From owner-freebsd-stable@FreeBSD.ORG Fri Nov 30 14:21:23 2007 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id DB2A316A417; Fri, 30 Nov 2007 14:21:23 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost1.sentex.ca (smarthost1.sentex.ca [64.7.153.18]) by mx1.freebsd.org (Postfix) with ESMTP id 960F813C474; Fri, 30 Nov 2007 14:21:22 +0000 (UTC) (envelope-from mike@sentex.net) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smarthost1.sentex.ca (8.13.8/8.13.8) with ESMTP id lAUELLdx030922; Fri, 30 Nov 2007 09:21:21 -0500 (EST) (envelope-from mike@sentex.net) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.13.8/8.13.3) with ESMTP id lAUELKx0097603 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Fri, 30 Nov 2007 09:21:21 -0500 (EST) (envelope-from mike@sentex.net) Message-Id: <200711301421.lAUELKx0097603@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Fri, 30 Nov 2007 09:19:21 -0500 To: Pete French , kris@freebsd.org From: Mike Tancsa In-Reply-To: References: <474FD17D.7080209@FreeBSD.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Cc: freebsd-stable@freebsd.org Subject: Re: Also seeing 2 x quad-core system slower that 2 x dual core X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 30 Nov 2007 14:21:24 -0000 At 07:10 AM 11/30/2007, Pete French wrote: > > Check dmesg for the APIC numbers corresponding to the CPUs you want to > > disable and add the corresponding entries to /boot/loader.conf, e.g.: > >O.K., I did that, got it running on 4 CPU's only, and the problem >is still there - so it's not the number of CPU's after all. Which >is good in a way in that it is not completely defeating common sense, >but means I shall go and look elsewhere for the problem. This sort of comparison is difficult as the disk layout can be different from machine to machine. Perhaps if you dedicate a drive (e.g ad8) to the test and do something like newfs -U -O2 /dev/ad8 mount /dev/ad8 /mnt sh /tmp/script-to-create-gig-of-files-on-mnt shutdown -r now mount /dev/ad8 /mnt time rm -R /mnt/testfiles and then repeat varying the number of CPUs, You should be able to get a rough sense of the impact of more CPUs if there is one. ---Mike