From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751792AbcEII4N (ORCPT ); Mon, 9 May 2016 04:56:13 -0400 Received: from mx2.suse.de ([195.135.220.15]:49219 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750934AbcEII4M (ORCPT ); Mon, 9 May 2016 04:56:12 -0400 Message-ID: <1462784169.3803.146.camel@suse.de> Subject: Re: sched: tweak select_idle_sibling to look for idle threads From: Mike Galbraith To: Peter Zijlstra Cc: Matt Fleming , Chris Mason , Ingo Molnar , linux-kernel@vger.kernel.org Date: Mon, 09 May 2016 10:56:09 +0200 In-Reply-To: <20160509083330.GH3430@twins.programming.kicks-ass.net> References: <1462086753.9717.29.camel@suse.de> <20160502084615.GB3430@twins.programming.kicks-ass.net> <1462200604.3736.42.camel@suse.de> <20160502145817.GW3408@twins.programming.kicks-ass.net> <20160502154725.ckiewczbdubudyc7@floor.masoncoding.com> <20160503143225.GG3448@twins.programming.kicks-ass.net> <20160503151153.wp6jcnjadmw2ypmx@floor.masoncoding.com> <20160504103701.GO3430@twins.programming.kicks-ass.net> <20160505220306.GO2839@codeblueprint.co.uk> <1462560878.5119.9.camel@suse.de> <20160509083330.GH3430@twins.programming.kicks-ass.net> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.16.5 Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2016-05-09 at 10:33 +0200, Peter Zijlstra wrote: > On Fri, May 06, 2016 at 08:54:38PM +0200, Mike Galbraith wrote: > > master > > Throughput 2722.43 MB/sec 4 clients 4 procs max_latency=2.400 ms > > > echo NO_IDLE_SIBLING > /sys/kernel/debug/sched_features > > Throughput 3484.18 MB/sec 4 clients 4 procs max_latency=3.430 ms > > Yeah, I know about that bump, I just haven't managed to find a way to > preserve that and keep all the other benchmarks ticking along :/ Yup, L3 ain't L2. I haven't come up with a good metric either. Poo. Until that comes along, microbenchmarks can bugger off, real boxen don't just play high speed ping-pong with themselves for a living ;-) -Mike