From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753357AbZIHMmP (ORCPT ); Tue, 8 Sep 2009 08:42:15 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752905AbZIHMmO (ORCPT ); Tue, 8 Sep 2009 08:42:14 -0400 Received: from mail.gmx.net ([213.165.64.20]:52584 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1751075AbZIHMmO (ORCPT ); Tue, 8 Sep 2009 08:42:14 -0400 X-Authenticated: #14349625 X-Provags-ID: V01U2FsdGVkX18/Sk4Lciw/shSLlpqr2LZzs8s3YXeVEZSSBxrYTz fSHPBSnKnRgdNT Subject: Re: [tip:sched/core] sched: Ensure that a child can't gain time over it's parent after fork() From: Mike Galbraith To: Jens Axboe Cc: Ingo Molnar , mingo@redhat.com, hpa@zytor.com, linux-kernel@vger.kernel.org, a.p.zijlstra@chello.nl, tglx@linutronix.de, linux-tip-commits@vger.kernel.org In-Reply-To: <20090908115300.GW18599@kernel.dk> References: <20090908114927.GA1649@elte.hu> <20090908115300.GW18599@kernel.dk> Content-Type: text/plain Date: Tue, 08 Sep 2009 14:42:13 +0200 Message-Id: <1252413733.11445.0.camel@marge.simson.net> Mime-Version: 1.0 X-Mailer: Evolution 2.24.1.1 Content-Transfer-Encoding: 7bit X-Y-GMX-Trusted: 0 X-FuHaFi: 0.57 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2009-09-08 at 13:53 +0200, Jens Axboe wrote: > On Tue, Sep 08 2009, Ingo Molnar wrote: > > > > * tip-bot for Mike Galbraith wrote: > > > > > Commit-ID: b5d9d734a53e0204aab0089079cbde2a1285a38f > > > Gitweb: http://git.kernel.org/tip/b5d9d734a53e0204aab0089079cbde2a1285a38f > > > Author: Mike Galbraith > > > AuthorDate: Tue, 8 Sep 2009 11:12:28 +0200 > > > Committer: Ingo Molnar > > > CommitDate: Tue, 8 Sep 2009 13:15:34 +0200 > > > > > > sched: Ensure that a child can't gain time over it's parent after fork() > > > > > This improves latencies of make -j kernel build workloads. > > > > > > Reported-by: Jens Axboe > > > > Jens, mind trying this patch (or latest tip:master) with your make > > -j4 workload, does it improve in any way? > > Already booted it, makes no real difference... I'm trying with NO_HZ > disabled now to see if that changes the picture. Oh well, BTTODB. Thanks for trying it out. -Mike