Re: REGRESSION on linux-next (next-20250919)

[Sean Christopherson <seanjc@google.com>]

On Thu, Oct 09, 2025, Dapeng Mi wrote:
> 
> On 10/7/2025 2:22 PM, Borah, Chaitanya Kumar wrote:
> > Hi,
> >
> > On 10/6/2025 1:33 PM, Borah, Chaitanya Kumar wrote:
> >> Thank you for your responses.
> >>
> >> Following change fixes the issue for us.
> >>
> >>
> >> diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
> >> index 40ac4cb44ed2..487ad19a236e 100644
> >> --- a/arch/x86/kvm/pmu.c
> >> +++ b/arch/x86/kvm/pmu.c
> >> @@ -108,16 +108,18 @@ void kvm_init_pmu_capability(const struct 
> >> kvm_pmu_ops *pmu_ops)
> >>          bool is_intel = boot_cpu_data.x86_vendor == X86_VENDOR_INTEL;
> >>          int min_nr_gp_ctrs = pmu_ops->MIN_NR_GP_COUNTERS;
> >>
> >> -       perf_get_x86_pmu_capability(&kvm_host_pmu);
> >> -
> >>          /*
> >>           * Hybrid PMUs don't play nice with virtualization without careful
> >>           * configuration by userspace, and KVM's APIs for reporting 
> >> supported
> >>           * vPMU features do not account for hybrid PMUs.  Disable vPMU 
> >> support
> >>           * for hybrid PMUs until KVM gains a way to let userspace opt-in.
> >>           */
> >> -       if (cpu_feature_enabled(X86_FEATURE_HYBRID_CPU))
> >> +       if (cpu_feature_enabled(X86_FEATURE_HYBRID_CPU)) {
> >>                  enable_pmu = false;
> >> +               memset(&kvm_host_pmu, 0, sizeof(kvm_host_pmu));
> >> +       } else {
> >> +               perf_get_x86_pmu_capability(&kvm_host_pmu);
> >> +       }
> > Can we expect a formal patch soon?
> 
> I'd like to post a patch to fix this tomorrow if Sean has no bandwidth on
> this. Thanks.

Sorry, my bad, I was waiting for you to post a patch, but that wasn't at all
clear.  So yeah, go ahead and post one :-)