Wow, Do you have the FSC 810 patched with a original Tyan Bios?? Is that the way, we should go for our customers rainer?? Or should we wait for a official patch for the 1.05 FSC bios revision?? friendly regards Andreas -----Ursprüngliche Nachricht----- Von: Thomas Renninger [mailto:trenn@suse.de] Gesendet: Freitag, 17. Dezember 2004 19:05 An: Andi Kleen Cc: Rainer Koenig; suse-amd64@suse.com; Andreas Wahlert Betreff: Re: [suse-amd64] Dual Opteron crashes with Kernel Panic Andi Kleen wrote:
On Thu, Dec 16, 2004 at 11:27:45AM +0100, Rainer Koenig wrote:
Hi Andreas,
Andreas Wahlert <Andreas.Wahlert@gmx.de> writes:
I'm in a serious discussion with the FSC Celsius man in Augsburg. perhaps there are any results today or monday.
Can you give me more contact info about that Celsius man in Augsburg? Per definition it should be me... but I just read about this because your problem is also showing up at other customers of our CELSIUS V810 and that other customer pointed us to this thread. Sorry that this problem didn't get my attention before.
And yes, I can perfectly reproduce it here on my V810 as well.
I could reproduce it here, too: oops at the exactly same function: RIP: 0010:[<ffffffff8011a8fe>] <ffffffff8011a8fe>{generic_set_all+318} The oops does not apear on a SLES9-SP1 installation However, I could solve the problem by installing a current BIOS version: It's a tyan S2885 board. The previos BIOS version was from 01.2004, the new one (v. 2885_202) is from 19.05.2004. what the update fixes (some mtrr enhancements are mentioned as well) see: http://www.tyan.com/support/html/b_s2885.html I got some scsi controller Problems on SLES9 (9.2 boots properly) now after the update, I cannot say anything on this for now, I will investigate further on this on Monday. Thomas
It probably depends on the amount of memory installed and the BIOS version. Basically the crash happens when the kernel tries to duplicate MTRR setup done by the BIOS to the other CPU, so likely there is some issue in the original MTRR setup.
I haven't heard a report for it from the V810 at suse, perhaps it doesn't show it.
There is a bug open for the issue, but I haven't had time to look at it in detail yet.
-Andi