From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265196AbUAJIAw (ORCPT ); Sat, 10 Jan 2004 03:00:52 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265198AbUAJIAw (ORCPT ); Sat, 10 Jan 2004 03:00:52 -0500 Received: from moutng.kundenserver.de ([212.227.126.177]:30943 "EHLO moutng.kundenserver.de") by vger.kernel.org with ESMTP id S265196AbUAJIAu (ORCPT ); Sat, 10 Jan 2004 03:00:50 -0500 Message-ID: <3FFFB1AD.5050307@boergens.de> Date: Sat, 10 Jan 2004 09:02:53 +0100 From: Kevin Boergens User-Agent: Mozilla/5.0 (X11; U; Linux i686; de-AT; rv:1.5) Gecko/20031007 X-Accept-Language: de, en-us, en MIME-Version: 1.0 To: Kernel Mailing List Subject: cs5530_set_xfer_mode Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Provags-ID: kundenserver.de abuse@kundenserver.de auth:e39b2c33c20867b2edbd9f79b3c25ad5 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hello! Maybe this is a question to stupid for this list, but I asked for it in de.comp.os.unix.linux.hardware and on the Debian user mailing list, and although many people tried, nobody could help me. I have a server with a National Semiconductor CS5530 southbridge (http://www.national.com/pf/CS/CS5530.html) and after booting it only worked in MDMA2. I tried to force it with hdparm -d1X66 but that caused only nasty messages on syslog, like: *********************************************************************** Jan 5 16:46:14 server kernel: hda: timeout waiting for DMA Jan 5 16:46:14 server kernel: ide_dmaproc: chipset supported ide_dma_timeout fu nc only: 14 Jan 5 16:46:14 server kernel: hda: status error: status=0x58 { DriveReady SeekC omplete DataRequest } Jan 5 16:46:14 server kernel: hda: drive not ready for command ********************************************************************** I tried this with the Debian kernel images 2.4.18-bf2.4 and 2.2.20-idepci and with several disks, all the same. I would have given up hope already, if I had not found out that the S.u.S.E. 7.3 kernel 2.4.10-4GB has no problem with this chipset at all. While booting, the kernel says: hda WDC[...] ide0 at 0x[...] on irq14 hda: cs5530_set_xfer_mode(UDMA2) And everything works as fast as I want. My questions: 1) What's the differences in the kernels to cause this behavior? 2) How do I get my Debian kernel to behave alike? Any help GREATLY appreciated, Kevin -- http://www.boergens.de