From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755822AbZBCTPi (ORCPT ); Tue, 3 Feb 2009 14:15:38 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755291AbZBCTPR (ORCPT ); Tue, 3 Feb 2009 14:15:17 -0500 Received: from mx2.compro.net ([216.54.166.4]:13516 "EHLO mx2.compro.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755265AbZBCTPP (ORCPT ); Tue, 3 Feb 2009 14:15:15 -0500 X-IronPort-AV: E=Sophos;i="4.37,373,1231131600"; d="scan'208";a="3348592" Message-ID: <498897C1.2090802@compro.net> Date: Tue, 03 Feb 2009 14:15:13 -0500 From: Mark Hounschell Reply-To: markh@compro.net Organization: Compro Computer Svcs. User-Agent: Thunderbird 2.0.0.19 (X11/20081227) MIME-Version: 1.0 To: Borislav Petkov CC: Mark Hounschell , Andreas Herrmann , john stultz , linux-kernel , Thomas Gleixner Subject: Re: PROBLEM: Can't boot a (HZ = 1000) kernel using an AMD Phenom-II processor References: <49798F33.8070906@cfl.rr.com> <20090123130416.GA27616@alberich.amd.com> <497A48D5.9040904@cfl.rr.com> <497E0072.7000500@compro.net> <20090127163843.GA1628@aftab> <497FA725.7020503@cfl.rr.com> <497FA7F2.3060709@cfl.rr.com> <498416CE.4040101@cfl <20090203151137.GB12017@aftab> In-Reply-To: <20090203151137.GB12017@aftab> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Borislav Petkov wrote: > Hi, > > On Tue, Feb 03, 2009 at 04:17:55AM -0500, Mark Hounschell wrote: >> Borislav Petkov wrote: >>> On Sat, Jan 31, 2009 at 04:15:58AM -0500, Mark Hounschell wrote: >>> >>> [..] >>> >>>> So were you able to reproduce this with my .config. Has anyone booted a Hz=1000 >>>> kernel on an AMD Phenom-II? Is there something wrong with my .config? Is there >>>> anything else I can do? >>> Hi, >>> >>> sorry for the delay - I've been busy last week and couldn't get to your >>> problem. So, I couldn't reproduce the bug here with a Phenom II and your >>> kernel config. However, we have a working theory, so can you please send >>> me the output of >>> >>> xxd /sys/firmware/acpi/tables/APIC >>> >> markh@harley:~> xxd /sys/firmware/acpi/tables/APIC >> 0000000: 4150 4943 8400 0000 014c 5258 3738 3020 APIC.....LRX780 >> 0000010: 4157 5244 4143 5049 312e 3042 4157 5244 AWRDACPI1.0BAWRD >> 0000020: 0000 0000 0000 e0fe 0100 0000 0008 0000 ................ >> 0000030: 0100 0000 0008 0101 0100 0000 0008 0202 ................ >> 0000040: 0100 0000 0008 0303 0100 0000 010c 0400 ................ >> 0000050: 0000 c0fe 0000 0000 020a 0000 0200 0000 ................ >> 0000060: 0000 020a 0009 0900 0000 0f00 0406 0005 ................ >> 0000070: 0001 0406 0105 0001 0406 0205 0001 0406 ................ >> 0000080: 0305 0001 > > looks good at a glance... > >>> Also, can you boot a 64-bit, HZ1000 kernel on your machine? >>> >> No, a 64 bit kernel @1000Hz doesn't boot either. As with the 32 bit kernel, >> setting Hz to 250 allows it to boot. The messages are slightly different when it >> hangs on the 64bit kernel at 1000Hz. > > That's not cool - as I said before, I can't reproduce this behavior here > which means that the problem might be chipset-related. What you could do > now is to boot the kernel with "debug initcall_debug log_buf_len=10M" > and try to capture the whole boot log with a serial console - or if that > doesn't work - make a screenshot with a camera of some sort. This might > shed some light on the question where exactly your machine hangs. > It will be easier to capture the process on video then rip it out into a bunch jpegs than it will be to connect the serial port. It's internal and I don't have the adapter to bring it out the back of the case. I'll see what I can do. It might be a couple of days but I'll get back to ya. Thanks Mark