Re: Help SMP/dual PII333MHz on GA686DLX board

Alan Cox (alan@lxorguk.ukuu.org.uk)
Sun, 22 Mar 1998 15:26:37 +0000 (GMT)

> > Congratulations, send the bill for your hair damage to the supplier. You
> > have yet another SMP box with faulty bios. There is a patch for 2.0.x
> > on www.uk.linux.org and there are people working on generic MTRR handling
> > for 2.1.x
> >
> I have one dumb question about that. How is NT able to work with it if the
> BIOS has problems? Or does it explain why NT has bad SMP performance(What
> I read in several magazines). I have a SMP box but I never tried to run NT
> on it.

I've no idea if NT does handle this. My first guess would be that the NT
programmers assume all BIOS writers are stupid. Thats basically what the
MTRR stuff does. It assumes the BIOS writer is too daft to have read
the MP spec and copies the MTRR settings from the boot processor to the
application processor(s).

Making the kernel always init the keyboard controller is a similar "Assume
all bios writers are stupid" thing we need to cover. It certainly appears
MS lose95 does this too.

Unfortunately it appears more and more that BIOS vendors "testing" procedure
consists of checking Win95 and NT boot, not of checking against specifications.

Alan

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu