Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753488AbXE3Mn4 (ORCPT ); Wed, 30 May 2007 08:43:56 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751871AbXE3Mnu (ORCPT ); Wed, 30 May 2007 08:43:50 -0400 Received: from linux.dunaweb.hu ([62.77.196.1]:39301 "EHLO linux.dunaweb.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751513AbXE3Mnt (ORCPT ); Wed, 30 May 2007 08:43:49 -0400 Message-ID: <465D7177.4000804@dunaweb.hu> Date: Wed, 30 May 2007 14:43:35 +0200 From: Zoltan Boszormenyi User-Agent: Thunderbird 1.5.0.10 (X11/20070302) MIME-Version: 1.0 To: Justin Piszcz Cc: linux-kernel , support@mailbox.cps.intel.com Subject: Re: Kernel 2.6.21.3 does not work with 8GB of RAM on Intel 965WH References: <465D43C0.60609@dunaweb.hu> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1869 Lines: 59 Justin Piszcz ?rta: > > > On Wed, 30 May 2007, Zoltan Boszormenyi wrote: > >> Hi, >> >>> Look at how slow the raid benchmarks are in dmesg with 8GB of memory! >>> >>> [ 59.592560] raid6: using algorithm sse2x4 (476 MB/s) >>> [ 59.597558] raid5: using function: generic_sse (56.000 MB/sec) >>> >>> Yikes! >>> >>> With mem=4096M: >>> [ 60.336352] raid6: using algorithm sse2x4 (6804 MB/s) >>> [ 60.341345] raid5: using function: generic_sse (8552.000 MB/sec) >>> >>> What is going on here? >>> >>> Justin. >> >> I saw a similar effect on a P5 board that could >> only cache 64MB of memory a long time ago. >> Putting more memory into it made the machine slower >> and e.g. made the sym2 driver giving up at initialization. >> Maybe it's a modern reincarnation of the problem? >> >> Best regards, >> Zolt?n B?sz?rm?nyi >> > > I see, so Intel 965 motherboards do not work with 8GB of memory? Or > is this a Linux issue? It was certainly a mainboard problem, the same system worked well on another board with a more capable L2 cache controller. Or on PPro/P-II boards. > I have an Intel Case ID open 225446 but so far we have not gotten > anywhere with this problem. Currently I have 8GB in the machine and > boot with mem=4096M. Was there any type of fix? Any recommendations > as to what I can do? Continue to speak with Intel? Or perhaps > someone with more knowledge of the kernel and chime in as to what > exactly is happening here? > > Justin. I think continue to talk Intel and with the kernel guys, I don't have experience with machines with so much memory (yet). Best regards, Zolt?n B?sz?rm?nyi - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/