From owner-freebsd-current@FreeBSD.ORG Tue Jan 6 21:14:20 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2838116A4CE for ; Tue, 6 Jan 2004 21:14:20 -0800 (PST) Received: from gw.catspoiler.org (217-ip-163.nccn.net [209.79.217.163]) by mx1.FreeBSD.org (Postfix) with ESMTP id 98DCF43D2D for ; Tue, 6 Jan 2004 21:14:18 -0800 (PST) (envelope-from truckman@FreeBSD.org) Received: from FreeBSD.org (mousie.catspoiler.org [192.168.101.2]) by gw.catspoiler.org (8.12.9p2/8.12.9) with ESMTP id i075E87E015640; Tue, 6 Jan 2004 21:14:12 -0800 (PST) (envelope-from truckman@FreeBSD.org) Message-Id: <200401070514.i075E87E015640@gw.catspoiler.org> Date: Tue, 6 Jan 2004 21:14:08 -0800 (PST) From: Don Lewis To: doconnor@gsoft.com.au In-Reply-To: <200401071515.22366.doconnor@gsoft.com.au> MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii cc: freebsd-current@FreeBSD.org cc: yurtesen@ispro.net.tr Subject: Re: FreeBSD Crashes with AMD X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 07 Jan 2004 05:14:20 -0000 On 7 Jan, Daniel O'Connor wrote: > On Wednesday 07 January 2004 14:29, Andre Guibert de Bruet wrote: >> I'm not aware of any memory test utility that runs under FreeBSD. If one >> is out there, it would most certainly require some type of kld to allow it >> to have direct access to random parts of system memory. I consider it is >> bad mojo to run hardware tests on a machine booted in multi-user mode. In >> order to get a definitive answer (and eliminate false-positives which >> could show up due to kernel bugs) and save yourself a lot of time and >> hassles, boot their floppy or bootable cd. > > I wonder if you could squish it into the loader? > > That would be kind of neat if you want to do a weekly memory test :) > > You could alter loader.conf to run the command, then have the loader alter it > back just before running the memory test (once around then reboot) memtest86 takes quite a while to run it's full suite of tests, and some of the memory errors that I have encountered were only found by some of the longer pattern-sensitivity tests. It would be better to just use ECC RAM and watch for ECC errors. Unfortunately FreeBSD is missing the code to do the latter. It sounds like recent versions of Linux have something called MCE which handles this.