"H. Peter Anvin" <hpa@transmeta.com> said:
> Horst von Brand wrote:
> > > ... right, and the process is almost certainly CPL 0 only, so it needs
> > > to be in kernel space anyhow.
> > What still worries me a bit is that intel places strict restrictions on
> > this. On the other hand, this is kernel bloat IMVHO: It is functionality
> > that should be used during boot (if at all), and then jettisoned.
> Loadable kernel modules are a beautiful thing.
Very true.
> The restrictions Intel place on this makes this probably a bad idea in
> case of the P6. Another issue is that, of course, other processors
> might have other constraints on how this operation is performed.
To download firmware into the CPU is conceptually redefining the CPU, so
this should happen (and intel says so!) _before_ the OS starts. In any
case, the whole CPU detection and bug workaround habilitation should be
done afterwards in all cases. That is why I advocate doing this f.ex. from
lilo, not from the OS.
-- Horst von Brand vonbrand@sleipnir.valparaiso.cl Casilla 9G, Viņa del Mar, Chile +56 32 672616- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Feb 19 2000 - 01:17:29 EST