KVM: x86/pmu: Add hardware Topdown metrics support
Subsystems (5)
v9 cover KVM: x86/pmu: Add hardware Topdown metrics support
Previous revisions were missing two things that are now addressed:
- A previous comment incorrectly claimed that MSR_PERF_METRICS bits [63:32] are ignored on hardware that only implements level-1 metrics; patch 10/12 fixes this by rejecting such writes.
- Patch 11/12 adds support for RDPMC Metrics Clear Mode. Without this feature, for user programs using RDPMC directly, the counter has to be reset explicitly using ioctl:
ioctl(perf_fd, PERF_EVENT_IOC_RESET, 0);
Tests: besides the tests mentioned on v8 cover letter, tested RDPMC Metrics Clear on a DMR machine, and tested Topdown metrics on nested setup on top of the vPMU support series [1].
perf tool misidentifies "cpu/slots,metrics_clear=1/" as not being the slots event, so it inserts an additional slots event into the group; event scheduling then fails because the group needs fixed counter 3 (SLOTS) twice:
$ perf stat -e "{cpu/slots,metrics_clear=1/,cpu/topdown-retiring/}" -- sleep 1
WARNING: events were regrouped to match PMUs
<not counted> slots
<not supported> cpu/slots,metrics_clear=1/
<not counted> cpu/topdown-retiring/A fix for this perf tool bug is posted at [2].
New patches:
patch 8/12: Add INTEL_TD_METRIC_FIELD_BITS for code cleanup. patch 9/12: perf/x86: Expose number of Topdown metric events to KVM. patch 10/12: Reject writes to reserved MSR_PERF_METRICS bits. patch 11/12: Support RDPMC Metrics Clear Mode.
Minor non-logical changes to existing patches:
patch 5/12: Remove the incorrect comment about writing to PERF_METRICS. patch 7/12: Remove references to Metrics Clear from the changelog.
Patch 11/12 could be folded into patch 5/12, but I kept them separate because patches 1-8 were already well reviewed, and keeping them unchanged should make reviewing v9 easier.
Rebased to kvm-x86/next: 70c944caf570fda2d79baa71435589a8db39f048
v8: https://lore.kernel.org/kvm/[email protected]/
QEMU: https://lore.kernel.org/qemu-devel/[email protected]/
[1] https://lore.kernel.org/kvm/[email protected]/
[2] https://lore.kernel.org/linux-perf-users/[email protected]/T/#t
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 (9):
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 perf/x86: Add INTEL_TD_METRIC_FIELD_{BITS,MASK} constants perf/x86: Expose number of Topdown metric events to KVM
KVM: x86/pmu: Reject writes to reserved MSR_PERF_METRICS bits
KVM: x86/pmu: Support RDPMC Metrics Clear Mode
arch/x86/events/core.c | 1 + arch/x86/events/intel/core.c | 8 +- arch/x86/include/asm/kvm-x86-pmu-ops.h | 2 +- arch/x86/include/asm/kvm_host.h | 5 +- arch/x86/include/asm/msr-index.h | 2 + arch/x86/include/asm/perf_event.h | 5 + arch/x86/kvm/msrs.c | 10 +- arch/x86/kvm/pmu.c | 37 +++-- 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 | 132 +++++++++++++++--- arch/x86/kvm/vmx/pmu_intel.h | 12 +- arch/x86/kvm/vmx/vmx.c | 19 ++- 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 | 97 ++++++++++++- 18 files changed, 303 insertions(+), 67 deletions(-)
Signature · 1 line
2.55.0
PATCH v9 1/12 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.
--- arch/x86/kvm/vmx/pmu_intel.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)
arch/x86/kvm/vmx/pmu_intel.c +8 -6
@@ -468,11 +468,8 @@ 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 ,
@@ -481,8 +478,13 @@ 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
PATCH v9 2/12 KVM: x86/pmu: Support Intel fixed counter 3 on mediated vPMU
In reply to Zide ChenFrom: 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.
--- 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
@@ -563,7 +563,7 @@ 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
@@ -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 ,
@@ -2672,7 +2672,7 @@ 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
@@ -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_MAX_NR_INTEL_FIXED_COUNTERS counters starting at +* KVM_FIXED_PMC_BASE_IDX <=> 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
PATCH v9 3/12 KVM: x86/pmu: Rename and move vcpu_get_perf_capabilities() to pmu.h
In reply to Zide ChenThis 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.
--- 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
@@ -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
@@ -193,13 +193,13 @@ 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 ;
@@ -554,7 +554,7 @@ 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
@@ -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
PATCH v9 4/12 KVM: x86/pmu: Snapshot host IA32_PERF_CAPABILITIES in kvm_host
In reply to Zide ChenFrom: 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
@@ -295,6 +295,7 @@ 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
@@ -7949,14 +7949,10 @@ 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);
@@ -7969,11 +7965,11 @@ 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
@@ -7037,6 +7037,10 @@ 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
PATCH v9 5/12 KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
In reply to Zide ChenFrom: 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].
---
v9:
- Remove the incorrect comment about writing to MSR_PERF_METRICS. v8:
- Intercept global ctrl MSR if host supports PerfMetrics but not enabled in the guest.
--- 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 | 29 +++++++++++++++++++++++++++++ arch/x86/kvm/vmx/vmx.c | 7 +++++++ 8 files changed, 57 insertions(+), 1 deletion(-)
arch/x86/include/asm/kvm_host.h +1 -0
@@ -584,6 +584,7 @@ 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
@@ -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
@@ -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. */
@@ -2609,6 +2609,10 @@ 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
@@ -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
@@ -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
@@ -724,6 +724,8 @@ 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 +29 -0
@@ -192,6 +192,8 @@ static bool intel_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)case MSR_CORE_PERF_GLOBAL_OVF_CTRL :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 ;
@@ -349,6 +351,9 @@ 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 ;
@@ -398,6 +403,9 @@ 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 : + pmu->perf_metrics = data; +break ;case MSR_IA32_PEBS_ENABLE :if (data & pmu->pebs_enable_rsvd)return 1 ;
@@ -580,6 +588,8 @@ 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)
@@ -636,6 +646,9 @@ 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); }
@@ -806,6 +819,9 @@ 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)
@@ -834,6 +850,19 @@ 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
@@ -4289,6 +4289,10 @@ 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)
@@ -7994,6 +7998,9 @@ 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
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
arch/x86/kvm/pmu.h static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu) · line 285
Quoted 5 lines · Zide Chen
arch/x86/kvm/pmu.h +4
-0+static inline bool kvm_vcpu_has_perf_metrics (struct kvm_vcpu *vcpu) +{ +return kvm_vcpu_get_perf_caps (vcpu) &PERF_CAP_PERF_METRICS ; +}
The SDM, volume 3, section 22.3.9.3, says:
Quoted 3 lines
IA32_PERF_GLOBAL_CTRL. EN_PERF_METRICS[48]: If this bit is set *and* fixed-function performance-monitoring counter 3 is enabled, built-in performance metrics are enabled. [emphasis mine]
and
Quoted 2 lines
The values in
MSR_PERF_METRICSare derived from fixed-function performance-monitoring counter 3.
Hence, a dependency on fixed counter 3 belongs in this predicate. Perhaps something like:
static inline bool kvm_pmu_has_fixed_counter(struct kvm_pmu *pmu,
unsigned int idx)
{
return idx < pmu->nr_arch_fixed_counters;
}
static inline bool kvm_vcpu_has_perf_metrics(struct kvm_vcpu *vcpu)
{
return (kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PERF_METRICS) &&
kvm_pmu_has_fixed_counter(vcpu_to_pmu(vcpu),
INTEL_PMC_IDX_FIXED_SLOTS -
INTEL_PMC_IDX_FIXED);
}with intel_pmu_refresh() calling kvm_vcpu_has_perf_metrics() instead of testing perf_capabilities directly. (nr_arch_fixed_counters has already been computed at that point; all_valid_pmc_idx has not.)
KVM only supports contiguous fixed counters today (see patch 2), so a simple comparison against nr_arch_fixed_counters suffices. However, if KVM ever honors the CPUID.0AH:ECX fixed counter bitmap, a separate helper keeps that assumption in one place.
As it stands, if userspace sets PERF_CAP_PERF_METRICS for a version 1 vPMU, MSR_PERF_METRICS is passed through (kvm_vcpu_has_mediated_pmu() only requires a non-zero version), but intel_pmu_refresh() bails out before it gets to the PERF_METRICS code. With patch 11, that leaves perf_metrics_rsvd at ~0ull, so a value the guest writes directly to hardware can be saved with KVM_GET_MSRS, but not restored with KVM_SET_MSRS.
Since nr_arch_fixed_counters is 0 for a v1 vPMU, the above handles that case too. It also takes care of the METRICS_CLEAR_EN unmasking without fixed counter 3 that Sashiko flagged on patch 12, and it obviates the NULL pmc check in intel_emulate_rdpmc().
[...]
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 3 lines
arch/x86/kvm/vmx/pmu_intel.c +3
-0+case MSR_PERF_METRICS : + pmu->perf_metrics = data; +break ;
On a part with only four metrics (e.g. Ice Lake), userspace can stuff bits 63:32 into pmu->perf_metrics here, and the unchecked wrmsrq() in intel_mediated_pmu_load() will then #GP on every VM-entry. Patch 11 isn't an optional follow-on; it needs to be folded into this patch.
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
Quoted 35 lines · Jim Mattson
arch/x86/kvm/pmu.h static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu) · line 285
Quoted 5 lines · Zide Chen
arch/x86/kvm/pmu.h +4
-0+static inline bool kvm_vcpu_has_perf_metrics (struct kvm_vcpu *vcpu) +{ +return kvm_vcpu_get_perf_caps (vcpu) &PERF_CAP_PERF_METRICS ; +}The SDM, volume 3, section 22.3.9.3, says:
Quoted 3 lines
IA32_PERF_GLOBAL_CTRL. EN_PERF_METRICS[48]: If this bit is set *and* fixed-function performance-monitoring counter 3 is enabled, built-in performance metrics are enabled. [emphasis mine]and
Quoted 2 lines
The values in
MSR_PERF_METRICSare derived from fixed-function performance-monitoring counter 3.Hence, a dependency on fixed counter 3 belongs in this predicate. Perhaps something like:
static inline bool kvm_pmu_has_fixed_counter(struct kvm_pmu *pmu, unsigned int idx) { return idx < pmu->nr_arch_fixed_counters; } static inline bool kvm_vcpu_has_perf_metrics(struct kvm_vcpu *vcpu) { return (kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PERF_METRICS) && kvm_pmu_has_fixed_counter(vcpu_to_pmu(vcpu), INTEL_PMC_IDX_FIXED_SLOTS - INTEL_PMC_IDX_FIXED); }
Quoted 3 lines
with
intel_pmu_refresh()callingkvm_vcpu_has_perf_metrics()instead of testingperf_capabilitiesdirectly. (nr_arch_fixed_countershas already been computed at that point;all_valid_pmc_idxhas not.)
Thanks for the detailed analysis. That makes total sense.
Initially I didn't gate PERF_METRICS on fixed counter 3 out of concern about policing userspace configurations. Additionally, as in the quoted SDM text above, it means PERF_METRICS can't function without fixed counter 3, but strictly speaking, the SDM doesn't prevent other operations, for example, writing the METRICS_CLEAR_EN bit.
But having IA32_PERF_CAPABILITIES[15] set in the absence of FC3 is not going to happen natively, so I think it doesn't break any rules if KVM chooses to gate PERF_METRICS on FC3. And as you pointed out, it's a simple and effective way to protect KVM invariants and migration integrity.
I will add the FC3 check to kvm_vcpu_has_perf_metrics().
Quoted 5 lines
KVM only supports contiguous fixed counters today (see patch 2), so a simple comparison against
nr_arch_fixed_counterssuffices. However, if KVM ever honors the CPUID.0AH:ECX fixed counter bitmap, a separate helper keeps that assumption in one place.
Before the pmu v5 series is merged:
https://lore.kernel.org/kvm/[email protected]/
I have to continue use nr_arch_fixed_counters in this series.
Quoted 9 lines
As it stands, if userspace sets
PERF_CAP_PERF_METRICSfor a version 1 vPMU,MSR_PERF_METRICSis passed through (kvm_vcpu_has_mediated_pmu()only requires a non-zero version), butintel_pmu_refresh()bails out before it gets to thePERF_METRICScode. With patch 11, that leavesperf_metrics_rsvdat ~0ull, so a value the guest writes directly to hardware can be saved withKVM_GET_MSRS, but not restored withKVM_SET_MSRS.
Quoted 4 lines
Since
nr_arch_fixed_countersis 0 for a v1 vPMU, the above handles that case too. It also takes care of theMETRICS_CLEAR_ENunmasking without fixed counter 3 that Sashiko flagged on patch 12, and it obviates the NULL pmc check inintel_emulate_rdpmc().
Yes, with FC3 gating, these issues can be easily resolved.
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 408
Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +3
-0+case MSR_PERF_METRICS : + pmu->perf_metrics = data; +break ;On a part with only four metrics (e.g. Ice Lake), userspace can stuff bits 63:32 into
pmu->perf_metricshere, and the uncheckedwrmsrq()inintel_mediated_pmu_load()will then #GP on every VM-entry. Patch 11 isn't an optional follow-on; it needs to be folded into this patch.
Yes, agree.
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
Quoted 51 lines · Zide
Quoted 35 lines · Jim Mattson
arch/x86/kvm/pmu.h static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu) · line 285
Quoted 5 lines · Zide Chen
arch/x86/kvm/pmu.h +4
-0+static inline bool kvm_vcpu_has_perf_metrics (struct kvm_vcpu *vcpu) +{ +return kvm_vcpu_get_perf_caps (vcpu) &PERF_CAP_PERF_METRICS ; +}The SDM, volume 3, section 22.3.9.3, says:
Quoted 3 lines
IA32_PERF_GLOBAL_CTRL. EN_PERF_METRICS[48]: If this bit is set *and* fixed-function performance-monitoring counter 3 is enabled, built-in performance metrics are enabled. [emphasis mine]and
Quoted 2 lines
The values in
MSR_PERF_METRICSare derived from fixed-function performance-monitoring counter 3.Hence, a dependency on fixed counter 3 belongs in this predicate. Perhaps something like:
static inline bool kvm_pmu_has_fixed_counter(struct kvm_pmu *pmu, unsigned int idx) { return idx < pmu->nr_arch_fixed_counters; } static inline bool kvm_vcpu_has_perf_metrics(struct kvm_vcpu *vcpu) { return (kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PERF_METRICS) && kvm_pmu_has_fixed_counter(vcpu_to_pmu(vcpu), INTEL_PMC_IDX_FIXED_SLOTS - INTEL_PMC_IDX_FIXED); }Quoted 3 lines
with
intel_pmu_refresh()callingkvm_vcpu_has_perf_metrics()instead of testingperf_capabilitiesdirectly. (nr_arch_fixed_countershas already been computed at that point;all_valid_pmc_idxhas not.)Thanks for the detailed analysis. That makes total sense.
Initially I didn't gate
PERF_METRICSon fixed counter 3 out of concern about policing userspace configurations. Additionally, as in the quoted SDM text above, it meansPERF_METRICScan't function without fixed counter 3, but strictly speaking, the SDM doesn't prevent other operations, for example, writing theMETRICS_CLEAR_ENbit.
Quoted 6 lines
But having IA32_PERF_CAPABILITIES[15] set in the absence of FC3 is not going to happen natively, so I think it doesn't break any rules if KVM chooses to gate
PERF_METRICSon FC3. And as you pointed out, it's a simple and effective way to protect KVM invariants and migration integrity.I will add the FC3 check to
kvm_vcpu_has_perf_metrics().
If userspace sets IA32_PERF_CAPABILITIES[15] but not FC3 (for some reason), I think architecturally it should still be valid to read/write PERF_METRICS. KVM disallowing access to PERF_METRICS while IA32_PERF_CAPABILITIES[15] is set would not be correct architecturally, right?
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
Quoted 57 lines · Yosry Ahmed
Quoted 56 lines · Zide
Quoted 35 lines · Jim Mattson
arch/x86/kvm/pmu.h static inline u64 kvm_vcpu_get_perf_caps(struct kvm_vcpu *vcpu) · line 285
Quoted 5 lines · Zide Chen
arch/x86/kvm/pmu.h +4
-0+static inline bool kvm_vcpu_has_perf_metrics (struct kvm_vcpu *vcpu) +{ +return kvm_vcpu_get_perf_caps (vcpu) &PERF_CAP_PERF_METRICS ; +}The SDM, volume 3, section 22.3.9.3, says:
Quoted 3 lines
IA32_PERF_GLOBAL_CTRL. EN_PERF_METRICS[48]: If this bit is set *and* fixed-function performance-monitoring counter 3 is enabled, built-in performance metrics are enabled. [emphasis mine]and
Quoted 2 lines
The values in
MSR_PERF_METRICSare derived from fixed-function performance-monitoring counter 3.Hence, a dependency on fixed counter 3 belongs in this predicate. Perhaps something like:
static inline bool kvm_pmu_has_fixed_counter(struct kvm_pmu *pmu, unsigned int idx) { return idx < pmu->nr_arch_fixed_counters; } static inline bool kvm_vcpu_has_perf_metrics(struct kvm_vcpu *vcpu) { return (kvm_vcpu_get_perf_caps(vcpu) & PERF_CAP_PERF_METRICS) && kvm_pmu_has_fixed_counter(vcpu_to_pmu(vcpu), INTEL_PMC_IDX_FIXED_SLOTS - INTEL_PMC_IDX_FIXED); }Quoted 3 lines
with
intel_pmu_refresh()callingkvm_vcpu_has_perf_metrics()instead of testingperf_capabilitiesdirectly. (nr_arch_fixed_countershas already been computed at that point;all_valid_pmc_idxhas not.)Thanks for the detailed analysis. That makes total sense.
Initially I didn't gate
PERF_METRICSon fixed counter 3 out of concern about policing userspace configurations. Additionally, as in the quoted SDM text above, it meansPERF_METRICScan't function without fixed counter 3, but strictly speaking, the SDM doesn't prevent other operations, for example, writing theMETRICS_CLEAR_ENbit.But having IA32_PERF_CAPABILITIES[15] set in the absence of FC3 is not going to happen natively, so I think it doesn't break any rules if KVM chooses to gate
PERF_METRICSon FC3. And as you pointed out, it's a simple and effective way to protect KVM invariants and migration integrity.
Quoted 7 lines
Quoted 1 line
I will add the FC3 check to
kvm_vcpu_has_perf_metrics().If userspace sets IA32_PERF_CAPABILITIES[15] but not FC3 (for some reason), I think architecturally it should still be valid to read/write
PERF_METRICS. KVM disallowing access toPERF_METRICSwhile IA32_PERF_CAPABILITIES[15] is set would not be correct architecturally, right?
Right. Per SDM 22.8, MSR_PERF_METRICS is available as long as IA32_PERF_CAPABILITIES[15] is set, so disallowing access to it when bit 15 is exposed seems architecturally incorrect. However, TMA metrics cannot be enabled or computed without FC3 (SDM 22.3.9.3).
Instead of gating kvm_vcpu_has_perf_metrics() on FC3, seems it's better off to do some changes to fix the invariants where they belong:
1. Untie FC3 from kvm_vcpu_has_perf_metrics() so the MSR remains accessible.
2. In intel_pmu_refresh(), compute perf_metrics_rsvd before the PMU v1 early return so save/restore works properly.
3. Explicitly check for FC3 when unmasking GLOBAL_CTRL[48] (TMA enable) and FIXED_CTR_CTRL[14] (METRICS_CLEAR_EN), and guard pmc in RDPMC emulation.
This seems satisfy the SDM while keeping migration and invariants intact.
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
In reply to Zide Chenarch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 809
Quoted 9 lines
arch/x86/kvm/pmu.c +1
-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)
The name is confusing, this doesn't control the intercept of PERF_METRICS (and it wouldn't be correct to use it for that AFAICT).
Quoted 5 lines
arch/x86/kvm/pmu.c +4
-0+{ +return (kvm_host.perf_capabilities &PERF_CAP_PERF_METRICS ) && + !kvm_vcpu_has_perf_metrics (vcpu); +}
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
Quoted 8 lines · Yosry Ahmed
Quoted 8 lines
arch/x86/kvm/pmu.c +0
-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; }
Quoted 4 lines
arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 809
Quoted 1 line
arch/x86/kvm/pmu.c +1
-0+static bool kvm_need_perf_metrics_intercept (struct kvm_vcpu *vcpu)The name is confusing, this doesn't control the intercept of
PERF_METRICS(and it wouldn't be correct to use it for that AFAICT).
Right, these two APIs do not directly determine whether the mentioned resources need interception:
kvm_need_any_pmc_intercept()kvm_need_perf_metrics_intercept()
This is different from kvm_need_perf_global_ctrl_intercept() and kvm_need_rdpmc_intercept().
How about change them to something like these?
kvm_pmu_has_unexposed_pmcs()kvm_pmu_has_unexposed_perf_metrics()
Or merge them into one single API? kvm_pmu_has_unexposed_resources()
Quoted 6 lines
Quoted 6 lines
arch/x86/kvm/pmu.c +4
-0+{ +return (kvm_host.perf_capabilities &PERF_CAP_PERF_METRICS ) && + !kvm_vcpu_has_perf_metrics (vcpu); +}
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
Quoted 17 lines · Zide
Quoted 8 lines · Yosry Ahmed
Quoted 8 lines
arch/x86/kvm/pmu.c +0
-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; }Quoted 4 lines
arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 809
Quoted 1 line
arch/x86/kvm/pmu.c +1
-0+static bool kvm_need_perf_metrics_intercept (struct kvm_vcpu *vcpu)The name is confusing, this doesn't control the intercept of
PERF_METRICS(and it wouldn't be correct to use it for that AFAICT).
Quoted 5 lines
Right, these two APIs do not directly determine whether the mentioned resources need interception:
kvm_need_any_pmc_intercept()kvm_need_perf_metrics_intercept()
Jim suggested folding kvm_need_perf_metrics_intercept() into kvm_need_any_pmc_intercept(). The name of kvm_need_any_pmc_intercept() might be confusing as well, but at least it's one less interface to worry about?
Quoted 10 lines
This is different from
kvm_need_perf_global_ctrl_intercept()andkvm_need_rdpmc_intercept().How about change them to something like these?
kvm_pmu_has_unexposed_pmcs()kvm_pmu_has_unexposed_perf_metrics()Or merge them into one single API?
kvm_pmu_has_unexposed_resources()
I don't like this name any better, but I don't have better suggestions. Since we already have kvm_need_any_pmc_intercept(), let's just use it for now?
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
Quoted 36 lines · Yosry Ahmed
Quoted 17 lines · Zide
Quoted 8 lines · Yosry Ahmed
Quoted 8 lines
arch/x86/kvm/pmu.c +0
-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; }Quoted 4 lines
arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 809
Quoted 1 line
arch/x86/kvm/pmu.c +1
-0+static bool kvm_need_perf_metrics_intercept (struct kvm_vcpu *vcpu)The name is confusing, this doesn't control the intercept of
PERF_METRICS(and it wouldn't be correct to use it for that AFAICT).Quoted 5 lines
Right, these two APIs do not directly determine whether the mentioned resources need interception:
kvm_need_any_pmc_intercept()kvm_need_perf_metrics_intercept()Jim suggested folding
kvm_need_perf_metrics_intercept()intokvm_need_any_pmc_intercept(). The name ofkvm_need_any_pmc_intercept()might be confusing as well, but at least it's one less interface to worry about?Quoted 6 lines
This is different from
kvm_need_perf_global_ctrl_intercept()andkvm_need_rdpmc_intercept().How about change them to something like these?
Quoted 8 lines
Quoted 4 lines
kvm_pmu_has_unexposed_pmcs()kvm_pmu_has_unexposed_perf_metrics()Or merge them into one single API?
kvm_pmu_has_unexposed_resources()I don't like this name any better, but I don't have better suggestions. Since we already have
kvm_need_any_pmc_intercept(), let's just use it for now?
How about kvm_pmu_incomplete_passthrough()?
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
Quoted 41 lines · Jim Mattson
Quoted 39 lines · Yosry Ahmed
Quoted 17 lines · Zide
Quoted 8 lines · Yosry Ahmed
Quoted 8 lines
arch/x86/kvm/pmu.c +0
-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; }Quoted 4 lines
arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 809
Quoted 1 line
arch/x86/kvm/pmu.c +1
-0+static bool kvm_need_perf_metrics_intercept (struct kvm_vcpu *vcpu)The name is confusing, this doesn't control the intercept of
PERF_METRICS(and it wouldn't be correct to use it for that AFAICT).Quoted 5 lines
Right, these two APIs do not directly determine whether the mentioned resources need interception:
kvm_need_any_pmc_intercept()kvm_need_perf_metrics_intercept()Jim suggested folding
kvm_need_perf_metrics_intercept()intokvm_need_any_pmc_intercept(). The name ofkvm_need_any_pmc_intercept()might be confusing as well, but at least it's one less interface to worry about?Quoted 9 lines
This is different from
kvm_need_perf_global_ctrl_intercept()andkvm_need_rdpmc_intercept().How about change them to something like these?
kvm_pmu_has_unexposed_pmcs()kvm_pmu_has_unexposed_perf_metrics()
Quoted 7 lines
Quoted 5 lines
Quoted 1 line
Or merge them into one single API?
kvm_pmu_has_unexposed_resources()I don't like this name any better, but I don't have better suggestions. Since we already have
kvm_need_any_pmc_intercept(), let's just use it for now?How about
kvm_pmu_incomplete_passthrough()?
Better than kvm_need_any_pmc_intercept() imo.
Re: [PATCH v9 05/12] KVM: x86/pmu: Support PERF_METRICS MSR in mediated vPMU
Quoted 44 lines · Yosry Ahmed
Quoted 43 lines · Jim Mattson
Quoted 41 lines · Yosry Ahmed
Quoted 17 lines · Zide
Quoted 8 lines · Yosry Ahmed
Quoted 8 lines
arch/x86/kvm/pmu.c +0
-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; }Quoted 4 lines
arch/x86/kvm/pmu.c static bool kvm_need_any_pmc_intercept(struct kvm_vcpu *vcpu) · line 809
Quoted 1 line
arch/x86/kvm/pmu.c +1
-0+static bool kvm_need_perf_metrics_intercept (struct kvm_vcpu *vcpu)The name is confusing, this doesn't control the intercept of
PERF_METRICS(and it wouldn't be correct to use it for that AFAICT).Quoted 5 lines
Right, these two APIs do not directly determine whether the mentioned resources need interception:
kvm_need_any_pmc_intercept()kvm_need_perf_metrics_intercept()Jim suggested folding
kvm_need_perf_metrics_intercept()intokvm_need_any_pmc_intercept(). The name ofkvm_need_any_pmc_intercept()might be confusing as well, but at least it's one less interface to worry about?Quoted 10 lines
This is different from
kvm_need_perf_global_ctrl_intercept()andkvm_need_rdpmc_intercept().How about change them to something like these?
kvm_pmu_has_unexposed_pmcs()kvm_pmu_has_unexposed_perf_metrics()Or merge them into one single API?
kvm_pmu_has_unexposed_resources()
Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
I don't like this name any better, but I don't have better suggestions. Since we already have
kvm_need_any_pmc_intercept(), let's just use it for now?How about
kvm_pmu_incomplete_passthrough()?Better than
kvm_need_any_pmc_intercept()imo.
OK, I can rename it in v3.
PATCH v9 6/12 KVM: x86/pmu: Move RDPMC emulation into per-vendor callbacks
In reply to Zide ChenThe 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
@@ -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
@@ -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
@@ -24,8 +24,8 @@#define KVM_FIXED_PMC_BASE_IDX INTEL_PMC_IDX_FIXEDstruct 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
@@ -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
@@ -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)
@@ -866,7 +865,7 @@ 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
PATCH v9 7/12 KVM: x86/pmu: Emulate RDPMC on performance metrics
In reply to Zide ChenIf 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.
---
v9:
- Remove references to Metrics Clear from the changelog. v8:
- Move kvm_need_perf_metrics_intercept() to patch 5/9.
--- 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
@@ -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
@@ -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
Re: [PATCH v9 07/12] KVM: x86/pmu: Emulate RDPMC on performance metrics
Quoted 10 lines · Zide Chen
If the host has the
PERF_METRICScapability but it's not present on the guest, RDPMC interception must be enabled and KVM should inject an #GP when the guest attempts aPERF_METRICSRDPMC.If the guest has
PERF_METRICSbut RDPMC interception is enabled for other reasons, KVM needs to emulate RDPMC with type0x2000.Signed-off-by Zide Chen <[email protected]>Reviewed-by Dapeng Mi <[email protected]>
PATCH v9 8/12 KVM: selftests: Add PERF_METRICS and fixed counter 3 tests
In reply to Zide ChenAdd 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.
--- 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 | 97 ++++++++++++++++++- 3 files changed, 96 insertions(+), 5 deletions(-)
tools/arch/x86/include/asm/msr-index.h +1 -0
@@ -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
@@ -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 +92 -5
145 lines not on this page · raw message
Signature · 1 line
2.55.0
[email protected]
AI reviewRe: [PATCH v9 08/12] 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] selftests: unconditional MSR read causes assertion failure on older kernels --
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] selftests: unconditional MSR read causes assertion failure on older kernels --
commit c5ae582e90bae306393c4bf2f37f04a11e37ba5f
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 Level 1 metrics. It also extends the PMU selftest to cover Intel fixed counter 3 by bumping the maximum number of fixed counters and validating functionality.
Quoted 3 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0
-0
[ ... ]
Quoted 1 line
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0
-0@@ -653,8 +657,85 @@ static void test_fixed_counters(u8 pmu_version, u64 perf_capabilities)
[ ... ]
tools/testing/selftests/kvm/x86/pmu_counters_test.c static void test_fixed_counters(u8 pmu_version, u64 perf_capabilities) · line 738
Quoted 3 lines
static void
test_intel_counters(void)
{tools/testing/selftests/kvm/x86/pmu_counters_test.c +1
-0+u64 advertised_perf_caps =kvm_get_feature_msr (MSR_IA32_PERF_CAPABILITIES );
[Severity: Medium]
Will this unconditional MSR read cause an assertion failure on older kernels?
When running the selftest on a host kernel prior to 5.9, KVM does not expose MSR_IA32_PERF_CAPABILITIES as a feature MSR. Calling kvm_get_feature_msr() unconditionally in test_intel_counters() causes the underlying ioctl to fail and return 0, which triggers an assertion failure rather than skipping the feature gracefully.
Other MSR accesses in this file appear to be guarded by checking kvm_has_perf_caps first. Could we apply a similar check before reading advertised_perf_caps so the test remains compatible with older environments?
Quoted 3 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0
-0u8 pmu_version =kvm_cpu_property (X86_PROPERTY_PMU_VERSION );unsigned int i;u8 v;
[ ... ]
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=8
Re: [PATCH v9 08/12] KVM: selftests: Add PERF_METRICS and fixed counter 3 tests
Quoted 32 lines · On 9/18/2026 3:08 PM, [email protected] wrote:
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] selftests: unconditional MSR read causes assertion failure on older kernels --commit
c5ae582e90bae306393c4bf2f37f04a11e37ba5f
Author: Zide Chen <[email protected]>KVM: selftests: Add
PERF_METRICSand fixed counter 3 testsThis commit adds a test case to exercise
IA32_PERF_METRICSfor Topdown Level 1 metrics. It also extends the PMU selftest to cover Intel fixed counter 3 by bumping the maximum number of fixed counters and validating functionality.Quoted 3 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0
-0[ ... ]
Quoted 1 line
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0
-0@@ -653,8 +657,85 @@ static void test_fixed_counters(u8 pmu_version, u64 perf_capabilities)[ ... ]
tools/testing/selftests/kvm/x86/pmu_counters_test.c static void test_fixed_counters(u8 pmu_version, u64 perf_capabilities) · line 738
Quoted 3 lines
static void
test_intel_counters(void)
{tools/testing/selftests/kvm/x86/pmu_counters_test.c +1
-0+u64 advertised_perf_caps =kvm_get_feature_msr (MSR_IA32_PERF_CAPABILITIES );[Severity: Medium]
Will this unconditional MSR read cause an assertion failure on older kernels?When running the selftest on a host kernel prior to 5.9, KVM does not expose
MSR_IA32_PERF_CAPABILITIESas a feature MSR. Callingkvm_get_feature_msr()unconditionally intest_intel_counters()causes the underlying ioctl to fail and return 0, which triggers an assertion failure rather than skipping the feature gracefully.
Quoted 3 lines
Other MSR accesses in this file appear to be guarded by checking
kvm_has_perf_capsfirst. Could we apply a similar check before readingadvertised_perf_capsso the test remains compatible with older environments?
Yes, you are right. It should be guarded with kvm_has_perf_caps.
Quoted 5 lines
Quoted 3 lines
tools/testing/selftests/kvm/x86/pmu_counters_test.c +0
-0u8 pmu_version =kvm_cpu_property (X86_PROPERTY_PMU_VERSION );unsigned int i;u8 v;[ ... ]
PATCH v9 9/12 perf/x86: Add INTEL_TD_METRIC_FIELD_{BITS,MASK} constants
In reply to Zide ChenReplace the hard coded field width (8) and the equivalent 0xff mask used for PERF_METRICS MSR field access with named constants INTEL_TD_METRIC_FIELD_{BITS,MASK}.
INTEL_TD_METRIC_FIELD_BITS will be used by a subsequent KVM patch to validat reserved bits in the PERF_METRICS MSR.
No functional change intended.
---
v9: new patch.
--- arch/x86/events/intel/core.c | 8 +++++--- arch/x86/include/asm/perf_event.h | 4 ++++ 2 files changed, 9 insertions(+), 3 deletions(-)
arch/x86/events/intel/core.c +5 -3
@@ -3159,6 +3159,7 @@ DEFINE_STATIC_CALL(intel_pmu_set_topdown_event_period, x86_perf_event_set_periodstatic inline u64 icl_get_metrics_event_value (u64 metric,u64 slots,int idx) { +int shift = (idx -INTEL_PMC_IDX_METRIC_BASE ) *INTEL_TD_METRIC_FIELD_BITS ;u32 val;/*
@@ -3166,8 +3167,8 @@ static inline u64 icl_get_metrics_event_value(u64 metric, u64 slots, int idx)* summing up to 0xff. * slots-in-metric = (Metric / 0xff) * slots */ - val = (metric >> ((idx -INTEL_PMC_IDX_METRIC_BASE ) *8 )) &0xff ; -return mul_u64_u32_div (slots, val,0xff ); + val = (metric >> shift) &INTEL_TD_METRIC_FIELD_MASK ; +return mul_u64_u32_div (slots, val,INTEL_TD_METRIC_FIELD_MASK ); } static u64 icl_get_topdown_value(struct perf_event *event,
@@ -4742,7 +4743,8 @@ static int core_pmu_hw_config(struct perf_event *event) }#define INTEL_TD_METRIC_AVAILABLE_MAX (INTEL_TD_METRIC_RETIRING + \ - ((x86_pmu.num_topdown_events - 1) << 8)) + ((x86_pmu.num_topdown_events - 1) << \ + INTEL_TD_METRIC_FIELD_BITS))static bool is_available_metric_event (struct perf_event *event) {
arch/x86/include/asm/perf_event.h +4 -0
@@ -418,6 +418,10 @@ static inline bool use_fixed_pseudo_encoding(u64 code)#define INTEL_TD_METRIC_MAX INTEL_TD_METRIC_MEM_BOUND#define INTEL_TD_METRIC_NUM 8 +/* Width, in bits, of each metric's field within the PERF_METRICS MSR. */ +#define INTEL_TD_METRIC_FIELD_BITS 8 +#define INTEL_TD_METRIC_FIELD_MASK GENMASK_ULL(INTEL_TD_METRIC_FIELD_BITS - 1, 0) +#define INTEL_TD_CFG_METRIC_CLEAR_BIT 0#define INTEL_TD_CFG_METRIC_CLEAR BIT_ULL(INTEL_TD_CFG_METRIC_CLEAR_BIT)
Signature · 1 line
2.55.0
[email protected]
AI reviewRe: [PATCH v9 09/12] perf/x86: Add INTEL_TD_METRIC_FIELD_{BITS,MASK} constants
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Low] perf/x86: conflation of UMASK shift and MSR metric field width --
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] perf/x86: conflation of UMASK shift and MSR metric field width
--
commit ca73afb21aaf968ca13fd3d0ba413055c3dc1861
Author: Zide Chen <[email protected]>
perf/x86: Add INTEL_TD_METRIC_FIELD_{BITS,MASK} constants
Replace the hard coded field width (8) and the equivalent 0xff mask used for PERF_METRICS MSR field access with named constants INTEL_TD_METRIC_FIELD_{BITS,MASK}.
Quoted 4 lines
arch/x86/events/intel/core.c +0
-0
[ ... ]
arch/x86/events/intel/core.c static int core_pmu_hw_config(struct perf_event *event) · line 4747
Quoted 7 lines
arch/x86/events/intel/core.c +2
-1@@ -4742,7 +4743,8 @@ static int core_pmu_hw_config(struct perf_event *event) }#define INTEL_TD_METRIC_AVAILABLE_MAX (INTEL_TD_METRIC_RETIRING + \ - ((x86_pmu.num_topdown_events - 1) << 8)) + ((x86_pmu.num_topdown_events - 1) << \ + INTEL_TD_METRIC_FIELD_BITS))
[Severity: Low]
Is it safe to replace the UMASK shift here with the hardware MSR metric field width?
The original << 8 was used to shift the metric ID into the UMASK field (bits 8-15) of the software perf event configuration (attr.config). INTEL_TD_METRIC_FIELD_BITS represents the hardware width of metrics within the PERF_METRICS MSR.
When evaluated in is_available_metric_event() during perf_event_open(), this currently functions because both values are 8. However, conflating a software API layout offset with a hardware register field width could cause this macro to calculate an invalid configuration mask if the MSR field width changes in future hardware.
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=9
Re: [PATCH v9 09/12] perf/x86: Add INTEL_TD_METRIC_FIELD_{BITS,MASK} constants
Quoted 23 lines · On 9/18/2026 2:59 PM, [email protected] wrote:
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] perf/x86: conflation of UMASK shift and MSR metric field width
--commit
ca73afb21aaf968ca13fd3d0ba413055c3dc1861
Author: Zide Chen <[email protected]>perf/x86: Add INTEL_TD_METRIC_FIELD_{BITS,MASK} constants
Replace the hard coded field width (8) and the equivalent
0xffmask used forPERF_METRICSMSR field access with named constants INTEL_TD_METRIC_FIELD_{BITS,MASK}.Quoted 4 lines
arch/x86/events/intel/core.c +0
-0[ ... ]
Quoted 3 lines
arch/x86/events/intel/core.c +0
-0@@ -4742,7 +4743,8 @@ static int core_pmu_hw_config(struct perf_event *event) }
Quoted 8 lines
arch/x86/events/intel/core.c static int core_pmu_hw_config(struct perf_event *event) · line 4747
Quoted 4 lines
arch/x86/events/intel/core.c +2
-1#define INTEL_TD_METRIC_AVAILABLE_MAX (INTEL_TD_METRIC_RETIRING + \ - ((x86_pmu.num_topdown_events - 1) << 8)) + ((x86_pmu.num_topdown_events - 1) << \ + INTEL_TD_METRIC_FIELD_BITS))[Severity: Low]
Is it safe to replace the UMASK shift here with the hardware MSR metric field width?
It's safe since x86_pmu.num_topdown_events is guaranteed to be <= 8, and it's called when is_metric_event(event) is true.
Quoted 11 lines
The original << 8 was used to shift the metric ID into the UMASK field (bits 8-15) of the software perf event configuration (attr.config).
INTEL_TD_METRIC_FIELD_BITSrepresents the hardware width of metrics within thePERF_METRICSMSR.When evaluated in
is_available_metric_event()duringperf_event_open(), this currently functions because both values are 8. However, conflating a software API layout offset with a hardware register field width could cause this macro to calculate an invalid configuration mask if the MSR field width changes in future hardware.
Re: [PATCH v9 09/12] perf/x86: Add INTEL_TD_METRIC_FIELD_{BITS,MASK} constants
In reply to Zide ChenQuoted 42 lines · Zide Chen
Replace the hard coded field width (8) and the equivalent
0xffmask used forPERF_METRICSMSR field access with named constants INTEL_TD_METRIC_FIELD_{BITS,MASK}.
INTEL_TD_METRIC_FIELD_BITSwill be used by a subsequent KVM patch to validat reserved bits in thePERF_METRICSMSR.No functional change intended.
Signed-off-by Zide Chen <[email protected]>Cc Peter Zijlstra <[email protected]>---
v9: new patch.--- arch/x86/events/intel/core.c | 8 +++++--- arch/x86/include/asm/perf_event.h | 4 ++++ 2 files changed, 9 insertions(+), 3 deletions(-)arch/x86/events/intel/core.c +3
-2@@ -3159,6 +3159,7 @@ DEFINE_STATIC_CALL(intel_pmu_set_topdown_event_period, x86_perf_event_set_periodstatic inline u64 icl_get_metrics_event_value (u64 metric,u64 slots,int idx) { +int shift = (idx -INTEL_PMC_IDX_METRIC_BASE ) *INTEL_TD_METRIC_FIELD_BITS ;u32 val;/* @@ -3166,8 +3167,8 @@ static inline u64 icl_get_metrics_event_value(u64 metric, u64 slots, int idx)* summing up to 0xff. * slots-in-metric = (Metric / 0xff) * slots */ - val = (metric >> ((idx -INTEL_PMC_IDX_METRIC_BASE ) *8 )) &0xff ; -return mul_u64_u32_div (slots, val,0xff ); + val = (metric >> shift) &INTEL_TD_METRIC_FIELD_MASK ; +return mul_u64_u32_div (slots, val,INTEL_TD_METRIC_FIELD_MASK ); } static u64 icl_get_topdown_value(struct perf_event *event,
arch/x86/events/intel/core.c static int core_pmu_hw_config(struct perf_event *event) · line 4747
Quoted 6 lines
arch/x86/events/intel/core.c +2
-1@@ -4742,7 +4743,8 @@ static int core_pmu_hw_config(struct perf_event *event) }#define INTEL_TD_METRIC_AVAILABLE_MAX (INTEL_TD_METRIC_RETIRING + \ - ((x86_pmu.num_topdown_events - 1) << 8)) + ((x86_pmu.num_topdown_events - 1) << \ + INTEL_TD_METRIC_FIELD_BITS))
Sashiko comments,
"
Is it safe to replace the UMASK shift here with the hardware MSR metric field width?
The original << 8 was used to shift the metric ID into the UMASK field (bits 8-15) of the software perf event configuration (attr.config). INTEL_TD_METRIC_FIELD_BITS represents the hardware width of metrics within the PERF_METRICS MSR.
When evaluated in is_available_metric_event() during perf_event_open(), this currently functions because both values are 8. However, conflating a software API layout offset with a hardware register field width could cause this macro to calculate an invalid configuration mask if the MSR field width changes in future hardware.
"
It looks reasonable, we either don't change this, or define a macro to represent the UMASK shift. Thanks.
Quoted 18 lines
arch/x86/events/intel/core.c +0
-0static bool is_available_metric_event (struct perf_event *event) {arch/x86/include/asm/perf_event.h +4
-0@@ -418,6 +418,10 @@ static inline bool use_fixed_pseudo_encoding(u64 code)#define INTEL_TD_METRIC_MAX INTEL_TD_METRIC_MEM_BOUND#define INTEL_TD_METRIC_NUM 8 +/* Width, in bits, of each metric's field within the PERF_METRICS MSR. */ +#define INTEL_TD_METRIC_FIELD_BITS 8 +#define INTEL_TD_METRIC_FIELD_MASK GENMASK_ULL(INTEL_TD_METRIC_FIELD_BITS - 1, 0) +#define INTEL_TD_CFG_METRIC_CLEAR_BIT 0#define INTEL_TD_CFG_METRIC_CLEAR BIT_ULL(INTEL_TD_CFG_METRIC_CLEAR_BIT)
Re: [PATCH v9 09/12] perf/x86: Add INTEL_TD_METRIC_FIELD_{BITS,MASK} constants
Quoted 67 lines · Dapeng
Quoted 42 lines · Zide Chen
Replace the hard coded field width (8) and the equivalent
0xffmask used forPERF_METRICSMSR field access with named constants INTEL_TD_METRIC_FIELD_{BITS,MASK}.
INTEL_TD_METRIC_FIELD_BITSwill be used by a subsequent KVM patch to validat reserved bits in thePERF_METRICSMSR.No functional change intended.
Signed-off-by Zide Chen <[email protected]>Cc Peter Zijlstra <[email protected]>---
v9: new patch.--- arch/x86/events/intel/core.c | 8 +++++--- arch/x86/include/asm/perf_event.h | 4 ++++ 2 files changed, 9 insertions(+), 3 deletions(-)arch/x86/events/intel/core.c +3
-2@@ -3159,6 +3159,7 @@ DEFINE_STATIC_CALL(intel_pmu_set_topdown_event_period, x86_perf_event_set_periodstatic inline u64 icl_get_metrics_event_value (u64 metric,u64 slots,int idx) { +int shift = (idx -INTEL_PMC_IDX_METRIC_BASE ) *INTEL_TD_METRIC_FIELD_BITS ;u32 val;/* @@ -3166,8 +3167,8 @@ static inline u64 icl_get_metrics_event_value(u64 metric, u64 slots, int idx)* summing up to 0xff. * slots-in-metric = (Metric / 0xff) * slots */ - val = (metric >> ((idx -INTEL_PMC_IDX_METRIC_BASE ) *8 )) &0xff ; -return mul_u64_u32_div (slots, val,0xff ); + val = (metric >> shift) &INTEL_TD_METRIC_FIELD_MASK ; +return mul_u64_u32_div (slots, val,INTEL_TD_METRIC_FIELD_MASK ); } static u64 icl_get_topdown_value(struct perf_event *event,arch/x86/events/intel/core.c static int core_pmu_hw_config(struct perf_event *event) · line 4747
Quoted 6 lines
arch/x86/events/intel/core.c +2
-1@@ -4742,7 +4743,8 @@ static int core_pmu_hw_config(struct perf_event *event) }#define INTEL_TD_METRIC_AVAILABLE_MAX (INTEL_TD_METRIC_RETIRING + \ - ((x86_pmu.num_topdown_events - 1) << 8)) + ((x86_pmu.num_topdown_events - 1) << \ + INTEL_TD_METRIC_FIELD_BITS))Sashiko comments,
"
Is it safe to replace the UMASK shift here with the hardware MSR metric field width?
The original << 8 was used to shift the metric ID into the UMASK field (bits 8-15) of the software perf event configuration (attr.config).INTEL_TD_METRIC_FIELD_BITSrepresents the hardware width of metrics within thePERF_METRICSMSR.
When evaluated inis_available_metric_event()duringperf_event_open(), this currently functions because both values are 8. However, conflating a software API layout offset with a hardware register field width could cause this macro to calculate an invalid configuration mask if the MSR field width changes in future hardware.
Quoted 4 lines
"
It looks reasonable, we either don't change this, or define a macro to represent the UMASK shift. Thanks.
Yes, I can revert this change.
PATCH v9 10/12 perf/x86: Expose number of Topdown metric events to KVM
In reply to Zide ChenThe number of Topdown metric events varies by CPU: ICL/TGL/RKL support 4, while SPR/GNR/DMR and some hybrid CPUs support 8.
Add num_topdown_events to struct x86_pmu_capability and populate it in perf_get_x86_pmu_capability() so that KVM can determine how many PERF_METRICS fields are valid on the host CPU and correctly identify the reserved bits in IA32_PERF_METRICS.
---
v9: new patch
--- arch/x86/events/core.c | 1 + arch/x86/include/asm/perf_event.h | 1 + 2 files changed, 2 insertions(+)
arch/x86/events/core.c +1 -0
@@ -3164,6 +3164,7 @@ void perf_get_x86_pmu_capability(struct x86_pmu_capability *cap) cap->events_mask_len = x86_pmu.events_mask_len; cap->pebs_ept = x86_pmu.pebs_ept; cap->mediated = !!(pmu.capabilities &PERF_PMU_CAP_MEDIATED_VPMU ); + cap->num_topdown_events = x86_pmu.num_topdown_events; }EXPORT_SYMBOL_FOR_KVM (perf_get_x86_pmu_capability);
arch/x86/include/asm/perf_event.h +1 -0
@@ -306,6 +306,7 @@ struct x86_pmu_capability {int bit_width_fixed;unsigned int events_mask;int events_mask_len; +int num_topdown_events; unsigned int pebs_ept :1; unsigned int mediated :1; };
Signature · 1 line
2.55.0
Re: [PATCH v9 10/12] perf/x86: Expose number of Topdown metric events to KVM
Quoted 41 lines · Zide Chen
The number of Topdown metric events varies by CPU: ICL/TGL/RKL support 4, while SPR/GNR/DMR and some hybrid CPUs support 8.
Add
num_topdown_eventsto structx86_pmu_capabilityand populate it inperf_get_x86_pmu_capability()so that KVM can determine how manyPERF_METRICSfields are valid on the host CPU and correctly identify the reserved bits inIA32_PERF_METRICS.Signed-off-by Zide Chen <[email protected]>Cc Peter Zijlstra <[email protected]>---
v9: new patch--- arch/x86/events/core.c | 1 + arch/x86/include/asm/perf_event.h | 1 + 2 files changed, 2 insertions(+)arch/x86/events/core.c +1
-0@@ -3164,6 +3164,7 @@ void perf_get_x86_pmu_capability(struct x86_pmu_capability *cap) cap->events_mask_len = x86_pmu.events_mask_len; cap->pebs_ept = x86_pmu.pebs_ept; cap->mediated = !!(pmu.capabilities &PERF_PMU_CAP_MEDIATED_VPMU ); + cap->num_topdown_events = x86_pmu.num_topdown_events; }EXPORT_SYMBOL_FOR_KVM (perf_get_x86_pmu_capability);arch/x86/include/asm/perf_event.h +1
-0@@ -306,6 +306,7 @@ struct x86_pmu_capability {int bit_width_fixed;unsigned int events_mask;int events_mask_len; +int num_topdown_events; unsigned int pebs_ept :1; unsigned int mediated :1; };
PATCH v9 11/12 KVM: x86/pmu: Reject writes to reserved MSR_PERF_METRICS bits
In reply to Zide ChenWhen the host CPU supports only 4 Topdown metrics, bits [63:32] of MSR_PERF_METRICS are reserved.
Derive pmu->perf_metrics_rsvd from the number of supported Topdown metrics and use it to validate writes. Reject any attempts to set reserved bits in MSR_PERF_METRICS.
---
v9: new patch.
--- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/pmu.c | 1 + arch/x86/kvm/vmx/pmu_intel.c | 18 +++++++++++++++++- 3 files changed, 19 insertions(+), 1 deletion(-)
arch/x86/include/asm/kvm_host.h +1 -0
@@ -585,6 +585,7 @@ struct kvm_pmu {u64 reserved_bits;u64 raw_event_mask;u64 perf_metrics; +u64 perf_metrics_rsvd;struct kvm_pmc gp_counters[KVM_MAX_NR_GP_COUNTERS ];struct kvm_pmc fixed_counters[KVM_MAX_NR_FIXED_COUNTERS ];
arch/x86/kvm/pmu.c +1 -0
@@ -1006,6 +1006,7 @@ void kvm_pmu_refresh(struct kvm_vcpu *vcpu) pmu->fixed_ctr_ctrl_rsvd = ~0ull ; pmu->pebs_enable_rsvd = ~0ull ; pmu->pebs_data_cfg_rsvd = ~0ull ; + pmu->perf_metrics_rsvd = ~0ull ;bitmap_zero (pmu->all_valid_pmc_idx,X86_PMC_IDX_MAX );if (!vcpu->kvm->arch.enable_pmu)
arch/x86/kvm/vmx/pmu_intel.c +17 -1
@@ -417,6 +417,9 @@ 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 : +if (data & pmu->perf_metrics_rsvd) +return 1 ; + pmu->perf_metrics = data;break ;case MSR_IA32_PEBS_ENABLE :
@@ -601,9 +604,22 @@ 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 ) +if (perf_capabilities &PERF_CAP_PERF_METRICS ) { pmu->global_ctrl_rsvd &= ~GLOBAL_CTRL_EN_PERF_METRICS ; +/* +* PERF_METRICS has one 8-bit field per metric. At 8 metrics, +* skip computing the mask and set perf_metrics_rsvd to 0 +* directly, since BIT_ULL(64) is undefined. +*/ +if (kvm_pmu_cap.num_topdown_events <INTEL_TD_METRIC_NUM ) + pmu->perf_metrics_rsvd = + ~(BIT_ULL (kvm_pmu_cap.num_topdown_events * +INTEL_TD_METRIC_FIELD_BITS ) -1 ); +else + pmu->perf_metrics_rsvd =0 ; + } +/* * GLOBAL_STATUS and GLOBAL_OVF_CONTROL (a.k.a. GLOBAL_STATUS_RESET) * share reserved bit definitions. The kernel just happens to use
Signature · 1 line
2.55.0
Re: [PATCH v9 11/12] KVM: x86/pmu: Reject writes to reserved MSR_PERF_METRICS bits
LGTM. Thanks.
Quoted 78 lines · Zide Chen
When the host CPU supports only 4 Topdown metrics, bits [63:32] of
MSR_PERF_METRICSare reserved.Derive
pmu->perf_metrics_rsvdfrom the number of supported Topdown metrics and use it to validate writes. Reject any attempts to set reserved bits inMSR_PERF_METRICS.Signed-off-by Zide Chen <[email protected]>---
v9: new patch.--- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/pmu.c | 1 + arch/x86/kvm/vmx/pmu_intel.c | 18 +++++++++++++++++- 3 files changed, 19 insertions(+), 1 deletion(-)arch/x86/include/asm/kvm_host.h +1
-0@@ -585,6 +585,7 @@ struct kvm_pmu {u64 reserved_bits;u64 raw_event_mask;u64 perf_metrics; +u64 perf_metrics_rsvd;struct kvm_pmc gp_counters[KVM_MAX_NR_GP_COUNTERS ];struct kvm_pmc fixed_counters[KVM_MAX_NR_FIXED_COUNTERS ];arch/x86/kvm/pmu.c +1
-0@@ -1006,6 +1006,7 @@ void kvm_pmu_refresh(struct kvm_vcpu *vcpu) pmu->fixed_ctr_ctrl_rsvd = ~0ull ; pmu->pebs_enable_rsvd = ~0ull ; pmu->pebs_data_cfg_rsvd = ~0ull ; + pmu->perf_metrics_rsvd = ~0ull ;bitmap_zero (pmu->all_valid_pmc_idx,X86_PMC_IDX_MAX );if (!vcpu->kvm->arch.enable_pmu)arch/x86/kvm/vmx/pmu_intel.c +17
-1@@ -417,6 +417,9 @@ 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 : +if (data & pmu->perf_metrics_rsvd) +return 1 ; + pmu->perf_metrics = data;break ;case MSR_IA32_PEBS_ENABLE :@@ -601,9 +604,22 @@ 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 ) +if (perf_capabilities &PERF_CAP_PERF_METRICS ) { pmu->global_ctrl_rsvd &= ~GLOBAL_CTRL_EN_PERF_METRICS ; +/* +* PERF_METRICS has one 8-bit field per metric. At 8 metrics, +* skip computing the mask and set perf_metrics_rsvd to 0 +* directly, since BIT_ULL(64) is undefined. +*/ +if (kvm_pmu_cap.num_topdown_events <INTEL_TD_METRIC_NUM ) + pmu->perf_metrics_rsvd = + ~(BIT_ULL (kvm_pmu_cap.num_topdown_events * +INTEL_TD_METRIC_FIELD_BITS ) -1 ); +else + pmu->perf_metrics_rsvd =0 ; + } +/* * GLOBAL_STATUS and GLOBAL_OVF_CONTROL (a.k.a. GLOBAL_STATUS_RESET) * share reserved bit definitions. The kernel just happens to use
Re: [PATCH v9 11/12] KVM: x86/pmu: Reject writes to reserved MSR_PERF_METRICS bits
In reply to Zide Chenarch/x86/kvm/vmx/pmu_intel.c static void intel_pmu_refresh(struct kvm_vcpu *vcpu) · line 620
Quoted 7 lines · Zide Chen
arch/x86/kvm/vmx/pmu_intel.c +6
-0+if (kvm_pmu_cap.num_topdown_events <INTEL_TD_METRIC_NUM ) + pmu->perf_metrics_rsvd = + ~(BIT_ULL (kvm_pmu_cap.num_topdown_events * +INTEL_TD_METRIC_FIELD_BITS ) -1 ); +else + pmu->perf_metrics_rsvd =0 ;
perf only sets x86_pmu.num_topdown_events for the models it knows about (Ice Lake and intel_pmu_init_glc() users). On a non-hybrid part that perf treats as "generic_arch_v5+", PERF_CAPABILITIES.PERF_METRICS_AVAILABLE can be set while num_topdown_events is 0. (intel_pmu_init() even sets GLOBAL_CTRL_EN_PERF_METRICS in intel_ctrl in that case.)
vmx_get_perf_capabilities() will still advertise PERF_CAP_PERF_METRICS, but perf_metrics_rsvd will be ~0ull. The guest writes PERF_METRICS directly, and the hardware updates it on its own, so KVM_GET_MSRS can return a non-zero value that KVM_SET_MSRS will then reject. That breaks save/restore. (The selftest's write of 0xdeaddead would also fail.)
Shouldn't vmx_get_perf_capabilities() withhold PERF_CAP_PERF_METRICS (and PERF_CAP_RDPMC_METRICS_CLEAR) when kvm_pmu_cap.num_topdown_events is 0? The SDM, volume 3, section 22.8, says:
Quoted 3 lines
IA32_PERF_CAPABILITIES.PERF_METRICS_AVAILABLE[15]: If set, indicates that the architecture provides built in support for [four] TMA L1 metrics through the
PERF_METRICSMSR.
So, four could be a fallback, but if the part actually implements eight, the upper fields would break the round trip in the same way.
Also, since the guest's own writes go straight to hardware, this check really exists to keep the wrmsrq() in intel_mediated_pmu_load() from faulting on a bogus userspace value. The commit message should say so.
Re: [PATCH v9 11/12] KVM: x86/pmu: Reject writes to reserved MSR_PERF_METRICS bits
Quoted 21 lines · Jim Mattson
arch/x86/kvm/vmx/pmu_intel.c static void intel_pmu_refresh(struct kvm_vcpu *vcpu) · line 620
Quoted 7 lines · Zide Chen
arch/x86/kvm/vmx/pmu_intel.c +6
-0+if (kvm_pmu_cap.num_topdown_events <INTEL_TD_METRIC_NUM ) + pmu->perf_metrics_rsvd = + ~(BIT_ULL (kvm_pmu_cap.num_topdown_events * +INTEL_TD_METRIC_FIELD_BITS ) -1 ); +else + pmu->perf_metrics_rsvd =0 ;perf only sets x86_pmu.num_topdown_events for the models it knows about (Ice Lake and
intel_pmu_init_glc()users). On a non-hybrid part that perf treats as "generic_arch_v5+", PERF_CAPABILITIES.PERF_METRICS_AVAILABLE can be set whilenum_topdown_eventsis 0. (intel_pmu_init()even setsGLOBAL_CTRL_EN_PERF_METRICSinintel_ctrlin that case.)
vmx_get_perf_capabilities()will still advertisePERF_CAP_PERF_METRICS, butperf_metrics_rsvdwill be ~0ull. The guest writesPERF_METRICSdirectly, and the hardware updates it on its own, soKVM_GET_MSRScan return a non-zero value thatKVM_SET_MSRSwill then reject. That breaks save/restore. (The selftest's write of0xdeaddeadwould also fail.)
Quoted 3 lines
Shouldn't
vmx_get_perf_capabilities()withholdPERF_CAP_PERF_METRICS(andPERF_CAP_RDPMC_METRICS_CLEAR) when kvm_pmu_cap.num_topdown_events is 0?
Yes, this is safer. Will updated vmx_get_perf_capabilities() to gate advertising both PERF_CAP_PERF_METRICS and PERF_CAP_RDPMC_METRICS_CLEAR on num_topdown_events.
Quoted 11 lines
The SDM, volume 3, section 22.8, says:
Quoted 3 lines
IA32_PERF_CAPABILITIES.PERF_METRICS_AVAILABLE[15]: If set, indicates that the architecture provides built in support for [four] TMA L1 metrics through the
PERF_METRICSMSR.So, four could be a fallback, but if the part actually implements eight, the upper fields would break the round trip in the same way.
Also, since the guest's own writes go straight to hardware, this check really exists to keep the
wrmsrq()inintel_mediated_pmu_load()from faulting on a bogus userspace value. The commit message should say so.
Agreed. I will update the commit message to clarify that because guest writes are passed through to hardware, the primary purpose of this check in intel_pmu_set_msr() is to prevent userspace (e.g. via KVM_SET_MSRS) from stuffing invalid metric values that would cause the unchecked wrmsrq() in intel_mediated_pmu_load() to #GP on VM-entry.
PATCH v9 12/12 KVM: x86/pmu: Support RDPMC Metrics Clear Mode
In reply to Zide ChenAdd PERF_CAP_RDPMC_METRICS_CLEAR (IA32_PERF_CAPABILITIES bit 19) support. When set, and IA32_FIXED_CTR_CTRL.METRICS_CLEAR_EN (bit 14) is also set, RDPMC of PERF_METRICS clears both PERF_METRICS and fixed counter 3 (SLOTS) after the read.
Advertise the capability to guests only when the mediated PMU is enabled and the host reports it. Unmask FIXED_CTR_CTRL bit 14 in intel_pmu_refresh(), scoped to fixed counter 3 only via intel_fixed_bits_by_idx().
For the mediated PMU, METRICS_CLEAR_EN takes effect natively via passthrough. But kvm_need_rdpmc_intercept() can still force interception for unrelated reasons, so intel_emulate_rdpmc() also replicates the clear-on-read behavior for consistency.
---
v9: new patch.
--- arch/x86/include/asm/msr-index.h | 1 + arch/x86/kvm/vmx/pmu_intel.c | 28 ++++++++++++++++++++++++++++ arch/x86/kvm/vmx/vmx.c | 6 +++++- 3 files changed, 34 insertions(+), 1 deletion(-)
arch/x86/include/asm/msr-index.h +1 -0
@@ -333,6 +333,7 @@#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_RDPMC_METRICS_CLEAR BIT_ULL(19)#define PERF_CAP_PEBS_MASK (PERF_CAP_PEBS_TRAP | PERF_CAP_ARCH_REG | \ PERF_CAP_PEBS_FORMAT | PERF_CAP_PEBS_BASELINE | \ PERF_CAP_PEBS_TIMING_INFO)
arch/x86/kvm/vmx/pmu_intel.c +28 -0
@@ -36,6 +36,9 @@#define INTEL_RDPMC_TYPE_MASK GENMASK(31, 16)#define INTEL_RDPMC_INDEX_MASK GENMASK(15, 0) +/* Index of the SLOTS fixed counter relative to the first fixed counter. */ +#define INTEL_FIXED_SLOTS_IDX (INTEL_PMC_IDX_FIXED_SLOTS - INTEL_PMC_IDX_FIXED) +#define MSR_PMC_FULL_WIDTH_BIT (MSR_IA32_PMC0 - MSR_IA32_PERFCTR0)static struct lbr_desc *vcpu_to_lbr_desc (struct kvm_vcpu *vcpu)
@@ -86,6 +89,14 @@ static void reprogram_fixed_counters(struct kvm_pmu *pmu, u64 data) } } +static bool intel_pmu_metrics_clear_enabled (struct kvm_pmu *pmu) +{ +u8 fixed_ctr_ctrl =fixed_ctrl_field (pmu->fixed_ctr_ctrl, +INTEL_FIXED_SLOTS_IDX ); + +return fixed_ctr_ctrl &INTEL_FIXED_3_METRICS_CLEAR ; +} +static int intel_emulate_rdpmc (struct kvm_vcpu *vcpu,unsigned int idx,u64 *data) {
@@ -138,6 +149,18 @@ static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,return 1 ; *data = pmu->perf_metrics; + +/* +* Per the SDM, RDPMC of PERF_METRICS clears both PERF_METRICS +* and the SLOTS fixed counter when Metrics Clear Mode is +* enabled via INTEL_FIXED_3_METRICS_CLEAR in FIXED_CTR_CTRL. +*/ + pmc =kvm_pmc_idx_to_pmc (pmu,INTEL_PMC_IDX_FIXED_SLOTS ); +if (pmc &&intel_pmu_metrics_clear_enabled (pmu)) { + pmu->perf_metrics =0 ; +pmc_write_counter (pmc,0 ); + } +return 0 ;default :return 1 ;
@@ -601,6 +624,11 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)INTEL_FIXED_0_USER | INTEL_FIXED_0_ENABLE_PMI); +if (perf_capabilities &PERF_CAP_RDPMC_METRICS_CLEAR ) + pmu->fixed_ctr_ctrl_rsvd &= + ~intel_fixed_bits_by_idx (INTEL_FIXED_SLOTS_IDX , +INTEL_FIXED_3_METRICS_CLEAR ); + 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;
arch/x86/kvm/vmx/vmx.c +5 -1
@@ -2616,6 +2616,9 @@ int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)if (!cpuid_model_is_consistent (vcpu))return 1 ; } +if ((data &PERF_CAP_RDPMC_METRICS_CLEAR ) && + !(data &PERF_CAP_PERF_METRICS )) +return 1 ; ret =kvm_set_msr_common (vcpu, msr_info);break ;
@@ -7999,7 +8002,8 @@ static __init u64 vmx_get_perf_capabilities(void) }if (enable_mediated_pmu) - perf_cap |= kvm_host.perf_capabilities &PERF_CAP_PERF_METRICS ; + perf_cap |= kvm_host.perf_capabilities & + (PERF_CAP_PERF_METRICS |PERF_CAP_RDPMC_METRICS_CLEAR );return perf_cap; }
Signature · 1 line
2.55.0
[email protected]
AI reviewRe: [PATCH v9 12/12] KVM: x86/pmu: Support RDPMC Metrics Clear Mode
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider: - [High] pmu: event filter mistakenly strips METRICS_CLEAR_EN from fixed_ctr_ctrl_hw - [Medium] pmu…
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] pmu: event filter mistakenly strips METRICS_CLEAR_EN from fixed_ctr_ctrl_hw
- [Medium] pmu: METRICS_CLEAR_EN unmasked without validating fixed counter bounds
- [Low] pmu: emulation incorrectly ties PERF_METRICS clearing to SLOTS PMC existence --
commit 8ba3725c32dbde4a1c3935cb433e898e06425b83
Author: Zide Chen <[email protected]>
KVM: x86/pmu: Support RDPMC Metrics Clear Mode
This commit adds support for PERF_CAP_RDPMC_METRICS_CLEAR (bit 19 of IA32_PERF_CAPABILITIES). It advertises this capability to guests when the mediated PMU is enabled and the host reports it, unmasking the relevant control bit. It also updates intel_emulate_rdpmc() to replicate the clear-on-read behavior for consistency.
Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0
[ ... ]
Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2
-0@@ -138,6 +149,18 @@ static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,return 1 ; *data = pmu->perf_metrics; + +/*
arch/x86/kvm/vmx/pmu_intel.c static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, · line 163
Quoted 10 lines
arch/x86/kvm/vmx/pmu_intel.c +10
-0+* Per the SDM, RDPMC of PERF_METRICS clears both PERF_METRICS +* and the SLOTS fixed counter when Metrics Clear Mode is +* enabled via INTEL_FIXED_3_METRICS_CLEAR in FIXED_CTR_CTRL. +*/ + pmc =kvm_pmc_idx_to_pmc (pmu,INTEL_PMC_IDX_FIXED_SLOTS ); +if (pmc &&intel_pmu_metrics_clear_enabled (pmu)) { + pmu->perf_metrics =0 ; +pmc_write_counter (pmc,0 ); + } +
[Severity: Low]
Does tying the clearing of PERF_METRICS to the existence of the SLOTS PMC in intel_emulate_rdpmc() violate the hardware specification?
If the guest executes RDPMC natively and triggers an interception, and the SLOTS counter is not exposed to the guest (making kvm_pmc_idx_to_pmc return NULL), but METRICS_CLEAR_EN is enabled, this code silently skips clearing PERF_METRICS.
Should the clearing of pmu->perf_metrics be performed independently of whether the software PMC object for the SLOTS counter exists?
Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0return 0 ;default :return 1 ;
[ ... ]
arch/x86/kvm/vmx/pmu_intel.c static void intel_pmu_refresh(struct kvm_vcpu *vcpu) · line 631
Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +5
-0@@ -601,6 +624,11 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)INTEL_FIXED_0_USER | INTEL_FIXED_0_ENABLE_PMI); +if (perf_capabilities &PERF_CAP_RDPMC_METRICS_CLEAR ) + pmu->fixed_ctr_ctrl_rsvd &= + ~intel_fixed_bits_by_idx (INTEL_FIXED_SLOTS_IDX , +INTEL_FIXED_3_METRICS_CLEAR ); +
[Severity: Medium]
Does intel_pmu_refresh() unconditionally unmask the METRICS_CLEAR_EN bit without verifying if the guest topology actually exposes the SLOTS counter?
If a VM is configured with 3 or fewer fixed counters, this unmasking would allow the unprivileged guest OS to write to bit 14 in MSR_CORE_PERF_FIXED_CTR_CTRL without triggering the expected #GP.
Should this be gated by checking if pmu->nr_arch_fixed_counters > INTEL_FIXED_SLOTS_IDX?
[Severity: High]
Does this patch inadvertently cause event filters to strip the METRICS_CLEAR_EN bit in kvm_mediated_pmu_refresh_event_filter()?
When a host-level event filter denies the SLOTS counter (counter 3) from counting, KVM applies a 0xf hardware mask that zeroes all 4 configuration bits for the counter:
arch/x86/kvm/pmu.c:kvm_mediated_pmu_refresh_event_filter() {
...
} else {
u64 mask = intel_fixed_bits_by_idx(pmc->idx - KVM_FIXED_PMC_BASE_IDX, 0xf);
pmu->fixed_ctr_ctrl_hw &= ~mask;
if (allowed)
pmu->fixed_ctr_ctrl_hw |= pmu->fixed_ctr_ctrl & mask;
}
}Since this mask covers the entire 4-bit configuration block, it will also clear bit 14 (METRICS_CLEAR_EN) from hardware. If the guest sets METRICS_CLEAR_EN and then executes RDPMC natively, it will lead to un-cleared stale metrics.
Should the mask be updated to avoid interfering with this newly introduced feature bit?
Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0counter_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;
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=12
Re: [PATCH v9 12/12] KVM: x86/pmu: Support RDPMC Metrics Clear Mode
Quoted 43 lines · On 9/18/2026 3:22 PM, [email protected] wrote:
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] pmu: event filter mistakenly stripsMETRICS_CLEAR_ENfromfixed_ctr_ctrl_hw
- [Medium] pmu:METRICS_CLEAR_ENunmasked without validating fixed counter bounds
- [Low] pmu: emulation incorrectly tiesPERF_METRICSclearing to SLOTS PMC existence --commit
8ba3725c32dbde4a1c3935cb433e898e06425b83
Author: Zide Chen <[email protected]>KVM: x86/pmu: Support RDPMC Metrics Clear Mode
This commit adds support for
PERF_CAP_RDPMC_METRICS_CLEAR(bit 19 ofIA32_PERF_CAPABILITIES). It advertises this capability to guests when the mediated PMU is enabled and the host reports it, unmasking the relevant control bit. It also updatesintel_emulate_rdpmc()to replicate the clear-on-read behavior for consistency.Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0[ ... ]
Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2
-0@@ -138,6 +149,18 @@ static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,return 1 ; *data = pmu->perf_metrics; + +/* arch/x86/kvm/vmx/pmu_intel.c static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, · line 163
Quoted 10 lines
arch/x86/kvm/vmx/pmu_intel.c +10
-0+* Per the SDM, RDPMC of PERF_METRICS clears both PERF_METRICS +* and the SLOTS fixed counter when Metrics Clear Mode is +* enabled via INTEL_FIXED_3_METRICS_CLEAR in FIXED_CTR_CTRL. +*/ + pmc =kvm_pmc_idx_to_pmc (pmu,INTEL_PMC_IDX_FIXED_SLOTS ); +if (pmc &&intel_pmu_metrics_clear_enabled (pmu)) { + pmu->perf_metrics =0 ; +pmc_write_counter (pmc,0 ); + } +[Severity: Low]
Does tying the clearing ofPERF_METRICSto the existence of the SLOTS PMC inintel_emulate_rdpmc()violate the hardware specification?
Quoted 7 lines
If the guest executes RDPMC natively and triggers an interception, and the SLOTS counter is not exposed to the guest (making
kvm_pmc_idx_to_pmcreturn NULL), butMETRICS_CLEAR_ENis enabled, this code silently skips clearingPERF_METRICS.Should the clearing of
pmu->perf_metricsbe performed independently of whether the software PMC object for the SLOTS counter exists?
This is a defensive check against invalid guest CPUID configuration. If the SLOTS counter is not available, clearing PERF_METRICS is meaningless and messy.
Quoted 18 lines
Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0return 0 ;default :return 1 ;[ ... ]
arch/x86/kvm/vmx/pmu_intel.c static void intel_pmu_refresh(struct kvm_vcpu *vcpu) · line 631
Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +5
-0@@ -601,6 +624,11 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)INTEL_FIXED_0_USER | INTEL_FIXED_0_ENABLE_PMI); +if (perf_capabilities &PERF_CAP_RDPMC_METRICS_CLEAR ) + pmu->fixed_ctr_ctrl_rsvd &= + ~intel_fixed_bits_by_idx (INTEL_FIXED_SLOTS_IDX , +INTEL_FIXED_3_METRICS_CLEAR ); +[Severity: Medium]
Doesintel_pmu_refresh()unconditionally unmask theMETRICS_CLEAR_ENbit without verifying if the guest topology actually exposes the SLOTS counter?
Quoted 6 lines
If a VM is configured with 3 or fewer fixed counters, this unmasking would allow the unprivileged guest OS to write to bit 14 in
MSR_CORE_PERF_FIXED_CTR_CTRLwithout triggering the expected #GP.Should this be gated by checking if
pmu->nr_arch_fixed_counters>INTEL_FIXED_SLOTS_IDX?
KVM is not intended to check against all possible invalid configurations, as long as it protects host integrity, I think this is fine.
Quoted 19 lines
[Severity: High]
Does this patch inadvertently cause event filters to strip theMETRICS_CLEAR_ENbit inkvm_mediated_pmu_refresh_event_filter()?When a host-level event filter denies the SLOTS counter (counter 3) from counting, KVM applies a
0xfhardware mask that zeroes all 4 configuration bits for the counter:arch/x86/kvm/pmu.c:kvm_mediated_pmu_refresh_event_filter() {
...} else { u64 mask = intel_fixed_bits_by_idx(pmc->idx - KVM_FIXED_PMC_BASE_IDX, 0xf); pmu->fixed_ctr_ctrl_hw &= ~mask; if (allowed) pmu->fixed_ctr_ctrl_hw |= pmu->fixed_ctr_ctrl & mask; } }
Quoted 6 lines
Since this mask covers the entire 4-bit configuration block, it will also clear bit 14 (
METRICS_CLEAR_EN) from hardware. If the guest setsMETRICS_CLEAR_ENand then executes RDPMC natively, it will lead to un-cleared stale metrics.Should the mask be updated to avoid interfering with this newly introduced feature bit?
I don't think this needs a change.
The clear/re-apply here happen while the vCPU isn't running, so there's no window for a native RDPMC to see it cleared: if allowed, the guest's METRICS_CLEAR_EN bit is copied through; if not allowed, counter 3 isn't counting anyway.
Quoted 5 lines
Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0counter_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;
Re: [PATCH v9 12/12] KVM: x86/pmu: Support RDPMC Metrics Clear Mode
In reply to [email protected]Quoted 67 lines · On 9/19/2026 4:22 AM, [email protected] wrote:
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] pmu: event filter mistakenly stripsMETRICS_CLEAR_ENfromfixed_ctr_ctrl_hw
- [Medium] pmu:METRICS_CLEAR_ENunmasked without validating fixed counter bounds
- [Low] pmu: emulation incorrectly tiesPERF_METRICSclearing to SLOTS PMC existence --commit
8ba3725c32dbde4a1c3935cb433e898e06425b83
Author: Zide Chen <[email protected]>KVM: x86/pmu: Support RDPMC Metrics Clear Mode
This commit adds support for
PERF_CAP_RDPMC_METRICS_CLEAR(bit 19 ofIA32_PERF_CAPABILITIES). It advertises this capability to guests when the mediated PMU is enabled and the host reports it, unmasking the relevant control bit. It also updatesintel_emulate_rdpmc()to replicate the clear-on-read behavior for consistency.Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0[ ... ]
Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2
-0@@ -138,6 +149,18 @@ static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,return 1 ; *data = pmu->perf_metrics; + +/* arch/x86/kvm/vmx/pmu_intel.c static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, · line 163
Quoted 10 lines
arch/x86/kvm/vmx/pmu_intel.c +10
-0+* Per the SDM, RDPMC of PERF_METRICS clears both PERF_METRICS +* and the SLOTS fixed counter when Metrics Clear Mode is +* enabled via INTEL_FIXED_3_METRICS_CLEAR in FIXED_CTR_CTRL. +*/ + pmc =kvm_pmc_idx_to_pmc (pmu,INTEL_PMC_IDX_FIXED_SLOTS ); +if (pmc &&intel_pmu_metrics_clear_enabled (pmu)) { + pmu->perf_metrics =0 ; +pmc_write_counter (pmc,0 ); + } +[Severity: Low]
Does tying the clearing ofPERF_METRICSto the existence of the SLOTS PMC inintel_emulate_rdpmc()violate the hardware specification?If the guest executes RDPMC natively and triggers an interception, and the SLOTS counter is not exposed to the guest (making
kvm_pmc_idx_to_pmcreturn NULL), butMETRICS_CLEAR_ENis enabled, this code silently skips clearingPERF_METRICS.Should the clearing of
pmu->perf_metricsbe performed independently of whether the software PMC object for the SLOTS counter exists?Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0return 0 ;default :return 1 ;[ ... ]
arch/x86/kvm/vmx/pmu_intel.c static void intel_pmu_refresh(struct kvm_vcpu *vcpu) · line 631
Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +5
-0@@ -601,6 +624,11 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)INTEL_FIXED_0_USER | INTEL_FIXED_0_ENABLE_PMI); +if (perf_capabilities &PERF_CAP_RDPMC_METRICS_CLEAR ) + pmu->fixed_ctr_ctrl_rsvd &= + ~intel_fixed_bits_by_idx (INTEL_FIXED_SLOTS_IDX , +INTEL_FIXED_3_METRICS_CLEAR ); +[Severity: Medium]
Doesintel_pmu_refresh()unconditionally unmask theMETRICS_CLEAR_ENbit without verifying if the guest topology actually exposes the SLOTS counter?
Quoted 6 lines
If a VM is configured with 3 or fewer fixed counters, this unmasking would allow the unprivileged guest OS to write to bit 14 in
MSR_CORE_PERF_FIXED_CTR_CTRLwithout triggering the expected #GP.Should this be gated by checking if
pmu->nr_arch_fixed_counters>INTEL_FIXED_SLOTS_IDX?
This sounds like a real risk. Should we mask out PERF_METRICS from guest PERF_CAPABILITIES in intel_pmu_refresh() if guest doesn't advertise fixed counter 3?
Quoted 20 lines
[Severity: High]
Does this patch inadvertently cause event filters to strip theMETRICS_CLEAR_ENbit inkvm_mediated_pmu_refresh_event_filter()?When a host-level event filter denies the SLOTS counter (counter 3) from counting, KVM applies a
0xfhardware mask that zeroes all 4 configuration bits for the counter:arch/x86/kvm/pmu.c:kvm_mediated_pmu_refresh_event_filter() {
...} else { u64 mask = intel_fixed_bits_by_idx(pmc->idx - KVM_FIXED_PMC_BASE_IDX, 0xf); pmu->fixed_ctr_ctrl_hw &= ~mask; if (allowed) pmu->fixed_ctr_ctrl_hw |= pmu->fixed_ctr_ctrl & mask; } }
Quoted 6 lines
Since this mask covers the entire 4-bit configuration block, it will also clear bit 14 (
METRICS_CLEAR_EN) from hardware. If the guest setsMETRICS_CLEAR_ENand then executes RDPMC natively, it will lead to un-cleared stale metrics.Should the mask be updated to avoid interfering with this newly introduced feature bit?
This is another tricky corner case. The best way should be disabling PERF_METRICS along with fixed counter 3. But Intel HW doesn't provide a way to disable PERF_METRICS except the EN_PERF_METRICS bit in PERF_GLOBAL_CTRL. Since PERF_GLOBAL_CTRL could be in pass-through mode and guest could directly overwrite PERF_GLOBAL_CTRL, it becomes meaningless for KVM clearing EN_PERF_METRICS bit for guest PERF_GLOBAL_CTRL.
For the filter mask, it may be good enough to only mask out (INTEL_FIXED_0_KERNEL | INTEL_FIXED_0_USER).
Thanks.
Quoted 4 lines
Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0counter_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;
Re: [PATCH v9 12/12] KVM: x86/pmu: Support RDPMC Metrics Clear Mode
Quoted 73 lines · Dapeng
Quoted 71 lines · On 9/19/2026 4:22 AM, [email protected] wrote:
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] pmu: event filter mistakenly stripsMETRICS_CLEAR_ENfromfixed_ctr_ctrl_hw
- [Medium] pmu:METRICS_CLEAR_ENunmasked without validating fixed counter bounds
- [Low] pmu: emulation incorrectly tiesPERF_METRICSclearing to SLOTS PMC existence --commit
8ba3725c32dbde4a1c3935cb433e898e06425b83
Author: Zide Chen <[email protected]>KVM: x86/pmu: Support RDPMC Metrics Clear Mode
This commit adds support for
PERF_CAP_RDPMC_METRICS_CLEAR(bit 19 ofIA32_PERF_CAPABILITIES). It advertises this capability to guests when the mediated PMU is enabled and the host reports it, unmasking the relevant control bit. It also updatesintel_emulate_rdpmc()to replicate the clear-on-read behavior for consistency.Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0[ ... ]
Quoted 6 lines
arch/x86/kvm/vmx/pmu_intel.c +2
-0@@ -138,6 +149,18 @@ static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx,return 1 ; *data = pmu->perf_metrics; + +/* arch/x86/kvm/vmx/pmu_intel.c static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, · line 163
Quoted 10 lines
arch/x86/kvm/vmx/pmu_intel.c +10
-0+* Per the SDM, RDPMC of PERF_METRICS clears both PERF_METRICS +* and the SLOTS fixed counter when Metrics Clear Mode is +* enabled via INTEL_FIXED_3_METRICS_CLEAR in FIXED_CTR_CTRL. +*/ + pmc =kvm_pmc_idx_to_pmc (pmu,INTEL_PMC_IDX_FIXED_SLOTS ); +if (pmc &&intel_pmu_metrics_clear_enabled (pmu)) { + pmu->perf_metrics =0 ; +pmc_write_counter (pmc,0 ); + } +[Severity: Low]
Does tying the clearing ofPERF_METRICSto the existence of the SLOTS PMC inintel_emulate_rdpmc()violate the hardware specification?If the guest executes RDPMC natively and triggers an interception, and the SLOTS counter is not exposed to the guest (making
kvm_pmc_idx_to_pmcreturn NULL), butMETRICS_CLEAR_ENis enabled, this code silently skips clearingPERF_METRICS.Should the clearing of
pmu->perf_metricsbe performed independently of whether the software PMC object for the SLOTS counter exists?Quoted 3 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0return 0 ;default :return 1 ;[ ... ]
arch/x86/kvm/vmx/pmu_intel.c static void intel_pmu_refresh(struct kvm_vcpu *vcpu) · line 631
Quoted 9 lines
arch/x86/kvm/vmx/pmu_intel.c +5
-0@@ -601,6 +624,11 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)INTEL_FIXED_0_USER | INTEL_FIXED_0_ENABLE_PMI); +if (perf_capabilities &PERF_CAP_RDPMC_METRICS_CLEAR ) + pmu->fixed_ctr_ctrl_rsvd &= + ~intel_fixed_bits_by_idx (INTEL_FIXED_SLOTS_IDX , +INTEL_FIXED_3_METRICS_CLEAR ); +[Severity: Medium]
Doesintel_pmu_refresh()unconditionally unmask theMETRICS_CLEAR_ENbit without verifying if the guest topology actually exposes the SLOTS counter?If a VM is configured with 3 or fewer fixed counters, this unmasking would allow the unprivileged guest OS to write to bit 14 in
MSR_CORE_PERF_FIXED_CTR_CTRLwithout triggering the expected #GP.
Quoted 6 lines
Quoted 2 lines
Should this be gated by checking if
pmu->nr_arch_fixed_counters>INTEL_FIXED_SLOTS_IDX?This sounds like a real risk. Should we mask out
PERF_METRICSfrom guestPERF_CAPABILITIESinintel_pmu_refresh()if guest doesn't advertise fixed counter 3?
Yes, will checks the availability of guest FC3 inkvm_vcpu_has_perf_metrics() as well.
Quoted 34 lines
Quoted 20 lines
[Severity: High]
Does this patch inadvertently cause event filters to strip theMETRICS_CLEAR_ENbit inkvm_mediated_pmu_refresh_event_filter()?When a host-level event filter denies the SLOTS counter (counter 3) from counting, KVM applies a
0xfhardware mask that zeroes all 4 configuration bits for the counter:arch/x86/kvm/pmu.c:kvm_mediated_pmu_refresh_event_filter() {
...} else { u64 mask = intel_fixed_bits_by_idx(pmc->idx - KVM_FIXED_PMC_BASE_IDX, 0xf); pmu->fixed_ctr_ctrl_hw &= ~mask; if (allowed) pmu->fixed_ctr_ctrl_hw |= pmu->fixed_ctr_ctrl & mask; } }Quoted 6 lines
Since this mask covers the entire 4-bit configuration block, it will also clear bit 14 (
METRICS_CLEAR_EN) from hardware. If the guest setsMETRICS_CLEAR_ENand then executes RDPMC natively, it will lead to un-cleared stale metrics.Should the mask be updated to avoid interfering with this newly introduced feature bit?
This is another tricky corner case. The best way should be disabling
PERF_METRICSalong with fixed counter 3. But Intel HW doesn't provide a way to disablePERF_METRICSexcept theEN_PERF_METRICSbit inPERF_GLOBAL_CTRL. Since PERF_GLOBAL_CTRL could be in pass-through mode and guest could directly overwritePERF_GLOBAL_CTRL, it becomes meaningless for KVM clearingEN_PERF_METRICSbit for guest PERF_GLOBAL_CTRL.
Quoted 2 lines
For the filter mask, it may be good enough to only mask out (
INTEL_FIXED_0_KERNEL|INTEL_FIXED_0_USER).
Yes, the 0xf mask erroneously strips bit 14 (METRICS_CLEAR_EN). Since event filtering is only intended to prevent counter increments, mask out only (INTEL_FIXED_0_KERNEL | INTEL_FIXED_0_USER) is sufficient, which also leaves the PMI enable bit intact.
This also matches the GP counter handling, which only masks out the necessary enable bit and keep all other bits unchanged.
@@ -557,7 +574,9 @@ static void kvm_mediated_pmu_refresh_event_filter(struct kvm_pmc *pmc)
pmc->eventsel_hw |= pmc->eventsel &
ARCH_PERFMON_EVENTSEL_ENABLE;
} else {- u64 mask = intel_fixed_bits_by_idx(pmc->idx -KVM_FIXED_PMC_BASE_IDX, 0xf);
+ u64 mask = intel_fixed_bits_by_idx(pmc->idx -KVM_FIXED_PMC_BASE_IDX,
+ INTEL_FIXED_0_KERNEL |
+ INTEL_FIXED_0_USER);
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
arch/x86/kvm/vmx/pmu_intel.c +0
-0counter_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;
Re: [PATCH v9 12/12] KVM: x86/pmu: Support RDPMC Metrics Clear Mode
In reply to Zide Chenarch/x86/kvm/vmx/pmu_intel.c static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, · line 159
Quoted 3 lines · Zide Chen
arch/x86/kvm/vmx/pmu_intel.c +2
-0+ pmc =kvm_pmc_idx_to_pmc (pmu,INTEL_PMC_IDX_FIXED_SLOTS ); +if (pmc &&intel_pmu_metrics_clear_enabled (pmu)) {
If kvm_vcpu_has_perf_metrics() requires fixed counter 3 (see my reply to 5/12), pmc can't be NULL here.
[...]
arch/x86/kvm/vmx/pmu_intel.c static void intel_pmu_refresh(struct kvm_vcpu *vcpu) · line 630
Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +4
-0+if (perf_capabilities &PERF_CAP_RDPMC_METRICS_CLEAR ) + pmu->fixed_ctr_ctrl_rsvd &= + ~intel_fixed_bits_by_idx (INTEL_FIXED_SLOTS_IDX , +INTEL_FIXED_3_METRICS_CLEAR );
This should be predicated on the guest actually having fixed counter 3 (and PERF_METRICS). intel_pmu_enable_fixed_counter_bits() only unmasks bits for counters below nr_arch_fixed_counters.
[...]
arch/x86/kvm/vmx/vmx.c int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 2621
Quoted 3 lines
arch/x86/kvm/vmx/vmx.c +3
-0+if ((data &PERF_CAP_RDPMC_METRICS_CLEAR ) && + !(data &PERF_CAP_PERF_METRICS )) +return 1 ;
If IA32_FIXED_CTR_CTRL.METRICS_CLEAR_EN[14] is only unmasked when the vCPU has PERF_METRICS, is this check still necessary? KVM generally doesn't police userspace's feature combinations unless it has to.
Re: [PATCH v9 12/12] KVM: x86/pmu: Support RDPMC Metrics Clear Mode
Quoted 7 lines · Jim Mattson
arch/x86/kvm/vmx/pmu_intel.c static int intel_emulate_rdpmc(struct kvm_vcpu *vcpu, unsigned int idx, · line 159
Quoted 3 lines · Zide Chen
arch/x86/kvm/vmx/pmu_intel.c +2
-0+ pmc =kvm_pmc_idx_to_pmc (pmu,INTEL_PMC_IDX_FIXED_SLOTS ); +if (pmc &&intel_pmu_metrics_clear_enabled (pmu)) {If
kvm_vcpu_has_perf_metrics()requires fixed counter 3 (see my reply to 5/12), pmc can't be NULL here.
Agreed. If kvm_vcpu_has_perf_metrics() requires FC3, then the NULL check can be dropped.
Quoted 10 lines
[...]
arch/x86/kvm/vmx/pmu_intel.c static void intel_pmu_refresh(struct kvm_vcpu *vcpu) · line 630
Quoted 4 lines
arch/x86/kvm/vmx/pmu_intel.c +4
-0+if (perf_capabilities &PERF_CAP_RDPMC_METRICS_CLEAR ) + pmu->fixed_ctr_ctrl_rsvd &= + ~intel_fixed_bits_by_idx (INTEL_FIXED_SLOTS_IDX , +INTEL_FIXED_3_METRICS_CLEAR );This should be predicated on the guest actually having fixed counter 3 (and
PERF_METRICS).intel_pmu_enable_fixed_counter_bits()only unmasks bits for counters belownr_arch_fixed_counters.
Yes, since now FC3 check is added to kvm_vcpu_has_perf_metrics(), then it makes sense to gate the unmasking on it as well.
Quoted 10 lines
[...]
arch/x86/kvm/vmx/vmx.c int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) · line 2621
Quoted 3 lines
arch/x86/kvm/vmx/vmx.c +3
-0+if ((data &PERF_CAP_RDPMC_METRICS_CLEAR ) && + !(data &PERF_CAP_PERF_METRICS )) +return 1 ;If IA32_FIXED_CTR_CTRL.METRICS_CLEAR_EN[14] is only unmasked when the vCPU has
PERF_METRICS, is this check still necessary? KVM generally doesn't police userspace's feature combinations unless it has to.
Agreed, dropped this check from vmx_set_msr(). Since intel_pmu_refresh() gates unmasking METRICS_CLEAR_EN on kvm_vcpu_has_perf_metrics(vcpu), userspace setting RDPMC_METRICS_CLEAR without PERF_METRICS will simply have no effect and bit 14 remains reserved for the guest.