is it possible for a CPU (which already has high priority deadline
> activity going on) to be busy enough to be not able to run the
> schedutil kthread for sometime ? That would be the only case I believe
> where it would be better to let some other CPU go and change the
> frequency for this one as we better run faster while we have the high
> load going on.

Shouldn't happen. This kthreads are "special" and will preempt any other
DL task (stop class win of course).