Re: Nasty oops (plural)

Keith Rohrer (
Tue, 25 Feb 1997 02:16:53 -0600 (CST)

> check your memory timings in your chipset set up first -- could be related
> to your faster cpu writing too fast/ slow vice versa etc.
> I had this same problem once --- that is how i fixed it (hard to remember
> now though)
Odd, though, that it should show up in a 2x40 -> 3x40 upgrade...

> On Tue, 25 Feb 1997, Christopher Wiles wrote:
> > All,
> > After I upgraded my trusty 486/80 to a 486/120, I've been having random
> > segfaults when I attempt to execute a program.
To see if it's memory in some way shape or form, repeatedly rebuild the
kernel (or otherwise floor gcc). I had a 486/100 that would overheat
after 4-5 hours running, causing invalid instruction errors and other
similar, non-repeatable errors; adding a heat sink and fan made it
work just fine (until the fan died). I did have to add a wait state
when I went from 3x33 to 2x50 (I went back because the onboard EIDE
chip hated 50 MHz even at "slowest")...

> > I stress that each oops is _the_ _same_. It seems to be more frequent
> > with heavy load (where heavy is defined as greater than one), but the
> > machine will go for a week without a problem -- then I'll get a whole
> > bunch of them.
> > I'd really like someone to tell me that it's bad RAM. I'd really not like
> > someone to tell me that the processer is hosed.
Linux idles the processor when Linux is idle, consuming less power and
generating less heat. If once the CPU is hot, the errors come in
droves, then you have inadequate cooling. If CPU temperature doesn't
have anything to do with it but load average does, it might be bad
RAM or too few wait states (or too slow a refresh period).