Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932637AbXAQTuA (ORCPT ); Wed, 17 Jan 2007 14:50:00 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932700AbXAQTt7 (ORCPT ); Wed, 17 Jan 2007 14:49:59 -0500 Received: from mx1.redhat.com ([66.187.233.31]:44698 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932637AbXAQTt7 (ORCPT ); Wed, 17 Jan 2007 14:49:59 -0500 Date: Wed, 17 Jan 2007 14:46:01 -0500 (EST) From: Chip Coldwell To: Chip Coldwell cc: Andi Kleen , Chris Wedgwood , Christoph Anton Mitterer , Robert Hancock , linux-kernel@vger.kernel.org, knweiss@gmx.de, andersen@codepoet.org, krader@us.ibm.com, lfriedman@nvidia.com, linux-nforce-bugs@nvidia.com Subject: Re: data corruption with nvidia chipsets and IDE/SATA drives (k8 cpu errata needed?) In-Reply-To: Message-ID: References: <45AD2D00.2040904@scientia.net> <20070116203143.GA4213@tuatara.stupidest.org> <200701170829.54540.ak@suse.de> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1444 Lines: 41 On Wed, 17 Jan 2007, Chip Coldwell wrote: > On Wed, 17 Jan 2007, Andi Kleen wrote: > >> On Wednesday 17 January 2007 07:31, Chris Wedgwood wrote: >>> On Tue, Jan 16, 2007 at 08:52:32PM +0100, Christoph Anton Mitterer wrote: >>>> I agree,... it seems drastic, but this is the only really secure >>>> solution. >>> >>> I'd like to here from Andi how he feels about this? It seems like a >>> somewhat drastic solution in some ways given a lot of hardware doesn't >>> seem to be affected (or maybe in those cases it's just really hard to >>> hit, I don't know). >> >> AMD is looking at the issue. Only Nvidia chipsets seem to be affected, >> although there were similar problems on VIA in the past too. >> Unless a good workaround comes around soon I'll probably default >> to iommu=soft on Nvidia. >> > > We've just verified that configuring the graphics aperture to be > write-combining instead of write-back using an MTRR also solves the > problem. It appears to be a cache incoherency issue in the graphics > aperture. I take it back. Further testing has revealed that this does not solve the problem. Chip -- Charles M. "Chip" Coldwell Senior Software Engineer Red Hat, Inc 978-392-2426 - 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/