From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Lezcano Subject: Re: [RFC PATCHC 3/3] sched/fair: use the idle state info to choose the idlest cpu Date: Fri, 18 Apr 2014 15:04:46 +0200 Message-ID: <535122EE.7030803@linaro.org> References: <1396009796-31598-1-git-send-email-daniel.lezcano@linaro.org> <1396009796-31598-4-git-send-email-daniel.lezcano@linaro.org> <534FDCDC.6000806@linaro.org> <534FFBB3.8050601@linaro.org> <20140418093845.GD11096@twins.programming.kicks-ass.net> <535116FC.5070700@linaro.org> <20140418125344.GX11182@twins.programming.kicks-ass.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mail-wi0-f175.google.com ([209.85.212.175]:64228 "EHLO mail-wi0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751159AbaDRNEW (ORCPT ); Fri, 18 Apr 2014 09:04:22 -0400 Received: by mail-wi0-f175.google.com with SMTP id cc10so692639wib.2 for ; Fri, 18 Apr 2014 06:04:21 -0700 (PDT) In-Reply-To: <20140418125344.GX11182@twins.programming.kicks-ass.net> Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: Peter Zijlstra Cc: Nicolas Pitre , linux-kernel@vger.kernel.org, mingo@elte.hu, rjw@rjwysocki.net, linux-pm@vger.kernel.org, alex.shi@linaro.org, vincent.guittot@linaro.org, morten.rasmussen@arm.com On 04/18/2014 02:53 PM, Peter Zijlstra wrote: > On Fri, Apr 18, 2014 at 02:13:48PM +0200, Daniel Lezcano wrote: >> On 04/18/2014 11:38 AM, Peter Zijlstra wrote: >>> On Thu, Apr 17, 2014 at 12:21:28PM -0400, Nicolas Pitre wrote: >>>> CPU topology is needed to properly describe scheduling domains. W= hether >>>> we balance across domains or pack using as few domains as possible= is a >>>> separate issue. In other words, you shouldn't have to care in thi= s >>>> patch series. >>>> >>>> And IMHO coupled C-state is a low-level mechanism that should rema= in >>>> private to cpuidle which the scheduler shouldn't be aware of. >>> >>> I'm confused.. why wouldn't you want to expose these? >> >> The couple C-state is used as a mechanism for cpuidle to sync the cp= us when >> entering a specific c-state. This mechanism is usually used to handl= e the >> cluster power down. It is only used for a two drivers (soon three) b= ut it is >> not the only mechanism used for syncing the cpus. There are also the= MCPM >> (tc2), the hand made sync when the hardware allows it (ux500), and a= n >> abstraction from the firmware (mwait), transparent to the kernel. >> >> Taking into account the couple c-state only does not make sense beca= use of >> the other mechanisms above. This is why it should stay inside the cp= uidle >> framework. >> >> The extension of the cpu topology will provide a generic way to desc= ribe and >> abstracting such dependencies. >> >> Does it answer your question ? > > I suppose so; its still a bit like we won't but we will :-) > > So we _will_ actually expose coupled C states through the topology bi= ts, > that's good. Ah, ok. I think I understood where the confusion is coming from. A couple of definitions for the same thing :) 1. Coupled C-states : *mechanism* implemented in the cpuidle framework:= =20 drivers/cpuidle/coupled.c 2. Coupled C-states : *constraint* to reach a cluster power down state,= =20 will be described through the topology and could be implemented by=20 different mechanism (MCPM, handmade sync, cpuidle-coupled-c-state,=20 firmware). We want to expose 2. not 1. to the scheduler. --=20 Linaro.org =E2=94=82 Open source software fo= r ARM SoCs =46ollow Linaro: Facebook | Twitter | Blog