> Havard Eidnes <he@NetBSD.org> typed:
> : one of my i386 boxes continually run UPDATE builds of NetBSD-
> : current, itself running reasonably new -current code -- now most
> : recently updated to 4.99.40 as of three days ago.
> [...]
> : It's a 2x2.2GHz Xeon with 2GB of ECC memory.
>
> Multiprocessor corruption under i386 - been around ~forever (well,
> as long as we've had multiprocessor for i386 branch).
> I've had it on PentiumII-233MHz, Pentium4-3000, C2D-E6850's.

Hmm...  Well, I'm using the exact same setup as on the 2x2.2GHz Xeon
with a 4x700MHz Xeon system with 2GB of memory where I've not seen
this behaviour.  The 4x700MHz system runs 3.1_STABLE, though, and it
appears to occasionally have issues with process run-down (causing
either panics or kernel page faults), ref. PR #35150.

> Testing done using the stress test feature in mersenne.org program,
> with number of programs as the number of processors.
> Running it under MSWindows on the same systems work fine, however
> MSWindows has processor affinity.

OK.  Maybe I should test with that program as well.  However,
the name service for mersenne.org appears to be inoperational at
the moment -- ns1.sbsi-net.com doesn't respond, and ns2.* replies
with SERVFAIL.  Bah!

Regards,

- H=E5vard