On Wed, Nov 8, 2023 at 6:39 AM Sean Christopherson <seanjc@xxxxxxxxxx> wrote:
>
> On Tue, Nov 07, 2023, Jim Mattson wrote:
> > On Tue, Nov 7, 2023 at 4:31 PM Sean Christopherson <seanjc@xxxxxxxxxx> wrote:
> > >
> > > Set the eventsel for all fixed counters during PMU initialization, the
> > > eventsel is hardcoded and consumed if and only if the counter is supported,
> > > i.e. there is no reason to redo the setup every time the PMU is refreshed.
> > >
> > > Configuring all KVM-supported fixed counter also eliminates a potential
> > > pitfall if/when KVM supports discontiguous fixed counters, in which case
> > > configuring only nr_arch_fixed_counters will be insufficient (ignoring the
> > > fact that KVM will need many other changes to support discontiguous fixed
> > > counters).
> > >
> > > Signed-off-by: Sean Christopherson <seanjc@xxxxxxxxxx>
> > > ---
> > > arch/x86/kvm/vmx/pmu_intel.c | 14 ++++----------
> > > 1 file changed, 4 insertions(+), 10 deletions(-)
> > >
> > > diff --git a/arch/x86/kvm/vmx/pmu_intel.c b/arch/x86/kvm/vmx/pmu_intel.c
> > > index c4f2c6a268e7..5fc5a62af428 100644
> > > --- a/arch/x86/kvm/vmx/pmu_intel.c
> > > +++ b/arch/x86/kvm/vmx/pmu_intel.c
> > > @@ -409,7 +409,7 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
> > > * Note, reference cycles is counted using a perf-defined "psuedo-encoding",
> > > * as there is no architectural general purpose encoding for reference cycles.
> > > */
> > > -static void setup_fixed_pmc_eventsel(struct kvm_pmu *pmu)
> > > +static u64 intel_get_fixed_pmc_eventsel(int index)
> > > {
> > > const struct {
> > > u8 eventsel;
> > > @@ -419,17 +419,11 @@ static void setup_fixed_pmc_eventsel(struct kvm_pmu *pmu)
> > > [1] = { 0x3c, 0x00 }, /* CPU Cycles/ PERF_COUNT_HW_CPU_CYCLES. */
> > > [2] = { 0x00, 0x03 }, /* Reference Cycles / PERF_COUNT_HW_REF_CPU_CYCLES*/
> > > };
> > > - int i;
> > >
> > > BUILD_BUG_ON(ARRAY_SIZE(fixed_pmc_events) != KVM_PMC_MAX_FIXED);
> > >
> > > - for (i = 0; i < pmu->nr_arch_fixed_counters; i++) {
> > > - int index = array_index_nospec(i, KVM_PMC_MAX_FIXED);
> > > - struct kvm_pmc *pmc = &pmu->fixed_counters[index];
> > > -
> > > - pmc->eventsel = (fixed_pmc_events[index].unit_mask << 8) |
> > > - fixed_pmc_events[index].eventsel;
> > > - }
> > > + return (fixed_pmc_events[index].unit_mask << 8) |
> > > + fixed_pmc_events[index].eventsel;
> >
> > Can I just say that it's really confusing that the value returned by
> > intel_get_fixed_pmc_eventsel() is the concatenation of an 8-bit "unit
> > mask" and an 8-bit "eventsel"?
>
> Heh, blame the SDM for having an "event select" field in "event select" MSRs.
>
> Is this better?
>
> const struct {
> u8 event;
> u8 unit_mask;
> } fixed_pmc_events[] = {
> [0] = { 0xc0, 0x00 }, /* Instruction Retired / PERF_COUNT_HW_INSTRUCTIONS. */
> [1] = { 0x3c, 0x00 }, /* CPU Cycles/ PERF_COUNT_HW_CPU_CYCLES. */
> [2] = { 0x00, 0x03 }, /* Reference Cycles / PERF_COUNT_HW_REF_CPU_CYCLES*/
> };
Better. Thank you.
