From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754206AbXDRTlE (ORCPT ); Wed, 18 Apr 2007 15:41:04 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753881AbXDRTlE (ORCPT ); Wed, 18 Apr 2007 15:41:04 -0400 Received: from hera.kernel.org ([140.211.167.34]:55483 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754213AbXDRTlD (ORCPT ); Wed, 18 Apr 2007 15:41:03 -0400 From: Len Brown Organization: Intel Open Source Technology Center To: Michal Jaegermann Subject: Re: 2.6.20.6 vanilla does't boot Date: Wed, 18 Apr 2007 15:39:25 -0400 User-Agent: KMail/1.9.5 Cc: Denis Kirjanov , Marat Buharov , Marc Donner , linux-kernel@vger.kernel.org References: <20070415155016.GA14489@ellpspace.math.ualberta.ca> In-Reply-To: <20070415155016.GA14489@ellpspace.math.ualberta.ca> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200704181539.26105.lenb@kernel.org> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sunday 15 April 2007 11:50, Michal Jaegermann wrote: > On Fri, Apr 13, 2007 at 03:38:18PM +0400, Denis Kirjanov wrote: > > I updated the BIOS to the latest version, but the problem persists. > > Boots option pci = noacpi not solved the problem. Reporting bios bug > > disappears when setting pci = nommconf, But the kernel is still not > > loaded ( The 2.6.15 dmesg do not give any clues to why 2.6.20.6 would not boot. Can you boot 2.6.20.6 with "debug" and capture the serial console log, or take a photo of what is on the screen when the boot hangs? Did anything between 2.6.15 and 2.6.20.6 work properly? Can you bisect where the failure started? > On x86_64 hardware using ata_piix I was unable to boot kernels > based on the current 2.6.20.x either. Regardless of extra kernel > parameters used I was getting consistently 'hdc: lost interrupt', > and/or similar and after something like that the whole machine > was dead. The only differences were if I could still reboot > from a keyboard or if I really have to pull a plug. > > A kernel derived from 2.6.21-rc6-git1 (2.6.20-1.3053.fc7.x86_64 from > Fedora "rawhide" to be more precise) did boot on the hardware in > question, though; but only when I gave it 'acpi=off'. Without that > parameter it was getting stuck apparently when starting hotplug. > In that kernel case disks were accessed using pata_atiixp driver. If "acpi=off" is necessary to boot the latest kernel, please report an ACPI bug: http://bugzilla.kernel.org/enter_bug.cgi?product=ACPI Please mention in the bug report what the latest working kernel was. thanks, -Len