On 23 Feb 2001, at 22:18, Steinar H. Gunderson wrote:
> After being away for five days recently, I noticed that my computer
> (running Linux kernel 2.4.1, by the way -- 2.4.2 now) had rebooted.
> Just a few hours later, it rebooted again -- and that night, it rebooted
> _again_.
>
> If I turn off mprime (v20), the problem goes away -- the computer
> doesn't reboot, at least not the 36 hours I tested. After I start
> mprime, it reboots in just a couple of minutes now.
As other people have suggested, it sounds like an overheating
problem. Most programs do not use the FPU much, and the FPU is a
major contributor to power consumption within the CPU.
The fact that your heatsink feels cool(ish) suggests to me that the
thermal bond between the processor and the heat sink has failed.
(If the fan had failed, you'd toast your fingers making this test!)
Actually I'm not overwhelmingly surprised. When I installed my Athlon
650 (slot A variety) I noticed that the heatsink was supplied with a
nasty piece of thermal tape which you were supposed to stick the
heatsink to the CPU cartridge with. (Note, all Athlon processors are
supplied as OEM packages, you have to supply your own heatsink if
you're a system builder). I replaced this with a dab of thermal
grease & haven't had any obvious overheating problems in the approx.
11 months the system has been operating. But I rather suspect that
anyone using the thermal tape will be getting somewhat inferior
cooling, which might fail more or less completely when the tape
cracks - a known age-related problem.
Another possibility is that the chipset may be overheating. If the
chipset has a cooling fan (probably a small one like those used on
486s) is that running too? Again mprime drives the memory bus hard &
may trigger problems which remain hidden when the system is running a
"normal" (read, light) load. You could investigate this by running an
integer-only memory thrashing program like a prime number sieve. If
running that makes the system unstable, look for a problem in the
chipset (or memory - but serious memory problems tend to show up even
under fairly light loads!). If not, it's much more likely a problem
in the processor.
Anyway, I'm afraid you're talking about a hardware problem rather
than software.
Regards
Brian Beesley
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
