From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759588AbYARK4S (ORCPT ); Fri, 18 Jan 2008 05:56:18 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753652AbYARK4E (ORCPT ); Fri, 18 Jan 2008 05:56:04 -0500 Received: from mx2.mail.elte.hu ([157.181.151.9]:42138 "EHLO mx2.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752946AbYARK4C (ORCPT ); Fri, 18 Jan 2008 05:56:02 -0500 Date: Fri, 18 Jan 2008 11:55:45 +0100 From: Ingo Molnar To: Colin Fowler Cc: linux-kernel@vger.kernel.org, Peter Zijlstra Subject: Re: Performance loss 2.6.22->22.6.23->2.6.24-rc7 on CPU intensive benchmark on 8 Core Xeon Message-ID: <20080118105545.GC12228@elte.hu> References: <20080114185520.GA26540@elte.hu> <20080115220641.GC2665@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.17 (2007-11-01) X-ELTE-VirusStatus: clean X-ELTE-SpamScore: -1.1 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-1.1 required=5.9 tests=BAYES_05 autolearn=no SpamAssassin version=3.2.3 -1.1 BAYES_05 BODY: Bayesian spam probability is 1 to 5% [score: 0.0187] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Colin Fowler wrote: > > there are a handful of 'scheduler feature bits' in > > /proc/sys/kernel/sched_features: > > > > enum { > > SCHED_FEAT_NEW_FAIR_SLEEPERS = 1, > > SCHED_FEAT_WAKEUP_PREEMPT = 2, > > SCHED_FEAT_START_DEBIT = 4, > > SCHED_FEAT_TREE_AVG = 8, > > SCHED_FEAT_APPROX_AVG = 16, > > }; > > > > Toggling SCHED_FEAT_NEW_FAIR_SLEEPERS to 0 or > SCHED_FEAT_WAKEUP_PREEMPT to 0 gives me results more inline with my > 2.6.22 results. Toggling them both to 0 gives me slightly better > results than 2.6.22! ok, but it would be nice to avoid having to turn these off. Could you try whether tuning the /proc/sys/kernel/*granularity* values (in particular wakeup_granularity) has any positive effect on your workload? also, could you run your workload as SCHED_BATCH [via schedtool -B ], does that improve the results as well on a default-tuned kernel? Ingo