Skip to contentSkip to conversation

KVM: x86/pmu: Add hardware Topdown metrics support

PATCH Superseded Zide Chen 5 participants 24 messages Aug 17 – Sep 16

Subsystems (4)

v8 of 9 Superseded Latest is v9 →

lore
Share & tools
lore

Zide Chen

author
#0 lore raw

v8 cover KVM: x86/pmu: Add hardware Topdown metrics support

The Topdown 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 large 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 a Sapphire Rapids. 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:

     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

Rebased to kvm-x86/next: 1b731e5ded480bd1

v8 changes:
- patch 5/8: Intercept global ctrl MSR if host supports PerfMetrics but not enabled in the guest.
- patch 3/9: Remove include "cpuid.h" from pmu_intel.h.
- Drop patch v7 7/9 to audit CPU RDPMC encodings.
v7 changes:
- Add "reviewed-by" tags.
- patch 5/8: Optimize bit 48 handling in global_status_rsvd; document the PERF_METRICS/FC3 write-order exception;
add nested_vmx_merge_msr_bitmaps_rw().
- patch 7/9: new patch to restrict RDPMC passthrough to known CPUs.
- patch 9/9: Use RDPMC instead of RDMSR in PERF_METRICS sanity test. v6 changes:
- patch 6/8: New patch to refactor rdpmc emulation code.
- patch 7/8: More strict handling of RDPMC ECX argument.
- patch 8/8: Move perf metrics out of test_arch_events().
- patch 2/8: Minor fix of comments.
v5 changes:
- patch 3,5,6/7: new patches to handle RDPMC on metrics.
- patch 6/7: remove host_initiated check.
v4 changes:
- patch 3/4: Remove WARN_ON_ONCE() and simply reject the guest accesses by checking host_initiated. (Sashiko)
- patch 3/4: Passthru MSR_PERF_METRICS only if has_mediated_pmu is true. (Sashiko)
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.

v7: https://lore.kernel.org/kvm/[email protected]/
v6: https://lore.kernel.org/kvm/[email protected]/T/#t
v5: https://lore.kernel.org/kvm/[email protected]/
v4: https://lore.kernel.org/kvm/[email protected]/
v3: https://lore.kernel.org/kvm/[email protected]/T/#t
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

Mingwei Zhang (1):
KVM: x86/pmu: Snapshot host IA32_PERF_CAPABILITIES in kvm_host

Zide Chen (5):
KVM: x86/pmu: Do not map fixed counters >= 3 to generic perf events
KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h
KVM: x86/pmu: Move RDPMC emulation into per-vendor callbacks
KVM: x86/pmu: Emulate RDPMC on performance metrics
KVM: selftests: Add PERF_METRICS and fixed counter 3 tests

 arch/x86/include/asm/kvm-x86-pmu-ops.h        |  2 +-
 arch/x86/include/asm/kvm_host.h               |  4 +-
 arch/x86/include/asm/msr-index.h              |  1 +
 arch/x86/kvm/msrs.c                           | 10 +-
 arch/x86/kvm/pmu.c                            | 36 +++++--
 arch/x86/kvm/pmu.h                            | 17 +++-
 arch/x86/kvm/svm/pmu.c                        | 13 ++-
 arch/x86/kvm/vmx/nested.c                     |  2 +
 arch/x86/kvm/vmx/pmu_intel.c                  | 93 ++++++++++++++-----
 arch/x86/kvm/vmx/pmu_intel.h                  | 12 +--
 arch/x86/kvm/vmx/vmx.c                        | 15 +--
 arch/x86/kvm/x86.c                            |  4 +
 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     | 90 +++++++++++++++++-
 15 files changed, 239 insertions(+), 64 deletions(-)
Signature · 1 line
2.55.0
author
#1 lore raw

PATCH v8 1/8 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.

The two BUILD_BUG_ON() checks are no longer needed, so drop them along with __always_inline.

---
v6:
- Re-arrange the code for early return. Clearer.
v2:
- Replace 3 in "if (index < 3)" with ARRAY_SIZE(fixed_pmc_perf_ids).

---
 arch/x86/kvm/vmx/pmu_intel.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)
arch/x86/kvm/vmx/pmu_intel.c +8 -6
index bfa8612fb450..df51a17b6e17 100644
@@ -465,11 +465,8 @@ 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,
@@ -478,8 +475,13 @@ static __always_inline u64 intel_get_fixed_pmc_eventsel(unsigned int index)
 	};
 	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);
+	/*
+	 * Fixed counters 3 and above don't have a corresponding generic
+	 * hardware perf event, and KVM does not intend to emulate them on
+	 * non-mediated vPMU.
+	 */
+	if (index >= ARRAY_SIZE(fixed_pmc_perf_ids))
+		return 0;
 
 	/*
 	 * Yell if perf reports support for a fixed counter but perf doesn't
Signature · 1 line
2.55.0
author
#2 lore raw

PATCH v8 2/8 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.

---
v6:
- Update comments to replace 2 with KVM_MAX_NR_INTEL_FIXED_COUNTERS - 1. 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/msrs.c             |  4 ++--
 arch/x86/kvm/pmu.c              | 19 ++++++++++++++++++-
 3 files changed, 21 insertions(+), 4 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -1
index 283847619ff8..8a4e542218a7 100644
@@ -562,7 +562,7 @@ 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/msrs.c +2 -2
index 66fa7140d65d..1fbd8f08a7f0 100644
@@ -217,7 +217,7 @@ 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,
@@ -2671,7 +2671,7 @@ 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;
arch/x86/kvm/pmu.c +18 -1
index a7d60c8785cd..913a0ecbf260 100644
@@ -19,6 +19,7 @@
 #include <linux/moduleparam.h>
 #include <asm/perf_event.h>
 #include <asm/cpu_device_id.h>
+#include <asm/cpuid/api.h>
 #include "x86.h"
 #include "cpuid.h"
 #include "lapic.h"
@@ -99,7 +100,8 @@ static const struct x86_cpu_id vmx_pebs_pdist_cpu[] = {
  *        all perf counters (both gp and fixed). The mapping relationship
  *        between pmc and perf counters is as the following:
  *        * Intel: [0 .. KVM_MAX_NR_INTEL_GP_COUNTERS-1] <=> gp counters
- *                 [KVM_FIXED_PMC_BASE_IDX .. KVM_FIXED_PMC_BASE_IDX + 2] <=> fixed
+ *                 [KVM_FIXED_PMC_BASE_IDX .. KVM_FIXED_PMC_BASE_IDX +
+ *                  KVM_MAX_NR_INTEL_FIXED_COUNTERS - 1] <=> fixed
  *        * AMD:   [0 .. AMD64_NUM_COUNTERS-1] and, for families 15H
  *          and later, [0 .. AMD64_NUM_COUNTERS_CORE-1] <=> gp counters
  */
@@ -134,6 +136,8 @@ 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
@@ -181,6 +185,19 @@ 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);
 
+	/*
+	 * Currently, KVM doesn't support non-contiguous fixed counters; make
+	 * sure only contiguous ones are retained in kvm_pmu_cap.
+	 */
+	if (kvm_host_pmu.version >= 5) {
+		cpuid(0xa, &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 =
Signature · 1 line
2.55.0
author
#3 lore raw

PATCH v8 3/8 KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h

In reply to Zide Chen

This is in preparation for it to be called from common x86 code, for example kvm_need_rdpmc_intercept(), to check the guest's PERF_METRICS capability.

Rename it to kvm_vcpu_get_perf_caps() to indicate that it's part of the common API, and shorten _capabilities to _caps.

No functional change intended.

---
v8:
- remove include "cpuid.h" from pmu_intel.h.
v5: new patch.

---
 arch/x86/kvm/pmu.h           |  8 ++++++++
 arch/x86/kvm/vmx/pmu_intel.c |  6 +++---
 arch/x86/kvm/vmx/pmu_intel.h | 12 +-----------
 3 files changed, 12 insertions(+), 14 deletions(-)
arch/x86/kvm/pmu.h +8 -0
index 090c9bbb74f4..82f955e5a450 100644
@@ -271,6 +271,14 @@ static inline bool kvm_pmu_is_fastpath_emulation_allowed(struct kvm_vcpu *vcpu)
 				  X86_PMC_IDX_MAX);
 }
 
+static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu)
+{
+	if (!guest_cpu_cap_has(vcpu, X86_FEATURE_PDCM))
+		return 0;
+
+	return vcpu->arch.perf_capabilities;
+}
+
 void kvm_pmu_deliver_pmi(struct kvm_vcpu *vcpu);
 int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned pmc, u64 *data);
 int kvm_pmu_check_rdpmc_early(struct kvm_vcpu *vcpu, unsigned int idx);
arch/x86/kvm/vmx/pmu_intel.c +3 -3
index df51a17b6e17..100c072b7fb3 100644
@@ -190,13 +190,13 @@ static bool intel_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
 	case MSR_CORE_PERF_FIXED_CTR_CTRL:
 		return kvm_pmu_has_perf_global_ctrl(pmu);
 	case MSR_IA32_PEBS_ENABLE:
-		ret = vcpu_get_perf_capabilities(vcpu) & PERF_CAP_PEBS_FORMAT;
+		ret = kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PEBS_FORMAT;
 		break;
 	case MSR_IA32_DS_AREA:
 		ret = guest_cpu_cap_has(vcpu, X86_FEATURE_DS);
 		break;
 	case MSR_PEBS_DATA_CFG:
-		perf_capabilities = vcpu_get_perf_capabilities(vcpu);
+		perf_capabilities = kvm_vcpu_get_perf_caps(vcpu);
 		ret = (perf_capabilities & PERF_CAP_PEBS_BASELINE) &&
 			((perf_capabilities & PERF_CAP_PEBS_FORMAT) > 3);
 		break;
@@ -551,7 +551,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 		pmu->raw_event_mask |= (HSW_IN_TX|HSW_IN_TX_CHECKPOINTED);
 	}
 
-	perf_capabilities = vcpu_get_perf_capabilities(vcpu);
+	perf_capabilities = kvm_vcpu_get_perf_caps(vcpu);
 	if (intel_pmu_lbr_is_compatible(vcpu) &&
 	    (perf_capabilities & PERF_CAP_LBR_FMT))
 		memcpy(&lbr_desc->records, &vmx_lbr_caps, sizeof(vmx_lbr_caps));
arch/x86/kvm/vmx/pmu_intel.h +1 -11
index 5d9357640aa1..dd447d8b6fdd 100644
@@ -4,19 +4,9 @@
 
 #include <linux/kvm_host.h>
 
-#include "cpuid.h"
-
-static inline u64 vcpu_get_perf_capabilities(struct kvm_vcpu *vcpu)
-{
-	if (!guest_cpu_cap_has(vcpu, X86_FEATURE_PDCM))
-		return 0;
-
-	return vcpu->arch.perf_capabilities;
-}
-
 static inline bool fw_writes_is_enabled(struct kvm_vcpu *vcpu)
 {
-	return (vcpu_get_perf_capabilities(vcpu) & PERF_CAP_FW_WRITES) != 0;
+	return (kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_FW_WRITES) != 0;
 }
 
 bool intel_pmu_lbr_is_enabled(struct kvm_vcpu *vcpu);
Signature · 1 line
2.55.0
#4 lore raw

Re: [PATCH v8 3/8] KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Low] Header files pmu.h and pmu_intel.h lack required includes, violating self-contained header…

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Header files pmu.h and pmu_intel.h lack required includes, violating self-contained header practices. --

--- Patch [3]: [PATCH v8 3/8] KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h --- commit 7eb009fb5328b80368e10acda6cfe621b3e1e1c0
Author: Zide Chen <[email protected]>

KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h

This commit renames vcpu_get_perf_capabilities() to kvm_vcpu_get_perf_caps() and moves it to the common x86 PMU header. This prepares the function to be called from common x86 code to check the guest's PERF_METRICS capability.

Quoted 4 lines
arch/x86/kvm/pmu.h +0 -0
index 090c9bbb74f48..82f955e5a4507 100644

[ ... ]

arch/x86/kvm/pmu.h static inline bool kvm_pmu_is_fastpath_emulation_allowed(struct kvm_vcpu *vcpu) · line 277

Quoted 8 lines
arch/x86/kvm/pmu.h +4 -0
@@ -271,6 +271,14 @@ static inline bool kvm_pmu_is_fastpath_emulation_allowed(struct kvm_vcpu *vcpu)
 				  X86_PMC_IDX_MAX);
 }
 
+static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu)
+{
+	if (!guest_cpu_cap_has(vcpu, X86_FEATURE_PDCM))
+		return 0;

[Severity: Low]
Does this file need to include cpuid.h? guest_cpu_cap_has() is used here, but pmu.h does not include the header that defines it.

Quoted 5 lines
arch/x86/kvm/pmu.h +4 -0
+
+	return vcpu->arch.perf_capabilities;
+}
+
 void kvm_pmu_deliver_pmi(struct kvm_vcpu *vcpu);

[ ... ]

Quoted 12 lines
arch/x86/kvm/vmx/pmu_intel.h +0 -4
index 5d9357640aa18..dd447d8b6fdde 100644
@@ -4,19 +4,9 @@
 
 #include <linux/kvm_host.h>
 
-#include "cpuid.h"
-
-static inline u64 vcpu_get_perf_capabilities(struct kvm_vcpu *vcpu)
-{

arch/x86/kvm/vmx/pmu_intel.h line 9

Quoted 10 lines
arch/x86/kvm/vmx/pmu_intel.h +1 -7
-	if (!guest_cpu_cap_has(vcpu, X86_FEATURE_PDCM))
-		return 0;
-
-	return vcpu->arch.perf_capabilities;
-}
-
 static inline bool fw_writes_is_enabled(struct kvm_vcpu *vcpu)
 {
-	return (vcpu_get_perf_capabilities(vcpu) & PERF_CAP_FW_WRITES) != 0;
+	return (kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_FW_WRITES) != 0;

[Severity: Low]
Should pmu_intel.h include pmu.h? It calls kvm_vcpu_get_perf_caps() which is now defined in pmu.h, but the include appears to be missing.

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

Re: [PATCH v8 3/8] KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h

Quoted 23 lines · On 8/17/2026 1:52 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Header files pmu.h and pmu_intel.h lack required includes, violating self-contained header practices. --

--- Patch [3]: [PATCH v8 3/8] KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h --- commit 7eb009fb5328b80368e10acda6cfe621b3e1e1c0
Author: Zide Chen <[email protected]>

KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h

This commit renames vcpu_get_perf_capabilities() to kvm_vcpu_get_perf_caps() and moves it to the common x86 PMU header. This prepares the function to be called from common x86 code to check the guest's PERF_METRICS capability.

Quoted 4 lines
arch/x86/kvm/pmu.h +0 -0
index 090c9bbb74f48..82f955e5a4507 100644

[ ... ]

Quoted 4 lines
arch/x86/kvm/pmu.h +0 -0
@@ -271,6 +271,14 @@ static inline bool kvm_pmu_is_fastpath_emulation_allowed(struct kvm_vcpu *vcpu)
 				  X86_PMC_IDX_MAX);
 }
 
Quoted 8 lines

arch/x86/kvm/pmu.h static inline bool kvm_pmu_is_fastpath_emulation_allowed(struct kvm_vcpu *vcpu) · line 277

Quoted 4 lines
arch/x86/kvm/pmu.h +4 -0
+static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu)
+{
+	if (!guest_cpu_cap_has(vcpu, X86_FEATURE_PDCM))
+		return 0;

[Severity: Low]
Does this file need to include cpuid.h? guest_cpu_cap_has() is used here, but pmu.h does not include the header that defines it.

Copied from my reply in v7 review:

It sounds reasonable, but I'd prefer to keep it as-is.

Adding cpuid.h to pmu.h would make pmu.h more self-contained, but it would also naturally lead to removing now-redundant cpuid.h includes from PMU-related files such as vmx/pmu_intel.c, svm/pmu.c, and potentially others.

I'd like to keep the current patch focused on the functional change and avoid mixing it with include dependency cleanup.

Quoted 37 lines
Quoted 5 lines
arch/x86/kvm/pmu.h +4 -0
+
+	return vcpu->arch.perf_capabilities;
+}
+
 void kvm_pmu_deliver_pmi(struct kvm_vcpu *vcpu);

[ ... ]

Quoted 12 lines
arch/x86/kvm/vmx/pmu_intel.h +0 -4
index 5d9357640aa18..dd447d8b6fdde 100644
@@ -4,19 +4,9 @@
 
 #include <linux/kvm_host.h>
 
-#include "cpuid.h"
-
-static inline u64 vcpu_get_perf_capabilities(struct kvm_vcpu *vcpu)
-{

arch/x86/kvm/vmx/pmu_intel.h line 9

Quoted 10 lines
arch/x86/kvm/vmx/pmu_intel.h +1 -7
-	if (!guest_cpu_cap_has(vcpu, X86_FEATURE_PDCM))
-		return 0;
-
-	return vcpu->arch.perf_capabilities;
-}
-
 static inline bool fw_writes_is_enabled(struct kvm_vcpu *vcpu)
 {
-	return (vcpu_get_perf_capabilities(vcpu) & PERF_CAP_FW_WRITES) != 0;
+	return (kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_FW_WRITES) != 0;

[Severity: Low]
Should pmu_intel.h include pmu.h? It calls kvm_vcpu_get_perf_caps() which is now defined in pmu.h, but the include appears to be missing.

Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.h +0 -0
 }
 
 bool intel_pmu_lbr_is_enabled(struct kvm_vcpu *vcpu);
author
#6 lore raw

PATCH v8 4/8 KVM: x86/pmu: Snapshot host IA32_PERF_CAPABILITIES in kvm_host

In reply to Zide Chen

From: Mingwei Zhang <[email protected]>

Cache the unadulterated snapshot of perf_capabilities so that KVM can compare guest vPMU capabilities against raw hardware capabilities.

For example, if the host supports PERF_METRICS but it is not configured for the guest, KVM can use it to determine that RDPMC accesses must be intercepted.

---
v5: new patch.

---
 arch/x86/include/asm/kvm_host.h | 1 +
 arch/x86/kvm/vmx/vmx.c          | 8 ++------
 arch/x86/kvm/x86.c              | 4 ++++
 3 files changed, 7 insertions(+), 6 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -0
index 8a4e542218a7..d893fa5be838 100644
@@ -294,6 +294,7 @@ struct kvm_host_values {
 	u64 xss;
 	u64 s_cet;
 	u64 arch_capabilities;
+	u64 perf_capabilities;
 };
 extern struct kvm_host_values kvm_host;
 
arch/x86/kvm/vmx/vmx.c +2 -6
index e3bfe6aca1a0..6e9d197982b7 100644
@@ -8023,14 +8023,10 @@ void vmx_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu)
 static __init u64 vmx_get_perf_capabilities(void)
 {
 	u64 perf_cap = PERF_CAP_FW_WRITES;
-	u64 host_perf_cap = 0;
 
 	if (!enable_pmu)
 		return 0;
 
-	if (boot_cpu_has(X86_FEATURE_PDCM))
-		rdmsrq(MSR_IA32_PERF_CAPABILITIES, host_perf_cap);
-
 	if (!cpu_feature_enabled(X86_FEATURE_ARCH_LBR) &&
 	    !enable_mediated_pmu) {
 		x86_perf_get_lbr(&vmx_lbr_caps);
@@ -8043,11 +8039,11 @@ static __init u64 vmx_get_perf_capabilities(void)
 		if (!vmx_lbr_caps.has_callstack)
 			memset(&vmx_lbr_caps, 0, sizeof(vmx_lbr_caps));
 		else if (vmx_lbr_caps.nr)
-			perf_cap |= host_perf_cap & PERF_CAP_LBR_FMT;
+			perf_cap |= kvm_host.perf_capabilities & PERF_CAP_LBR_FMT;
 	}
 
 	if (vmx_pebs_supported()) {
-		perf_cap |= host_perf_cap & PERF_CAP_PEBS_MASK;
+		perf_cap |= kvm_host.perf_capabilities & PERF_CAP_PEBS_MASK;
 
 		/*
 		 * Disallow adaptive PEBS as it is functionally broken, can be
arch/x86/kvm/x86.c +4 -0
index d94b59140c45..d349224d2734 100644
@@ -7083,6 +7083,10 @@ int kvm_x86_vendor_init(struct kvm_x86_init_ops *ops)
 	if (boot_cpu_has(X86_FEATURE_ARCH_CAPABILITIES))
 		rdmsrq(MSR_IA32_ARCH_CAPABILITIES, kvm_host.arch_capabilities);
 
+	if (boot_cpu_has(X86_FEATURE_PDCM))
+		rdmsrq_safe(MSR_IA32_PERF_CAPABILITIES,
+			    &kvm_host.perf_capabilities);
+
 	WARN_ON_ONCE(kvm_nr_uret_msrs);
 
 	r = ops->hardware_setup();
Signature · 1 line
2.55.0
author
#7 lore raw

PATCH v8 5/8 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.

If the host supports PERF_METRICS but it is not exposed to the guest, MSR_CORE_PERF_GLOBAL_CTRL must be intercepted so the guest cannot enable PERF_METRICS via GLOBAL_CTRL[48].

---
v8:
- Intercept global ctrl MSR if host supports PerfMetrics but not enabled in the guest.
v7:
- Clear bit 48 in global_ctrl_rsvd instead of explicitly clearing it in global_status_rsvd, allowing it to propagate through the assignment from global_ctrl_rsvd.
- Add comments documenting the all-zero PERF_METRICS/FC3 write-order exception.
- Add nested_vmx_merge_msr_bitmaps_rw(MSR_PERF_METRICS);
v5:
- Remove host_initiated check in set/get MSR handlers.
v4:
- Remove WARN_ON_ONCE() and simply reject the guest accesses by checking host_initiated. (Sashiko)
- Passthru MSR_PERF_METRICS only if has_mediated_pmu is true. (Sashiko)
- Remove the redundant !! in vcpu_has_perf_metrics().
v3:
- Replace WARN_ON() with WARN_ON_ONCE(). (Dapeng)
- Add comments to explain why we 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/kvm/msrs.c              |  6 +++++-
 arch/x86/kvm/pmu.c               |  7 +++++++
 arch/x86/kvm/pmu.h               |  5 +++++
 arch/x86/kvm/vmx/nested.c        |  2 ++
 arch/x86/kvm/vmx/pmu_intel.c     | 34 ++++++++++++++++++++++++++++++++
 arch/x86/kvm/vmx/vmx.c           |  7 +++++++
 8 files changed, 62 insertions(+), 1 deletion(-)
arch/x86/include/asm/kvm_host.h +1 -0
index d893fa5be838..7166efdf79ce 100644
@@ -583,6 +583,7 @@ 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/kvm/msrs.c +5 -1
index 1fbd8f08a7f0..2b2cc6bf3ce7 100644
@@ -219,7 +219,7 @@ 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. */
@@ -2608,6 +2608,10 @@ 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 >=
arch/x86/kvm/pmu.c +7 -0
index 913a0ecbf260..2fddcbc8924d 100644
@@ -806,9 +806,16 @@ static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu)
 	       pmu->nr_arch_fixed_counters != kvm_host_pmu.num_counters_fixed;
 }
 
+static bool kvm_need_perf_metrics_intercept(struct kvm_vcpu *vcpu)
+{
+	return (kvm_host.perf_capabilities & PERF_CAP_PERF_METRICS) &&
+		!kvm_vcpu_has_perf_metrics(vcpu);
+}
+
 bool kvm_need_perf_global_ctrl_intercept(struct kvm_vcpu *vcpu)
 {
 	return kvm_need_any_pmc_intercept(vcpu) ||
+	       kvm_need_perf_metrics_intercept(vcpu) ||
 	       !kvm_pmu_has_perf_global_ctrl(vcpu_to_pmu(vcpu));
 }
 EXPORT_SYMBOL_FOR_KVM_INTERNAL(kvm_need_perf_global_ctrl_intercept);
arch/x86/kvm/pmu.h +5 -0
index 82f955e5a450..be578013ed83 100644
@@ -279,6 +279,11 @@ static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu)
 	return vcpu->arch.perf_capabilities;
 }
 
+static inline bool kvm_vcpu_has_perf_metrics(struct kvm_vcpu *vcpu)
+{
+	return kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PERF_METRICS;
+}
+
 void kvm_pmu_deliver_pmi(struct kvm_vcpu *vcpu);
 int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned pmc, u64 *data);
 int kvm_pmu_check_rdpmc_early(struct kvm_vcpu *vcpu, unsigned int idx);
arch/x86/kvm/vmx/nested.c +2 -0
index 7ed79894d11d..e81b8a5423c8 100644
@@ -717,6 +717,8 @@ static void nested_vmx_merge_pmu_msr_bitmaps(struct kvm_vcpu *vcpu,
 	nested_vmx_merge_msr_bitmaps_rw(MSR_CORE_PERF_GLOBAL_CTRL);
 	nested_vmx_merge_msr_bitmaps_read(MSR_CORE_PERF_GLOBAL_STATUS);
 	nested_vmx_merge_msr_bitmaps_write(MSR_CORE_PERF_GLOBAL_OVF_CTRL);
+
+	nested_vmx_merge_msr_bitmaps_rw(MSR_PERF_METRICS);
 }
 
 /*
arch/x86/kvm/vmx/pmu_intel.c +34 -0
index 100c072b7fb3..1fd0480baf68 100644
@@ -189,6 +189,8 @@ 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 kvm_vcpu_has_perf_metrics(vcpu);
 	case MSR_IA32_PEBS_ENABLE:
 		ret = kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PEBS_FORMAT;
 		break;
@@ -346,6 +348,9 @@ 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:
+		msr_info->data = pmu->perf_metrics;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		msr_info->data = pmu->pebs_enable;
 		break;
@@ -395,6 +400,14 @@ 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:
+		/*
+		 * Bits [63:32] are ignored on hardware that supports only
+		 * level-1 metrics, but may be valid on hardware that supports
+		 * level-2 metrics. Preserve guest writes verbatim.
+		 */
+		pmu->perf_metrics = data;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		if (data & pmu->pebs_enable_rsvd)
 			return 1;
@@ -577,6 +590,8 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	counter_rsvd = ~((BIT_ULL(pmu->nr_arch_gp_counters) - 1) |
 			 ((BIT_ULL(pmu->nr_arch_fixed_counters) - 1) << KVM_FIXED_PMC_BASE_IDX));
 	pmu->global_ctrl_rsvd = counter_rsvd;
+	if (perf_capabilities & PERF_CAP_PERF_METRICS)
+		pmu->global_ctrl_rsvd &= ~GLOBAL_CTRL_EN_PERF_METRICS;
 
 	/*
 	 * GLOBAL_STATUS and GLOBAL_OVF_CONTROL (a.k.a. GLOBAL_STATUS_RESET)
@@ -633,6 +648,9 @@ 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 +821,9 @@ static void intel_mediated_pmu_load(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	u64 global_status, toggle;
 
+	if (kvm_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 +852,19 @@ 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 (kvm_vcpu_has_perf_metrics(vcpu)) {
+		pmu->perf_metrics = rdpmc(INTEL_PMC_FIXED_RDPMC_METRICS);
+		/*
+		 * The SDM requires restoring fixed counter 3 before
+		 * PERF_METRICS. However, this path writes 0 to PERF_METRICS
+		 * before fixed counter 3. For this all-zero case, the
+		 * resulting hardware state is therefore the same regardless
+		 * of write order.
+		 */
+		if (pmu->perf_metrics)
+			wrmsrq(MSR_PERF_METRICS, 0);
+	}
 }
 
 struct kvm_pmu_ops intel_pmu_ops __initdata = {
arch/x86/kvm/vmx/vmx.c +7 -0
index 6e9d197982b7..8cb9d642277a 100644
@@ -4277,6 +4277,10 @@ 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);
+
+	intercept = !has_mediated_pmu || !kvm_vcpu_has_perf_metrics(vcpu);
+	vmx_set_intercept_for_msr(vcpu, MSR_PERF_METRICS,
+				  MSR_TYPE_RW, intercept);
 }
 
 static void vmx_recalc_msr_intercepts(struct kvm_vcpu *vcpu)
@@ -8068,6 +8072,9 @@ static __init u64 vmx_get_perf_capabilities(void)
 		perf_cap &= ~PERF_CAP_PEBS_BASELINE;
 	}
 
+	if (enable_mediated_pmu)
+		perf_cap |= kvm_host.perf_capabilities & PERF_CAP_PERF_METRICS;
+
 	return perf_cap;
 }
 
Signature · 1 line
2.55.0
#8 lore raw

Re: [PATCH v8 5/8] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

Thank you for your contribution! Sashiko AI review found 4 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 4 potential issue(s) to consider:

New issues:
- [Critical] Unvalidated guest/userspace writes to MSR_PERF_METRICS cause host kernel WARN/Oops during PMU context load.
- [High] Missing RDPMC emulation for PERF_METRICS causes guest #GP when RDPMC is intercepted.
- [High] Missing RDPMC intercept allows information leak of host MSR_PERF_METRICS.

Pre-existing issues:
- [High] Intercepting MSR_CORE_PERF_GLOBAL_STATUS during mediated PMU operation breaks PMIs by returning stale overflow data to the guest. --

--- Patch [5]: [PATCH v8 5/8] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU --- commit 3c012e6d223702ac2057eaf2c6a15810c5334d8c
Author: Dapeng Mi <[email protected]>

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

This patch exposes the IA32_PERF_METRICS MSR to the guest when mediated vPMU is enabled. It allows host/guest PMU context switching to save and restore this MSR, and intercepts MSR_CORE_PERF_GLOBAL_CTRL if the host supports PERF_METRICS but it is not exposed to the guest.

Quoted 5 lines
arch/x86/kvm/pmu.c +0 -0
index 913a0ecbf2609..2fddcbc8924d4 100644

arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 814

Quoted 9 lines
arch/x86/kvm/pmu.c +6 -0
@@ -806,9 +806,16 @@ static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu)
 	       pmu->nr_arch_fixed_counters != kvm_host_pmu.num_counters_fixed;
 }
 
+static bool kvm_need_perf_metrics_intercept(struct kvm_vcpu *vcpu)
+{
+	return (kvm_host.perf_capabilities & PERF_CAP_PERF_METRICS) &&
+		!kvm_vcpu_has_perf_metrics(vcpu);
+}
+

[Severity: High]
Does this code miss checking for kvm_need_perf_metrics_intercept() in kvm_need_rdpmc_intercept()?

If mediated PMU is active and the PMU counter topology matches the host, RDPMC exiting is disabled. The guest can execute RDPMC with ECX=(1<<29) directly on the hardware. Since intel_mediated_pmu_load() skips clearing MSR_PERF_METRICS when the guest lacks the feature, the guest might observe the host's raw active MSR_PERF_METRICS value.

arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 820

Quoted 6 lines
arch/x86/kvm/pmu.c +1 -0
 bool kvm_need_perf_global_ctrl_intercept(struct kvm_vcpu *vcpu)
 {
 	return kvm_need_any_pmc_intercept(vcpu) ||
+	       kvm_need_perf_metrics_intercept(vcpu) ||
 	       !kvm_pmu_has_perf_global_ctrl(vcpu_to_pmu(vcpu));
 }

[Severity: High]
This is a pre-existing issue, but does forcing intercepts here break PMIs for guests lacking PERF_METRICS?

By returning true here, MSR_CORE_PERF_GLOBAL_STATUS and MSR_CORE_PERF_GLOBAL_OVF_CTRL are intercepted. During mediated PMU operation, hardware updates the physical GLOBAL_STATUS directly on overflow. Because KVM does not sync the physical MSR to pmu->global_status on every VM-exit (only on vcpu_put()), guest reads of the intercepted GLOBAL_STATUS could return stale cached data (0), potentially hiding hardware PMIs from the guest.

[ ... ]

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
index 100c072b7fb38..1fd0480baf680 100644

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 353

Quoted 7 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -346,6 +348,9 @@ 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:
+		msr_info->data = pmu->perf_metrics;
+		break;

[Severity: High]
Does KVM also need to emulate RDPMC for PERF_METRICS?

The SDM specifies that IA32_PERF_METRICS can be read using the RDPMC instruction with ECX = (1<<29) | 0. However, intel_rdpmc_ecx_to_pmc() only recognizes GP counters (type 0) and fixed counters (type 1<<30).

If KVM intercepts RDPMC, could a guest attempt to read PERF_METRICS via RDPMC fall through to the default case and return NULL, causing KVM to inject a #GP?

Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
 	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 410

Quoted 12 lines
arch/x86/kvm/vmx/pmu_intel.c +8 -0
@@ -395,6 +400,14 @@ 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:
+		/*
+		 * Bits [63:32] are ignored on hardware that supports only
+		 * level-1 metrics, but may be valid on hardware that supports
+		 * level-2 metrics. Preserve guest writes verbatim.
+		 */
+		pmu->perf_metrics = data;
+		break;

[Severity: Critical]
Is it safe to accept untrusted guest or userspace data here without validation?

If a malicious guest or host userspace (via KVM_SET_MSRS) injects a value with reserved bits set into pmu->perf_metrics, it could cause a host #GP during vcpu_load() when intel_mediated_pmu_load() blindly writes this value to the physical MSR using the unsafe wrmsrq() macro.

Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
 	case MSR_IA32_PEBS_ENABLE:
 		if (data & pmu->pebs_enable_rsvd)
 			return 1;

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static void intel_mediated_pmu_load(struct kvm_vcpu *vcpu) · line 826

Quoted 7 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -803,6 +821,9 @@ static void intel_mediated_pmu_load(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	u64 global_status, toggle;
 
+	if (kvm_vcpu_has_perf_metrics(vcpu))
+		wrmsrq(MSR_PERF_METRICS, pmu->perf_metrics);
+

[Severity: Critical]
Could this wrmsrq() cause a host oops or panic?

As mentioned above, if pmu->perf_metrics contains unvalidated data with reserved bits set, wrmsrq() (which lacks an exception table entry) will trigger a host #GP.

Quoted 2 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
 	rdmsrq(MSR_CORE_PERF_GLOBAL_STATUS, global_status);
 	toggle = pmu->global_status ^ global_status;
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=5
author
#9 lore raw

Re: [PATCH v8 5/8] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

Quoted 30 lines · On 8/17/2026 2:04 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 4 potential issue(s) to consider:

New issues:
- [Critical] Unvalidated guest/userspace writes to MSR_PERF_METRICS cause host kernel WARN/Oops during PMU context load.
- [High] Missing RDPMC emulation for PERF_METRICS causes guest #GP when RDPMC is intercepted.
- [High] Missing RDPMC intercept allows information leak of host MSR_PERF_METRICS.

Pre-existing issues:
- [High] Intercepting MSR_CORE_PERF_GLOBAL_STATUS during mediated PMU operation breaks PMIs by returning stale overflow data to the guest. --

--- Patch [5]: [PATCH v8 5/8] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU --- commit 3c012e6d223702ac2057eaf2c6a15810c5334d8c
Author: Dapeng Mi <[email protected]>

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

This patch exposes the IA32_PERF_METRICS MSR to the guest when mediated vPMU is enabled. It allows host/guest PMU context switching to save and restore this MSR, and intercepts MSR_CORE_PERF_GLOBAL_CTRL if the host supports PERF_METRICS but it is not exposed to the guest.

Quoted 8 lines
arch/x86/kvm/pmu.c +0 -0
index 913a0ecbf2609..2fddcbc8924d4 100644
@@ -806,9 +806,16 @@ static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu)
 	       pmu->nr_arch_fixed_counters != kvm_host_pmu.num_counters_fixed;
 }
 
Quoted 10 lines

arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 814

Quoted 6 lines
arch/x86/kvm/pmu.c +6 -0
+static bool kvm_need_perf_metrics_intercept(struct kvm_vcpu *vcpu)
+{
+	return (kvm_host.perf_capabilities & PERF_CAP_PERF_METRICS) &&
+		!kvm_vcpu_has_perf_metrics(vcpu);
+}
+

[Severity: High]
Does this code miss checking for kvm_need_perf_metrics_intercept() in kvm_need_rdpmc_intercept()?

Yes, it's included in patch 7/8.

Quoted 17 lines

If mediated PMU is active and the PMU counter topology matches the host, RDPMC exiting is disabled. The guest can execute RDPMC with ECX=(1<<29) directly on the hardware. Since intel_mediated_pmu_load() skips clearing MSR_PERF_METRICS when the guest lacks the feature, the guest might observe the host's raw active MSR_PERF_METRICS value.

arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 820

Quoted 6 lines
arch/x86/kvm/pmu.c +1 -0
 bool kvm_need_perf_global_ctrl_intercept(struct kvm_vcpu *vcpu)
 {
 	return kvm_need_any_pmc_intercept(vcpu) ||
+	       kvm_need_perf_metrics_intercept(vcpu) ||
 	       !kvm_pmu_has_perf_global_ctrl(vcpu_to_pmu(vcpu));
 }

[Severity: High]
This is a pre-existing issue, but does forcing intercepts here break PMIs for guests lacking PERF_METRICS?

Quoted 7 lines

By returning true here, MSR_CORE_PERF_GLOBAL_STATUS and MSR_CORE_PERF_GLOBAL_OVF_CTRL are intercepted. During mediated PMU operation, hardware updates the physical GLOBAL_STATUS directly on overflow. Because KVM does not sync the physical MSR to pmu->global_status on every VM-exit (only on vcpu_put()), guest reads of the intercepted GLOBAL_STATUS could return stale cached data (0), potentially hiding hardware PMIs from the guest.

False positive.

GLOBAL_STATUS is synced to pmu->global_status in intel_mediated_pmu_load() in every VM exit. So it won't return stale data to guest reads of the intercepted GLOBAL_STATUS.

Quoted 7 lines

[ ... ]

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
index 100c072b7fb38..1fd0480baf680 100644

[ ... ]

Quoted 10 lines

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 353

Quoted 7 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -346,6 +348,9 @@ 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:
+		msr_info->data = pmu->perf_metrics;
+		break;

[Severity: High]
Does KVM also need to emulate RDPMC for PERF_METRICS?

Yes, it's implemented in the following patches.>

Quoted 12 lines

The SDM specifies that IA32_PERF_METRICS can be read using the RDPMC instruction with ECX = (1<<29) | 0. However, intel_rdpmc_ecx_to_pmc() only recognizes GP counters (type 0) and fixed counters (type 1<<30).

If KVM intercepts RDPMC, could a guest attempt to read PERF_METRICS via RDPMC fall through to the default case and return NULL, causing KVM to inject a #GP?

Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
 	case MSR_IA32_PEBS_ENABLE:
 		msr_info->data = pmu->pebs_enable;
 		break;

[ ... ]

Quoted 16 lines

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 410

Quoted 12 lines
arch/x86/kvm/vmx/pmu_intel.c +8 -0
@@ -395,6 +400,14 @@ 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:
+		/*
+		 * Bits [63:32] are ignored on hardware that supports only
+		 * level-1 metrics, but may be valid on hardware that supports
+		 * level-2 metrics. Preserve guest writes verbatim.
+		 */
+		pmu->perf_metrics = data;
+		break;

[Severity: Critical]
Is it safe to accept untrusted guest or userspace data here without validation?

As the comments suggest, no #GP is caused.

Quoted 20 lines

If a malicious guest or host userspace (via KVM_SET_MSRS) injects a value with reserved bits set into pmu->perf_metrics, it could cause a host #GP during vcpu_load() when intel_mediated_pmu_load() blindly writes this value to the physical MSR using the unsafe wrmsrq() macro.

Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
 	case MSR_IA32_PEBS_ENABLE:
 		if (data & pmu->pebs_enable_rsvd)
 			return 1;

[ ... ]

arch/x86/kvm/vmx/pmu_intel.c static void intel_mediated_pmu_load(struct kvm_vcpu *vcpu) · line 826

Quoted 7 lines
arch/x86/kvm/vmx/pmu_intel.c +3 -0
@@ -803,6 +821,9 @@ static void intel_mediated_pmu_load(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	u64 global_status, toggle;
 
+	if (kvm_vcpu_has_perf_metrics(vcpu))
+		wrmsrq(MSR_PERF_METRICS, pmu->perf_metrics);
+

[Severity: Critical]
Could this wrmsrq() cause a host oops or panic?

Quoted 3 lines

As mentioned above, if pmu->perf_metrics contains unvalidated data with reserved bits set, wrmsrq() (which lacks an exception table entry) will trigger a host #GP.

No, no #GP in this case.

Quoted 3 lines
Quoted 2 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
 	rdmsrq(MSR_CORE_PERF_GLOBAL_STATUS, global_status);
 	toggle = pmu->global_status ^ global_status;
Reviewed-by on v8 5/8
#10 lore raw

Re: [PATCH v8 5/8] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

In reply to Zide Chen
Quoted 266 lines · 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.

If the host supports PERF_METRICS but it is not exposed to the guest, MSR_CORE_PERF_GLOBAL_CTRL must be intercepted so the guest cannot enable PERF_METRICS via GLOBAL_CTRL[48].

---
v8:
- Intercept global ctrl MSR if host supports PerfMetrics but not enabled in the guest.
v7:
- Clear bit 48 in global_ctrl_rsvd instead of explicitly clearing it in global_status_rsvd, allowing it to propagate through the assignment from global_ctrl_rsvd.
- Add comments documenting the all-zero PERF_METRICS/FC3 write-order exception.
- Add nested_vmx_merge_msr_bitmaps_rw(MSR_PERF_METRICS);
v5:
- Remove host_initiated check in set/get MSR handlers.
v4:
- Remove WARN_ON_ONCE() and simply reject the guest accesses by checking host_initiated. (Sashiko)
- Passthru MSR_PERF_METRICS only if has_mediated_pmu is true. (Sashiko)
- Remove the redundant !! in vcpu_has_perf_metrics().
v3:
- Replace WARN_ON() with WARN_ON_ONCE(). (Dapeng)
- Add comments to explain why we 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/kvm/msrs.c              |  6 +++++-
 arch/x86/kvm/pmu.c               |  7 +++++++
 arch/x86/kvm/pmu.h               |  5 +++++
 arch/x86/kvm/vmx/nested.c        |  2 ++
 arch/x86/kvm/vmx/pmu_intel.c     | 34 ++++++++++++++++++++++++++++++++
 arch/x86/kvm/vmx/vmx.c           |  7 +++++++
 8 files changed, 62 insertions(+), 1 deletion(-)
arch/x86/include/asm/kvm_host.h +1 -0
index d893fa5be838..7166efdf79ce 100644
@@ -583,6 +583,7 @@ 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/kvm/msrs.c +5 -1
index 1fbd8f08a7f0..2b2cc6bf3ce7 100644
@@ -219,7 +219,7 @@ 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. */
@@ -2608,6 +2608,10 @@ 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 >=
arch/x86/kvm/pmu.c +7 -0
index 913a0ecbf260..2fddcbc8924d 100644
@@ -806,9 +806,16 @@ static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu)
 	       pmu->nr_arch_fixed_counters != kvm_host_pmu.num_counters_fixed;
 }
 
+static bool kvm_need_perf_metrics_intercept(struct kvm_vcpu *vcpu)
+{
+	return (kvm_host.perf_capabilities & PERF_CAP_PERF_METRICS) &&
+		!kvm_vcpu_has_perf_metrics(vcpu);
+}
+
 bool kvm_need_perf_global_ctrl_intercept(struct kvm_vcpu *vcpu)
 {
 	return kvm_need_any_pmc_intercept(vcpu) ||
+	       kvm_need_perf_metrics_intercept(vcpu) ||
 	       !kvm_pmu_has_perf_global_ctrl(vcpu_to_pmu(vcpu));
 }
 EXPORT_SYMBOL_FOR_KVM_INTERNAL(kvm_need_perf_global_ctrl_intercept);
arch/x86/kvm/pmu.h +5 -0
index 82f955e5a450..be578013ed83 100644
@@ -279,6 +279,11 @@ static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu)
 	return vcpu->arch.perf_capabilities;
 }
 
+static inline bool kvm_vcpu_has_perf_metrics(struct kvm_vcpu *vcpu)
+{
+	return kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PERF_METRICS;
+}
+
 void kvm_pmu_deliver_pmi(struct kvm_vcpu *vcpu);
 int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned pmc, u64 *data);
 int kvm_pmu_check_rdpmc_early(struct kvm_vcpu *vcpu, unsigned int idx);
arch/x86/kvm/vmx/nested.c +2 -0
index 7ed79894d11d..e81b8a5423c8 100644
@@ -717,6 +717,8 @@ static void nested_vmx_merge_pmu_msr_bitmaps(struct kvm_vcpu *vcpu,
 	nested_vmx_merge_msr_bitmaps_rw(MSR_CORE_PERF_GLOBAL_CTRL);
 	nested_vmx_merge_msr_bitmaps_read(MSR_CORE_PERF_GLOBAL_STATUS);
 	nested_vmx_merge_msr_bitmaps_write(MSR_CORE_PERF_GLOBAL_OVF_CTRL);
+
+	nested_vmx_merge_msr_bitmaps_rw(MSR_PERF_METRICS);
 }
 
 /*
arch/x86/kvm/vmx/pmu_intel.c +34 -0
index 100c072b7fb3..1fd0480baf68 100644
@@ -189,6 +189,8 @@ 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 kvm_vcpu_has_perf_metrics(vcpu);
 	case MSR_IA32_PEBS_ENABLE:
 		ret = kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PEBS_FORMAT;
 		break;
@@ -346,6 +348,9 @@ 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:
+		msr_info->data = pmu->perf_metrics;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		msr_info->data = pmu->pebs_enable;
 		break;
@@ -395,6 +400,14 @@ 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:
+		/*
+		 * Bits [63:32] are ignored on hardware that supports only
+		 * level-1 metrics, but may be valid on hardware that supports
+		 * level-2 metrics. Preserve guest writes verbatim.
+		 */
+		pmu->perf_metrics = data;
+		break;
 	case MSR_IA32_PEBS_ENABLE:
 		if (data & pmu->pebs_enable_rsvd)
 			return 1;
@@ -577,6 +590,8 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	counter_rsvd = ~((BIT_ULL(pmu->nr_arch_gp_counters) - 1) |
 			 ((BIT_ULL(pmu->nr_arch_fixed_counters) - 1) << KVM_FIXED_PMC_BASE_IDX));
 	pmu->global_ctrl_rsvd = counter_rsvd;
+	if (perf_capabilities & PERF_CAP_PERF_METRICS)
+		pmu->global_ctrl_rsvd &= ~GLOBAL_CTRL_EN_PERF_METRICS;
 
 	/*
 	 * GLOBAL_STATUS and GLOBAL_OVF_CONTROL (a.k.a. GLOBAL_STATUS_RESET)
@@ -633,6 +648,9 @@ 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 +821,9 @@ static void intel_mediated_pmu_load(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	u64 global_status, toggle;
 
+	if (kvm_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 +852,19 @@ 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 (kvm_vcpu_has_perf_metrics(vcpu)) {
+		pmu->perf_metrics = rdpmc(INTEL_PMC_FIXED_RDPMC_METRICS);
+		/*
+		 * The SDM requires restoring fixed counter 3 before
+		 * PERF_METRICS. However, this path writes 0 to PERF_METRICS
+		 * before fixed counter 3. For this all-zero case, the
+		 * resulting hardware state is therefore the same regardless
+		 * of write order.
+		 */
+		if (pmu->perf_metrics)
+			wrmsrq(MSR_PERF_METRICS, 0);
+	}
 }
 
 struct kvm_pmu_ops intel_pmu_ops __initdata = {
arch/x86/kvm/vmx/vmx.c +7 -0
index 6e9d197982b7..8cb9d642277a 100644
@@ -4277,6 +4277,10 @@ 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);
+
+	intercept = !has_mediated_pmu || !kvm_vcpu_has_perf_metrics(vcpu);
+	vmx_set_intercept_for_msr(vcpu, MSR_PERF_METRICS,
+				  MSR_TYPE_RW, intercept);
 }
 
 static void vmx_recalc_msr_intercepts(struct kvm_vcpu *vcpu)
@@ -8068,6 +8072,9 @@ static __init u64 vmx_get_perf_capabilities(void)
 		perf_cap &= ~PERF_CAP_PEBS_BASELINE;
 	}
 
+	if (enable_mediated_pmu)
+		perf_cap |= kvm_host.perf_capabilities & PERF_CAP_PERF_METRICS;
+
 	return perf_cap;
 }
 
author
#11 lore raw

Re: [PATCH v8 5/8] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU

In reply to Zide Chen

arch/x86/kvm/vmx/pmu_intel.c static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 408

Quoted 10 lines · Zide Chen
arch/x86/kvm/vmx/pmu_intel.c +6 -0
@@ -395,6 +400,14 @@ 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:
+		/*
+		 * Bits [63:32] are ignored on hardware that supports only
+		 * level-1 metrics, but may be valid on hardware that supports
+		 * level-2 metrics. Preserve guest writes verbatim.
+		 */

That comment isn't correct. I've confirmed that MSR_PERF_METRICS[63:32] isn't simply ignored on hardware that implements only the four level-1 Topdown metrics. Those bits are reserved, and KVM should reject non-zero writes.

I'll post a fix in the next version. KVM will get the number of Topdown metric events supported by the host CPU from perf/x86 and derive the MSR_PERF_METRICS reserved-bit mask from that count, using it to reject invalid writes.

author
#12 lore raw

PATCH v8 6/8 KVM: x86/pmu: Move RDPMC emulation into per-vendor callbacks

In reply to Zide Chen

The current RDPMC emulation splits responsibility: rdpmc_ecx_to_pmc() in each vendor returns a kvm_pmc, then common code calls pmc_read_counter().

This design cannot support RDPMC reads that don't map to a counter, such as PERF_METRICS on Intel platforms.

Replace rdpmc_ecx_to_pmc() with emulate_rdpmc(), which takes full ownership of the emulation and writes the result directly into @data.

Opportunistically drop the redundant bitmask in intel_emulate_rdpmc() since pmc_read_counter() already applies the counter's bit-width mask.

No functional change intended.

---
v6: new patch.

---
 arch/x86/include/asm/kvm-x86-pmu-ops.h |  2 +-
 arch/x86/kvm/pmu.c                     |  9 +--------
 arch/x86/kvm/pmu.h                     |  4 ++--
 arch/x86/kvm/svm/pmu.c                 | 13 +++++++++----
 arch/x86/kvm/vmx/pmu_intel.c           | 25 ++++++++++++-------------
 5 files changed, 25 insertions(+), 28 deletions(-)
arch/x86/include/asm/kvm-x86-pmu-ops.h +1 -1
index 4a223c2793e3..4b50ed058aed 100644
@@ -13,7 +13,7 @@
  * KVM_X86_PMU_OP_OPTIONAL() can be used for those functions that can have
  * a NULL definition.
  */
-KVM_X86_PMU_OP(rdpmc_ecx_to_pmc)
+KVM_X86_PMU_OP(emulate_rdpmc)
 KVM_X86_PMU_OP(msr_idx_to_pmc)
 KVM_X86_PMU_OP_OPTIONAL(check_rdpmc_early)
 KVM_X86_PMU_OP(is_valid_msr)
arch/x86/kvm/pmu.c +1 -8
index 2fddcbc8924d..789624658e8f 100644
@@ -768,8 +768,6 @@ static int kvm_pmu_rdpmc_vmware(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
 int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
 {
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
-	struct kvm_pmc *pmc;
-	u64 mask = ~0ull;
 
 	if (!pmu->version)
 		return 1;
@@ -777,17 +775,12 @@ int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
 	if (is_vmware_backdoor_pmc(idx))
 		return kvm_pmu_rdpmc_vmware(vcpu, idx, data);
 
-	pmc = kvm_pmu_call(rdpmc_ecx_to_pmc)(vcpu, idx, &mask);
-	if (!pmc)
-		return 1;
-
 	if (!kvm_is_cr4_bit_set(vcpu, X86_CR4_PCE) &&
 	    (kvm_x86_call(get_cpl)(vcpu) != 0) &&
 	    kvm_is_cr0_bit_set(vcpu, X86_CR0_PE))
 		return 1;
 
-	*data = pmc_read_counter(pmc) & mask;
-	return 0;
+	return kvm_pmu_call(emulate_rdpmc)(vcpu, idx, data);
 }
 
 static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu)
arch/x86/kvm/pmu.h +2 -2
index be578013ed83..eb6bbc658f5b 100644
@@ -24,8 +24,8 @@
 #define KVM_FIXED_PMC_BASE_IDX INTEL_PMC_IDX_FIXED
 
 struct kvm_pmu_ops {
-	struct kvm_pmc *(*rdpmc_ecx_to_pmc)(struct kvm_vcpu *vcpu,
-		unsigned int idx, u64 *mask);
+	int (*emulate_rdpmc)(struct kvm_vcpu *vcpu, unsigned int idx,
+			     u64 *data);
 	struct kvm_pmc *(*msr_idx_to_pmc)(struct kvm_vcpu *vcpu, u32 msr);
 	int (*check_rdpmc_early)(struct kvm_vcpu *vcpu, unsigned int idx);
 	bool (*is_valid_msr)(struct kvm_vcpu *vcpu, u32 msr);
arch/x86/kvm/svm/pmu.c +9 -4
index c18286545a7a..0517fd4bbcd7 100644
@@ -84,10 +84,15 @@ static int amd_check_rdpmc_early(struct kvm_vcpu *vcpu, unsigned int idx)
 }
 
 /* idx is the ECX register of RDPMC instruction */
-static struct kvm_pmc *amd_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu,
-	unsigned int idx, u64 *mask)
+static int amd_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, u64 *data)
 {
-	return amd_pmu_get_pmc(vcpu_to_pmu(vcpu), idx);
+	struct kvm_pmc *pmc = amd_pmu_get_pmc(vcpu_to_pmu(vcpu), idx);
+
+	if (!pmc)
+		return 1;
+
+	*data = pmc_read_counter(pmc);
+	return 0;
 }
 
 static struct kvm_pmc *amd_msr_idx_to_pmc(struct kvm_vcpu *vcpu, u32 msr)
@@ -302,7 +307,7 @@ static bool amd_pmc_is_disabled_in_current_mode(struct kvm_pmc *pmc)
 }
 
 struct kvm_pmu_ops amd_pmu_ops __initdata = {
-	.rdpmc_ecx_to_pmc = amd_rdpmc_ecx_to_pmc,
+	.emulate_rdpmc = amd_emulate_rdpmc,
 	.msr_idx_to_pmc = amd_msr_idx_to_pmc,
 	.check_rdpmc_early = amd_check_rdpmc_early,
 	.is_valid_msr = amd_is_valid_msr,
arch/x86/kvm/vmx/pmu_intel.c +12 -13
index 1fd0480baf68..bbf4afcffafc 100644
@@ -85,14 +85,13 @@ static void reprogram_fixed_counters(struct kvm_pmu *pmu, u64 data)
 	}
 }
 
-static struct kvm_pmc *intel_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu,
-					    unsigned int idx, u64 *mask)
+static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,
+			       u64 *data)
 {
 	unsigned int type = idx & INTEL_RDPMC_TYPE_MASK;
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
-	struct kvm_pmc *counters;
+	struct kvm_pmc *counters, *pmc;
 	unsigned int num_counters;
-	u64 bitmask;
 
 	/*
 	 * The encoding of ECX for RDPMC is different for architectural versus
@@ -105,7 +104,9 @@ static struct kvm_pmc *intel_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu,
 	 * as KVM doesn't support such PMUs.
 	 */
 	if (WARN_ON_ONCE(!pmu->version))
-		return NULL;
+		return 1;
+
+	idx &= INTEL_RDPMC_INDEX_MASK;
 
 	/*
 	 * General Purpose (GP) PMCs are supported on all PMUs, and fixed PMCs
@@ -119,23 +120,21 @@ static struct kvm_pmc *intel_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu,
 	case INTEL_RDPMC_FIXED:
 		counters = pmu->fixed_counters;
 		num_counters = pmu->nr_arch_fixed_counters;
-		bitmask = pmu->counter_bitmask[KVM_PMC_FIXED];
 		break;
 	case INTEL_RDPMC_GP:
 		counters = pmu->gp_counters;
 		num_counters = pmu->nr_arch_gp_counters;
-		bitmask = pmu->counter_bitmask[KVM_PMC_GP];
 		break;
 	default:
-		return NULL;
+		return 1;
 	}
 
-	idx &= INTEL_RDPMC_INDEX_MASK;
 	if (idx >= num_counters)
-		return NULL;
+		return 1;
 
-	*mask &= bitmask;
-	return &counters[array_index_nospec(idx, num_counters)];
+	pmc = &counters[array_index_nospec(idx, num_counters)];
+	*data = pmc_read_counter(pmc);
+	return 0;
 }
 
 static inline struct kvm_pmc *get_fw_gp_pmc(struct kvm_pmu *pmu, u32 msr)
@@ -868,7 +867,7 @@ static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu)
 }
 
 struct kvm_pmu_ops intel_pmu_ops __initdata = {
-	.rdpmc_ecx_to_pmc = intel_rdpmc_ecx_to_pmc,
+	.emulate_rdpmc = intel_emulate_rdpmc,
 	.msr_idx_to_pmc = intel_msr_idx_to_pmc,
 	.is_valid_msr = intel_is_valid_msr,
 	.get_msr = intel_pmu_get_msr,
Signature · 1 line
2.55.0
Reviewed-by on v8 6/8
#13 lore raw

Re: [PATCH v8 6/8] KVM: x86/pmu: Move RDPMC emulation into per-vendor callbacks

Quoted 20 lines · Zide Chen

The current RDPMC emulation splits responsibility: rdpmc_ecx_to_pmc() in each vendor returns a kvm_pmc, then common code calls pmc_read_counter().

This design cannot support RDPMC reads that don't map to a counter, such as PERF_METRICS on Intel platforms.

Replace rdpmc_ecx_to_pmc() with emulate_rdpmc(), which takes full ownership of the emulation and writes the result directly into @data.

Opportunistically drop the redundant bitmask in intel_emulate_rdpmc() since pmc_read_counter() already applies the counter's bit-width mask.

No functional change intended.

---
v6: new patch.

Quoted 8 lines
---
 arch/x86/include/asm/kvm-x86-pmu-ops.h |  2 +-
 arch/x86/kvm/pmu.c                     |  9 +--------
 arch/x86/kvm/pmu.h                     |  4 ++--
 arch/x86/kvm/svm/pmu.c                 | 13 +++++++++----
 arch/x86/kvm/vmx/pmu_intel.c           | 25 ++++++++++++-------------
 5 files changed, 25 insertions(+), 28 deletions(-)
Quoted 163 lines
arch/x86/include/asm/kvm-x86-pmu-ops.h +1 -1
index 4a223c2793e3..4b50ed058aed 100644
@@ -13,7 +13,7 @@
  * KVM_X86_PMU_OP_OPTIONAL() can be used for those functions that can have
  * a NULL definition.
  */
-KVM_X86_PMU_OP(rdpmc_ecx_to_pmc)
+KVM_X86_PMU_OP(emulate_rdpmc)
 KVM_X86_PMU_OP(msr_idx_to_pmc)
 KVM_X86_PMU_OP_OPTIONAL(check_rdpmc_early)
 KVM_X86_PMU_OP(is_valid_msr)
arch/x86/kvm/pmu.c +1 -8
index 2fddcbc8924d..789624658e8f 100644
@@ -768,8 +768,6 @@ static int kvm_pmu_rdpmc_vmware(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
 int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
 {
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
-	struct kvm_pmc *pmc;
-	u64 mask = ~0ull;
 
 	if (!pmu->version)
 		return 1;
@@ -777,17 +775,12 @@ int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
 	if (is_vmware_backdoor_pmc(idx))
 		return kvm_pmu_rdpmc_vmware(vcpu, idx, data);
 
-	pmc = kvm_pmu_call(rdpmc_ecx_to_pmc)(vcpu, idx, &mask);
-	if (!pmc)
-		return 1;
-
 	if (!kvm_is_cr4_bit_set(vcpu, X86_CR4_PCE) &&
 	    (kvm_x86_call(get_cpl)(vcpu) != 0) &&
 	    kvm_is_cr0_bit_set(vcpu, X86_CR0_PE))
 		return 1;
 
-	*data = pmc_read_counter(pmc) & mask;
-	return 0;
+	return kvm_pmu_call(emulate_rdpmc)(vcpu, idx, data);
 }
 
 static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu)
arch/x86/kvm/pmu.h +2 -2
index be578013ed83..eb6bbc658f5b 100644
@@ -24,8 +24,8 @@
 #define KVM_FIXED_PMC_BASE_IDX INTEL_PMC_IDX_FIXED
 
 struct kvm_pmu_ops {
-	struct kvm_pmc *(*rdpmc_ecx_to_pmc)(struct kvm_vcpu *vcpu,
-		unsigned int idx, u64 *mask);
+	int (*emulate_rdpmc)(struct kvm_vcpu *vcpu, unsigned int idx,
+			     u64 *data);
 	struct kvm_pmc *(*msr_idx_to_pmc)(struct kvm_vcpu *vcpu, u32 msr);
 	int (*check_rdpmc_early)(struct kvm_vcpu *vcpu, unsigned int idx);
 	bool (*is_valid_msr)(struct kvm_vcpu *vcpu, u32 msr);
arch/x86/kvm/svm/pmu.c +9 -4
index c18286545a7a..0517fd4bbcd7 100644
@@ -84,10 +84,15 @@ static int amd_check_rdpmc_early(struct kvm_vcpu *vcpu, unsigned int idx)
 }
 
 /* idx is the ECX register of RDPMC instruction */
-static struct kvm_pmc *amd_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu,
-	unsigned int idx, u64 *mask)
+static int amd_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, u64 *data)
 {
-	return amd_pmu_get_pmc(vcpu_to_pmu(vcpu), idx);
+	struct kvm_pmc *pmc = amd_pmu_get_pmc(vcpu_to_pmu(vcpu), idx);
+
+	if (!pmc)
+		return 1;
+
+	*data = pmc_read_counter(pmc);
+	return 0;
 }
 
 static struct kvm_pmc *amd_msr_idx_to_pmc(struct kvm_vcpu *vcpu, u32 msr)
@@ -302,7 +307,7 @@ static bool amd_pmc_is_disabled_in_current_mode(struct kvm_pmc *pmc)
 }
 
 struct kvm_pmu_ops amd_pmu_ops __initdata = {
-	.rdpmc_ecx_to_pmc = amd_rdpmc_ecx_to_pmc,
+	.emulate_rdpmc = amd_emulate_rdpmc,
 	.msr_idx_to_pmc = amd_msr_idx_to_pmc,
 	.check_rdpmc_early = amd_check_rdpmc_early,
 	.is_valid_msr = amd_is_valid_msr,
arch/x86/kvm/vmx/pmu_intel.c +12 -13
index 1fd0480baf68..bbf4afcffafc 100644
@@ -85,14 +85,13 @@ static void reprogram_fixed_counters(struct kvm_pmu *pmu, u64 data)
 	}
 }
 
-static struct kvm_pmc *intel_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu,
-					    unsigned int idx, u64 *mask)
+static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,
+			       u64 *data)
 {
 	unsigned int type = idx & INTEL_RDPMC_TYPE_MASK;
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
-	struct kvm_pmc *counters;
+	struct kvm_pmc *counters, *pmc;
 	unsigned int num_counters;
-	u64 bitmask;
 
 	/*
 	 * The encoding of ECX for RDPMC is different for architectural versus
@@ -105,7 +104,9 @@ static struct kvm_pmc *intel_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu,
 	 * as KVM doesn't support such PMUs.
 	 */
 	if (WARN_ON_ONCE(!pmu->version))
-		return NULL;
+		return 1;
+
+	idx &= INTEL_RDPMC_INDEX_MASK;
 
 	/*
 	 * General Purpose (GP) PMCs are supported on all PMUs, and fixed PMCs
@@ -119,23 +120,21 @@ static struct kvm_pmc *intel_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu,
 	case INTEL_RDPMC_FIXED:
 		counters = pmu->fixed_counters;
 		num_counters = pmu->nr_arch_fixed_counters;
-		bitmask = pmu->counter_bitmask[KVM_PMC_FIXED];
 		break;
 	case INTEL_RDPMC_GP:
 		counters = pmu->gp_counters;
 		num_counters = pmu->nr_arch_gp_counters;
-		bitmask = pmu->counter_bitmask[KVM_PMC_GP];
 		break;
 	default:
-		return NULL;
+		return 1;
 	}
 
-	idx &= INTEL_RDPMC_INDEX_MASK;
 	if (idx >= num_counters)
-		return NULL;
+		return 1;
 
-	*mask &= bitmask;
-	return &counters[array_index_nospec(idx, num_counters)];
+	pmc = &counters[array_index_nospec(idx, num_counters)];
+	*data = pmc_read_counter(pmc);
+	return 0;
 }
 
 static inline struct kvm_pmc *get_fw_gp_pmc(struct kvm_pmu *pmu, u32 msr)
@@ -868,7 +867,7 @@ static void intel_mediated_pmu_put(struct kvm_vcpu *vcpu)
 }
 
 struct kvm_pmu_ops intel_pmu_ops __initdata = {
-	.rdpmc_ecx_to_pmc = intel_rdpmc_ecx_to_pmc,
+	.emulate_rdpmc = intel_emulate_rdpmc,
 	.msr_idx_to_pmc = intel_msr_idx_to_pmc,
 	.is_valid_msr = intel_is_valid_msr,
 	.get_msr = intel_pmu_get_msr,
author
#14 lore raw

PATCH v8 7/8 KVM: x86/pmu: Emulate RDPMC on performance metrics

In reply to Zide Chen

If the host has the PERF_METRICS capability but it's not present on the guest, RDPMC interception must be enabled and KVM should inject an #GP when the guest attempts a PERF_METRICS RDPMC.

If the guest has PERF_METRICS but RDPMC interception is enabled for other reasons, KVM needs to emulate RDPMC with type 0x2000.

For simplicity, Metrics Clear Mode is not supported.

---
v8:
- Move kvm_need_perf_metrics_intercept() to patch 5/9. v6:
- Merge kvm_pmu_rdpmc_metrics() into intel_emulate_rdpmc().
- Reject non-zero index.
v5:
- new patch.

---
 arch/x86/kvm/pmu.c           |  1 +
 arch/x86/kvm/vmx/pmu_intel.c | 14 ++++++++++++++
 2 files changed, 15 insertions(+)
arch/x86/kvm/pmu.c +1 -0
index 789624658e8f..dab0bcabaa8f 100644
@@ -825,6 +825,7 @@ bool kvm_need_rdpmc_intercept(struct kvm_vcpu *vcpu)
 		return true;
 
 	return kvm_need_any_pmc_intercept(vcpu) ||
+	       kvm_need_perf_metrics_intercept(vcpu) ||
 	       pmu->counter_bitmask[KVM_PMC_GP] != (BIT_ULL(kvm_host_pmu.bit_width_gp) - 1) ||
 	       pmu->counter_bitmask[KVM_PMC_FIXED] != (BIT_ULL(kvm_host_pmu.bit_width_fixed) - 1);
 }
arch/x86/kvm/vmx/pmu_intel.c +14 -0
index bbf4afcffafc..3ca61e4b0ba6 100644
@@ -31,6 +31,7 @@
  */
 #define INTEL_RDPMC_GP		0
 #define INTEL_RDPMC_FIXED	INTEL_PMC_FIXED_RDPMC_BASE
+#define INTEL_RDPMC_METRICS	INTEL_PMC_FIXED_RDPMC_METRICS
 
 #define INTEL_RDPMC_TYPE_MASK	GENMASK(31, 16)
 #define INTEL_RDPMC_INDEX_MASK	GENMASK(15, 0)
@@ -125,6 +126,19 @@ static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,
 		counters = pmu->gp_counters;
 		num_counters = pmu->nr_arch_gp_counters;
 		break;
+	case INTEL_RDPMC_METRICS:
+		if (!kvm_vcpu_has_perf_metrics(vcpu))
+			return 1;
+
+		/*
+		 * The index in ECX[15:0] is implementation specific, but no
+		 * platform currently supports a non-zero index.
+		 */
+		if (idx)
+			return 1;
+
+		*data = pmu->perf_metrics;
+		return 0;
 	default:
 		return 1;
 	}
Signature · 1 line
2.55.0
#15 lore raw

Re: [PATCH v8 7/8] KVM: x86/pmu: Emulate RDPMC on performance metrics

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] Metrics Clear Mode is not emulated, causing guest application crashes (#GP) when RDPMC is…

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Metrics Clear Mode is not emulated, causing guest application crashes (#GP) when RDPMC is intercepted. --

--- Patch [7]: [PATCH v8 7/8] KVM: x86/pmu: Emulate RDPMC on performance metrics --- commit 5022d9b3f098d7ca5a383495c52fb9e9c119df3b
Author: Zide Chen <[email protected]>

KVM: x86/pmu: Emulate RDPMC on performance metrics

This commit adds emulation for RDPMC performance metrics when RDPMC interception is enabled. It handles type 0x2000 but skips Metrics Clear Mode for simplicity.

[ ... ]

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
index bbf4afcffafc8..3ca61e4b0ba68 100644

[ ... ]

Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +5 -0
@@ -125,6 +126,19 @@ static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,
 		counters = pmu->gp_counters;
 		num_counters = pmu->nr_arch_gp_counters;
 		break;
+	case INTEL_RDPMC_METRICS:
+		if (!kvm_vcpu_has_perf_metrics(vcpu))
+			return 1;
+
+		/*

arch/x86/kvm/vmx/pmu_intel.c static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, · line 143

Quoted 10 lines
arch/x86/kvm/vmx/pmu_intel.c +8 -0
+		 * The index in ECX[15:0] is implementation specific, but no
+		 * platform currently supports a non-zero index.
+		 */
+		if (idx)
+			return 1;
+
+		*data = pmu->perf_metrics;
+		return 0;
 	default:
 		return 1;

[Severity: Medium]
The commit message notes that Metrics Clear Mode is not supported for simplicity. However, if a guest application uses Metrics Clear Mode (setting bit 16 in ECX, making the type 0x20010000), it will fall through to this default case since it doesn't match INTEL_RDPMC_METRICS.

Returning 1 here causes KVM to inject a #GP. Does this cause unexpected crashes (like a SIGSEGV) for unprivileged guest userspace applications that legitimately try to use this architecturally valid mode when RDPMC is intercepted?

Quoted 1 line
arch/x86/kvm/vmx/pmu_intel.c +0 -0
 	}
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=7
author
#16 lore raw

Re: [PATCH v8 7/8] KVM: x86/pmu: Emulate RDPMC on performance metrics

Quoted 47 lines · On 8/17/2026 2:01 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Metrics Clear Mode is not emulated, causing guest application crashes (#GP) when RDPMC is intercepted. --

--- Patch [7]: [PATCH v8 7/8] KVM: x86/pmu: Emulate RDPMC on performance metrics --- commit 5022d9b3f098d7ca5a383495c52fb9e9c119df3b
Author: Zide Chen <[email protected]>

KVM: x86/pmu: Emulate RDPMC on performance metrics

This commit adds emulation for RDPMC performance metrics when RDPMC interception is enabled. It handles type 0x2000 but skips Metrics Clear Mode for simplicity.

[ ... ]

Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0 -0
index bbf4afcffafc8..3ca61e4b0ba68 100644

[ ... ]

Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +5 -0
@@ -125,6 +126,19 @@ static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,
 		counters = pmu->gp_counters;
 		num_counters = pmu->nr_arch_gp_counters;
 		break;
+	case INTEL_RDPMC_METRICS:
+		if (!kvm_vcpu_has_perf_metrics(vcpu))
+			return 1;
+
+		/*

arch/x86/kvm/vmx/pmu_intel.c static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, · line 143

Quoted 10 lines
arch/x86/kvm/vmx/pmu_intel.c +8 -0
+		 * The index in ECX[15:0] is implementation specific, but no
+		 * platform currently supports a non-zero index.
+		 */
+		if (idx)
+			return 1;
+
+		*data = pmu->perf_metrics;
+		return 0;
 	default:
 		return 1;

[Severity: Medium]
The commit message notes that Metrics Clear Mode is not supported for simplicity. However, if a guest application uses Metrics Clear Mode (setting bit 16 in ECX, making the type 0x20010000), it will fall through to this default case since it doesn't match INTEL_RDPMC_METRICS.

Quoted 4 lines

Returning 1 here causes KVM to inject a #GP. Does this cause unexpected crashes (like a SIGSEGV) for unprivileged guest userspace applications that legitimately try to use this architecturally valid mode when RDPMC is intercepted?

This is a false positive. Setting bit 16 in ECX is not indicating Metrics Clear mode. Whether or not clear metrics, depends on IA32_FIXED_CTR_CTRL.METRICS_CLEAR_EN[bit 14].

BTW, IA32_PERF_CAPABILITIES.RDPMC_METRICS_CLEAR[bit 19] is not exposed.

Quoted 2 lines
Quoted 1 line
arch/x86/kvm/vmx/pmu_intel.c +0 -0
 	}
#17 lore raw

Re: [PATCH v8 7/8] KVM: x86/pmu: Emulate RDPMC on performance metrics

In reply to Zide Chen
Quoted 8 lines · Zide Chen

If the host has the PERF_METRICS capability but it's not present on the guest, RDPMC interception must be enabled and KVM should inject an #GP when the guest attempts a PERF_METRICS RDPMC.

If the guest has PERF_METRICS but RDPMC interception is enabled for other reasons, KVM needs to emulate RDPMC with type 0x2000.

For simplicity, Metrics Clear Mode is not supported.

But is the bit cleared in PERF_CAPABILITIES? And would the enable bit fault?

It would be better to just support it, metrics clear is actually useful.

-Andi

author
#18 lore raw

Re: [PATCH v8 7/8] KVM: x86/pmu: Emulate RDPMC on performance metrics

Quoted 5 lines · Andi Kleen
Quoted 4 lines · Zide Chen

If the host has the PERF_METRICS capability but it's not present on the guest, RDPMC interception must be enabled and KVM should inject an #GP when the guest attempts a PERF_METRICS RDPMC.

Quoted 7 lines
Quoted 4 lines

If the guest has PERF_METRICS but RDPMC interception is enabled for other reasons, KVM needs to emulate RDPMC with type 0x2000.

For simplicity, Metrics Clear Mode is not supported.

But is the bit cleared in PERF_CAPABILITIES? And would the enable bit fault?

Currently KVM doesn't advertise
IA32_PERF_CAPABILITIES.RDPMC_METRICS_CLEAR[19].

Quoted 1 line

It would be better to just support it, metrics clear is actually useful.

Sure, I'll add it.

Quoted 1 line

-Andi

#19 lore raw

Re: [PATCH v8 7/8] KVM: x86/pmu: Emulate RDPMC on performance metrics

Quoted 11 lines · Zide
Quoted 8 lines · Andi Kleen
Quoted 7 lines · Zide Chen

If the host has the PERF_METRICS capability but it's not present on the guest, RDPMC interception must be enabled and KVM should inject an #GP when the guest attempts a PERF_METRICS RDPMC.

If the guest has PERF_METRICS but RDPMC interception is enabled for other reasons, KVM needs to emulate RDPMC with type 0x2000.

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

For simplicity, Metrics Clear Mode is not supported.

But is the bit cleared in PERF_CAPABILITIES? And would the enable bit fault?

Currently KVM doesn't advertise
IA32_PERF_CAPABILITIES.RDPMC_METRICS_CLEAR[19].

Thanks.

How about the fault for the enable bit?

Quoted 3 lines
Quoted 1 line

It would be better to just support it, metrics clear is actually useful.

Sure, I'll add it.

I guess it could be a follow-on patch, no need to hold up this one.

-Andi

author
#20 lore raw

Re: [PATCH v8 7/8] KVM: x86/pmu: Emulate RDPMC on performance metrics

Quoted 17 lines · Andi Kleen
Quoted 16 lines · Zide
Quoted 12 lines · Andi Kleen
Quoted 8 lines · Zide Chen

If the host has the PERF_METRICS capability but it's not present on the guest, RDPMC interception must be enabled and KVM should inject an #GP when the guest attempts a PERF_METRICS RDPMC.

If the guest has PERF_METRICS but RDPMC interception is enabled for other reasons, KVM needs to emulate RDPMC with type 0x2000.

For simplicity, Metrics Clear Mode is not supported.

But is the bit cleared in PERF_CAPABILITIES? And would the enable bit fault?

Quoted 6 lines
Quoted 2 lines

Currently KVM doesn't advertise
IA32_PERF_CAPABILITIES.RDPMC_METRICS_CLEAR[19].

Thanks.

How about the fault for the enable bit?

Yes, KVM always injects a #GP if the guest writes
IA32_FIXED_CTR_CTRL.METRICS_CLEAR_EN[bit 14].

Quoted 6 lines
Quoted 4 lines
Quoted 2 lines

It would be better to just support it, metrics clear is actually useful.

Sure, I'll add it.

I guess it could be a follow-on patch, no need to hold up this one.

OK.

author
#21 lore raw

PATCH v8 8/8 KVM: selftests: Add PERF_METRICS and fixed counter 3 tests

In reply to Zide Chen

Add a test case 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.

---
v7:
- Use RDPMC instead of RDMSR in PERF_METRICS sanity test. v6:
- Move perf metrics test out of test_arch_events(); it doesn't belong there, and this also avoids redundant runs of perf metrics test.
- Correct the +/-3 error margin.
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     | 93 ++++++++++++++++++-
 3 files changed, 92 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 +88 -5

141 lines not on this page · raw message

Signature · 1 line
2.55.0
#22 lore raw

Re: [PATCH v8 8/8] 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: - [Medium] Unhandled #GP / Selftest Crash when writing 0xdeaddead to MSR_PERF_METRICS. --

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Unhandled #GP / Selftest Crash when writing 0xdeaddead to MSR_PERF_METRICS. --

--- Patch [8]: [PATCH v8 8/8] KVM: selftests: Add PERF_METRICS and fixed counter 3 tests --- commit 19d6869fac1fbdcc63ff54623403cec33b922848
Author: Zide Chen <[email protected]>

KVM: selftests: Add PERF_METRICS and fixed counter 3 tests

This commit adds a test case to exercise IA32_PERF_METRICS for Topdown (TMA) Level 1 metrics and extends the PMU selftest to cover Intel fixed counter 3.

Quoted 3 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0 -0

[ ... ]

Quoted 2 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +2 -0
+static void __guest_test_perf_metrics(void)
+{

[ ... ]

tools/testing/selftests/kvm/x86/pmu_counters_test.c static void test_fixed_counters(u8 pmu_version, u64 perf_capabilities) · line 708

Quoted 7 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +7 -0
+	/* Sanity check after PERF_METRICS disabled. */
+	__asm__ __volatile__("loop ." : "+c"((int){NUM_LOOPS}));
+	GUEST_ASSERT_EQ(rdmsr(MSR_PERF_METRICS), metrics);
+	wrmsr(MSR_PERF_METRICS, 0xdeaddead);
+
+	guest_test_rdpmc(INTEL_RDPMC_METRICS, true, 0xdeaddead);
+}

[Severity: Medium]
Will this wrmsr(MSR_PERF_METRICS, 0xdeaddead) cause an unhandled #GP and crash the selftest on hosts with Mediated PMU pass-through?

Writing an architecturally invalid non-zero value to MSR_PERF_METRICS generates a #GP on real hardware. Under Mediated PMU, KVM configures the VMCS to pass-through MSR accesses directly to hardware, so this fault would be injected into the guest.

If the hardware silently discards the write instead of faulting, will the subsequent guest_test_rdpmc() assertion fail because it expects to read back 0xdeaddead, while the hardware holds the previous value?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=8
author
#23 lore raw

Re: [PATCH v8 8/8] KVM: selftests: Add PERF_METRICS and fixed counter 3 tests

Quoted 33 lines · On 8/17/2026 1:57 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Unhandled #GP / Selftest Crash when writing 0xdeaddead to MSR_PERF_METRICS. --

--- Patch [8]: [PATCH v8 8/8] KVM: selftests: Add PERF_METRICS and fixed counter 3 tests --- commit 19d6869fac1fbdcc63ff54623403cec33b922848
Author: Zide Chen <[email protected]>

KVM: selftests: Add PERF_METRICS and fixed counter 3 tests

This commit adds a test case to exercise IA32_PERF_METRICS for Topdown (TMA) Level 1 metrics and extends the PMU selftest to cover Intel fixed counter 3.

Quoted 3 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0 -0

[ ... ]

Quoted 2 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +2 -0
+static void __guest_test_perf_metrics(void)
+{

[ ... ]

tools/testing/selftests/kvm/x86/pmu_counters_test.c static void test_fixed_counters(u8 pmu_version, u64 perf_capabilities) · line 708

Quoted 7 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +7 -0
+	/* Sanity check after PERF_METRICS disabled. */
+	__asm__ __volatile__("loop ." : "+c"((int){NUM_LOOPS}));
+	GUEST_ASSERT_EQ(rdmsr(MSR_PERF_METRICS), metrics);
+	wrmsr(MSR_PERF_METRICS, 0xdeaddead);
+
+	guest_test_rdpmc(INTEL_RDPMC_METRICS, true, 0xdeaddead);
+}

[Severity: Medium]
Will this wrmsr(MSR_PERF_METRICS, 0xdeaddead) cause an unhandled #GP and crash the selftest on hosts with Mediated PMU pass-through?

Quoted 4 lines

Writing an architecturally invalid non-zero value to MSR_PERF_METRICS generates a #GP on real hardware. Under Mediated PMU, KVM configures the VMCS to pass-through MSR accesses directly to hardware, so this fault would be injected into the guest.

0xdeaddead is a valid non-zero value. So this statement is supposed to work well.

Quoted 5 lines

If the hardware silently discards the write instead of faulting, will the subsequent guest_test_rdpmc() assertion fail because it expects to read back 0xdeaddead, while the hardware holds the previous value?