From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Peter Gervai" Subject: Re: Hard freeze / interrupt-related death / instability Date: Wed, 9 Apr 2008 00:02:09 +0200 Message-ID: References: <1207687673.3091.12.camel@linux-2bdv.site> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1207687673.3091.12.camel@linux-2bdv.site> Content-Disposition: inline Sender: linux-acpi-owner@vger.kernel.org To: trenn@suse.de Cc: linux-acpi , linux-ide List-Id: linux-ide@vger.kernel.org On Tue, Apr 8, 2008 at 10:47 PM, Thomas Renninger wrote: > Hi, > > I expect this is more a ata/ahci or chipset problem (this is ICH7, > right?), Yes. > maybe ide list has an idea or has seen something similar already? Good point, let's see. > Two more little comments below. > > Then one page started to suggest to play with acpi switches on boot, and I went > > on. The results: > > > > #irqpoll - dies eventually > > #noapic irqpoll pci=nommconf - dies eventually > > #pci=noacpi - dies eventually > > #noacpi - dies eventually, but it took longest (3 days) > > > > #acpi=off - uniprocessor, ok > > #noapic acpi=off - uniprocessor, ok > What means uniprocessor? > Also added nosmp or maxcpus=1? Oh, maybe forgot to explain this one (copypaste from my grub menu.lst :)). It means the machine didn't freeze but the kernel completely ignored the second core of the dual core cpu, basically seem to have ignored SMP capability altogether. I did not use nosmp or maxcpu since I do want to use them all. :) > Maybe pci=nomsi helps? I note that, and try it the next time. Thanks, Peter