From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755744AbaHHIjC (ORCPT ); Fri, 8 Aug 2014 04:39:02 -0400 Received: from mail-we0-f178.google.com ([74.125.82.178]:54762 "EHLO mail-we0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752939AbaHHIi6 (ORCPT ); Fri, 8 Aug 2014 04:38:58 -0400 Date: Fri, 8 Aug 2014 10:38:53 +0200 From: Ingo Molnar To: Rik van Riel Cc: Fengguang Wu , Dave Hansen , Ingo Molnar , LKML , lkp@01.org, Peter Zijlstra Subject: Re: [sched/numa] 096aa33863a: -21.4% hackbench.throughput, -20.2% netperf.Throughput_Mbps Message-ID: <20140808083853.GA8919@gmail.com> References: <20140807105311.GA17655@localhost> <53E3C270.1010302@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <53E3C270.1010302@redhat.com> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Rik van Riel wrote: > On 08/07/2014 06:53 AM, Fengguang Wu wrote: > > Hi Rik, > > > > We noticed the below performance regression in commit > > 096aa33863a5e48de52d2ff30e0801b7487944f4 ("sched/numa: Decay > > ->wakee_flips instead of zeroing") > > > > b1ad065e65f5610 096aa33863a5e48de52d2ff30 testbox/testcase/testparams > > --------------- ------------------------- --------------------------- > > 122361 ± 0% -21.4% 96140 ± 0% lkp-snb01/hackbench/50%-process-pipe > > 122361 ± 0% -21.4% 96140 ± 0% TOTAL hackbench.throughput > > I guess the performance of that benchmark depends on it > "slipping under the wire" after each time the kernel > zeroes out ->wakee_flips. > > Depending on repeatedly pulling the wakee back to the same > node as the waker suggests something else in the kernel may > be pulling the wakee to another place in the system repeatedly, > as well, just at a lower frequency (load balancer?). > > I have also noticed that select_idle_sibling often fails to > find an idle sibling within the LLC domain, even when it > exists. Fixing that bug sometimes results in lower performance. > > It appears that some of the performance results of the scheduler > appear on the code acting in an opposite way to its documented > intention. > > It may be best to revert 096aa33863a for now... Mind sending a revert patch, with an explanation, a Reported-by, etc? Thanks, Ingo