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, 6 Feb 2003 17:44:34 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 6 Feb 2003 17:44:34 -0500 Received: from 60.54.252.64.snet.net ([64.252.54.60]:40109 "EHLO hotmale.blue-labs.org") by vger.kernel.org with ESMTP id ; Thu, 6 Feb 2003 17:44:32 -0500 Message-ID: <3E42E7B7.2080900@blue-labs.org> Date: Thu, 06 Feb 2003 17:54:47 -0500 From: David Ford User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3b) Gecko/20030131 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Andrew Morton CC: linux-kernel@vger.kernel.org, "Adam J. Richter" Subject: Re: 2.5.59 OOPS w/ fdisk & devfs References: <3E42D05E.8080907@blue-labs.org> <20030206135138.3c083007.akpm@digeo.com> In-Reply-To: <20030206135138.3c083007.akpm@digeo.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Well, don't jump too fast :) I have been having extreme hardware issues with this box for the past several hours. I thought it was 2.5.59 giving me grief but an old kernel I had laying around presented similar issues. I'm busy swapping cables, memory, etc to see who is being the bad guy. So far, I've had issues with reiserfs, raw dd, mkreiserfs, tar, cp, mv (insert a lot of shell commands), interpreters, you name it. Nothing seems common except that it all occurs when hda data is going somewhere else. I just swapped my udma66 cable on hda and I'm trying to stress test hda now. David Andrew Morton wrote: >David Ford wrote: > > >>Unable to handle kernel paging request at virtual address 6b6b6b87 >>... >>EIP is at _devfs_unhook+0x2b/0x70 >> >> > > >Look. devfs is sick. Richard has disappeared. Al did some work on it and >also disappeared. Adam laid it on the ground and drove a truck over it, and >I had that patch in -mm for two or three weeks and had one single, sole, sad, >sorry report from a tester. > >