From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754072AbdL1TDy (ORCPT ); Thu, 28 Dec 2017 14:03:54 -0500 Received: from forward100j.mail.yandex.net ([5.45.198.240]:51281 "EHLO forward100j.mail.yandex.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751229AbdL1TDw (ORCPT ); Thu, 28 Dec 2017 14:03:52 -0500 X-Greylist: delayed 315 seconds by postgrey-1.27 at vger.kernel.org; Thu, 28 Dec 2017 14:03:52 EST Authentication-Results: mxback16j.mail.yandex.net; dkim=pass header.i=@yandex.com.tr From: Ozgur Envelope-From: okaratas@yandex.com.tr To: Pavel Machek , Mike Galbraith Cc: Thomas Gleixner , kernel list , "mingo@redhat.com" , "hpa@zytor.com" , "x86@kernel.org" , "peterz@infradead.org" In-Reply-To: <20171228185354.GA3631@amd> References: <20171228121024.GA18152@amd> <1514467413.6886.35.camel@gmx.de> <20171228185354.GA3631@amd> Subject: Re: niced tasks on SMT system MIME-Version: 1.0 Message-Id: <1150011514487512@web36o.yandex.ru> X-Mailer: Yamail [ http://yandex.ru ] 5.0 Date: Thu, 28 Dec 2017 21:58:32 +0300 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=utf-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 28.12.2017, 21:54, "Pavel Machek" : > Hi! > >>  > Ok, so I'm compiling, and I'd like to run a flight simulator. >>  > >>  > Flightgear normally does 20fps on my system... kinda low but playable. >>  > >>  > I have reniced make -j 5 fo kernel running. Scheduler gives 100% of >>  > one of CPUs to Flightgear (good), but the smt sibling is used by the >>  > compilation, and I'm down to 9 fps. Not good. >>  > >>  > Even with single-threaded make, I have 10-13fps. >>  > >>  > Is there way to learn which CPUs are SMT siblings? >> >>  cat /sys/devices/system/cpu/cpu{N}/topology/thread_siblings > > Thanks for a hint. > > Well, something is definitely wrong there: > > pavel@duo:/data/l/linux-n900$ cat > /sys/devices/system/cpu/cpu0/topology/thread_siblings > 03 Ops, do you fly on N900 via flightgear? :) I have two N900 and one run gentoo. > I believe that means that cpu0 & cpu1 are sharing physical cpu. Yet, > when I run flightgear and "nice while1". flightgear goes to CPU#0 and > while1 to CPU#1. Would not it be nice if while1 went preferably to > CPUs #2 and #3? Ok. after a while while1 moved to cpu#2, good. > >>  > Is there way to disable SMT during runtime? >> >>  You could offline them, but wouldn't it be better to tell each which >>  CPUs they can use, or perhaps partition your box with cpusets? > > Let me try offlining first... and yes, SMT seems to be problem here. > > | CPU0 | CPU1 | CPU2 | CPU3 | | > | flightgear | off | off | off | 21 fps | > | flightgear | while1 | off | off | 17 fps | > | flightgear+while1 | off | off | off | 10 fps | > | flightgear | off | while1 | off | 21 fps | > >>  > Can I do something to improve Flightgear performance and still do >>  > compilation? >> >>  Sure, run everything associated with your game as RT, and everything >>  not game gets the leftover cycles.  If there are none, box will >>   throttle RT to save itself from it's psycho :) driver and you'll know >>  that you need a bigger box. (it's likely your phone) > >>  Oh yeah, echo NO_RT_RUNTIME_SHARE > /sys/kernel/debug/sched_features >>  before you try that, otherwise the throttle won't help. > > Sure, but that is not a problem I have. > > Scheduler already (correctly) decides that the game is important, and > gives game all the time it wants (100% of one CPU). That is not a problem. > > Problem seems to be that it schedules other tasks to SMT sibling, and > that slows down the game quite significantly. > > Best regards, > >                                                                         Pavel > -- > (english) http://www.livejournal.com/~pavelmachek > (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html