Skip to contentSkip to conversation

KVM: x86/pmu: Disable vPMU if EVENTSEL_GUESTONLY bit doesn't exist

PATCH Superseded Like Xu 2 participants 2 messages Mar 7, 2023 – Apr 7, 2023

Subsystems (3)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

Like Xu

author
#0 lore raw

PATCH 1/1 KVM: x86/pmu: Disable vPMU if EVENTSEL_GUESTONLY bit doesn't exist

From: Like Xu <[email protected]>

Unlike Intel's msr atomic_switch mechanism, AMD supports guest pmu basic counter feature by setting the GUESTONLY bit on the host, so the presence or absence of this bit determines whether vPMU is emulatable (e.g. in nested virtualization). Since on AMD, writing reserved bits of EVENTSEL register does not bring #GP, KVM needs to update the global enable_pmu value by checking the persistence of this GUESTONLY bit.

---
 arch/x86/kvm/svm/svm.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)
arch/x86/kvm/svm/svm.c +14 -1
index dd21e8b1a259..f41d96e638ef 100644
@@ -4866,6 +4866,16 @@ static __init void svm_adjust_mmio_mask(void)
 	kvm_mmu_set_mmio_spte_mask(mask, mask, PT_WRITABLE_MASK | PT_USER_MASK);
 }
 
+static __init bool pmu_has_guestonly_mode(void)
+{
+	u64 value;
+
+	wrmsrl(MSR_F15H_PERF_CTL0, AMD64_EVENTSEL_GUESTONLY);
+	rdmsrl(MSR_F15H_PERF_CTL0, value);
+
+	return value == AMD64_EVENTSEL_GUESTONLY;
+}
+
 static __init void svm_set_cpu_caps(void)
 {
 	kvm_set_cpu_caps();
@@ -4911,8 +4921,11 @@ static __init void svm_set_cpu_caps(void)
 		kvm_cpu_cap_set(X86_FEATURE_VIRT_SSBD);
 
 	/* AMD PMU PERFCTR_CORE CPUID */
-	if (enable_pmu && boot_cpu_has(X86_FEATURE_PERFCTR_CORE))
+	if (enable_pmu && boot_cpu_has(X86_FEATURE_PERFCTR_CORE) &&
+	    pmu_has_guestonly_mode())
 		kvm_cpu_cap_set(X86_FEATURE_PERFCTR_CORE);
+	else
+		enable_pmu = false;
 
 	/* CPUID 0x8000001F (SME/SEV features) */
 	sev_set_cpu_caps();

base-commit: 13738a3647368f7f600b30d241779bcd2a3ebbfd

Signature · 1 line
2.39.2
#1 lore raw
Quoted 19 lines · Like Xu

From: Like Xu <[email protected]>

Unlike Intel's msr atomic_switch mechanism, AMD supports guest pmu basic counter feature by setting the GUESTONLY bit on the host, so the presence or absence of this bit determines whether vPMU is emulatable (e.g. in nested virtualization). Since on AMD, writing reserved bits of EVENTSEL register does not bring #GP, KVM needs to update the global enable_pmu value by checking the persistence of this GUESTONLY bit.

---
 arch/x86/kvm/svm/svm.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)
arch/x86/kvm/svm/svm.c +0 -0
index dd21e8b1a259..f41d96e638ef 100644

arch/x86/kvm/svm/svm.c static __init void svm_adjust_mmio_mask(void) · line 4874

Quoted 9 lines
arch/x86/kvm/svm/svm.c +6 -0
@@ -4866,6 +4866,16 @@ static __init void svm_adjust_mmio_mask(void)
 	kvm_mmu_set_mmio_spte_mask(mask, mask, PT_WRITABLE_MASK | PT_USER_MASK);
 }
 
+static __init bool pmu_has_guestonly_mode(void)
+{
+	u64 value;
+
+	wrmsrl(MSR_F15H_PERF_CTL0, AMD64_EVENTSEL_GUESTONLY);
+	rdmsrl(MSR_F15H_PERF_CTL0, value);

Preemption needs to be disabled to ensure a stable CPU. I gotta imagine KVM should be restoring the original value too.

I'm guessing I'm not going to like the answer, but is there really no better way to probe for GUESTONLY support?

Quoted 8 lines
arch/x86/kvm/svm/svm.c +4 -0
+
+	return value == AMD64_EVENTSEL_GUESTONLY;
+}
+
 static __init void svm_set_cpu_caps(void)
 {
 	kvm_set_cpu_caps();

arch/x86/kvm/svm/svm.c static __init void svm_set_cpu_caps(void) · line 4928

Quoted 9 lines
arch/x86/kvm/svm/svm.c +4 -1
@@ -4911,8 +4921,11 @@ static __init void svm_set_cpu_caps(void)
 		kvm_cpu_cap_set(X86_FEATURE_VIRT_SSBD);
 
 	/* AMD PMU PERFCTR_CORE CPUID */
-	if (enable_pmu && boot_cpu_has(X86_FEATURE_PERFCTR_CORE))
+	if (enable_pmu && boot_cpu_has(X86_FEATURE_PERFCTR_CORE) &&
+	    pmu_has_guestonly_mode())
 		kvm_cpu_cap_set(X86_FEATURE_PERFCTR_CORE);
+	else
+		enable_pmu = false;

Why does lack of AMD64_EVENTSEL_GUESTONLY disable the PMU, but if and only if X86_FEATURE_PERFCTR_CORE? E.g. why does the behavior not also apply to legacy perfmon support?

Quoted 8 lines
arch/x86/kvm/svm/svm.c +0 -0
 	/* CPUID 0x8000001F (SME/SEV features) */
 	sev_set_cpu_caps();
 

base-commit: 13738a3647368f7f600b30d241779bcd2a3ebbfd

Signature · 1 line
2.39.2