Re: [Qemu-devel] [RFC PATCH] powerpc: add PVR mask support

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Thu, 2013-08-15 at 08:03 +0200, Alexander Graf wrote:

> >> How does the user select that he wants a v2.3 p7 cpu with this
> patch?
> > 
> > Why would he want that? The behaviour would not change because of
> the
> > version - all definitions use the same POWERPC_FAMILY(POWER7) and
> PVR is
> > not virtualized anyway.
> 
> Quite frankly I don't know what to say here. Are you trying to play
> dumb or are you just one of those totally sloppy people who don't care
> about anything outside of their own scope of work?

Can you stop the bloody personal attacks on Alexey ? It's becoming
tiresome.

He makes a very valid point. The ability to specify a specific revision
of the processor is pointless for pretty much any use case we have in
mind at the moment, and is even more pointless as long as we emulate
them all exactly the same way.

Besides, we can probably still organize the table from "more precise" to
"less precise" entries and match that way if you *really* want to have
specific entries for obscure revision of the chip.

Ben.

> With HV KVM we can not trap PVR, yes. With PR KVM we do trap PVR and
> we emulate it. With TCG we do trap PVR and we emulate it.
> 
> > May be (may be) ppc_cpu_class_by_name() needs to be fixed to try to
> find
> > the PPC CPU class with the biggest mask to choose (for example)
> > 004a0201/ffffffff rather than more common 004a0000/ffff0000 (if
> 004a0201 is
> > added to the list separately from the common definition for some
> reason).
> 
> I think the class split as Linux has it should work just fine, no?
> 
> 
> Alex