From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Lezcano Subject: Re: [PATCH V3 2/6] sched: idle: cpuidle: Check the latency req before idle Date: Wed, 12 Nov 2014 14:53:10 +0100 Message-ID: <54636646.6080709@linaro.org> References: <1415370687-18688-1-git-send-email-daniel.lezcano@linaro.org> <1415370687-18688-3-git-send-email-daniel.lezcano@linaro.org> <20141110124111.GN3337@twins.programming.kicks-ass.net> <5460D5EF.2000805@linaro.org> <20141110152803.GX10501@worktop.programming.kicks-ass.net> <5460E0A5.9040508@linaro.org> <20141110161530.GB10501@worktop.programming.kicks-ass.net> <5460F386.1050109@linaro.org> <20141110194820.GD10501@worktop.programming.kicks-ass.net> <54613A5F.7060107@linaro.org> <20141111102058.GI10501@worktop.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-f182.google.com ([209.85.212.182]:63557 "EHLO mail-wi0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751797AbaKLNxQ (ORCPT ); Wed, 12 Nov 2014 08:53:16 -0500 Received: by mail-wi0-f182.google.com with SMTP id d1so4907981wiv.15 for ; Wed, 12 Nov 2014 05:53:14 -0800 (PST) In-Reply-To: <20141111102058.GI10501@worktop.programming.kicks-ass.net> Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: Peter Zijlstra Cc: rjw@rjwysocki.net, preeti@linux.vnet.ibm.com, nicolas.pitre@linaro.org, linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, linaro-kernel@lists.linaro.org, patches@linaro.org, lenb@kernel.org On 11/11/2014 11:20 AM, Peter Zijlstra wrote: > On Mon, Nov 10, 2014 at 11:21:19PM +0100, Daniel Lezcano wrote: >> On 11/10/2014 08:48 PM, Peter Zijlstra wrote: >>> On Mon, Nov 10, 2014 at 06:19:02PM +0100, Daniel Lezcano wrote: >>>>> I really don't get why the governors should know about this thoug= h, its >>>>> just another state, they should iterate all states and pick the b= est, >>>>> given the power usage this state should really never be eligible = unless >>>>> we're QoS forced or whatnot. >>>> >>>> The governors just don't use the poll state at all, except for a c= ouple of >>>> cases in menu.c defined above in the previous email. What is the r= ational of >>>> adding a state in the cpuidle driver and do everything we can to a= void using >>>> it ? From my POV, the poll state is a special state, we should rem= ove from >>>> the driver's idle states like the arch_cpu_idle() is a specific id= le state >>>> only used in idle.c (but which may overlap with an idle state in d= ifferent >>>> archs eg. cpu_do_idle() and the 0th idle state). >>> >>> So I disagree, I think poll-idle is an idle mode just like all the >>> others. It should be an available state to the governor and it shou= ld >>> treat it like any other. >> >> The governors are just ignoring it, except for a small timer optimiz= ation in >> menu.c (and I am still not convinced it is worth to have it). I don'= t see >> the point to add a state we don't want to use. > > The ignoring it is _wrong_. Make that go away and you'll get rid of m= ost > of the STATE_START crap. > > The governors are the place where we combine the QoS constraints with > idle predictors and pick an idle state, polling is a valid state to > pick, and given QoS constraints it might be the only state to pick. Well, I understand your point of view but I still disagree. IMO, the=20 poll loop shouldn't be considered as a valid idle state for different=20 reasons: 0. thermal issue if wrongly selected from any of the governor 1. a polling loop does not have a valid time measurements even if the=20 TIME_VALID flag has been added 2. entering the idle governors is not free, especially the menu=20 governor, which is contradictory with zero latency req 3. what is the meaning of having a zero latency (target + exit) idle=20 state ? We know it will always succeed if the other fails 4. IIUC, you are suggesting to add the poll loop for all the cpuidle=20 drivers. This is a *lot* of changes, I am not afraid about the work to=20 do but there is a significant code impact and the entire behavior of th= e=20 cpuidle framework for all the arch will be changed. So given the points above, why not have one poll function, generic, and= =20 if we fail to find an idle state or if the req is zero, then fallback t= o=20 the poll loop ? >> Eg. on my server it was called 2 times over 1313856 times. >> >>> I don't tihnk the whole ARCH_HAS_CPU_RELAX thing makes any kind of >>> sense, _every_ arch has some definition of it, the generic polling = loop >>> is always a valid idle implementation. >>> >>> What we can do is always populate the idle state table with it befo= re >>> calling the regular drivers. >> >> I am not sure to understand. You want to add the poll idle loop in a= ll the >> drivers ? >> >> What about "safe_halt()" ? (arch_cpu_idle() for x86). It is also an = idle >> state. Why not add it in the idle state table also ? > > Because the latter is actually arch specific, whereas the idle pollin= g > thing is not. > You can _always_ poll idle, its generic, its valid, and > its guaranteed the most responsive method. I agree with this point but this kind of loop is hardware optimized for= =20 x86. On the other arch, where today this loop is never used, if we=20 change the behavior of the cpuidle framework and introduces this loop,=20 it may be selected and stay on this state for a long time (resulting=20 from a bad prediction), I am afraid that can lead to some thermal issue= s. > The arch drivers get to add arch specific idle states; if a x86 cpuid= le > driver wants to add hlt they can. --=20 Linaro.org =E2=94=82 Open source software fo= r ARM SoCs =46ollow Linaro: Facebook | Twitter | Blog