Re: [PATCH v4 27/30] KVM: x86: Add KVM_VCPU_TSC_EFFECTIVE_FREQ attribute

[David Woodhouse <dwmw2@xxxxxxxxxxxxx>]

On Fri, 2026-05-22 at 05:49 -0700, Sean Christopherson wrote:
> 
> Oh, that's just an oversight, definitely not intentional.  Easy enough to fix:

Want me to roll that into the series? As you eloquently put it the
other day, what's one more patch...?

> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 1616b2eec6e7..cd4a244ca0c5 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -2235,7 +2235,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long 
> ext)
>                 r = tdp_enabled;
>                 break;
>         case KVM_CAP_X86_APIC_BUS_CYCLES_NS:
> -               r = APIC_BUS_CYCLE_NS_DEFAULT;
> +               r = kvm ? kvm->arch.apic_bus_cycle_ns : 
> APIC_BUS_CYCLE_NS_DEFAULT;
>                 break;
>         case KVM_CAP_EXIT_HYPERCALL:
>                 r = KVM_EXIT_HYPERCALL_VALID_MASK;

Please tell me that can never be zero. Because we divide by it when
reading HV_X64_MSR_APIC_FREQUENCY.

 ... checks ... it does look like it's initialised to
APIC_BUS_CYCLE_NS_DEFAULT in kvm_arch_init_vm(), and we don't allow
userspace to set it to zero.

Attachment: smime.p7s
Description: S/MIME cryptographic signature