From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 8 Nov 2001 22:09:45 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 8 Nov 2001 22:09:35 -0500 Received: from ns1.hicom.net ([208.245.180.8]:38411 "EHLO ns1.hicom.net") by vger.kernel.org with ESMTP id ; Thu, 8 Nov 2001 22:09:22 -0500 Message-ID: <3BEB4929.4CE5E403@pobox.com> Date: Thu, 08 Nov 2001 22:10:33 -0500 From: Tom Zych Organization: Detachment 2702 X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.2.18pre21 i586) X-Accept-Language: en MIME-Version: 1.0 To: linux-kernel@vger.kernel.org CC: lung@theuw.net, crazed_cowboy@stones.com, kernel@corrosive.freeserve.co.uk, freedman@ccmr.cornell.edu Subject: ECS K7S5A bus problem may be causing file corruption Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org (Apologies for munged address, I don't have a good spam solution yet. No need to CC me.) There appears to be a marginal bus problem with some ECS K7S5A boards, which may be contributing to some of the file corruption problems people have reported. Try running Memtest86 ( http://www.memtest86.com/ ) with the CPU (and possibly FSB) speed set to 100, then 133. If 100 is ok and 133 isn't, check out http://65.66.90.193/K7S5A_Corruption_Fix_V131.pdf (warning: pdf) for details. My own machine exhibits completely reproducible errors using either or both DIMMS when the CPU/FSB speed is 133/133, and no errors at 100/100 or 100/133. Memtest86 is run from a boot floppy and doesn't use the hard drives, so this is unrelated to the kernel / IDE. A couple of people have had trouble under Linux but not Windows, so this may not be the only issue. It should be eliminated on a given machine before testing kernel patches, though. Hope this helps, -- Tom Zych freethinker_DEATHTOSPAM@pobox.com