Skip to contentSkip to conversation

KVM: x86/pmu: Add hardware Topdown metrics support

PATCH Superseded Zide Chen 3 participants 10 messages Jun 15 – Jun 18

Subsystems (5)
x86 KVM Kselftest perf events +1 more subsystems

v3 of 9 Superseded Latest is v9 →

lore
Share & tools
lore

Zide Chen

author
#0 lore raw

v3 cover KVM: x86/pmu: Add hardware Topdown metrics support

The Top-Down Microarchitecture Analysis (TMA) method is a structured approach for identifying performance bottlenecks in out-of-order processors.

Currently, guests support the TMA method by collecting Topdown events using GP counters, which may trigger multiplexing. To free up scarce GP counters, eliminate multiplexing-induced skew, and obtain coherent Topdown metric ratios, it is desirable to expose fixed counter 3 and the IA32_PERF_METRICS MSR to guests.

Several attempts have been made to virtualize this under the legacy vPMU model [1][2][3], but they were unsuccessful. With the new mediated vPMU, enabling TMA support in guests becomes much simpler. It avoids invasive changes to the perf core, eliminates CPU pinning and fixed-counter affinity issues, and reduces the latge overhead of trapping and emulating MSR accesses.

[1] https://lore.kernel.org/kvm/[email protected]/
[2] https://lore.kernel.org/all/[email protected]/T/
[3] https://lwn.net/ml/linux-kernel/[email protected]/

Tested on an SPR. Without this series, only raw topdown.*_slots events work in the guest, and metric events (e.g. cpu/topdown-bad-spec/) are not available.

With this series, metric events are visible in the guest. Run this command on both host and guest:

$ perf stat --topdown --no-metric-only -- taskset -c 2 perf bench sched messaging

Host results:

# Running 'sched/messaging' benchmark:
# 20 sender and receiver processes per group
# 10 groups == 400 processes run

     Total time: 1.500 [sec]

Performance counter stats for 'taskset -c 2 perf bench sched messaging':

     4,266,060,558      TOPDOWN.SLOTS:u              #     32.0 %  tma_frontend_bound
                                                     #      5.2 %  tma_bad_speculation
       588,397,905      topdown-retiring:u           #     13.8 %  tma_retiring
                                                     #     49.0 %  tma_backend_bound
     1,376,283,990      topdown-fe-bound:u
     2,096,827,304      topdown-be-bound:u
       217,425,841      topdown-bad-spec:u
         5,050,520      INT_MISC.UOP_DROPPING:u

Only minor changes in v3.

Rebased to kvm-x86/next: c1f730330292

v3 changes:
- patch 2/4: Move the non-contiguous counter filter code to pmu.c (Dapeng)
- patch 3/4: Replace WARN_ON() with WARN_ON_ONCE(). (Dapeng)
- patch 4/4: Change abs() with explicit bounds (sum >= 0xfd && sum <= 0x102).
- Minor comment cleanups.

v2 changes:
- As suggested by Dapeng, implement a new selftest patch.
- Don't advertise fixed counter 3 if the host doesn't support it.
- Minor change in patch 1 to remove a magic number.

v2:
https://lore.kernel.org/kvm/[email protected]/T/#u
v1:
https://lore.kernel.org/kvm/[email protected]/T/#t
QEMU:
https://lore.kernel.org/qemu-devel/[email protected]/

Dapeng Mi (2):
KVM: x86/pmu: Support Intel fixed counter 3 on mediated vPMU
KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

Zide Chen (2):
KVM: x86/pmu: Do not map fixed counters >= 3 to generic perf events
KVM: selftests: Add perf_metrics and fixed counter 3 tests

 arch/x86/include/asm/kvm_host.h               |  3 +-
 arch/x86/include/asm/msr-index.h              |  1 +
 arch/x86/include/asm/perf_event.h             |  1 +
 arch/x86/kvm/pmu.c                            | 18 +++++
 arch/x86/kvm/vmx/pmu_intel.c                  | 62 ++++++++++++----
 arch/x86/kvm/vmx/pmu_intel.h                  |  5 ++
 arch/x86/kvm/vmx/vmx.c                        |  6 ++
 arch/x86/kvm/x86.c                            | 10 ++-
 tools/arch/x86/include/asm/msr-index.h        |  1 +
 tools/testing/selftests/kvm/include/x86/pmu.h |  3 +
 .../selftests/kvm/x86/pmu_counters_test.c     | 72 +++++++++++++++++--
 11 files changed, 161 insertions(+), 21 deletions(-)
Signature · 1 line
2.54.0
author
#1 lore raw

PATCH v3 1/4 KVM: x86/pmu: Do not map fixed counters >= 3 to generic perf events

Only fixed counters 0..2 have matching generic cross-platform hardware perf events (INSTRUCTIONS, CPU_CYCLES, REF_CPU_CYCLES). Therefore, perf_get_hw_event_config() is only applicable to these counters.

KVM does not intend to emulate fixed counters >= 3 on legacy (non-mediated) vPMU, while for mediated vPMU, KVM does not care what the fixed counter event mappings are. Therefore, return 0 for their eventsel.

Also remove __always_inline as BUILD_BUG_ON() is no longer needed.

---
v2:
- Replace 3 in "if (index < 3)" with ARRAY_SIZE(fixed_pmc_perf_ids).

---
 arch/x86/kvm/vmx/pmu_intel.c | 26 ++++++++++++++------------
 1 file changed, 14 insertions(+), 12 deletions(-)
arch/x86/kvm/vmx/pmu_intel.c +14 -12
index a73a9515d96c..59b7a90c79e1 100644
@@ -464,28 +464,30 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
  * different perf_event is already utilizing the requested counter, but the end
  * result is the same (ignoring the fact that using a general purpose counter
  * will likely exacerbate counter contention).
- *
- * Forcibly inlined to allow asserting on @index at build time, and there should
- * never be more than one user.
  */
-static __always_inline u64 intel_get_fixed_pmc_eventsel(unsigned int index)
+static u64 intel_get_fixed_pmc_eventsel(unsigned int index)
 {
 	const enum perf_hw_id fixed_pmc_perf_ids[] = {
 		[0] = PERF_COUNT_HW_INSTRUCTIONS,
 		[1] = PERF_COUNT_HW_CPU_CYCLES,
 		[2] = PERF_COUNT_HW_REF_CPU_CYCLES,
 	};
-	u64 eventsel;
-
-	BUILD_BUG_ON(ARRAY_SIZE(fixed_pmc_perf_ids) != KVM_MAX_NR_INTEL_FIXED_COUNTERS);
-	BUILD_BUG_ON(index >= KVM_MAX_NR_INTEL_FIXED_COUNTERS);
+	u64 eventsel = 0;
 
 	/*
-	 * Yell if perf reports support for a fixed counter but perf doesn't
-	 * have a known encoding for the associated general purpose event.
+	 * Fixed counters 3 and above don't have corresponding generic hardware
+	 * perf event, and KVM does not intend to emulate them on non-mediated
+	 * vPMU.
 	 */
-	eventsel = perf_get_hw_event_config(fixed_pmc_perf_ids[index]);
-	WARN_ON_ONCE(!eventsel && index < kvm_pmu_cap.num_counters_fixed);
+	if (index < ARRAY_SIZE(fixed_pmc_perf_ids)) {
+		/*
+		 * Yell if perf reports support for a fixed counter but perf
+		 * doesn't have a known encoding for the associated general
+		 * purpose event.
+		 */
+		eventsel = perf_get_hw_event_config(fixed_pmc_perf_ids[index]);
+		WARN_ON_ONCE(!eventsel && index < kvm_pmu_cap.num_counters_fixed);
+	}
 	return eventsel;
 }
 
Signature · 1 line
2.54.0
author
#2 lore raw

PATCH v3 2/4 KVM: x86/pmu: Support Intel fixed counter 3 on mediated vPMU

In reply to Zide Chen

From: Dapeng Mi <[email protected]>

Starting with Ice Lake, Intel introduced fixed counter 3, which counts TOPDOWN.SLOTS - the number of available slots for an unhalted logical processor. It serves as the denominator for top-level metrics in the Top-down Microarchitecture Analysis method.

Emulating this counter on legacy vPMU would require introducing a new generic perf encoding for the Intel-specific TOPDOWN.SLOTS event in order to call perf_get_hw_event_config(). This is undesirable as it would pollute the generic perf event encoding.

Moreover, KVM does not intend to emulate IA32_PERF_METRICS in the legacy vPMU model, and without IA32_PERF_METRICS, emulating this counter has little practical value. Therefore, expose fixed counter 3 to guests only when mediated vPMU is enabled.

---
v3:
- Move the non-contiguous counter filter code to pmu.c v2:
- Don't advertise fixed counter 3 to userspace if the host doesn't support it.

---
 arch/x86/include/asm/kvm_host.h |  2 +-
 arch/x86/kvm/pmu.c              | 18 ++++++++++++++++++
 arch/x86/kvm/x86.c              |  4 ++--
 3 files changed, 21 insertions(+), 3 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -1
index 3886b536c8a5..754103e7ab4d 100644
@@ -577,7 +577,7 @@ struct kvm_pmc {
 #define KVM_MAX_NR_GP_COUNTERS		KVM_MAX(KVM_MAX_NR_INTEL_GP_COUNTERS, \
 						KVM_MAX_NR_AMD_GP_COUNTERS)
 
-#define KVM_MAX_NR_INTEL_FIXED_COUNTERS	3
+#define KVM_MAX_NR_INTEL_FIXED_COUNTERS	4
 #define KVM_MAX_NR_AMD_FIXED_COUNTERS	0
 #define KVM_MAX_NR_FIXED_COUNTERS	KVM_MAX(KVM_MAX_NR_INTEL_FIXED_COUNTERS, \
 						KVM_MAX_NR_AMD_FIXED_COUNTERS)
arch/x86/kvm/pmu.c +18 -0
index b92dd2e58335..0faf580782d5 100644
@@ -122,6 +122,8 @@ void kvm_init_pmu_capability(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;
+	union cpuid10_edx edx;
+	u32 eax, ebx, ecx;
 
 	/*
 	 * Hybrid PMUs don't play nice with virtualization without careful
@@ -169,6 +171,22 @@ void kvm_init_pmu_capability(struct kvm_pmu_ops *pmu_ops)
 	kvm_pmu_cap.num_counters_fixed = min(kvm_pmu_cap.num_counters_fixed,
 					     KVM_MAX_NR_FIXED_COUNTERS);
 
+	/*
+	 * Intel platforms may support non-contiguous fixed counters, e.g., some
+	 * E-core based server processors don't implement fixed counter 3.
+	 *
+	 * Before KVM supports non-contiguous fixed counters, make sure only
+	 * contiguous ones are retained in kvm_pmu_cap.
+	 */
+	if (kvm_host_pmu.version >= 5) {
+		cpuid(10, &eax, &ebx, &ecx, &edx.full);
+		if (kvm_pmu_cap.num_counters_fixed > edx.split.num_counters_fixed)
+			kvm_pmu_cap.num_counters_fixed = edx.split.num_counters_fixed;
+	}
+
+	if (!enable_mediated_pmu && kvm_pmu_cap.num_counters_fixed > 3)
+		kvm_pmu_cap.num_counters_fixed = 3;
+
 	kvm_pmu_eventsel.INSTRUCTIONS_RETIRED =
 		perf_get_hw_event_config(PERF_COUNT_HW_INSTRUCTIONS);
 	kvm_pmu_eventsel.BRANCH_INSTRUCTIONS_RETIRED =
arch/x86/kvm/x86.c +2 -2
index cf122b8c3210..b9cca855bc10 100644
@@ -350,7 +350,7 @@ static const u32 msrs_to_save_base[] = {
 
 static const u32 msrs_to_save_pmu[] = {
 	MSR_ARCH_PERFMON_FIXED_CTR0, MSR_ARCH_PERFMON_FIXED_CTR1,
-	MSR_ARCH_PERFMON_FIXED_CTR0 + 2,
+	MSR_ARCH_PERFMON_FIXED_CTR2, MSR_ARCH_PERFMON_FIXED_CTR3,
 	MSR_CORE_PERF_FIXED_CTR_CTRL, MSR_CORE_PERF_GLOBAL_STATUS,
 	MSR_CORE_PERF_GLOBAL_CTRL,
 	MSR_IA32_PEBS_ENABLE, MSR_IA32_DS_AREA, MSR_PEBS_DATA_CFG,
@@ -7742,7 +7742,7 @@ static void kvm_init_msr_lists(void)
 {
 	unsigned i;
 
-	BUILD_BUG_ON_MSG(KVM_MAX_NR_FIXED_COUNTERS != 3,
+	BUILD_BUG_ON_MSG(KVM_MAX_NR_FIXED_COUNTERS != 4,
 			 "Please update the fixed PMCs in msrs_to_save_pmu[]");
 
 	num_msrs_to_save = 0;
Signature · 1 line
2.54.0
author
#3 lore raw

PATCH v3 3/4 KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

In reply to Zide Chen

From: Dapeng Mi <[email protected]>

Bit 15 in IA32_PERF_CAPABILITIES indicates that the CPU provides built-in support for Topdown Microarchitecture Analysis (TMA) L1 metrics via the IA32_PERF_METRICS MSR.

Expose this capability only when mediated vPMU is enabled, as emulating IA32_PERF_METRICS in the legacy vPMU model is impractical.

Pass IA32_PERF_METRICS through to the guest only when mediated vPMU is enabled and bit 15 is set in guest IA32_PERF_CAPABILITIES. Allow kvm_pmu_{get,set}_msr() to handle this MSR for host accesses.

Save and restore this MSR on host/guest PMU context switches so that host PMU activity does not clobber the guest value, and guest state is not leaked into the host.

---
v3:
- Replace WARN_ON() with WARN_ON_ONCE(). (Dapeng)
- Add comments to explain why don't validate writes on PERF_METRICS.

---
 arch/x86/include/asm/kvm_host.h   |  1 +
 arch/x86/include/asm/msr-index.h  |  1 +
 arch/x86/include/asm/perf_event.h |  1 +
 arch/x86/kvm/vmx/pmu_intel.c      | 36 +++++++++++++++++++++++++++++++
 arch/x86/kvm/vmx/pmu_intel.h      |  5 +++++
 arch/x86/kvm/vmx/vmx.c            |  6 ++++++
 arch/x86/kvm/x86.c                |  6 +++++-
 7 files changed, 55 insertions(+), 1 deletion(-)
arch/x86/include/asm/kvm_host.h +1 -0
index 754103e7ab4d..9b03475843f3 100644
@@ -597,6 +597,7 @@ struct kvm_pmu {
 	u64 global_status_rsvd;
 	u64 reserved_bits;
 	u64 raw_event_mask;
+	u64 perf_metrics;
 	struct kvm_pmc gp_counters[KVM_MAX_NR_GP_COUNTERS];
 	struct kvm_pmc fixed_counters[KVM_MAX_NR_FIXED_COUNTERS];
 
arch/x86/include/asm/msr-index.h +1 -0
index 18c4be75e927..fdcaeb6c8352 100644
@@ -331,6 +331,7 @@
 #define PERF_CAP_PEBS_FORMAT		0xf00
 #define PERF_CAP_FW_WRITES		BIT_ULL(13)
 #define PERF_CAP_PEBS_BASELINE		BIT_ULL(14)
+#define PERF_CAP_PERF_METRICS		BIT_ULL(15)
 #define PERF_CAP_PEBS_TIMING_INFO	BIT_ULL(17)
 #define PERF_CAP_PEBS_MASK		(PERF_CAP_PEBS_TRAP | PERF_CAP_ARCH_REG | \
 					 PERF_CAP_PEBS_FORMAT | PERF_CAP_PEBS_BASELINE | \
arch/x86/include/asm/perf_event.h +1 -0
index 1eb13673e889..bc2e1cbcd9b9 100644
@@ -447,6 +447,7 @@ static inline bool is_topdown_idx(int idx)
 #define GLOBAL_STATUS_ARCH_PEBS_THRESHOLD_BIT	54
 #define GLOBAL_STATUS_ARCH_PEBS_THRESHOLD	BIT_ULL(GLOBAL_STATUS_ARCH_PEBS_THRESHOLD_BIT)
 #define GLOBAL_STATUS_PERF_METRICS_OVF_BIT	48
+#define GLOBAL_STATUS_PERF_METRICS_OVF		BIT_ULL(GLOBAL_STATUS_PERF_METRICS_OVF_BIT)
 
 #define GLOBAL_CTRL_EN_PERF_METRICS		BIT_ULL(48)
 /*
arch/x86/kvm/vmx/pmu_intel.c +36 -0
index 59b7a90c79e1..19daee29b731 100644
@@ -188,6 +188,8 @@ static bool intel_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
 	switch (msr) {
 	case MSR_CORE_PERF_FIXED_CTR_CTRL:
 		return kvm_pmu_has_perf_global_ctrl(pmu);
+	case MSR_PERF_METRICS:
+		return vcpu_has_perf_metrics(vcpu);
 	case MSR_IA32_PEBS_ENABLE:
 		ret = vcpu_get_perf_capabilities(vcpu) & PERF_CAP_PEBS_FORMAT;
 		break;
@@ -345,6 +347,10 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_CORE_PERF_FIXED_CTR_CTRL:
 		msr_info->data = pmu->fixed_ctr_ctrl;
 		break;
+	case MSR_PERF_METRICS:
+		WARN_ON_ONCE(!msr_info->host_initiated);
+		msr_info->data = pmu->perf_metrics;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		msr_info->data = pmu->pebs_enable;
 		break;
@@ -394,6 +400,15 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		if (pmu->fixed_ctr_ctrl != data)
 			reprogram_fixed_counters(pmu, data);
 		break;
+	case MSR_PERF_METRICS:
+		WARN_ON_ONCE(!msr_info->host_initiated);
+
+		/*
+		 * If TMA level 2 is not supported, bits [63:32] are reserved
+		 * and ignored on write, so no validation is needed here.
+		 */
+		pmu->perf_metrics = data;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		if (data & pmu->pebs_enable_rsvd)
 			return 1;
@@ -589,6 +604,11 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 		pmu->global_status_rsvd &=
 				~MSR_CORE_PERF_GLOBAL_OVF_CTRL_TRACE_TOPA_PMI;
 
+	if (perf_capabilities & PERF_CAP_PERF_METRICS) {
+		pmu->global_ctrl_rsvd &= ~GLOBAL_CTRL_EN_PERF_METRICS;
+		pmu->global_status_rsvd &= ~GLOBAL_STATUS_PERF_METRICS_OVF;
+	}
+
 	if (perf_capabilities & PERF_CAP_PEBS_FORMAT) {
 		if (perf_capabilities & PERF_CAP_PEBS_BASELINE) {
 			pmu->pebs_enable_rsvd = counter_rsvd;
@@ -632,6 +652,9 @@ static void intel_pmu_init(struct kvm_vcpu *vcpu)
 
 static void intel_pmu_reset(struct kvm_vcpu *vcpu)
 {
+	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
+
+	pmu->perf_metrics = 0;
 	intel_pmu_release_guest_lbr_event(vcpu);
 }
 
@@ -803,6 +826,13 @@ static void intel_mediated_pmu_load(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	u64 global_status, toggle;
 
+	/*
+	 * PERF_METRICS MSR must be restored closely after fixed counter 3
+	 * (kvm_pmu_load_guest_pmcs()).
+	 */
+	if (vcpu_has_perf_metrics(vcpu))
+		wrmsrq(MSR_PERF_METRICS, pmu->perf_metrics);
+
 	rdmsrq(MSR_CORE_PERF_GLOBAL_STATUS, global_status);
 	toggle = pmu->global_status ^ global_status;
 	if (global_status & toggle)
@@ -831,6 +861,12 @@ static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu)
 	 */
 	if (pmu->fixed_ctr_ctrl_hw)
 		wrmsrq(MSR_CORE_PERF_FIXED_CTR_CTRL, 0);
+
+	if (vcpu_has_perf_metrics(vcpu)) {
+		pmu->perf_metrics = rdpmc(INTEL_PMC_FIXED_RDPMC_METRICS);
+		if (pmu->perf_metrics)
+			wrmsrq(MSR_PERF_METRICS, 0);
+	}
 }
 
 struct kvm_pmu_ops intel_pmu_ops __initdata = {
arch/x86/kvm/vmx/pmu_intel.h +5 -0
index 5d9357640aa1..2ec547223b09 100644
@@ -40,4 +40,9 @@ struct lbr_desc {
 
 extern struct x86_pmu_lbr vmx_lbr_caps;
 
+static inline bool vcpu_has_perf_metrics(struct kvm_vcpu *vcpu)
+{
+	return !!(vcpu_get_perf_capabilities(vcpu) & PERF_CAP_PERF_METRICS);
+}
+
 #endif /* __KVM_X86_VMX_PMU_INTEL_H */
arch/x86/kvm/vmx/vmx.c +6 -0
index c548f22375ad..7cffe1619a1f 100644
@@ -4264,6 +4264,9 @@ static void vmx_recalc_pmu_msr_intercepts(struct kvm_vcpu *vcpu)
 				  MSR_TYPE_RW, intercept);
 	vmx_set_intercept_for_msr(vcpu, MSR_CORE_PERF_GLOBAL_OVF_CTRL,
 				  MSR_TYPE_RW, intercept);
+
+	vmx_set_intercept_for_msr(vcpu, MSR_PERF_METRICS, MSR_TYPE_RW,
+				  !vcpu_has_perf_metrics(vcpu));
 }
 
 static void vmx_recalc_msr_intercepts(struct kvm_vcpu *vcpu)
@@ -8088,6 +8091,9 @@ static __init u64 vmx_get_perf_capabilities(void)
 		perf_cap &= ~PERF_CAP_PEBS_BASELINE;
 	}
 
+	if (enable_mediated_pmu)
+		perf_cap |= host_perf_cap & PERF_CAP_PERF_METRICS;
+
 	return perf_cap;
 }
 
arch/x86/kvm/x86.c +5 -1
index b9cca855bc10..2ac297ba3598 100644
@@ -352,7 +352,7 @@ static const u32 msrs_to_save_pmu[] = {
 	MSR_ARCH_PERFMON_FIXED_CTR0, MSR_ARCH_PERFMON_FIXED_CTR1,
 	MSR_ARCH_PERFMON_FIXED_CTR2, MSR_ARCH_PERFMON_FIXED_CTR3,
 	MSR_CORE_PERF_FIXED_CTR_CTRL, MSR_CORE_PERF_GLOBAL_STATUS,
-	MSR_CORE_PERF_GLOBAL_CTRL,
+	MSR_CORE_PERF_GLOBAL_CTRL, MSR_PERF_METRICS,
 	MSR_IA32_PEBS_ENABLE, MSR_IA32_DS_AREA, MSR_PEBS_DATA_CFG,
 
 	/* This part of MSRs should match KVM_MAX_NR_INTEL_GP_COUNTERS. */
@@ -7679,6 +7679,10 @@ static void kvm_probe_msr_to_save(u32 msr_index)
 		     intel_pt_validate_hw_cap(PT_CAP_num_address_ranges) * 2))
 			return;
 		break;
+	case MSR_PERF_METRICS:
+		if (!(kvm_caps.supported_perf_cap & PERF_CAP_PERF_METRICS))
+			return;
+		break;
 	case MSR_ARCH_PERFMON_PERFCTR0 ...
 	     MSR_ARCH_PERFMON_PERFCTR0 + KVM_MAX_NR_GP_COUNTERS - 1:
 		if (msr_index - MSR_ARCH_PERFMON_PERFCTR0 >=
Signature · 1 line
2.54.0
#4 lore raw

Re: [PATCH V3 3/4] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider: - [High] Guest-reachable WARN_ON_ONCE assertions in MSR_PERF_METRICS accessors allow unprivileged guests to…

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] Guest-reachable WARN_ON_ONCE assertions in MSR_PERF_METRICS accessors allow unprivileged guests to trigger host-side denial of service.
- [High] Software updates to MSR_PERF_METRICS via KVM ioctls or VCPU reset are silently clobbered by stale hardware MSR values during vcpu_put().
- [High] MSR_PERF_METRICS is passed through to the guest without context switching if the guest has a PMU version of 0, bypassing host isolation. --

commit 65a5a5edf8809e596749d317875bbc842f4eb7fc
Author: Dapeng Mi <[email protected]>

KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

This commit exposes the IA32_PERF_METRICS MSR to guests when mediated vPMU is enabled. It allows host access via kvm_pmu_{get,set}_msr() and handles save/restore during PMU context switches.

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
index 05a59f4acfdd3..61102aa5e2f9e 100644

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 341

Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
@@ -335,6 +337,10 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_CORE_PERF_FIXED_CTR_CTRL:
 		msr_info->data = pmu->fixed_ctr_ctrl;
 		break;
+	case MSR_PERF_METRICS:
+		WARN_ON_ONCE(!msr_info->host_initiated);

[Severity: High]
Could this WARN_ON_ONCE() be triggered by an unprivileged guest?

If a guest forces KVM's instruction emulator to handle the MSR access, for example by executing RDMSR over an MMIO page, the emulator might process the instruction and invoke KVM's MSR handlers with host_initiated = false.

This would unconditionally trigger the warning and allow the guest to spam the host kernel log or cause a host panic if panic_on_warn is enabled.

Quoted 5 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
+		msr_info->data = pmu->perf_metrics;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		msr_info->data = pmu->pebs_enable;
 		break;

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 394

Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
@@ -384,6 +390,15 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		if (pmu->fixed_ctr_ctrl != data)
 			reprogram_fixed_counters(pmu, data);
 		break;
+	case MSR_PERF_METRICS:
+		WARN_ON_ONCE(!msr_info->host_initiated);

[Severity: High]
Can a malicious guest also trigger this WARN_ON_ONCE() by executing WRMSR over an MMIO page to force instruction emulation?

Quoted 8 lines
arch/x86/kvm/vmx/pmu_intel.c +7 -0
+
+		/*
+		 * If TMA level 2 is not supported, bits [63:32] are reserved
+		 * and ignored on write, so no validation is needed here.
+		 */
+		pmu->perf_metrics = data;
+		break;
 	case MSR_IA32_PEBS_ENABLE:

[ ... ]

Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -622,6 +642,9 @@ static void intel_pmu_init(struct kvm_vcpu *vcpu)
 
 static void intel_pmu_reset(struct kvm_vcpu *vcpu)
 {
+	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
+
+	pmu->perf_metrics = 0;
 	intel_pmu_release_guest_lbr_event(vcpu);
 }

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu) · line 856

Quoted 7 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -821,6 +851,12 @@ static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu)
 	 */
 	if (pmu->fixed_ctr_ctrl_hw)
 		wrmsrq(MSR_CORE_PERF_FIXED_CTR_CTRL, 0);
+
+	if (vcpu_has_perf_metrics(vcpu)) {
+		pmu->perf_metrics = rdpmc(INTEL_PMC_FIXED_RDPMC_METRICS);

[Severity: High]
Does this rdpmc() call silently clobber software updates made to pmu->perf_metrics?

When the VCPU is loaded during a KVM_SET_MSRS ioctl or VCPU reset, updates to pmu->perf_metrics in intel_pmu_set_msr() or intel_pmu_reset() modify the software struct but do not synchronously write to the active hardware MSR.

Later, when the VCPU is descheduled, intel_mediated_pmu_put() reads the physical MSR (which still holds the old pre-update value) and overwrites the software state.

Will this permanently discard the new software state and break live migration or VCPU resets?

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
+		if (pmu->perf_metrics)
+			wrmsrq(MSR_PERF_METRICS, 0);
+	}
 }

[ ... ]

arch/x86/kvm/vmx/vmx.c static void vmx_recalc_pmu_msr_intercepts(struct kvm_vcpu *vcpu) · line 4252

Quoted 11 lines
arch/x86/kvm/vmx/vmx.c +3 -0
index 9324d60839419..475e72c733cee 100644
@@ -4247,6 +4247,9 @@ static void vmx_recalc_pmu_msr_intercepts(struct kvm_vcpu *vcpu)
 				  MSR_TYPE_RW, intercept);
 	vmx_set_intercept_for_msr(vcpu, MSR_CORE_PERF_GLOBAL_OVF_CTRL,
 				  MSR_TYPE_RW, intercept);
+
+	vmx_set_intercept_for_msr(vcpu, MSR_PERF_METRICS, MSR_TYPE_RW,
+				  !vcpu_has_perf_metrics(vcpu));

[Severity: High]
Could clearing the intercept here allow the guest direct, unintercepted access to the host's physical MSR_PERF_METRICS register without a context switch?

If enable_mediated_pmu is true, KVM exposes PERF_CAP_PERF_METRICS. A VMM could configure a VCPU with MSR_IA32_PERF_CAPABILITIES.PERF_METRICS=1 but leave the PMU version at 0.

In this scenario, vcpu_has_perf_metrics() would return true, clearing the intercept. However, because the PMU version is 0, kvm_vcpu_has_mediated_pmu() returns false.

This causes the context-switch handlers kvm_mediated_pmu_load() and kvm_mediated_pmu_put() to exit early.

Does this bypass host isolation and allow the guest to leak or clobber host PMU state?

Quoted 3 lines
arch/x86/kvm/vmx/vmx.c +0 -0
 }
 
 static void vmx_recalc_msr_intercepts(struct kvm_vcpu *vcpu)
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=3
author
#5 lore raw

Re: [PATCH V3 3/4] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

Quoted 32 lines · On 6/15/2026 6:26 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] Guest-reachable WARN_ON_ONCE assertions in MSR_PERF_METRICS accessors allow unprivileged guests to trigger host-side denial of service.
- [High] Software updates to MSR_PERF_METRICS via KVM ioctls or VCPU reset are silently clobbered by stale hardware MSR values during vcpu_put().
- [High] MSR_PERF_METRICS is passed through to the guest without context switching if the guest has a PMU version of 0, bypassing host isolation. --

commit 65a5a5edf8809e596749d317875bbc842f4eb7fc
Author: Dapeng Mi <[email protected]>

KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

This commit exposes the IA32_PERF_METRICS MSR to guests when mediated vPMU is enabled. It allows host access via kvm_pmu_{get,set}_msr() and handles save/restore during PMU context switches.

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
index 05a59f4acfdd3..61102aa5e2f9e 100644

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 341

Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
@@ -335,6 +337,10 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_CORE_PERF_FIXED_CTR_CTRL:
 		msr_info->data = pmu->fixed_ctr_ctrl;
 		break;
+	case MSR_PERF_METRICS:
+		WARN_ON_ONCE(!msr_info->host_initiated);

[Severity: High]
Could this WARN_ON_ONCE() be triggered by an unprivileged guest?

Quoted 3 lines

If a guest forces KVM's instruction emulator to handle the MSR access, for example by executing RDMSR over an MMIO page, the emulator might process the instruction and invoke KVM's MSR handlers with host_initiated = false.

Yes, in theory, it's possible for the RDMSR/WRMSR to be handled in the emulator path and WARN_ON_ONCE() could be unexpectedly triggered.

Since WARN_ON() is not intended to handle untrusted input, it's better to use a defensive check and reject non-host-initiated accesses:

arch/x86/kvm/vmx/pmu_intel.c +0 -0
index 19daee29b731..6073227bdd20 100644
@@ -348,7 +348,8 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu,

struct msr_data *msr_info)

                msr_info->data = pmu->fixed_ctr_ctrl;
                break;
        case MSR_PERF_METRICS:
-               WARN_ON_ONCE(!msr_info->host_initiated);
+               if (!msr_info->host_initiated)
+                       return 1;
                msr_info->data = pmu->perf_metrics;
                break;
        case MSR_IA32_PEBS_ENABLE:
@@ -401,7 +402,8 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu,
struct msr_data *msr_info)
                        reprogram_fixed_counters(pmu, data);
                break;
        case MSR_PERF_METRICS:

- WARN_ON_ONCE(!msr_info->host_initiated);
+ if (!msr_info->host_initiated)
+ return 1;

Quoted 11 lines

This would unconditionally trigger the warning and allow the guest to spam the host kernel log or cause a host panic if panic_on_warn is enabled.

Quoted 5 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
+		msr_info->data = pmu->perf_metrics;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		msr_info->data = pmu->pebs_enable;
 		break;

[ ... ]

Quoted 10 lines

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 394

Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
@@ -384,6 +390,15 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		if (pmu->fixed_ctr_ctrl != data)
 			reprogram_fixed_counters(pmu, data);
 		break;
+	case MSR_PERF_METRICS:
+		WARN_ON_ONCE(!msr_info->host_initiated);

[Severity: High]
Can a malicious guest also trigger this WARN_ON_ONCE() by executing WRMSR over an MMIO page to force instruction emulation?

Ditto.

Quoted 35 lines
Quoted 8 lines
arch/x86/kvm/vmx/pmu_intel.c +7 -0
+
+		/*
+		 * If TMA level 2 is not supported, bits [63:32] are reserved
+		 * and ignored on write, so no validation is needed here.
+		 */
+		pmu->perf_metrics = data;
+		break;
 	case MSR_IA32_PEBS_ENABLE:

[ ... ]

Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -622,6 +642,9 @@ static void intel_pmu_init(struct kvm_vcpu *vcpu)
 
 static void intel_pmu_reset(struct kvm_vcpu *vcpu)
 {
+	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
+
+	pmu->perf_metrics = 0;
 	intel_pmu_release_guest_lbr_event(vcpu);
 }

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu) · line 856

Quoted 7 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -821,6 +851,12 @@ static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu)
 	 */
 	if (pmu->fixed_ctr_ctrl_hw)
 		wrmsrq(MSR_CORE_PERF_FIXED_CTR_CTRL, 0);
+
+	if (vcpu_has_perf_metrics(vcpu)) {
+		pmu->perf_metrics = rdpmc(INTEL_PMC_FIXED_RDPMC_METRICS);

[Severity: High]
Does this rdpmc() call silently clobber software updates made to pmu->perf_metrics?

Quoted 3 lines

When the VCPU is loaded during a KVM_SET_MSRS ioctl or VCPU reset, updates to pmu->perf_metrics in intel_pmu_set_msr() or intel_pmu_reset() modify the software struct but do not synchronously write to the active hardware MSR.

This is safe.

After intel_pmu_set_msr() or intel_pmu_reset(), intel_mediated_pmu_load() is called to sync the state to hardware before the next intel_mediated_pmu_put().

Quoted 39 lines

Later, when the VCPU is descheduled, intel_mediated_pmu_put() reads the physical MSR (which still holds the old pre-update value) and overwrites the software state.

Will this permanently discard the new software state and break live migration or VCPU resets?

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
+		if (pmu->perf_metrics)
+			wrmsrq(MSR_PERF_METRICS, 0);
+	}
 }

[ ... ]

arch/x86/kvm/vmx/vmx.c static void vmx_recalc_pmu_msr_intercepts(struct kvm_vcpu *vcpu) · line 4252

Quoted 11 lines
arch/x86/kvm/vmx/vmx.c +3 -0
index 9324d60839419..475e72c733cee 100644
@@ -4247,6 +4247,9 @@ static void vmx_recalc_pmu_msr_intercepts(struct kvm_vcpu *vcpu)
 				  MSR_TYPE_RW, intercept);
 	vmx_set_intercept_for_msr(vcpu, MSR_CORE_PERF_GLOBAL_OVF_CTRL,
 				  MSR_TYPE_RW, intercept);
+
+	vmx_set_intercept_for_msr(vcpu, MSR_PERF_METRICS, MSR_TYPE_RW,
+				  !vcpu_has_perf_metrics(vcpu));

[Severity: High]
Could clearing the intercept here allow the guest direct, unintercepted access to the host's physical MSR_PERF_METRICS register without a context switch?

If enable_mediated_pmu is true, KVM exposes PERF_CAP_PERF_METRICS. A VMM could configure a VCPU with MSR_IA32_PERF_CAPABILITIES.PERF_METRICS=1 but leave the PMU version at 0.

In this scenario, vcpu_has_perf_metrics() would return true, clearing the intercept. However, because the PMU version is 0, kvm_vcpu_has_mediated_pmu() returns false.

Quoted 5 lines

This causes the context-switch handlers kvm_mediated_pmu_load() and kvm_mediated_pmu_put() to exit early.

Does this bypass host isolation and allow the guest to leak or clobber host PMU state?

Yes, you are right.

arch/x86/kvm/vmx/vmx.c +0 -0
index 7cffe1619a1f..28027d302d90 100644
@@ -4266,7 +4266,7 @@ static void vmx_recalc_pmu_msr_intercepts(struct

kvm_vcpu *vcpu)
MSR_TYPE_RW, intercept);

vmx_set_intercept_for_msr(vcpu, MSR_PERF_METRICS, MSR_TYPE_RW,
- !vcpu_has_perf_metrics(vcpu));
+ !has_mediated_pmu ||
!vcpu_has_perf_metrics(vcpu));
}

static void vmx_recalc_msr_intercepts(struct kvm_vcpu *vcpu)

Quoted 4 lines
Quoted 3 lines
arch/x86/kvm/vmx/vmx.c +0 -0
 }
 
 static void vmx_recalc_msr_intercepts(struct kvm_vcpu *vcpu)
#6 lore raw

Re: [PATCH V3 3/4] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

Quoted 57 lines · Zide
Quoted 30 lines · On 6/15/2026 6:26 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] Guest-reachable WARN_ON_ONCE assertions in MSR_PERF_METRICS accessors allow unprivileged guests to trigger host-side denial of service.
- [High] Software updates to MSR_PERF_METRICS via KVM ioctls or VCPU reset are silently clobbered by stale hardware MSR values during vcpu_put().
- [High] MSR_PERF_METRICS is passed through to the guest without context switching if the guest has a PMU version of 0, bypassing host isolation. --

commit 65a5a5edf8809e596749d317875bbc842f4eb7fc
Author: Dapeng Mi <[email protected]>

KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

This commit exposes the IA32_PERF_METRICS MSR to guests when mediated vPMU is enabled. It allows host access via kvm_pmu_{get,set}_msr() and handles save/restore during PMU context switches.

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
index 05a59f4acfdd3..61102aa5e2f9e 100644

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 341

Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
@@ -335,6 +337,10 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_CORE_PERF_FIXED_CTR_CTRL:
 		msr_info->data = pmu->fixed_ctr_ctrl;
 		break;
+	case MSR_PERF_METRICS:
+		WARN_ON_ONCE(!msr_info->host_initiated);

[Severity: High]
Could this WARN_ON_ONCE() be triggered by an unprivileged guest?

Quoted 3 lines

If a guest forces KVM's instruction emulator to handle the MSR access, for example by executing RDMSR over an MMIO page, the emulator might process the instruction and invoke KVM's MSR handlers with host_initiated = false.

Yes, in theory, it's possible for the RDMSR/WRMSR to be handled in the emulator path and WARN_ON_ONCE() could be unexpectedly triggered.

Since WARN_ON() is not intended to handle untrusted input, it's better to use a defensive check and reject non-host-initiated accesses:

arch/x86/kvm/vmx/pmu_intel.c +2 -1
index 19daee29b731..6073227bdd20 100644
@@ -348,7 +348,8 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu,
 struct msr_data *msr_info)
                msr_info->data = pmu->fixed_ctr_ctrl;
                break;
        case MSR_PERF_METRICS:
-               WARN_ON_ONCE(!msr_info->host_initiated);
+               if (!msr_info->host_initiated)
+                       return 1;
                msr_info->data = pmu->perf_metrics;
                break;

case MSR_IA32_PEBS_ENABLE:

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, · line 407

Quoted 7 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -1
@@ -401,7 +402,8 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu,
 struct msr_data *msr_info)
                        reprogram_fixed_counters(pmu, data);
                break;
        case MSR_PERF_METRICS:
-               WARN_ON_ONCE(!msr_info->host_initiated);
+               if (!msr_info->host_initiated)
+                       return 1;

Per my understanding, WARN_ON_ONCE() only prints the warning message once and it should not spam Host log too much, but it's fine for me to remove them. Thanks.

Quoted 129 lines
Quoted 13 lines

This would unconditionally trigger the warning and allow the guest to spam the host kernel log or cause a host panic if panic_on_warn is enabled.

Quoted 5 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
+		msr_info->data = pmu->perf_metrics;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		msr_info->data = pmu->pebs_enable;
 		break;

[ ... ]

Quoted 9 lines

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 394

Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2 -0
@@ -384,6 +390,15 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		if (pmu->fixed_ctr_ctrl != data)
 			reprogram_fixed_counters(pmu, data);
 		break;
+	case MSR_PERF_METRICS:
+		WARN_ON_ONCE(!msr_info->host_initiated);

[Severity: High]
Can a malicious guest also trigger this WARN_ON_ONCE() by executing WRMSR over an MMIO page to force instruction emulation?

Ditto.

Quoted 32 lines
Quoted 8 lines
arch/x86/kvm/vmx/pmu_intel.c +7 -0
+
+		/*
+		 * If TMA level 2 is not supported, bits [63:32] are reserved
+		 * and ignored on write, so no validation is needed here.
+		 */
+		pmu->perf_metrics = data;
+		break;
 	case MSR_IA32_PEBS_ENABLE:

[ ... ]

Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -622,6 +642,9 @@ static void intel_pmu_init(struct kvm_vcpu *vcpu)
 
 static void intel_pmu_reset(struct kvm_vcpu *vcpu)
 {
+	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
+
+	pmu->perf_metrics = 0;
 	intel_pmu_release_guest_lbr_event(vcpu);
 }

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu) · line 856

Quoted 7 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -821,6 +851,12 @@ static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu)
 	 */
 	if (pmu->fixed_ctr_ctrl_hw)
 		wrmsrq(MSR_CORE_PERF_FIXED_CTR_CTRL, 0);
+
+	if (vcpu_has_perf_metrics(vcpu)) {
+		pmu->perf_metrics = rdpmc(INTEL_PMC_FIXED_RDPMC_METRICS);

[Severity: High]
Does this rdpmc() call silently clobber software updates made to pmu->perf_metrics?

Quoted 3 lines

When the VCPU is loaded during a KVM_SET_MSRS ioctl or VCPU reset, updates to pmu->perf_metrics in intel_pmu_set_msr() or intel_pmu_reset() modify the software struct but do not synchronously write to the active hardware MSR.

This is safe.

After intel_pmu_set_msr() or intel_pmu_reset(), intel_mediated_pmu_load() is called to sync the state to hardware before the next intel_mediated_pmu_put().

Quoted 37 lines

Later, when the VCPU is descheduled, intel_mediated_pmu_put() reads the physical MSR (which still holds the old pre-update value) and overwrites the software state.

Will this permanently discard the new software state and break live migration or VCPU resets?

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
+		if (pmu->perf_metrics)
+			wrmsrq(MSR_PERF_METRICS, 0);
+	}
 }

[ ... ]

arch/x86/kvm/vmx/vmx.c static void vmx_recalc_pmu_msr_intercepts(struct kvm_vcpu *vcpu) · line 4252

Quoted 11 lines
arch/x86/kvm/vmx/vmx.c +3 -0
index 9324d60839419..475e72c733cee 100644
@@ -4247,6 +4247,9 @@ static void vmx_recalc_pmu_msr_intercepts(struct kvm_vcpu *vcpu)
 				  MSR_TYPE_RW, intercept);
 	vmx_set_intercept_for_msr(vcpu, MSR_CORE_PERF_GLOBAL_OVF_CTRL,
 				  MSR_TYPE_RW, intercept);
+
+	vmx_set_intercept_for_msr(vcpu, MSR_PERF_METRICS, MSR_TYPE_RW,
+				  !vcpu_has_perf_metrics(vcpu));

[Severity: High]
Could clearing the intercept here allow the guest direct, unintercepted access to the host's physical MSR_PERF_METRICS register without a context switch?

If enable_mediated_pmu is true, KVM exposes PERF_CAP_PERF_METRICS. A VMM could configure a VCPU with MSR_IA32_PERF_CAPABILITIES.PERF_METRICS=1 but leave the PMU version at 0.

In this scenario, vcpu_has_perf_metrics() would return true, clearing the intercept. However, because the PMU version is 0, kvm_vcpu_has_mediated_pmu() returns false.

Quoted 5 lines

This causes the context-switch handlers kvm_mediated_pmu_load() and kvm_mediated_pmu_put() to exit early.

Does this bypass host isolation and allow the guest to leak or clobber host PMU state?

Yes, you are right.

arch/x86/kvm/vmx/vmx.c +1 -1
index 7cffe1619a1f..28027d302d90 100644
@@ -4266,7 +4266,7 @@ static void vmx_recalc_pmu_msr_intercepts(struct
 kvm_vcpu *vcpu)
                                  MSR_TYPE_RW, intercept);
 
        vmx_set_intercept_for_msr(vcpu, MSR_PERF_METRICS, MSR_TYPE_RW,
-                                 !vcpu_has_perf_metrics(vcpu));
+                                 !has_mediated_pmu ||
 !vcpu_has_perf_metrics(vcpu));
 }

static void vmx_recalc_msr_intercepts(struct kvm_vcpu *vcpu)

Quoted 3 lines
Quoted 3 lines
arch/x86/kvm/vmx/vmx.c +0 -0
 }
 
 static void vmx_recalc_msr_intercepts(struct kvm_vcpu *vcpu)
author
#7 lore raw

PATCH v3 4/4 KVM: selftests: Add perf_metrics and fixed counter 3 tests

In reply to Zide Chen

Add a selftest to exercise IA32_PERF_METRICS, i.e. architectural support for Topdown (TMA) Level 1 metrics, enumerated by IA32_PERF_CAPABILITIES[15].

Only check for non-zero metrics, as they are derived and depend on the workload, CPU model, and host scheduling, making precise expectations fragile.

Extend the PMU selftest to cover Intel fixed counter 3 by bumping MAX_NR_FIXED_COUNTERS to 4 and validating basic functionality.

---
v3:
- Slightly reword comment to explain the sum of topdown metrics is close to 100%.
- Change abs() with explicit bounds (sum >= 0xfd && sum <= 0x102) for better readability.
v2:
- New patch.
---

---
 tools/arch/x86/include/asm/msr-index.h        |  1 +
 tools/testing/selftests/kvm/include/x86/pmu.h |  3 +
 .../selftests/kvm/x86/pmu_counters_test.c     | 72 +++++++++++++++++--
 3 files changed, 71 insertions(+), 5 deletions(-)
tools/arch/x86/include/asm/msr-index.h +1 -0
index eff29645719b..e7745e2cd543 100644
@@ -331,6 +331,7 @@
 #define PERF_CAP_PEBS_FORMAT		0xf00
 #define PERF_CAP_FW_WRITES		BIT_ULL(13)
 #define PERF_CAP_PEBS_BASELINE		BIT_ULL(14)
+#define PERF_CAP_PERF_METRICS		BIT_ULL(15)
 #define PERF_CAP_PEBS_TIMING_INFO	BIT_ULL(17)
 #define PERF_CAP_PEBS_MASK		(PERF_CAP_PEBS_TRAP | PERF_CAP_ARCH_REG | \
 					 PERF_CAP_PEBS_FORMAT | PERF_CAP_PEBS_BASELINE | \
tools/testing/selftests/kvm/include/x86/pmu.h +3 -0
index 608ed83d7c6a..6c19503e0bb7 100644
@@ -52,6 +52,9 @@
 /* Fixed PMC controls, Intel only. */
 #define FIXED_PMC_GLOBAL_CTRL_ENABLE(_idx)	BIT_ULL((32 + (_idx)))
 
+/* PERF_METRICS enable, Intel only. */
+#define PERF_METRICS_GLOBAL_CTRL_ENABLE		BIT_ULL(48)
+
 #define FIXED_PMC_KERNEL			BIT_ULL(0)
 #define FIXED_PMC_USER				BIT_ULL(1)
 #define FIXED_PMC_ANYTHREAD			BIT_ULL(2)
tools/testing/selftests/kvm/x86/pmu_counters_test.c +67 -5

127 lines not on this page · raw message

Signature · 1 line
2.54.0
#8 lore raw

Re: [PATCH V3 4/4] KVM: selftests: Add perf_metrics and fixed counter 3 tests

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] A malicious or misconfigured guest can trigger a host-side WARN_ON_ONCE by forcing KVM to emulate an RDMSR or WRMSR to MSR_PERF_METRICS, leading to log spam or a host kernel panic. --

commit 2e12d71150f3902b3e1f245bbcb48d6fe2235b4a
Author: Zide Chen <[email protected]>

KVM: selftests: Add perf_metrics and fixed counter 3 tests

This commit adds a selftest to exercise IA32_PERF_METRICS support for Topdown (TMA) Level 1 metrics. It checks for non-zero metrics derived from the workload and extends the PMU selftest to cover Intel fixed counter 3 by bumping MAX_NR_FIXED_COUNTERS to 4.

Quoted 4 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0 -0
index dc6afac3aa919..d4688a7d1e556 100644

[ ... ]

Quoted 7 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +3 -0
@@ -318,6 +322,56 @@ static void guest_test_arch_event(u8 idx)
 				FIXED_PMC_GLOBAL_CTRL_ENABLE(i));
 }
 
+static void guest_test_perf_metrics(void)
+{
+	int retiring, bad_spec, fe_bound, be_bound, sum;

tools/testing/selftests/kvm/x86/pmu_counters_test.c static void guest_test_arch_event(u8 idx) · line 337

Quoted 10 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +10 -0
+	u64 global_ctrl, metrics;
+
+	if ((guest_get_pmu_version() < 2) ||	/* Does guest have GLOBAL_CTRL? */
+	    !this_cpu_has(X86_FEATURE_PDCM) ||
+	    !(rdmsr(MSR_IA32_PERF_CAPABILITIES) & PERF_CAP_PERF_METRICS))
+		return;
+
+	wrmsr(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+	wrmsr(MSR_CORE_PERF_FIXED_CTR3, 0);
+	wrmsr(MSR_PERF_METRICS, 0);

[Severity: High]
This is a pre-existing issue, but since this test now exercises MSR_PERF_METRICS, is it possible for a guest to trigger a host-side WARN_ON_ONCE by forcing KVM to emulate RDMSR or WRMSR to MSR_PERF_METRICS?

Looking at KVM's emulation path in arch/x86/kvm/vmx/pmu_intel.c, intel_pmu_get_msr() and intel_pmu_set_msr() both contain a WARN_ON_ONCE:

intel_pmu_get_msr() {
    ...
    case MSR_PERF_METRICS:
        WARN_ON_ONCE(!msr_info->host_initiated);
        msr_info->data = pmu->perf_metrics;
        break;
}

If a malicious guest forces instruction emulation for RDMSR or WRMSR targeting this MSR (for example, by using the KVM_FEP forced emulation prefix, or by executing the instruction from a page that triggers an EPT violation on fetch), the instruction is processed by KVM's software emulator.

The emulator calls kvm_emulate_msr_read() or kvm_emulate_msr_write(), which explicitly set msr_info.host_initiated = false.

Could this trigger the WARN_ON_ONCE in intel_pmu_get_msr() or intel_pmu_set_msr() and allow an unprivileged guest to spam the host kernel log or panic the host if panic_on_warn is enabled?

Quoted 10 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +10 -0
+
+	/* Enable fixed ctr3 (TOPDOWN.SLOTS) and PERF_METRICS. */
+	wrmsr(MSR_CORE_PERF_FIXED_CTR_CTRL, FIXED_PMC_CTRL(3, FIXED_PMC_KERNEL));
+	global_ctrl = FIXED_PMC_GLOBAL_CTRL_ENABLE(3) |
+		      PERF_METRICS_GLOBAL_CTRL_ENABLE;
+
+	GUEST_RUN_PAYLOAD(MSR_CORE_PERF_GLOBAL_CTRL, global_ctrl, "");
+
+	/* Check test results. */
+	metrics = rdmsr(MSR_PERF_METRICS);
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=4
author
#9 lore raw

Re: [PATCH V3 4/4] KVM: selftests: Add perf_metrics and fixed counter 3 tests

Quoted 60 lines · On 6/15/2026 6:26 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] A malicious or misconfigured guest can trigger a host-side WARN_ON_ONCE by forcing KVM to emulate an RDMSR or WRMSR to MSR_PERF_METRICS, leading to log spam or a host kernel panic. --

commit 2e12d71150f3902b3e1f245bbcb48d6fe2235b4a
Author: Zide Chen <[email protected]>

KVM: selftests: Add perf_metrics and fixed counter 3 tests

This commit adds a selftest to exercise IA32_PERF_METRICS support for Topdown (TMA) Level 1 metrics. It checks for non-zero metrics derived from the workload and extends the PMU selftest to cover Intel fixed counter 3 by bumping MAX_NR_FIXED_COUNTERS to 4.

Quoted 4 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0 -0
index dc6afac3aa919..d4688a7d1e556 100644

[ ... ]

Quoted 7 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +3 -0
@@ -318,6 +322,56 @@ static void guest_test_arch_event(u8 idx)
 				FIXED_PMC_GLOBAL_CTRL_ENABLE(i));
 }
 
+static void guest_test_perf_metrics(void)
+{
+	int retiring, bad_spec, fe_bound, be_bound, sum;

tools/testing/selftests/kvm/x86/pmu_counters_test.c static void guest_test_arch_event(u8 idx) · line 337

Quoted 10 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +10 -0
+	u64 global_ctrl, metrics;
+
+	if ((guest_get_pmu_version() < 2) ||	/* Does guest have GLOBAL_CTRL? */
+	    !this_cpu_has(X86_FEATURE_PDCM) ||
+	    !(rdmsr(MSR_IA32_PERF_CAPABILITIES) & PERF_CAP_PERF_METRICS))
+		return;
+
+	wrmsr(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+	wrmsr(MSR_CORE_PERF_FIXED_CTR3, 0);
+	wrmsr(MSR_PERF_METRICS, 0);

[Severity: High]
This is a pre-existing issue, but since this test now exercises MSR_PERF_METRICS, is it possible for a guest to trigger a host-side WARN_ON_ONCE by forcing KVM to emulate RDMSR or WRMSR to MSR_PERF_METRICS?

Looking at KVM's emulation path in arch/x86/kvm/vmx/pmu_intel.c, intel_pmu_get_msr() and intel_pmu_set_msr() both contain a WARN_ON_ONCE:

intel_pmu_get_msr() {
    ...
    case MSR_PERF_METRICS:
        WARN_ON_ONCE(!msr_info->host_initiated);
        msr_info->data = pmu->perf_metrics;
        break;
}

If a malicious guest forces instruction emulation for RDMSR or WRMSR targeting this MSR (for example, by using the KVM_FEP forced emulation prefix, or by executing the instruction from a page that triggers an EPT violation on fetch), the instruction is processed by KVM's software emulator.

Quoted 6 lines

The emulator calls kvm_emulate_msr_read() or kvm_emulate_msr_write(), which explicitly set msr_info.host_initiated = false.

Could this trigger the WARN_ON_ONCE in intel_pmu_get_msr() or intel_pmu_set_msr() and allow an unprivileged guest to spam the host kernel log or panic the host if panic_on_warn is enabled?

Yes, it's good to change this in patch 3/4 to avoid the WARN_ON_ONCE().

case MSR_PERF_METRICS:
- WARN_ON_ONCE(!msr_info->host_initiated);
+ if (!msr_info->host_initiated)
+ return 1;

Quoted 12 lines
Quoted 11 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +10 -0
+
+	/* Enable fixed ctr3 (TOPDOWN.SLOTS) and PERF_METRICS. */
+	wrmsr(MSR_CORE_PERF_FIXED_CTR_CTRL, FIXED_PMC_CTRL(3, FIXED_PMC_KERNEL));
+	global_ctrl = FIXED_PMC_GLOBAL_CTRL_ENABLE(3) |
+		      PERF_METRICS_GLOBAL_CTRL_ENABLE;
+
+	GUEST_RUN_PAYLOAD(MSR_CORE_PERF_GLOBAL_CTRL, global_ctrl, "");
+
+	/* Check test results. */
+	metrics = rdmsr(MSR_PERF_METRICS);