Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S266193AbUI0Sa5 (ORCPT ); Mon, 27 Sep 2004 14:30:57 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S267170AbUI0Saz (ORCPT ); Mon, 27 Sep 2004 14:30:55 -0400 Received: from 64.221.211.203.ptr.us.xo.net ([64.221.211.203]:1937 "EHLO mail.pathscale.com") by vger.kernel.org with ESMTP id S266193AbUI0SaN (ORCPT ); Mon, 27 Sep 2004 14:30:13 -0400 Subject: Re: AMD64 and NFORCE3 250GB very slow and USB hungs From: "Bryan O'Sullivan" To: Michael Thonke Cc: linux-kernel@vger.kernel.org In-Reply-To: <41569DE5.4080206@web.de> References: <41569DE5.4080206@web.de> Content-Type: text/plain Date: Mon, 27 Sep 2004 11:30:12 -0700 Message-Id: <1096309812.26765.7.camel@serpentine.internal.keyresearch.com> Mime-Version: 1.0 X-Mailer: Evolution 1.5.9.1 (1.5.9.1-2) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1197 Lines: 30 On Sun, 2004-09-26 at 12:45 +0200, Michael Thonke wrote: > I bought just a new AMD64 system with an NForce 3 based mainboard (MSI > K8N Neo2 Platinum). There's a known problem with NForce2 and NForce3 chipsets that affects at least IDE interrupt handling, the effect being that the system hangs if there's "too much" disk activity. I've verified that this occurs with 2.6.8.1, but haven't tried more recent snapshots. The symptom is that the system hangs hard during boot. A tolerable workaround appears to be to drop the IDE UDMA level down to 3 using hdparm. If you're using a Fedora Core distro and edit /etc/ sysconfig/harddisks to do this, it gets set up early enough during boot that the system rarely hangs. I haven't had time to look into this deeper, but I did file a bug against it: http://bugzilla.kernel.org/show_bug.cgi?id=3398 It should probably be owned by Jens, since it seems to be IDE-specific, but Andi has it for now.