Re: [kvm-unit-tests PATCH v3 12/13] x86/pmu: Add assignment framework for Intel-specific HW resources

[Sean Christopherson <seanjc@google.com>]

On Fri, Oct 21, 2022, Like Xu wrote:
> On 6/10/2022 6:44 am, Sean Christopherson wrote:
> > On Fri, Aug 19, 2022, Like Xu wrote:
> > > @@ -142,12 +148,22 @@ static void global_disable(pmu_counter_t *cnt)
> > >   			~(1ull << cnt->idx));
> > >   }
> > > +static inline uint32_t get_gp_counter_msr(unsigned int i)
> > 
> > Rather than helpers, what about macros?  The problem with "get" is that it sounds
> > like the helper is actually reading the counter/MSR.  E.g. see MSR_IA32_MCx_CTL()
> > 
> > Something like this?
> > 
> >    MSR_PERF_GP_CTRx()
> 
> The base address msr is different for intel and amd (including K7), and
> using different macros in the same location sounds like it would require a
> helper.

I wasn't thinking different macros, I was thinking a macro that consumes the
gp_counter_base.

But actually, there's no need to use a macro, it's really just the naming
convention that I think we should use.  It obviously violates the preferred style,
but in this case I think the deviation is a net postive.

static inline uint32_t MSR_PERF_GP_CTRx(unsigned int i)
{
	return gp_counter_base + i;
}