Hi all I am almost certain this is not a Suse issue, but I thought I might inquire just in case. As stated before, I have a Tyan K8W motherboard with 4Gb of DDR400 memory running Suse 9.0 AMD64 with the patch kernel. The BIOS registers 4Gb of memory, but the operating system only shows ~2.5 Gb of addressable memory (and starts swopping when it hits this limit). Any clue what is causing this? Since the Opteron memory controller is integral to the CPU, is this a fundamental problem with 244s or could it still be in the BIOS? Additionally, if I enable ECC in the BIOS I encounter unpredictable failures. How can this be? I thought Opterons ONLY work with registered ECC memory. Could BIOS settings be incompattible with the memory controller? Has anybody with a K8W updated to the 2885101i BIOS beta revision? Does it do any of the things it purports to fix: * Add DDR400 display support. * Add AMD Errata 86 fix * Open HT frequency options * Update CPU module to fix the CPU APIC Version error * Replace IORR with MTRR for AGP Aperture memory * Update Silicon Image Option ROM to version 5.0.31 (from tyan ) Thanks for your patience! Eugene PS. Thanks also to previous correspondents regarding the ATI driver issue (it doesnt yet exist for 64bit linux) _________________________________________________________________ Need info in a hurry? Use MSN Search! http://search.msn.co.za/
On Wed, 28 Jan 2004 23:59:39 +0000 "Eugene de Villiers" <edevilliers@hotmail.com> wrote:
Since the Opteron memory controller is integral to the CPU, is this a fundamental problem with 244s or could it still be in the BIOS?
Do a head -30 /var/log/boot.msg The kernel prints the e820 memory map supplied by the BIOS at the beginning. If the memory is included there (and marked usable) it could be an OS problem. If not it's definitely BIOS.
Has anybody with a K8W updated to the 2885101i BIOS beta revision? Does it do any of the things it purports to fix: * Add DDR400 display support. * Add AMD Errata 86 fix * Open HT frequency options * Update CPU module to fix the CPU APIC Version error * Replace IORR with MTRR for AGP Aperture memory
This fix together with the latest update kernel should increase stability with a lot of RAM. -Andi
The BIOS registers 4Gb of memory, but the operating system only shows ~2.5 Gb of addressable memory (and starts swopping when it hits this limit). Any clue what is causing this?
Yes, it's eating it all for the AGP aperture and PCI memory mapping and stuff because the BIOS is using some pretty dumb layouts. This is fixed by the "memory hole" option in the 0.01b BIOS, and I've been told that this will make it into some future 1.xx official release at some undetermined future time. FYI, I've been running 0.01b for a while now (about two months) and have had zero problems and zero instability. I use a 2GB memory hole (physical at 0-2GB, then 4GB-18GB) and I see every byte of my 16GB of RAM.
Additionally, if I enable ECC in the BIOS I encounter unpredictable failures. How can this be? I thought Opterons ONLY work with registered ECC memory. Could BIOS settings be incompattible with the memory controller?
You should also be aware that there is apparantly (I'm told and have not verified this as I have no 244s) a bug in the Tyan BIOS that sets the memory timings for DDR400 even 'tho your 244s aren't rated to do that. You can use my northbridge/8131/8151 register dumping tool here: http://viz.cacr.caltech.edu/dl/regdump-8xxx to see what state your BIOS is leaving things in. It prints one bitfield per line, so things like diff become useful to see what changes from BIOS to BIOS. Yes, it took a lot of beer, but I really was insane enough to type in all the bitfield descriptions. I've haven't worked up the insanity to do the 8111 yet 'cause it's monstrously huge. I'd be thankful if you'd forward me a copy of its output if you have the time (along with the exact BIOS version number you're running).
Thanks for your patience!
Good luck! -mcq
participants (3)
-
Andi Kleen -
Eugene de Villiers -
John McCorquodale