On Thu, 03 Jun 2004 19:36:59 +0200 (CEST) Havard Eidnes <he@netbsd.org> wro=
te:
> > IPIs can get lost appearently.
> > I don't fully understand how this can happen, but changing
> > the code to be more conservative helped on my dual-Opteron.
> >
> > Could you try the i386 equivalent of the patch I appended
> > to the PR database?
>=20
> For your collective convenience, an equivalent patch for i386 is
> attached below.  I'm presently running one SMP box with the patch,
> FWIW, though it didn't trigger this problem frequently in the past.
>=20
> > Would also be interesting whether this has performance
> > implications.
>=20
> Do you have a suspicion whether it would make matters better or worse?
>=20
> Regards,
>=20
> - H=E5vard

I rebooted my machine this morning and put the latest BIOS rev onto it.
Dell listed the update a critical.  One of the features that this rev
seems to have added is "Sequential Memory Access" (that's from memory)
under the CPU related section.  I enabled this.

Since then, I've been pounding on the machine in an attempt to get it
to crash again.  So far, so good.  Perhaps this was just a BIOS related
problem.  Or, enabling "Sequential Memory Access" is a workaround.

I'll reboot it in a little while and turn off the sequential memory access
and see what happens.

Paul
--=20
Paul Dokas                                            dokas@cs.umn.edu
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Don Juan Matus:  "an enigma wrapped in mystery wrapped in a tortilla."