Memory/cpu problem with Asus K7M motherboard/Athlon 750

[Mark Steele]

>First, there is a known incompatibility between Linux 2.2.x & many ACPI
>bioses.  The first thing I would check is to see if ACPI is enabled on your
>750 & disabled on the 650.  For 2.2.x, you might try disabling ACPI in the
>bios & see if this fixes the 64M limit.
>
>Depending upon you situation, you might try the latest 2.3.x kernel.  There
>are a number of improvements which the Athlon will be able to showcase.
>
>
>Nathan

Well I've installed kernel 2.3.40, and the issue seems to be resolved.

After a bit of further testing, I found that turning off ACPI had no effect
on wether or not kernel 2.2.14 'saw' all of the RAM. Enabling or disabling
ECC parity checking in the BIOS also did not change the results.

Kernel 2.3.40 seems to function properly regardless of wether ECC/ACPI
are enabled or disabled in the BIOS.

In regards to Mark Hanh:

>now, the fact that the 750 is newer leads me to wonder whether 
>its bios is inadequate (ie, doesn't properly handle features 17 and 24, 
>whatever they are.)

>> is that it's some kind of memory problem, but I'm completely baffled.

>why? how do the dimms look? it's probably possible that the 750's 
>memory is flakey, but ecc is correcting the problems, imposing some 
>performance penalty. seems like quite a stretch to me, though. 
>are the machines stable otherwise?
>regards, mark hahn

The dimms 'look' just fine :). My first guess was that the memory was 
flackey. I had been previously using non-ECC memory, so I switched.
Seeing the same problem leads me to believe it's not the RAM.

I had also replaced the motherboard to no effect, so it has to be the
kernel as far as I can figure out.

With the experimental kernel, everything seems to be operating normally,
a few quick checks seem to indicate that the machine is stable, and I'll
be running some benchmarks to see how it's performing, but preliminary
tests seem to indicate that it's blazing fast :)

I'm unfortunately not very familiar with programming/debugging in C, 
so I'm fairly inept at figuring out exactly what went wrong, but if
someone wants to try to figure it out I'm more than willing to give
access to the box if you need a test environment or a live demonstration
of the problem.

Also, if you know what's causing the problem, or how to fix it, please
email me, as I'm not on the mailing list.

Mark Steele
Systems Administrator
Belmont Web Services
msteele en belent.com

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo en vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/