From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751773AbXCTVGp (ORCPT ); Tue, 20 Mar 2007 17:06:45 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753898AbXCTVGp (ORCPT ); Tue, 20 Mar 2007 17:06:45 -0400 Received: from www17.your-server.de ([213.133.104.17]:2600 "EHLO www17.your-server.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751773AbXCTVGp (ORCPT ); Tue, 20 Mar 2007 17:06:45 -0400 Message-ID: <46004D2F.4020901@m3y3r.de> Date: Tue, 20 Mar 2007 22:07:59 +0100 From: Thomas Meyer User-Agent: Thunderbird 1.5.0.10 (X11/20070302) MIME-Version: 1.0 To: Linux Kernel Mailing List CC: Pekka Enberg Subject: Re: Bug in pci_restore_msi_state References: <45FBE22A.8030304@m3y3r.de> <84144f020703170618x73808b4fmf5b420f4c2e02818@mail.gmail.com> In-Reply-To: <84144f020703170618x73808b4fmf5b420f4c2e02818@mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Authenticated-Sender: thomas@m3y3r.de Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Pekka Enberg schrieb: > On 3/17/07, Thomas Meyer wrote: >> Hello everybody. >> >> I get this bug after suspending to disk twice: >> >> http://m3y3r.de/bilder/Bug-pci_restore_msi_state.png >> >> This happens with current git head >> cd05a1f818073a623455a58e756c5b419fc98db9. > > If you know a kernel that works, please consider doing git bisect: > > http://www.kernel.org/pub/software/scm/git/docs/howto/isolate-bugs-with-bisect.txt > > Ok. The error happens in the call of pcibios_enable_device in the function do_pci_enable_device (drivers/pci/pci.c): Before the call of pcibios_enable_device: do_pci_enable_device: pci_dev= c1a40000 do_pci_enable_device: irq= 218 do_pci_enable_device: msi_enabled= 1 after the call: ACPI: PCI Interrupt 0000:00:1f.2[B] -> GSI 19 (level, low) -> IRQ 19 do_pci_enable_device: pci_dev= c1a40000 do_pci_enable_device: irq= 19 do_pci_enable_device: msi_enabled= 1 So the function acpi_pci_irq_enable in drivers/acpi/pci_irq.c needs to be fixed: a.) acpi_pci_irq_enable should care about msi setups and set the flag msi_enabled to 0 or b.) acpi_pci_irq_enable should care about msi setups and should touch the dev->irq field. Please comment on this. with kind regards thomas