Deadlocks and Machine Check Exception on Athlon64

Deadlocks and Machine Check Exception on Athlon64

[Ronny V. Vindenes]

My machine has been locking up [1] every now and then the last couple of
weeks and yesterday it locked during boot and I managed to copy this
message:

CPU 0: Machine Check Exception : 0000000000000004
Bank 4: b200000000070f0f
kernel panic: CPU context corrupt
In interrupt handler - not syncing

I tried to decode it with davej's parsemce, but it didn't produce any
usable output.

The machine is an MSI K8T Neo-FIS2R (flashed with 1.2 bios), with
Athlon64 3200+ and 2x512MB PC3200 DDR RAM, running 32bit 2.6.3-rc3-mm1
kernel. Any suggestions on what could be wrong?

[1] The lock-ups mostly happen during browsing with epiphany, I compile
a lot of big projects almost every day so it appears that cpu load or
memory use has no influence on the lock-ups.

-- 
Ronny V. Vindenes <s864@ii.uib.no>

Re: Deadlocks and Machine Check Exception on Athlon64

[Tom Vier]

On Thu, Feb 19, 2004 at 11:31:50PM +0100, Ronny V. Vindenes wrote:
> CPU 0: Machine Check Exception : 0000000000000004
> Bank 4: b200000000070f0f
> kernel panic: CPU context corrupt
> In interrupt handler - not syncing

i've gotten the same machine check a couple times. it only happened the
first few days i had my machine. i updated to the latest tyan bios which
fixes some bugs. i haven't had it happen since.

if i'm reading it right (which is hard on this small screen, slow cpu
laptop), page 153 of the opteron kernel/bios guide says this is an ecc data
cache exception (which is always fatal/nonrecoverable). this sounds like
errata 86, which is fixed in the tyan bios update i installed.

does msi say their 2.1 bios fixes #86?

-- 
Tom Vier <tmv@comcast.net>
DSA Key ID 0xE6CB97DA