From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262580AbUA3Knr (ORCPT ); Fri, 30 Jan 2004 05:43:47 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262603AbUA3Knq (ORCPT ); Fri, 30 Jan 2004 05:43:46 -0500 Received: from h24-77-134-228.wp.shawcable.net ([24.77.134.228]:3332 "EHLO polgara") by vger.kernel.org with ESMTP id S262580AbUA3Kno (ORCPT ); Fri, 30 Jan 2004 05:43:44 -0500 From: Mathieu Allard Reply-To: mathieu@opalescent.ca To: linux-kernel@vger.kernel.org Subject: crash when using dma for cdrom drive Date: Fri, 30 Jan 2004 04:43:39 -0600 User-Agent: KMail/1.5.4 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200401300443.39841.mathieu@opalescent.ca> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I have an MSIDDR motherboard, model k7t266 pro2 ver. 2.0. When I enable dma for my cdrom drive using hdparm /dev/hdc, dma is turned on. However, when I try and access the drive, my computer crashes. In a console this is the text that I see : dma timeout error : status=0x50 {Drive Ready SeekComplete} ide-scsi : abort called for 1629 bad : scheduling while atomic ! call trace : ******( a whole bunch of information is here) and then something like hdc dma reset complete. I have tested this with a different cdrom drive and a different ide cable and the same thing happens. The cdrom is the only device on the ide chain. I am running a vanilla linux kernel 2.6.1 and am using the via chipset driver. Debian unstable is my distribution. I have searched google to no avail. I have looked at various bios options but none seem interesting. Any help would be appreciated with this problem. Thanks in advance. Mathieu