Message-ID: <54DE043A.9040605@linux.intel.com>
Date: Fri, 13 Feb 2015 06:03:38 -0800
From: Arjan van de Ven <arjan@linux.intel.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.4.0
MIME-Version: 1.0
To: Peter Zijlstra <peterz@infradead.org>,
        Preeti U Murthy <preeti@linux.vnet.ibm.com>
CC: tglx@linutronix.de, linux-kernel@vger.kernel.org, jacob.jun.pan@intel.com,
        fweisbec@gmail.com, frederic@kernel.org, daniel.lezcano@linaro.org,
        amit.kucheria@linaro.org, edubezval@gmail.com, viresh.kumar@linaro.org,
        rui.zhang@intel.com
Subject: Re: [PATCH V2] idle/intel_powerclamp: Redesign idle injection to
 use bandwidth control mechanism
References: <20150209044852.6231.66456.stgit@preeti.in.ibm.com> <20150213133158.GP2896@worktop.programming.kicks-ass.net>
In-Reply-To: <20150213133158.GP2896@worktop.programming.kicks-ass.net>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 800
Lines: 22


> Also, exposing these and root_task_group is of course vile. Not to
> mention you change the user (cgroup) interface without mention.
>
> In any case, I cannot see how this could ever work. Bandwidth is shared
> across CPUs; nothing will even attempt to get CPUs to idle at the same
> time.

idle injection is only worth it if you can get package C states,
e.g. all cpus in the system are idle at the same time.

(and in powerclamp, the package C state %age is the target of the control
loop, since that's pretty much the amount of power reduced)

>
>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/