Re: [PATCH v11 07/11] KVM: x86: Expose MSR_IA32_PERF_CAPABILITIES for LBR record format

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 2020/5/19 18:53, Peter Zijlstra wrote:
On Thu, May 14, 2020 at 04:30:50PM +0800, Like Xu wrote:
@@ -203,6 +206,12 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
  	case MSR_CORE_PERF_GLOBAL_OVF_CTRL:
  		msr_info->data = pmu->global_ovf_ctrl;
  		return 0;
+	case MSR_IA32_PERF_CAPABILITIES:
+		if (!msr_info->host_initiated &&
+			!guest_cpuid_has(vcpu, X86_FEATURE_PDCM))
+			return 1;
I know this is KVM code, so maybe they feel differently, but I find the
above indentation massively confusing. Consider using: "set cino=:0(0"
if you're a vim user.
Nice tip and I'll apply it. Thanks.

+		msr_info->data = vcpu->arch.perf_capabilities;
+		return 0;
  	default:
  		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0))) {
  			u64 val = pmc_read_counter(pmc);
@@ -261,6 +270,16 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
  			return 0;
  		}
  		break;
+	case MSR_IA32_PERF_CAPABILITIES:
+		if (!msr_info->host_initiated ||
+			!guest_cpuid_has(vcpu, X86_FEATURE_PDCM))
+			return 1;
Idem.

+		if (!(data & ~vmx_get_perf_capabilities()))
+			return 1;
+		if ((data ^ vmx_get_perf_capabilities()) & PERF_CAP_LBR_FMT)
+			return 1;
+		vcpu->arch.perf_capabilities = data;
+		return 0;
  	default:
  		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0))) {
  			if (!msr_info->host_initiated)




[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux