From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S268142AbUH2AZm (ORCPT ); Sat, 28 Aug 2004 20:25:42 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S268145AbUH2AZm (ORCPT ); Sat, 28 Aug 2004 20:25:42 -0400 Received: from mustang.oldcity.dca.net ([216.158.38.3]:18120 "HELO mustang.oldcity.dca.net") by vger.kernel.org with SMTP id S268142AbUH2AZk (ORCPT ); Sat, 28 Aug 2004 20:25:40 -0400 Subject: Re: Scheduler fairness problem on 2.6 series (Attn: Nick Piggin and others) From: Lee Revell To: Peter Williams Cc: spaminos-ker@yahoo.com, linux-kernel In-Reply-To: <41312174.40707@bigpond.net.au> References: <20040828015937.50607.qmail@web13902.mail.yahoo.com> <41312174.40707@bigpond.net.au> Content-Type: text/plain Message-Id: <1093739136.7078.1.camel@krustophenia.net> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.6 Date: Sat, 28 Aug 2004 20:25:37 -0400 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sat, 2004-08-28 at 20:21, Peter Williams wrote: > spaminos-ker@yahoo.com wrote: > > --- Peter Williams wrote: > > ----------------- > > => started at: kernel_fpu_begin+0x21/0x60 > > => ended at: _mmx_memcpy+0x131/0x180 > > =======> > > 00000001 0.000ms (+0.000ms): kernel_fpu_begin (_mmx_memcpy) > > 00000001 0.730ms (+0.730ms): sub_preempt_count (_mmx_memcpy) > > 00000001 0.730ms (+0.000ms): _mmx_memcpy (check_preempt_timing) > > 00000001 0.730ms (+0.000ms): kernel_fpu_begin (_mmx_memcpy) > > > > As far as I can see sub_preempt_count() is part of the latency measuring > component of the voluntary preempt patch so, like you, I'm not sure > whether this report makes any sense. Is this an SMP machine? There were problems with that version of the voluntary preemption patches on SMP. The latest version, Q3, should fix these. Lee