On Thu, 2004-07-22 at 08:04 +0200, Dominik Brodowski wrote:
> AFAICS in some specs, there are identical-looking (CPUID, string) Dothans
> with different voltage requirements, at least for the B1 stepping.

Yeah, I worked that out from a closer reading of the docs.  There really
doesn't seem to be any non-hacky way to work out which CPU you have, and
they seem to have non-overlapping specs (ie, you can't run a VID#D part
with VID#A voltages).  The hacky way to work it out would be to read
back the MSR and see what freq/voltage pair you have (which would only
work at max speed).

>  Also, as
> the usage of the ACPI P-State library in speedstep-centrino seems to work 
> flawlessly on Dothans [at least I haven't heard otherwise], I'm tempted to 
> remove the "(EXPERIMENTAL)" mark on it and possibly make it even the default. 
> What do you think?

Seems like a reasonable idea, though it would be nice for Dothan users
to not have an ACPI dependency.  I put together this patch to clean
things up a bit for Dothan users and to reduce the flow of mail I'm
getting.  I've tested it as much as I can without a Dothan on hand.

	J