Re: [PATCH V6 2/5] perf/x86/kvm: Avoid unnecessary work in guest filtering

[Liang, Kan]

On 2/4/2019 2:43 PM, Borislav Petkov wrote:
> On Mon, Feb 04, 2019 at 01:57:49PM -0500, Liang, Kan wrote:
> > You mean a given microcode revision X applying to multiple stepping,
> > right?
>
> Yes, the range thing. You specify a range of steppings:
>
> kabylake mobile  with steppings 9-12
> kabylake desktop with steppings 10-13
>
> > I don't think so. I still think the KABYLAKE case is an uncommon case.
>
> You mean it is uncommon because there are already *two* models which
> need it or because this is only kabylake and it won't happen in the
> future?

The latter.
As you can see in the patch series, for other platforms (from core to 
atom), the microcode revision is always different on different stepping 
of the same model.
I believe the kabylake thing is one off.

> > Can we do something as below just for this case?
>
> Of course not. _FOUR is just silly and if a *fifth* stepping appears,
> you need to go fixup again.
>
> If anything and if this kabylake thing is one off, I'd prefer we keep it
> as is.

OK. I will keep it as is on V7.

Thanks,
Kan