From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262170AbVFUSWR (ORCPT ); Tue, 21 Jun 2005 14:22:17 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261824AbVFUSWR (ORCPT ); Tue, 21 Jun 2005 14:22:17 -0400 Received: from osten.wh.uni-dortmund.de ([129.217.129.130]:6291 "EHLO osten.wh.uni-dortmund.de") by vger.kernel.org with ESMTP id S262209AbVFUSWD (ORCPT ); Tue, 21 Jun 2005 14:22:03 -0400 Message-ID: <42B85AC9.4060708@web.de> Date: Tue, 21 Jun 2005 20:22:01 +0200 From: Alexander Fieroch User-Agent: Debian Thunderbird 1.0.2 (X11/20050402) X-Accept-Language: de-de, en-us, en MIME-Version: 1.0 Newsgroups: gmane.linux.kernel To: "Protasevich, Natalie" Cc: linux-kernel@vger.kernel.org, Alan Cox , bzolnier@gmail.com, axboe@suse.de, Bartlomiej Zolnierkiewicz Subject: Re: [2.6.12rc4] PROBLEM: "drive appears confused" and "irq 18: nobody cared!" References: <19D0D50E9B1D0A40A9F0323DBFA04ACCE04C08@USRV-EXCH4.na.uis.unisys.com> In-Reply-To: <19D0D50E9B1D0A40A9F0323DBFA04ACCE04C08@USRV-EXCH4.na.uis.unisys.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Protasevich, Natalie wrote: > I would try forcing legacy mode some way, say by tweaking the > line in do_ide_setup_pci_device(): > if ((dev->class >> 8) == PCI_CLASS_STORAGE_IDE && (dev->class & 5) != > 5) { > > To something like > > if ((dev->class >> 8) == PCI_CLASS_STORAGE_IDE) { I did this with kernel 2.6.12-git2. There are more much error messages like... "hda: lost interrupt" ...where linux hangs with a timeout. Finally linux falls in a loop of errors and is repeating the following tree lines continually: hda: lost interrupt hda: dma_timer_expiry: dmastatus == 0x64 hda: DMA interrupt recovery > Another thought is to dump PCI configuration space for ICH6 IDE, and > verify the values, especially INT_LN, INTR_PIN, and PCICMD which has > such info as whether IDE interrupt is enabled on the chip (bit 10), see > Intel's ICH6 spec, chapter 11 ( > ftp://download.intel.com/design/chipsets/datashts/30147302.pdf ). Ok, is there anything that I can do? > I posted reply to you last email but surprisingly didn't see it going > trough to the mailing list. Maybe, our mailer blocked it for some > reason. Curious. I also have recognized that some replys did not go through the mailing list, but I've cc'd all of you. Perhaps someone should check the filter? Regards, Alexander