[PATCH v9 00/16] KVM: arm64: PMU: Use multiple host PMUs

[PATCH v9 00/16] KVM: arm64: PMU: Use multiple host PMUs

[Akihiko Odaki]

On a heterogeneous arm64 system, KVM's PMU emulation is based on the
features of a single host PMU instance. When a vCPU is migrated to a
pCPU with an incompatible PMU, counters such as PMCCNTR_EL0 stop
incrementing.

Although this behavior is permitted by the architecture, Windows does
not handle it gracefully and may crash with a division-by-zero error.

The current workaround requires VMMs to pin vCPUs to a set of pCPUs
that share a compatible PMU. This is difficult to implement correctly in
QEMU/libvirt, where pinning occurs after vCPU initialization, and it
also restricts the guest to a subset of available pCPUs.

This patch introduces the KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY
attribute. If set, PMUv3 will be emulated without programmable event
counters. KVM will be able to run VCPUs on any physical CPUs with a
compatible hardware PMU.

This allows Windows guests to run reliably on heterogeneous systems
without crashing, even without vCPU pinning, and enables VMMs to
schedule vCPUs across all available pCPUs, making full use of the host
hardware.

A QEMU patch that demonstrates the usage of the new attribute is
available at:
https://lore.kernel.org/qemu-devel/20260706-kvm-v4-1-089cac8c25c3@rsg.ci.i.u-tokyo.ac.jp/
("[PATCH RFC v4] target/arm/kvm: Choose PMU backend")

Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
Changes in v9:
- Dropped patch
  "KVM: arm64: Disallow vPMU when pPMUs do not cover all CPUs".
- Added patch
  "KVM: arm64: PMU: Stop updating MDCR_EL2.HPMN".
- Added patch
  "KVM: arm64: selftests: Cover PMU state in MDCR_EL2".
- Added patch
  "arm64: errata: Require Apple IMPDEF PMUv3 traps on all CPUs".
- Added patch
  "KVM: arm64: Don't clear vcpu->cpu in kvm_arch_vcpu_put()".
- Noted that fixed-counters-only mode avoids relying on the default PMU.
- Clarified why fixed-counters-only mode hides PMCEID.
- Added patch "KVM: arm64: Serialize repeated vCPU initialization".
- Added patch "KVM: arm64: PMU: Freeze counter count after first run".
- Added patch
  "KVM: arm64: selftests: Test SET_NR_COUNTERS after first run".
- Added patch "KVM: arm64: PMU: Recreate events after MDCR_EL2 changes".
- Clarified why extra IMPDEF PMUv3 trap support on a late CPU is safe.
- Changed fixed-counters-only emulation to taint the kernel when it
  encounters a CPU without a registered pPMU; see the commit message.
- Added patch
  "KVM: arm64: PMU: Keep implemented counter mask EL-independent".
- Added a targeted bitfield helper type change to tools headers.
- Link to v8: https://patch.msgid.link/20260710-hybrid-v8-0-621409f3a592@rsg.ci.i.u-tokyo.ac.jp

Changes in v8:
- Added patch
  "KVM: arm64: Disallow vPMU when pPMUs do not cover all CPUs".
- Split patch "KVM: arm64: PMU: Introduce FIXED_COUNTERS_ONLY".
- Documented the FEAT_PMUv3_ICNTR support plan.
- Ensured KVM_ARM_VCPU_PMU_V3_FILTER and
  KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY are mutually exclusive.
- Ensured KVM_ARM_VCPU_PMU_V3_SET_PMU and
  KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY are mutually exclusive.
- Dropped the code to support KVM_ARM_VCPU_PMU_V3_SET_NR_COUNTERS for
  KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY.
- Added a function to de-duplicate
  "test_bit(KVM_ARCH_FLAG_PMU_V3_FIXED_COUNTERS_ONLY, ...)".
- Link to v7: https://lore.kernel.org/r/20260418-hybrid-v7-0-2bf39ad009bf@rsg.ci.i.u-tokyo.ac.jp

Changes in v7:
- Fixed the vCPU run hang in test_fixed_counters_only().
- Link to v6: https://lore.kernel.org/r/20260413-hybrid-v6-0-e79d760f7f1b@rsg.ci.i.u-tokyo.ac.jp

Changes in v6:
- Removed WARN_ON_ONCE() in kvm_pmu_create_perf_event(). It can be
  triggered in kvm_arch_vcpu_load() before it checks supported_cpus.
- Removed an extra lockdep assertion in kvm_arm_pmu_v3_get_attr().
- Fixed error messages in test_fixed_counters_only().
- Fixed the vCPU run in test_fixed_counters_only().
- Link to v5: https://lore.kernel.org/r/20260411-hybrid-v5-0-b043b4d9f49e@rsg.ci.i.u-tokyo.ac.jp

Changes in v5:
- Rebased.
- Fixed the order to clear KVM_ARCH_FLAG_PMU_V3_FIXED_COUNTERS_ONLY in
  kvm_arm_pmu_v3_set_pmu().
- Fixed the setting of KVM_ARM_VCPU_PMU_V3_IRQ in
  test_fixed_counters_only().
- Changed to WARN_ON_ONCE() when kvm_pmu_probe_armpmu() returns NULL in
  kvm_pmu_create_perf_event(), which is no longer supposed to happen.
- Link to v4: https://lore.kernel.org/r/20260317-hybrid-v4-0-bd62bcd48644@rsg.ci.i.u-tokyo.ac.jp

Changes in v4:
- Extracted kvm_pmu_enabled_counter_mask() into a separate patch.
- Added patch "KVM: arm64: PMU: Protect the list of PMUs with RCU".
- Merged KVM_REQ_CREATE_PMU into KVM_REQ_RELOAD_PMU.
- Added a check to avoid unnecessary KVM_REQ_RELOAD_PMU requests.
- Dropped the change to avoid setting kvm_arm_set_default_pmu() when
  KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY is not set.
- Link to v3: https://lore.kernel.org/r/20260225-hybrid-v3-0-46e8fe220880@rsg.ci.i.u-tokyo.ac.jp

Changes in v3:
- Renamed the attribute to KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY.
- Changed to request the creation of perf counters when loading vCPU.
- Link to v2: https://lore.kernel.org/r/20250806-hybrid-v2-0-0661aec3af8c@rsg.ci.i.u-tokyo.ac.jp

Changes in v2:
- Added the KVM_ARM_VCPU_PMU_V3_COMPOSITION attribute to opt in the
  feature.
- Added code to handle overflow.
- Link to v1: https://lore.kernel.org/r/20250319-hybrid-v1-1-4d1ada10e705@daynix.com

---
Akihiko Odaki (15):
      KVM: arm64: Serialize repeated vCPU initialization
      KVM: arm64: PMU: Stop updating MDCR_EL2.HPMN
      KVM: arm64: PMU: Freeze counter count after first run
      KVM: arm64: selftests: Test SET_NR_COUNTERS after first run
      KVM: arm64: PMU: Keep implemented counter mask EL-independent
      KVM: arm64: PMU: Recreate events after MDCR_EL2 changes
      tools headers: Use u* types for bitfield helpers
      KVM: arm64: selftests: Cover PMU state in MDCR_EL2
      arm64: errata: Require Apple IMPDEF PMUv3 traps on all CPUs
      KVM: arm64: PMU: Protect the list of PMUs with RCU
      KVM: arm64: PMU: Pass the pPMU to kvm_map_pmu_event()
      KVM: arm64: PMU: Pass the target CPU to kvm_pmu_probe_armpmu()
      KVM: arm64: PMU: Implement fixed-counters-only emulation
      KVM: arm64: PMU: Introduce FIXED_COUNTERS_ONLY
      KVM: arm64: selftests: Test PMU_V3_FIXED_COUNTERS_ONLY

Oliver Upton (1):
      KVM: arm64: Don't clear vcpu->cpu in kvm_arch_vcpu_put()

 Documentation/virt/kvm/devices/vcpu.rst            |  30 ++-
 arch/arm64/include/asm/kvm_host.h                  |   4 +
 arch/arm64/include/uapi/asm/kvm.h                  |   1 +
 arch/arm64/kernel/cpu_errata.c                     |   2 +-
 arch/arm64/kvm/arm.c                               |  18 +-
 arch/arm64/kvm/pmu-emul.c                          | 241 +++++++++++------
 arch/arm64/kvm/reset.c                             |  16 +-
 arch/arm64/kvm/sys_regs.c                          |  31 ++-
 include/kvm/arm_pmu.h                              |   5 +
 tools/arch/arm64/include/uapi/asm/kvm.h            |   1 +
 tools/include/linux/bitfield.h                     |  26 +-
 .../selftests/kvm/arm64/vpmu_counter_access.c      | 284 ++++++++++++++++++---
 12 files changed, 510 insertions(+), 149 deletions(-)
---
base-commit: cb9360ae6549120975f20887840fb33d47c13898
change-id: 20250224-hybrid-01d5ff47edd2

Best regards,
--  
Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>

[PATCH v9 01/16] KVM: arm64: Serialize repeated vCPU initialization

[Akihiko Odaki]

The initial KVM_ARM_VCPU_INIT path holds config_lock while resetting the
vCPU, but repeated initialization of an existing vCPU does not. As a
result, reset_mdcr() can race with VM-wide PMU configuration and
overwrite MDCR_EL2 with a stale counter count.

Take config_lock around the repeated-init checks and reset. The vCPU
ioctl already holds vcpu->mutex, matching the established lock ordering.

Fixes: c8823e51b534 ("KVM: arm64: Fix MDCR_EL2.HPMN reset value")
Closes: https://sashiko.dev/#/patchset/20260710-hybrid-v8-0-621409f3a592@rsg.ci.i.u-tokyo.ac.jp?part=1
Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/arm.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index 50adfff75be8..a06ea39f57b1 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -1652,29 +1652,26 @@ static int __kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 {
 	unsigned long features = init->features[0];
 	struct kvm *kvm = vcpu->kvm;
-	int ret = -EINVAL;
+	int ret;
 
-	mutex_lock(&kvm->arch.config_lock);
+	lockdep_assert_held(&kvm->arch.config_lock);
 
 	if (test_bit(KVM_ARCH_FLAG_VCPU_FEATURES_CONFIGURED, &kvm->arch.flags) &&
 	    kvm_vcpu_init_changed(vcpu, init))
-		goto out_unlock;
+		return -EINVAL;
 
 	bitmap_copy(kvm->arch.vcpu_features, &features, KVM_VCPU_MAX_FEATURES);
 
 	ret = kvm_setup_vcpu(vcpu);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	/* Now we know what it is, we can reset it. */
 	kvm_reset_vcpu(vcpu);
 
 	set_bit(KVM_ARCH_FLAG_VCPU_FEATURES_CONFIGURED, &kvm->arch.flags);
 	vcpu_set_flag(vcpu, VCPU_INITIALIZED);
-	ret = 0;
-out_unlock:
-	mutex_unlock(&kvm->arch.config_lock);
-	return ret;
+	return 0;
 }
 
 static int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
@@ -1690,6 +1687,8 @@ static int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 	if (ret)
 		return ret;
 
+	guard(mutex)(&vcpu->kvm->arch.config_lock);
+
 	if (!kvm_vcpu_initialized(vcpu))
 		return __kvm_vcpu_set_target(vcpu, init);
 

-- 
2.55.0

[PATCH v9 02/16] KVM: arm64: PMU: Stop updating MDCR_EL2.HPMN

[Akihiko Odaki]

KVM allows userspace to write any value to MDCR_EL2.HPMN. However,
kvm_arm_set_nr_counters() rewrites HPMN for every vCPU whenever
userspace changes the PMU or its counter count. This can discard a
value previously restored with KVM_SET_ONE_REG.

The architecture only defines HPMN's value on warm reset. Stop
rewriting it after vCPU initialization and update nr_pmu_counters
directly instead. reset_mdcr() continues to initialize HPMN from the
counter count current at KVM_ARM_VCPU_INIT.

Fixes: c8823e51b534 ("KVM: arm64: Fix MDCR_EL2.HPMN reset value")
Closes: https://sashiko.dev/#/patchset/20260706-hybrid-v8-0-de459617b59d%40rsg.ci.i.u-tokyo.ac.jp?part=6
Assisted-by: Codex:gpt-5.5
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/pmu-emul.c | 22 ++--------------------
 1 file changed, 2 insertions(+), 20 deletions(-)

diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index 98305bbfc095..b5df6843dbcd 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -1022,30 +1022,12 @@ u8 kvm_arm_pmu_get_max_counters(struct kvm *kvm)
 	return bitmap_weight(arm_pmu->cntr_mask, ARMV8_PMU_MAX_GENERAL_COUNTERS);
 }
 
-static void kvm_arm_set_nr_counters(struct kvm *kvm, unsigned int nr)
-{
-	kvm->arch.nr_pmu_counters = nr;
-
-	/* Reset MDCR_EL2.HPMN behind the vcpus' back... */
-	if (test_bit(KVM_ARM_VCPU_HAS_EL2, kvm->arch.vcpu_features)) {
-		struct kvm_vcpu *vcpu;
-		unsigned long i;
-
-		kvm_for_each_vcpu(i, vcpu, kvm) {
-			u64 val = __vcpu_sys_reg(vcpu, MDCR_EL2);
-			val &= ~MDCR_EL2_HPMN;
-			val |= FIELD_PREP(MDCR_EL2_HPMN, kvm->arch.nr_pmu_counters);
-			__vcpu_assign_sys_reg(vcpu, MDCR_EL2, val);
-		}
-	}
-}
-
 static void kvm_arm_set_pmu(struct kvm *kvm, struct arm_pmu *arm_pmu)
 {
 	lockdep_assert_held(&kvm->arch.config_lock);
 
 	kvm->arch.arm_pmu = arm_pmu;
-	kvm_arm_set_nr_counters(kvm, kvm_arm_pmu_get_max_counters(kvm));
+	kvm->arch.nr_pmu_counters = kvm_arm_pmu_get_max_counters(kvm);
 }
 
 /**
@@ -1111,7 +1093,7 @@ static int kvm_arm_pmu_v3_set_nr_counters(struct kvm_vcpu *vcpu, unsigned int n)
 	if (n > kvm_arm_pmu_get_max_counters(kvm))
 		return -EINVAL;
 
-	kvm_arm_set_nr_counters(kvm, n);
+	kvm->arch.nr_pmu_counters = n;
 	return 0;
 }
 

-- 
2.55.0

[PATCH v9 03/16] KVM: arm64: PMU: Freeze counter count after first run

[Akihiko Odaki]

PMU configuration is VM-scoped, but SET_NR_COUNTERS can be issued
through any vCPU. Checking only whether that vCPU's PMU has been
initialized allows userspace to change the counter count through an idle
sibling after another vCPU has run.

Reject the attribute once any vCPU has run. This keeps the VM-wide
implemented counter mask and the PMCR_EL0.N value stable after guest
execution begins.

Fixes: b7628c797376 ("KVM: arm64: Allow userspace to limit the number of PMU counters for EL2 VMs")
Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/pmu-emul.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index b5df6843dbcd..b4b877f31097 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -1087,6 +1087,9 @@ static int kvm_arm_pmu_v3_set_nr_counters(struct kvm_vcpu *vcpu, unsigned int n)
 {
 	struct kvm *kvm = vcpu->kvm;
 
+	if (kvm_vm_has_ran_once(kvm))
+		return -EBUSY;
+
 	if (!kvm->arch.arm_pmu)
 		return -EINVAL;
 

-- 
2.55.0

[PATCH v9 04/16] KVM: arm64: selftests: Test SET_NR_COUNTERS after first run

[Akihiko Odaki]

Add a regression test that runs one vCPU and then tries to change the
VM-scoped PMU counter count through an idle sibling. Verify that
SET_NR_COUNTERS fails with EBUSY once any vCPU has run.

Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 .../selftests/kvm/arm64/vpmu_counter_access.c      | 38 ++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c b/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
index 22223395969e..3b5bb421f401 100644
--- a/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
+++ b/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
@@ -461,6 +461,11 @@ static void run_vcpu(struct kvm_vcpu *vcpu, u64 pmcr_n)
 	}
 }
 
+static void guest_code_done(void)
+{
+	GUEST_DONE();
+}
+
 static void test_create_vpmu_vm_with_nr_counters(unsigned int nr_counters, bool expect_fail)
 {
 	struct kvm_vcpu *vcpu;
@@ -622,6 +627,37 @@ static bool kvm_supports_nr_counters_attr(void)
 	return supported;
 }
 
+static void test_set_nr_counters_after_vcpu_run(void)
+{
+	struct kvm_vcpu *running_vcpu, *stopped_vcpu;
+	unsigned int nr_counters = 0;
+	struct kvm_vcpu_init init;
+	struct kvm_vm *vm;
+	int ret;
+	u64 irq = 23;
+
+	vm = vm_create(2);
+	vm_ioctl(vm, KVM_ARM_PREFERRED_TARGET, &init);
+	init.features[0] |= BIT(KVM_ARM_VCPU_PMU_V3);
+	running_vcpu = aarch64_vcpu_add(vm, 0, &init, guest_code_done);
+	stopped_vcpu = aarch64_vcpu_add(vm, 1, &init, guest_code_done);
+	kvm_arch_vm_finalize_vcpus(vm);
+
+	vcpu_device_attr_set(running_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_IRQ, &irq);
+	vcpu_device_attr_set(running_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_INIT, NULL);
+	vcpu_run(running_vcpu);
+
+	ret = __vcpu_device_attr_set(stopped_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+				     KVM_ARM_VCPU_PMU_V3_SET_NR_COUNTERS,
+				     &nr_counters);
+	TEST_ASSERT(ret == -1 && errno == EBUSY,
+		    KVM_IOCTL_ERROR(KVM_SET_DEVICE_ATTR, ret));
+
+	kvm_vm_free(vm);
+}
+
 int main(void)
 {
 	u64 i, pmcr_n;
@@ -630,6 +666,8 @@ int main(void)
 	TEST_REQUIRE(kvm_supports_vgic_v3());
 	TEST_REQUIRE(kvm_supports_nr_counters_attr());
 
+	test_set_nr_counters_after_vcpu_run();
+
 	pmcr_n = get_pmcr_n_limit();
 	for (i = 0; i <= pmcr_n; i++) {
 		run_access_test(i);

-- 
2.55.0

[PATCH v9 05/16] KVM: arm64: PMU: Keep implemented counter mask EL-independent

[Akihiko Odaki]

kvm_pmu_implemented_counter_mask() derives the counter count from
kvm_vcpu_read_pmcr(). For a nested vCPU outside EL2, PMCR_EL0.N reflects
MDCR_EL2.HPMN instead of the VM-wide counter count.

Consequently, processing a PMU reload while the saved context is L2 can
clear and fail to recreate state for counters reserved for EL2.

Build the implemented counter mask from nr_pmu_counters instead. Keep
the EL-dependent HPMN restriction in
kvm_pmu_accessible_counter_mask(), where it controls guest access.

Fixes: 600f6fa5c90c ("KVM: arm64: Let kvm_vcpu_read_pmcr() return an EL-dependent value for PMCR_EL0.N")
Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/pmu-emul.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index b4b877f31097..5281d37634a0 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -317,12 +317,12 @@ u64 kvm_pmu_accessible_counter_mask(struct kvm_vcpu *vcpu)
 
 u64 kvm_pmu_implemented_counter_mask(struct kvm_vcpu *vcpu)
 {
-	u64 val = FIELD_GET(ARMV8_PMU_PMCR_N, kvm_vcpu_read_pmcr(vcpu));
+	u64 n = vcpu->kvm->arch.nr_pmu_counters;
 
-	if (val == 0)
+	if (n == 0)
 		return BIT(ARMV8_PMU_CYCLE_IDX);
 	else
-		return GENMASK(val - 1, 0) | BIT(ARMV8_PMU_CYCLE_IDX);
+		return GENMASK(n - 1, 0) | BIT(ARMV8_PMU_CYCLE_IDX);
 }
 
 static void kvm_pmc_enable_perf_event(struct kvm_pmc *pmc)

-- 
2.55.0

[PATCH v9 06/16] KVM: arm64: PMU: Recreate events after MDCR_EL2 changes

[Akihiko Odaki]

MDCR_EL2.HPMN changes which counters are reserved for EL2 and thus which
enable control, event filter, and overflow width apply. HPMD changes EL2
filtering, while HLP changes the overflow width and sample period.

The existing guest MDCR_EL2 handling only requests a PMU reload for
HPME. Reloading enables or disables existing perf events, but does not
rebuild events whose attributes have become stale. Generic userspace
writes through KVM_SET_ONE_REG do not request a reload at all.

Route guest writes, userspace writes, and reset through a common helper.
Mark the vCPU's perf events for recreation when HPMN, HPMD, or HLP
changes. Consume the marker on the vCPU thread before reprogramming the
counters, which also preserves counter values and avoids creating events
on an arbitrary ioctl thread.

Use a setter-only accessor so register restore gains these side effects
without changing generic reads or rejecting register values.

Fixes: fe827f916662 ("KVM: arm64: nv: Honor MDCR_EL2.HPME")
Fixes: 8a34979030f6 ("KVM: arm64: nv: Apply EL2 event filtering when in hyp context")
Fixes: 16535d55e91f ("KVM: arm64: nv: Honor MDCR_EL2.HLP")
Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/pmu-emul.c | 27 +++++++++++++++++++++++++++
 arch/arm64/kvm/sys_regs.c | 31 +++++++++++++++++++++----------
 include/kvm/arm_pmu.h     |  3 +++
 3 files changed, 51 insertions(+), 10 deletions(-)

diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index 5281d37634a0..9ac5f307ea31 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -17,6 +17,9 @@
 
 #define PERF_ATTR_CFG1_COUNTER_64BIT	BIT(0)
 
+#define MDCR_EL2_PMU_RECREATE_MASK	(MDCR_EL2_HPMN | MDCR_EL2_HPMD | \
+					 MDCR_EL2_HLP)
+
 static LIST_HEAD(arm_pmus);
 static DEFINE_MUTEX(arm_pmus_lock);
 
@@ -606,6 +609,21 @@ void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val)
 	}
 }
 
+void kvm_pmu_handle_mdcr(struct kvm_vcpu *vcpu, u64 old, u64 val)
+{
+	u64 changed = old ^ val;
+
+	/*
+	 * HPMN determines which counters HPMD and HLP apply to. Changes to
+	 * these fields require new perf event filters and sample periods.
+	 */
+	if (changed & MDCR_EL2_PMU_RECREATE_MASK)
+		vcpu->arch.pmu.events_need_recreate = true;
+
+	if (changed & (MDCR_EL2_HPME | MDCR_EL2_PMU_RECREATE_MASK))
+		kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu);
+}
+
 static bool kvm_pmu_counter_is_enabled(struct kvm_pmc *pmc)
 {
 	struct kvm_vcpu *vcpu = kvm_pmc_to_vcpu(pmc);
@@ -899,7 +917,16 @@ u64 kvm_pmu_get_pmceid(struct kvm_vcpu *vcpu, bool pmceid1)
 
 void kvm_vcpu_reload_pmu(struct kvm_vcpu *vcpu)
 {
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
 	u64 mask = kvm_pmu_implemented_counter_mask(vcpu);
+	int i;
+
+	if (pmu->events_need_recreate) {
+		for (i = 0; i < KVM_ARMV8_PMU_MAX_COUNTERS; i++)
+			kvm_pmu_stop_counter(kvm_vcpu_idx_to_pmc(vcpu, i));
+
+		pmu->events_need_recreate = false;
+	}
 
 	__vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, &=, mask);
 	__vcpu_rmw_sys_reg(vcpu, PMINTENSET_EL1, &=, mask);
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 5d5c579d4579..9e905d9be9f6 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -3020,17 +3020,22 @@ static bool access_mdcr(struct kvm_vcpu *vcpu,
 	}
 
 	__vcpu_assign_sys_reg(vcpu, MDCR_EL2, val);
-
-	/*
-	 * Request a reload of the PMU to enable/disable the counters
-	 * affected by HPME.
-	 */
-	if ((old ^ val) & MDCR_EL2_HPME)
-		kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu);
+	kvm_pmu_handle_mdcr(vcpu, old, val);
 
 	return true;
 }
 
+static int set_mdcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
+		    u64 val)
+{
+	u64 old = __vcpu_sys_reg(vcpu, MDCR_EL2);
+
+	__vcpu_assign_sys_reg(vcpu, MDCR_EL2, val);
+	kvm_pmu_handle_mdcr(vcpu, old, val);
+
+	return 0;
+}
+
 static bool access_ras(struct kvm_vcpu *vcpu,
 		       struct sys_reg_params *p,
 		       const struct sys_reg_desc *r)
@@ -3177,8 +3182,13 @@ static int set_imp_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r,
 
 static u64 reset_mdcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
-	__vcpu_assign_sys_reg(vcpu, r->reg, vcpu->kvm->arch.nr_pmu_counters);
-	return vcpu->kvm->arch.nr_pmu_counters;
+	u64 old = __vcpu_sys_reg(vcpu, r->reg);
+	u64 val = vcpu->kvm->arch.nr_pmu_counters;
+
+	__vcpu_assign_sys_reg(vcpu, r->reg, val);
+	kvm_pmu_handle_mdcr(vcpu, old, val);
+
+	return val;
 }
 
 /*
@@ -3741,7 +3751,8 @@ static const struct sys_reg_desc sys_reg_descs[] = {
 	EL2_REG_FILTERED(SCTLR2_EL2, access_vm_reg, reset_val, 0,
 			 sctlr2_el2_visibility),
 	EL2_REG_VNCR(HCR_EL2, reset_hcr, 0),
-	EL2_REG(MDCR_EL2, access_mdcr, reset_mdcr, 0),
+	SYS_REG_USER_FILTER(MDCR_EL2, access_mdcr, reset_mdcr, 0,
+			    NULL, set_mdcr, el2_visibility),
 	EL2_REG(CPTR_EL2, access_rw, reset_val, CPTR_NVHE_EL2_RES1),
 	EL2_REG_VNCR(HSTR_EL2, reset_val, 0),
 	EL2_REG_VNCR_FILT(HFGRTR_EL2, fgt_visibility),
diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h
index b5e5942204fc..1971f8794c9d 100644
--- a/include/kvm/arm_pmu.h
+++ b/include/kvm/arm_pmu.h
@@ -32,6 +32,7 @@ struct kvm_pmu {
 	struct kvm_pmc pmc[KVM_ARMV8_PMU_MAX_COUNTERS];
 	int irq_num;
 	bool created;
+	bool events_need_recreate;
 };
 
 struct arm_pmu_entry {
@@ -56,6 +57,7 @@ bool kvm_pmu_should_notify_user(struct kvm_vcpu *vcpu);
 bool kvm_pmu_update_run(struct kvm_vcpu *vcpu);
 void kvm_pmu_software_increment(struct kvm_vcpu *vcpu, u64 val);
 void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val);
+void kvm_pmu_handle_mdcr(struct kvm_vcpu *vcpu, u64 old, u64 val);
 void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data,
 				    u64 select_idx);
 void kvm_vcpu_reload_pmu(struct kvm_vcpu *vcpu);
@@ -133,6 +135,7 @@ static inline bool kvm_pmu_should_notify_user(struct kvm_vcpu *vcpu)
 static inline bool kvm_pmu_update_run(struct kvm_vcpu *vcpu) { return false; }
 static inline void kvm_pmu_software_increment(struct kvm_vcpu *vcpu, u64 val) {}
 static inline void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val) {}
+static inline void kvm_pmu_handle_mdcr(struct kvm_vcpu *vcpu, u64 old, u64 val) {}
 static inline void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu,
 						  u64 data, u64 select_idx) {}
 static inline int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu,

-- 
2.55.0

[PATCH v9 07/16] tools headers: Use u* types for bitfield helpers

[Akihiko Odaki]

Commit 9486aa387716 ("perf tools: Fix 64 bit integer format strings")
defined the tools-local u64 as uint64_t so that PRI*64 format macros
match its underlying C type. On architectures where uint64_t and __u64
are distinct types, however, ____MAKE_OP() still gives the u64_*
helpers an __u64 interface.

Teach ____MAKE_OP() to accept the helper name and storage type
separately. Keep endian helpers on __le* and __be*, and generate the
native helpers using the tools-local u* types.

Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 tools/include/linux/bitfield.h | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/tools/include/linux/bitfield.h b/tools/include/linux/bitfield.h
index ddf81f24956b..1c633343d576 100644
--- a/tools/include/linux/bitfield.h
+++ b/tools/include/linux/bitfield.h
@@ -142,32 +142,32 @@ static __always_inline u64 field_mask(u64 field)
 	return field / field_multiplier(field);
 }
 #define field_max(field)	((typeof(field))field_mask(field))
-#define ____MAKE_OP(type,base,to,from)					\
-static __always_inline __##type type##_encode_bits(base v, base field)	\
+#define ____MAKE_OP(name,type,base,to,from)				\
+static __always_inline type name##_encode_bits(base v, base field)	\
 {									\
 	if (__builtin_constant_p(v) && (v & ~field_mask(field)))	\
 		__field_overflow();					\
 	return to((v & field_mask(field)) * field_multiplier(field));	\
 }									\
-static __always_inline __##type type##_replace_bits(__##type old,	\
-					base val, base field)		\
+static __always_inline type name##_replace_bits(type old,		\
+						base val, base field)	\
 {									\
-	return (old & ~to(field)) | type##_encode_bits(val, field);	\
+	return (old & ~to(field)) | name##_encode_bits(val, field);	\
 }									\
-static __always_inline void type##p_replace_bits(__##type *p,		\
+static __always_inline void name##p_replace_bits(type *p,		\
 					base val, base field)		\
 {									\
-	*p = (*p & ~to(field)) | type##_encode_bits(val, field);	\
+	*p = (*p & ~to(field)) | name##_encode_bits(val, field);	\
 }									\
-static __always_inline base type##_get_bits(__##type v, base field)	\
+static __always_inline base name##_get_bits(type v, base field)	\
 {									\
 	return (from(v) & field)/field_multiplier(field);		\
 }
-#define __MAKE_OP(size)							\
-	____MAKE_OP(le##size,u##size,cpu_to_le##size,le##size##_to_cpu)	\
-	____MAKE_OP(be##size,u##size,cpu_to_be##size,be##size##_to_cpu)	\
-	____MAKE_OP(u##size,u##size,,)
-____MAKE_OP(u8,u8,,)
+#define __MAKE_OP(size)									\
+	____MAKE_OP(le##size,__le##size,u##size,cpu_to_le##size,le##size##_to_cpu)	\
+	____MAKE_OP(be##size,__be##size,u##size,cpu_to_be##size,be##size##_to_cpu)	\
+	____MAKE_OP(u##size,u##size,u##size,,)
+____MAKE_OP(u8,u8,u8,,)
 __MAKE_OP(16)
 __MAKE_OP(32)
 __MAKE_OP(64)

-- 
2.55.0

[PATCH v9 08/16] KVM: arm64: selftests: Cover PMU state in MDCR_EL2

[Akihiko Odaki]

Add MDCR_EL2 coverage to vpmu_counter_access when EL2 is available.

For each configured PMCR_EL0.N value, verify that an arbitrary HPMN
value, including one above the configured counter count, round-trips
through KVM_SET_ONE_REG and survives SET_NR_COUNTERS. Also verify that
HPME can be toggled without disturbing HPMN and that KVM_ARM_VCPU_INIT
resets HPMN from the configured count.

Assisted-by: Codex:gpt-5.5
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 .../selftests/kvm/arm64/vpmu_counter_access.c      | 90 ++++++++++++++++++++--
 1 file changed, 85 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c b/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
index 3b5bb421f401..73f04e4c6ca3 100644
--- a/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
+++ b/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
@@ -25,6 +25,14 @@
 /* The cycle counter bit position that's common among the PMU registers */
 #define ARMV8_PMU_CYCLE_IDX		31
 
+#ifndef MDCR_EL2_HPMN
+#define MDCR_EL2_HPMN			GENMASK_ULL(4, 0)
+#endif
+
+#ifndef MDCR_EL2_HPME
+#define MDCR_EL2_HPME			BIT_ULL(7)
+#endif
+
 struct vpmu_vm {
 	struct kvm_vm *vm;
 	struct kvm_vcpu *vcpu;
@@ -466,15 +474,12 @@ static void guest_code_done(void)
 	GUEST_DONE();
 }
 
-static void test_create_vpmu_vm_with_nr_counters(unsigned int nr_counters, bool expect_fail)
+static void set_nr_counters(struct kvm_vcpu *vcpu,
+			    unsigned int nr_counters, bool expect_fail)
 {
-	struct kvm_vcpu *vcpu;
 	unsigned int prev;
 	int ret;
 
-	create_vpmu_vm(guest_code);
-	vcpu = vpmu_vm.vcpu;
-
 	prev = get_pmcr_n(vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_PMCR_EL0)));
 
 	ret = __vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
@@ -486,6 +491,17 @@ static void test_create_vpmu_vm_with_nr_counters(unsigned int nr_counters, bool
 			    nr_counters, prev);
 	else
 		TEST_ASSERT(!ret, KVM_IOCTL_ERROR(KVM_SET_DEVICE_ATTR, ret));
+}
+
+static void test_create_vpmu_vm_with_nr_counters(unsigned int nr_counters,
+						 bool expect_fail)
+{
+	struct kvm_vcpu *vcpu;
+
+	create_vpmu_vm(guest_code);
+	vcpu = vpmu_vm.vcpu;
+
+	set_nr_counters(vcpu, nr_counters, expect_fail);
 
 	vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL, KVM_ARM_VCPU_PMU_V3_INIT, NULL);
 }
@@ -588,6 +604,69 @@ static void run_pmregs_validity_test(u64 pmcr_n)
 	destroy_vpmu_vm();
 }
 
+static void run_mdcr_el2_validity_test(u64 pmcr_n)
+{
+	struct kvm_vcpu_init init;
+	struct kvm_vcpu *vcpu;
+	u64 expected_mdcr, mdcr;
+
+	pr_debug("MDCR_EL2 test with pmcr_n %lu\n", pmcr_n);
+
+	create_vpmu_vm(guest_code);
+	if (!vm_supports_el2(vpmu_vm.vm)) {
+		destroy_vpmu_vm();
+		return;
+	}
+
+	vcpu = vpmu_vm.vcpu;
+
+	mdcr = vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2));
+	expected_mdcr = u64_replace_bits(mdcr, FIELD_MAX(MDCR_EL2_HPMN),
+					 MDCR_EL2_HPMN);
+	vcpu_set_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2),
+		     expected_mdcr);
+
+	mdcr = vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2));
+	TEST_ASSERT(mdcr == expected_mdcr,
+		    "MDCR_EL2 was not properly updated after HPMN write (expected 0x%lx, got 0x%lx)",
+		    expected_mdcr, mdcr);
+
+	set_nr_counters(vcpu, pmcr_n, false);
+
+	mdcr = vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2));
+	TEST_ASSERT(mdcr == expected_mdcr,
+		    "MDCR_EL2 changed after PMU_V3_SET_NR_COUNTERS (expected 0x%lx, got 0x%lx)",
+		    expected_mdcr, mdcr);
+
+	vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_INIT, NULL);
+
+	mdcr = vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2));
+	TEST_ASSERT(mdcr == expected_mdcr,
+		    "MDCR_EL2 changed after PMU_V3_INIT (expected 0x%lx, got 0x%lx)",
+		    expected_mdcr, mdcr);
+
+	expected_mdcr = mdcr ^ MDCR_EL2_HPME;
+	vcpu_set_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2),
+		     expected_mdcr);
+
+	mdcr = vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2));
+	TEST_ASSERT(mdcr == expected_mdcr,
+		    "MDCR_EL2 was not properly updated after HPME write (expected 0x%lx, got 0x%lx)",
+		    expected_mdcr, mdcr);
+
+	kvm_get_default_vcpu_target(vpmu_vm.vm, &init);
+	init.features[0] |= (1 << KVM_ARM_VCPU_PMU_V3);
+	aarch64_vcpu_setup(vcpu, &init);
+
+	mdcr = vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2));
+	TEST_ASSERT(FIELD_GET(MDCR_EL2_HPMN, mdcr) == pmcr_n,
+		    "MDCR_EL2.HPMN is not reset after INIT (expected %lu, got %lu)",
+		    pmcr_n, FIELD_GET(MDCR_EL2_HPMN, mdcr));
+
+	destroy_vpmu_vm();
+}
+
 /*
  * Create a guest with one vCPU, and attempt to set the PMCR_EL0.N for
  * the vCPU to @pmcr_n, which is larger than the host value.
@@ -672,6 +751,7 @@ int main(void)
 	for (i = 0; i <= pmcr_n; i++) {
 		run_access_test(i);
 		run_pmregs_validity_test(i);
+		run_mdcr_el2_validity_test(i);
 	}
 
 	for (i = pmcr_n + 1; i < ARMV8_PMU_MAX_COUNTERS; i++)

-- 
2.55.0

[PATCH v9 09/16] arm64: errata: Require Apple IMPDEF PMUv3 traps on all CPUs

[Akihiko Odaki]

KVM relies on ARM64_WORKAROUND_PMUV3_IMPDEF_TRAPS as a system-wide
precondition for PMUv3 emulation on affected Apple CPUs. A local CPU
erratum is too weak for that: once the workaround is enabled, a late CPU
that misses it may still be onlined.

Make the workaround an all-early-CPU capability instead. The matcher is
still local because it depends on MIDR_EL1, but the capability is only
advertised if all early CPUs match. A late CPU that misses an advertised
capability is therefore rejected.

Although kept in the errata capability table, this capability describes
optional trap support rather than a workaround required for safe CPU
operation. A late CPU may therefore match it when the system does not.
KVM treats an IMPDEF PMU as unimplemented unless the capability was
finalized system-wide, so the additional trap support remains unused.

Fixes: bed9b8ec8c71 ("KVM: arm64: Advertise PMUv3 if IMPDEF traps are present")
Assisted-by: Codex:gpt-5.5
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kernel/cpu_errata.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
index 1995e1198648..a5ae628204a1 100644
--- a/arch/arm64/kernel/cpu_errata.c
+++ b/arch/arm64/kernel/cpu_errata.c
@@ -1005,7 +1005,7 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
 	{
 		.desc = "Apple IMPDEF PMUv3 Traps",
 		.capability = ARM64_WORKAROUND_PMUV3_IMPDEF_TRAPS,
-		.type = ARM64_CPUCAP_LOCAL_CPU_ERRATUM,
+		.type = ARM64_CPUCAP_EARLY_LOCAL_CPU_FEATURE,
 		.matches = has_impdef_pmuv3,
 		.cpu_enable = cpu_enable_impdef_pmuv3_traps,
 	},

-- 
2.55.0

[PATCH v9 10/16] KVM: arm64: Don't clear vcpu->cpu in kvm_arch_vcpu_put()

[Akihiko Odaki]

From: Oliver Upton <oupton@kernel.org>

commit e9b152cb957c ("arm/arm64: kvm: Set vcpu->cpu to -1 on vcpu_put")
reset vcpu->cpu in order for the VGIC to determine if there was any vCPU
running at the time of access. The VGIC has gone through an entire
rewrite since then, and with commit 7d450e282171 ("KVM: arm/arm64:
vgic-new: Add userland access to VGIC dist registers") the user
accessors just grab all vCPU mutexes instead.

Drop this remaining vestige such that kvm_arch_vcpu_load() can properly
detect a CPU migration. While at it, rework kvm_reset_vcpu() to do a
much more pedantic check that the provided vCPU is actually what's
running on the present CPU.

Signed-off-by: Oliver Upton <oupton@kernel.org>
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/arm.c   |  1 -
 arch/arm64/kvm/reset.c | 16 +++++++++++-----
 2 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index a06ea39f57b1..72ce79ce4d5e 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -749,7 +749,6 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 	kvm_arm_vmid_clear_active();
 
 	vcpu_clear_on_unsupported_cpu(vcpu);
-	vcpu->cpu = -1;
 }
 
 static void __kvm_arm_vcpu_power_off(struct kvm_vcpu *vcpu)
diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index b963fd975aac..6bfb21f00444 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -190,7 +190,8 @@ static void kvm_vcpu_reset_sve(struct kvm_vcpu *vcpu)
 void kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_reset_state reset_state;
-	bool loaded;
+	struct kvm_vcpu *running;
+	bool loaded = false;
 	u32 pstate;
 
 	spin_lock(&vcpu->arch.mp_state_lock);
@@ -198,10 +199,16 @@ void kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 	vcpu->arch.reset_state.reset = false;
 	spin_unlock(&vcpu->arch.mp_state_lock);
 
-	preempt_disable();
-	loaded = (vcpu->cpu != -1);
-	if (loaded)
+	guard(preempt)();
+
+	running = kvm_get_running_vcpu();
+	if (running) {
+		if (KVM_BUG_ON(running != vcpu, vcpu->kvm))
+			return;
+
+		loaded = true;
 		kvm_arch_vcpu_put(vcpu);
+	}
 
 	if (!kvm_arm_vcpu_sve_finalized(vcpu)) {
 		if (vcpu_has_feature(vcpu, KVM_ARM_VCPU_SVE))
@@ -269,7 +276,6 @@ void kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 
 	if (loaded)
 		kvm_arch_vcpu_load(vcpu, smp_processor_id());
-	preempt_enable();
 }
 
 u32 kvm_get_pa_bits(struct kvm *kvm)

-- 
2.55.0

[PATCH v9 11/16] KVM: arm64: PMU: Protect the list of PMUs with RCU

[Akihiko Odaki]

Convert the list of PMUs to a RCU-protected list that has primitives to
avoid read-side contention.

Assisted-by: Codex:gpt-5.5
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/pmu-emul.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index 9ac5f307ea31..c5c505b6fc61 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -7,9 +7,9 @@
 #include <linux/cpu.h>
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
-#include <linux/list.h>
 #include <linux/perf_event.h>
 #include <linux/perf/arm_pmu.h>
+#include <linux/rculist.h>
 #include <linux/uaccess.h>
 #include <asm/kvm_emulate.h>
 #include <kvm/arm_pmu.h>
@@ -20,6 +20,10 @@
 #define MDCR_EL2_PMU_RECREATE_MASK	(MDCR_EL2_HPMN | MDCR_EL2_HPMD | \
 					 MDCR_EL2_HLP)
 
+/*
+ * arm_pmus is append-only. kvm_supports_guest_pmuv3() feeds persistent
+ * VM state, so a true result must remain valid after the check.
+ */
 static LIST_HEAD(arm_pmus);
 static DEFINE_MUTEX(arm_pmus_lock);
 
@@ -29,7 +33,6 @@ static bool kvm_pmu_counter_is_enabled(struct kvm_pmc *pmc);
 
 bool kvm_supports_guest_pmuv3(void)
 {
-	guard(mutex)(&arm_pmus_lock);
 	return !list_empty(&arm_pmus);
 }
 
@@ -807,7 +810,7 @@ void kvm_host_pmu_init(struct arm_pmu *pmu)
 		return;
 
 	entry->arm_pmu = pmu;
-	list_add_tail(&entry->entry, &arm_pmus);
+	list_add_tail_rcu(&entry->entry, &arm_pmus);
 }
 
 static struct arm_pmu *kvm_pmu_probe_armpmu(void)
@@ -816,7 +819,7 @@ static struct arm_pmu *kvm_pmu_probe_armpmu(void)
 	struct arm_pmu *pmu;
 	int cpu;
 
-	guard(mutex)(&arm_pmus_lock);
+	guard(rcu)();
 
 	/*
 	 * It is safe to use a stale cpu to iterate the list of PMUs so long as
@@ -836,7 +839,7 @@ static struct arm_pmu *kvm_pmu_probe_armpmu(void)
 	 * carried here.
 	 */
 	cpu = raw_smp_processor_id();
-	list_for_each_entry(entry, &arm_pmus, entry) {
+	list_for_each_entry_rcu(entry, &arm_pmus, entry) {
 		pmu = entry->arm_pmu;
 
 		if (cpumask_test_cpu(cpu, &pmu->supported_cpus))
@@ -1088,9 +1091,9 @@ static int kvm_arm_pmu_v3_set_pmu(struct kvm_vcpu *vcpu, int pmu_id)
 	int ret = -ENXIO;
 
 	lockdep_assert_held(&kvm->arch.config_lock);
-	mutex_lock(&arm_pmus_lock);
+	guard(rcu)();
 
-	list_for_each_entry(entry, &arm_pmus, entry) {
+	list_for_each_entry_rcu(entry, &arm_pmus, entry) {
 		arm_pmu = entry->arm_pmu;
 		if (arm_pmu->pmu.type == pmu_id) {
 			if (kvm_vm_has_ran_once(kvm) ||
@@ -1106,7 +1109,6 @@ static int kvm_arm_pmu_v3_set_pmu(struct kvm_vcpu *vcpu, int pmu_id)
 		}
 	}
 
-	mutex_unlock(&arm_pmus_lock);
 	return ret;
 }
 

-- 
2.55.0

[PATCH v9 12/16] KVM: arm64: PMU: Pass the pPMU to kvm_map_pmu_event()

[Akihiko Odaki]

Replace the VM argument with the pPMU used for event creation. The
current caller still passes the VM's default pPMU, but this removes the
implicit lookup from kvm_map_pmu_event() so later changes can map events
against the pPMU selected for an individual vCPU.

Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/pmu-emul.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index c5c505b6fc61..27db8877f96e 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -670,10 +670,8 @@ static bool kvm_pmc_counts_at_el2(struct kvm_pmc *pmc)
 	return kvm_pmc_read_evtreg(pmc) & ARMV8_PMU_INCLUDE_EL2;
 }
 
-static int kvm_map_pmu_event(struct kvm *kvm, unsigned int eventsel)
+static int kvm_map_pmu_event(struct arm_pmu *pmu, unsigned int eventsel)
 {
-	struct arm_pmu *pmu = kvm->arch.arm_pmu;
-
 	/*
 	 * The CPU PMU likely isn't PMUv3; let the driver provide a mapping
 	 * for the guest's PMUv3 event ID.
@@ -725,7 +723,7 @@ static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
 	 * Don't create an event if we're running on hardware that requires
 	 * PMUv3 event translation and we couldn't find a valid mapping.
 	 */
-	eventsel = kvm_map_pmu_event(vcpu->kvm, eventsel);
+	eventsel = kvm_map_pmu_event(vcpu->kvm->arch.arm_pmu, eventsel);
 	if (eventsel < 0)
 		return;
 

-- 
2.55.0

[PATCH v9 13/16] KVM: arm64: PMU: Pass the target CPU to kvm_pmu_probe_armpmu()

[Akihiko Odaki]

kvm_pmu_probe_armpmu() currently samples the current CPU internally,
which ties the helper to default PMU selection.

Move that policy to kvm_arm_set_default_pmu() by passing
raw_smp_processor_id() from the caller, and make the helper search for
the pPMU covering an explicit CPU. Move the helper above
kvm_pmu_create_perf_event() so later code can reuse it when creating
PMU events for a VCPU's current pCPU.

This preserves the existing default PMU selection behavior while
preparing fixed-counters-only mode to select a pPMU at runtime.

Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/kvm/pmu-emul.c | 72 +++++++++++++++++++++++------------------------
 1 file changed, 35 insertions(+), 37 deletions(-)

diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index 27db8877f96e..78611eaecca2 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -682,6 +682,23 @@ static int kvm_map_pmu_event(struct arm_pmu *pmu, unsigned int eventsel)
 	return eventsel;
 }
 
+static struct arm_pmu *kvm_pmu_probe_armpmu(int cpu)
+{
+	struct arm_pmu_entry *entry;
+	struct arm_pmu *pmu;
+
+	guard(rcu)();
+
+	list_for_each_entry_rcu(entry, &arm_pmus, entry) {
+		pmu = entry->arm_pmu;
+
+		if (cpumask_test_cpu(cpu, &pmu->supported_cpus))
+			return pmu;
+	}
+
+	return NULL;
+}
+
 /**
  * kvm_pmu_create_perf_event - create a perf event for a counter
  * @pmc: Counter context
@@ -811,42 +828,6 @@ void kvm_host_pmu_init(struct arm_pmu *pmu)
 	list_add_tail_rcu(&entry->entry, &arm_pmus);
 }
 
-static struct arm_pmu *kvm_pmu_probe_armpmu(void)
-{
-	struct arm_pmu_entry *entry;
-	struct arm_pmu *pmu;
-	int cpu;
-
-	guard(rcu)();
-
-	/*
-	 * It is safe to use a stale cpu to iterate the list of PMUs so long as
-	 * the same value is used for the entirety of the loop. Given this, and
-	 * the fact that no percpu data is used for the lookup there is no need
-	 * to disable preemption.
-	 *
-	 * It is still necessary to get a valid cpu, though, to probe for the
-	 * default PMU instance as userspace is not required to specify a PMU
-	 * type. In order to uphold the preexisting behavior KVM selects the
-	 * PMU instance for the core during vcpu init. A dependent use
-	 * case would be a user with disdain of all things big.LITTLE that
-	 * affines the VMM to a particular cluster of cores.
-	 *
-	 * In any case, userspace should just do the sane thing and use the UAPI
-	 * to select a PMU type directly. But, be wary of the baggage being
-	 * carried here.
-	 */
-	cpu = raw_smp_processor_id();
-	list_for_each_entry_rcu(entry, &arm_pmus, entry) {
-		pmu = entry->arm_pmu;
-
-		if (cpumask_test_cpu(cpu, &pmu->supported_cpus))
-			return pmu;
-	}
-
-	return NULL;
-}
-
 static u64 __compute_pmceid(struct arm_pmu *pmu, bool pmceid1)
 {
 	u32 hi[2], lo[2];
@@ -1072,7 +1053,24 @@ static void kvm_arm_set_pmu(struct kvm *kvm, struct arm_pmu *arm_pmu)
  */
 int kvm_arm_set_default_pmu(struct kvm *kvm)
 {
-	struct arm_pmu *arm_pmu = kvm_pmu_probe_armpmu();
+	/*
+	 * It is safe to use a stale cpu to iterate the list of PMUs so long as
+	 * the same value is used for the entirety of the loop. Given this, and
+	 * the fact that no percpu data is used for the lookup there is no need
+	 * to disable preemption.
+	 *
+	 * It is still necessary to get a valid cpu, though, to probe for the
+	 * default PMU instance as userspace is not required to specify a PMU
+	 * type. In order to uphold the preexisting behavior KVM selects the
+	 * PMU instance for the core during vcpu init. A dependent use
+	 * case would be a user with disdain of all things big.LITTLE that
+	 * affines the VMM to a particular cluster of cores.
+	 *
+	 * In any case, userspace should just do the sane thing and use the UAPI
+	 * to select a PMU type directly. But, be wary of the baggage being
+	 * carried here.
+	 */
+	struct arm_pmu *arm_pmu = kvm_pmu_probe_armpmu(raw_smp_processor_id());
 
 	if (!arm_pmu)
 		return -ENODEV;

-- 
2.55.0

[PATCH v9 14/16] KVM: arm64: PMU: Implement fixed-counters-only emulation

[Akihiko Odaki]

Add internal state for PMUv3 emulation without programmable event
counters. When fixed-counters-only mode is active, KVM reports no
programmable counters and hides PMCEID, avoiding event-counter state
whose behavior can depend on the selected hardware PMU.

The cycle counter still uses a host perf event. Unlike the normal PMU
path, fixed-counters-only mode may create that event from the hardware
PMU attached to the VCPU's current pCPU. If the VCPU later loads on a
pCPU that is not covered by the existing event's PMU, request a PMU
reload so the cycle counter can be recreated against the new pCPU's PMU.
Keep this affinity check limited to fixed-counters-only VMs; the normal
programmable-counter mode continues to use the VM-wide PMU and does not
need per-load reload decisions.

Registered pPMUs must cover every CPU on which fixed-counters-only
emulation runs. On ACPI systems, a CPU brought online after PMU probing
may have an unregistered PMU implementation. Full coverage is also not
guaranteed when ARM64_WORKAROUND_PMUV3_IMPDEF_TRAPS is present. Treat
uncovered CPUs as unsupported and add TAINT_CPU_OUT_OF_SPEC.

Add a separate internal flag for explicit userspace PMU selection. The
UAPI wiring added later will use it to keep explicit PMU selection and
fixed-counters-only mode mutually exclusive while still allowing
fixed-counters-only mode to replace the default PMU selected during
KVM_ARM_VCPU_INIT.

The UAPI wiring that sets the fixed-counters-only flag and records
explicit PMU selection is added later in the series.

Assisted-by: Codex:gpt-5.5
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 arch/arm64/include/asm/kvm_host.h |  4 +++
 arch/arm64/kvm/arm.c              |  2 ++
 arch/arm64/kvm/pmu-emul.c         | 67 ++++++++++++++++++++++++++++++++++-----
 include/kvm/arm_pmu.h             |  2 ++
 4 files changed, 67 insertions(+), 8 deletions(-)

diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index bae2c4f92ef5..3998e66c6571 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -367,6 +367,10 @@ struct kvm_arch {
 #define KVM_ARCH_FLAG_WRITABLE_IMP_ID_REGS		10
 	/* Unhandled SEAs are taken to userspace */
 #define KVM_ARCH_FLAG_EXIT_SEA				11
+	/* PMUv3 is emulated with an explicitly specified hardware PMU */
+#define KVM_ARCH_FLAG_PMU_V3_EXPLICIT			12
+	/* PMUv3 is emulated without programmable event counters */
+#define KVM_ARCH_FLAG_PMU_V3_FIXED_COUNTERS_ONLY	13
 	unsigned long flags;
 
 	/* VM-wide vCPU feature set */
diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index 72ce79ce4d5e..a866beb8d873 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -651,6 +651,7 @@ static bool kvm_vcpu_should_clear_twe(struct kvm_vcpu *vcpu)
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	struct kvm_s2_mmu *mmu;
+	int last_cpu = vcpu->cpu;
 	int *last_ran;
 
 	if (is_protected_kvm_enabled())
@@ -700,6 +701,7 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	if (has_vhe())
 		kvm_vcpu_load_vhe(vcpu);
 	kvm_arch_vcpu_load_fp(vcpu);
+	kvm_vcpu_load_pmu(vcpu, last_cpu);
 	kvm_vcpu_pmu_restore_guest(vcpu);
 	if (kvm_arm_is_pvtime_enabled(&vcpu->arch))
 		kvm_make_request(KVM_REQ_RECORD_STEAL, vcpu);
diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index 78611eaecca2..1ec09ef28467 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -86,6 +86,11 @@ u64 kvm_pmu_evtyper_mask(struct kvm *kvm)
 	return mask;
 }
 
+static bool kvm_pmu_fixed_counters_only(struct kvm *kvm)
+{
+	return test_bit(KVM_ARCH_FLAG_PMU_V3_FIXED_COUNTERS_ONLY, &kvm->arch.flags);
+}
+
 /**
  * kvm_pmc_is_64bit - determine if counter is 64bit
  * @pmc: counter context
@@ -333,7 +338,12 @@ u64 kvm_pmu_implemented_counter_mask(struct kvm_vcpu *vcpu)
 
 static void kvm_pmc_enable_perf_event(struct kvm_pmc *pmc)
 {
-	if (!pmc->perf_event) {
+	struct kvm_vcpu *vcpu = kvm_pmc_to_vcpu(pmc);
+
+	if (!pmc->perf_event ||
+	    (kvm_pmu_fixed_counters_only(vcpu->kvm) &&
+	     !cpumask_test_cpu(READ_ONCE(vcpu->cpu),
+			       &to_arm_pmu(pmc->perf_event->pmu)->supported_cpus))) {
 		kvm_pmu_create_perf_event(pmc);
 		return;
 	}
@@ -699,14 +709,10 @@ static struct arm_pmu *kvm_pmu_probe_armpmu(int cpu)
 	return NULL;
 }
 
-/**
- * kvm_pmu_create_perf_event - create a perf event for a counter
- * @pmc: Counter context
- */
-static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
+static void kvm_pmu_create_perf_event_with_pmu(struct kvm_pmc *pmc,
+					       struct arm_pmu *arm_pmu)
 {
 	struct kvm_vcpu *vcpu = kvm_pmc_to_vcpu(pmc);
-	struct arm_pmu *arm_pmu = vcpu->kvm->arch.arm_pmu;
 	struct perf_event *event;
 	struct perf_event_attr attr;
 	int eventsel;
@@ -740,7 +746,7 @@ static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
 	 * Don't create an event if we're running on hardware that requires
 	 * PMUv3 event translation and we couldn't find a valid mapping.
 	 */
-	eventsel = kvm_map_pmu_event(vcpu->kvm->arch.arm_pmu, eventsel);
+	eventsel = kvm_map_pmu_event(arm_pmu, eventsel);
 	if (eventsel < 0)
 		return;
 
@@ -785,6 +791,32 @@ static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
 	pmc->perf_event = event;
 }
 
+/**
+ * kvm_pmu_create_perf_event - create a perf event for a counter
+ * @pmc: Counter context
+ */
+static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
+{
+	struct kvm_vcpu *vcpu = kvm_pmc_to_vcpu(pmc);
+	struct arm_pmu *arm_pmu = vcpu->kvm->arch.arm_pmu;
+
+	if (kvm_pmu_fixed_counters_only(vcpu->kvm)) {
+		do {
+			arm_pmu = kvm_pmu_probe_armpmu(READ_ONCE(vcpu->cpu));
+
+			if (!arm_pmu) {
+				pr_warn_once("kvm: Unsupported PMU variation detected.\n");
+				add_taint(TAINT_CPU_OUT_OF_SPEC, LOCKDEP_STILL_OK);
+				return;
+			}
+
+			kvm_pmu_create_perf_event_with_pmu(pmc, arm_pmu);
+		} while (!cpumask_test_cpu(READ_ONCE(vcpu->cpu), &arm_pmu->supported_cpus));
+	} else {
+		kvm_pmu_create_perf_event_with_pmu(pmc, arm_pmu);
+	}
+}
+
 /**
  * kvm_pmu_set_counter_event_type - set selected counter to monitor some event
  * @vcpu: The vcpu pointer
@@ -870,6 +902,13 @@ u64 kvm_pmu_get_pmceid(struct kvm_vcpu *vcpu, bool pmceid1)
 	u64 val, mask = 0;
 	int base, i, nr_events;
 
+	/*
+	 * Hide the hardware PMU's event set to keep PMCEID stable across
+	 * physical CPU migration.
+	 */
+	if (kvm_pmu_fixed_counters_only(vcpu->kvm))
+		return 0;
+
 	if (!pmceid1) {
 		val = compute_pmceid0(cpu_pmu);
 		base = 0;
@@ -897,6 +936,15 @@ u64 kvm_pmu_get_pmceid(struct kvm_vcpu *vcpu, bool pmceid1)
 	return val & mask;
 }
 
+void kvm_vcpu_load_pmu(struct kvm_vcpu *vcpu, int last_cpu)
+{
+	if (!kvm_pmu_fixed_counters_only(vcpu->kvm) || vcpu->cpu == last_cpu || last_cpu == -1)
+		return;
+
+	if (kvm_pmu_probe_armpmu(vcpu->cpu) != kvm_pmu_probe_armpmu(last_cpu))
+		kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu);
+}
+
 void kvm_vcpu_reload_pmu(struct kvm_vcpu *vcpu)
 {
 	struct kvm_pmu *pmu = &vcpu->arch.pmu;
@@ -1017,6 +1065,9 @@ u8 kvm_arm_pmu_get_max_counters(struct kvm *kvm)
 {
 	struct arm_pmu *arm_pmu = kvm->arch.arm_pmu;
 
+	if (kvm_pmu_fixed_counters_only(kvm))
+		return 0;
+
 	/*
 	 * PMUv3 requires that all event counters are capable of counting any
 	 * event, though the same may not be true of non-PMUv3 hardware.
diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h
index 1971f8794c9d..0c31b72e3435 100644
--- a/include/kvm/arm_pmu.h
+++ b/include/kvm/arm_pmu.h
@@ -60,6 +60,7 @@ void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val);
 void kvm_pmu_handle_mdcr(struct kvm_vcpu *vcpu, u64 old, u64 val);
 void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data,
 				    u64 select_idx);
+void kvm_vcpu_load_pmu(struct kvm_vcpu *vcpu, int last_cpu);
 void kvm_vcpu_reload_pmu(struct kvm_vcpu *vcpu);
 int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu,
 			    struct kvm_device_attr *attr);
@@ -166,6 +167,7 @@ static inline u64 kvm_pmu_get_pmceid(struct kvm_vcpu *vcpu, bool pmceid1)
 static inline void kvm_pmu_update_vcpu_events(struct kvm_vcpu *vcpu) {}
 static inline void kvm_vcpu_pmu_restore_guest(struct kvm_vcpu *vcpu) {}
 static inline void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu) {}
+static inline void kvm_vcpu_load_pmu(struct kvm_vcpu *vcpu, int last_cpu) {}
 static inline void kvm_vcpu_reload_pmu(struct kvm_vcpu *vcpu) {}
 static inline u8 kvm_arm_pmu_get_pmuver_limit(void)
 {

-- 
2.55.0

[PATCH v9 15/16] KVM: arm64: PMU: Introduce FIXED_COUNTERS_ONLY

[Akihiko Odaki]

Introduce the KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY attribute to
create a "fixed-counters-only" PMU.

Much like KVM_ARM_VCPU_PMU_V3_IRQ and other read-write attributes, this
attribute provides a getter that facilitates kernel and userspace
debugging/testing.

Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 Documentation/virt/kvm/devices/vcpu.rst | 30 ++++++++++++++++++++++++++----
 arch/arm64/include/uapi/asm/kvm.h       |  1 +
 arch/arm64/kvm/pmu-emul.c               | 30 +++++++++++++++++++++++++++---
 tools/arch/arm64/include/uapi/asm/kvm.h |  1 +
 4 files changed, 55 insertions(+), 7 deletions(-)

diff --git a/Documentation/virt/kvm/devices/vcpu.rst b/Documentation/virt/kvm/devices/vcpu.rst
index 66e714f2fcfa..1d592e70bdea 100644
--- a/Documentation/virt/kvm/devices/vcpu.rst
+++ b/Documentation/virt/kvm/devices/vcpu.rst
@@ -74,7 +74,8 @@ irqchip.
 	 -ENODEV  PMUv3 not supported or GIC not initialized
 	 -ENXIO   PMUv3 not properly configured or in-kernel irqchip not
 	 	  configured as required prior to calling this attribute
-	 -EBUSY   PMUv3 already initialized or a VCPU has already run
+	 -EBUSY   PMUv3 already initialized, a VCPU has already run or
+		  FIXED_COUNTERS_ONLY has already been set
 	 -EINVAL  Invalid filter range
 	 =======  ======================================================
 
@@ -116,14 +117,14 @@ using event 0x11 (CPU_CYCLES).
 
 :Returns:
 
-	 =======  ====================================================
+	 =======  ===========================================================
 	 -EBUSY   PMUv3 already initialized, a VCPU has already run or
-                  an event filter has already been set
+                  an event filter or FIXED_COUNTERS_ONLY has already been set
 	 -EFAULT  Error accessing the PMU identifier
 	 -ENXIO   PMU not found
 	 -ENODEV  PMUv3 not supported or GIC not initialized
 	 -ENOMEM  Could not allocate memory
-	 =======  ====================================================
+	 =======  ===========================================================
 
 Request that the VCPU uses the specified hardware PMU when creating guest events
 for the purpose of PMU emulation. The PMU identifier can be read from the "type"
@@ -165,6 +166,27 @@ explicitly selected, or the number of counters is out of range for the
 selected PMU. Selecting a new PMU cancels the effect of setting this
 attribute.
 
+1.6 ATTRIBUTE: KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY
+------------------------------------------------------
+
+:Parameters: no additional parameter in kvm_device_attr.addr
+
+:Returns:
+
+	 =======  ==================================================
+	 -EBUSY   PMUv3 already initialized, a VCPU has already run,
+		  an event filter has already been set or
+		  a hardware PMU has already been specified
+	 -ENXIO   Attempted to get before setting
+	 -ENODEV  Attempted to set while PMUv3 not supported
+	 =======  ==================================================
+
+If set, KVM emulates PMUv3 without programmable event counters.
+
+When this attribute is enabled, the vCPU can run on any physical CPU
+that has a PMU, regardless of the underlying implementation. This
+attribute is VM-scoped.
+
 2. GROUP: KVM_ARM_VCPU_TIMER_CTRL
 =================================
 
diff --git a/arch/arm64/include/uapi/asm/kvm.h b/arch/arm64/include/uapi/asm/kvm.h
index 1c13bfa2d38a..39a1a1e412e6 100644
--- a/arch/arm64/include/uapi/asm/kvm.h
+++ b/arch/arm64/include/uapi/asm/kvm.h
@@ -437,6 +437,7 @@ enum {
 #define   KVM_ARM_VCPU_PMU_V3_FILTER		2
 #define   KVM_ARM_VCPU_PMU_V3_SET_PMU		3
 #define   KVM_ARM_VCPU_PMU_V3_SET_NR_COUNTERS	4
+#define   KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY	5
 #define KVM_ARM_VCPU_TIMER_CTRL		1
 #define   KVM_ARM_VCPU_TIMER_IRQ_VTIMER		0
 #define   KVM_ARM_VCPU_TIMER_IRQ_PTIMER		1
diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index 1ec09ef28467..ff5d5f66a9c0 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -1118,8 +1118,8 @@ int kvm_arm_set_default_pmu(struct kvm *kvm)
 	 * affines the VMM to a particular cluster of cores.
 	 *
 	 * In any case, userspace should just do the sane thing and use the UAPI
-	 * to select a PMU type directly. But, be wary of the baggage being
-	 * carried here.
+	 * to select a PMU type directly, or request fixed-counters-only
+	 * emulation. But, be wary of the baggage being carried here.
 	 */
 	struct arm_pmu *arm_pmu = kvm_pmu_probe_armpmu(raw_smp_processor_id());
 
@@ -1144,11 +1144,13 @@ static int kvm_arm_pmu_v3_set_pmu(struct kvm_vcpu *vcpu, int pmu_id)
 		arm_pmu = entry->arm_pmu;
 		if (arm_pmu->pmu.type == pmu_id) {
 			if (kvm_vm_has_ran_once(kvm) ||
+			    kvm_pmu_fixed_counters_only(kvm) ||
 			    (kvm->arch.pmu_filter && kvm->arch.arm_pmu != arm_pmu)) {
 				ret = -EBUSY;
 				break;
 			}
 
+			set_bit(KVM_ARCH_FLAG_PMU_V3_EXPLICIT, &kvm->arch.flags);
 			kvm_arm_set_pmu(kvm, arm_pmu);
 			cpumask_copy(kvm->arch.supported_cpus, &arm_pmu->supported_cpus);
 			ret = 0;
@@ -1159,6 +1161,22 @@ static int kvm_arm_pmu_v3_set_pmu(struct kvm_vcpu *vcpu, int pmu_id)
 	return ret;
 }
 
+static int kvm_arm_pmu_v3_set_pmu_fixed_counters_only(struct kvm_vcpu *vcpu)
+{
+	struct kvm *kvm = vcpu->kvm;
+
+	lockdep_assert_held(&kvm->arch.config_lock);
+
+	if (kvm_vm_has_ran_once(kvm) || kvm->arch.pmu_filter ||
+	    test_bit(KVM_ARCH_FLAG_PMU_V3_EXPLICIT, &kvm->arch.flags))
+		return -EBUSY;
+
+	set_bit(KVM_ARCH_FLAG_PMU_V3_FIXED_COUNTERS_ONLY, &kvm->arch.flags);
+	kvm->arch.nr_pmu_counters = 0;
+
+	return 0;
+}
+
 static int kvm_arm_pmu_v3_set_nr_counters(struct kvm_vcpu *vcpu, unsigned int n)
 {
 	struct kvm *kvm = vcpu->kvm;
@@ -1236,7 +1254,7 @@ int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 		     filter.action != KVM_PMU_EVENT_DENY))
 			return -EINVAL;
 
-		if (kvm_vm_has_ran_once(kvm))
+		if (kvm_vm_has_ran_once(kvm) || kvm_pmu_fixed_counters_only(kvm))
 			return -EBUSY;
 
 		if (!kvm->arch.pmu_filter) {
@@ -1281,6 +1299,8 @@ int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 
 		return kvm_arm_pmu_v3_set_nr_counters(vcpu, n);
 	}
+	case KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY:
+		return kvm_arm_pmu_v3_set_pmu_fixed_counters_only(vcpu);
 	case KVM_ARM_VCPU_PMU_V3_INIT:
 		return kvm_arm_pmu_v3_init(vcpu);
 	}
@@ -1307,6 +1327,9 @@ int kvm_arm_pmu_v3_get_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 		irq = vcpu->arch.pmu.irq_num;
 		return put_user(irq, uaddr);
 	}
+	case KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY:
+		if (kvm_pmu_fixed_counters_only(vcpu->kvm))
+			return 0;
 	}
 
 	return -ENXIO;
@@ -1320,6 +1343,7 @@ int kvm_arm_pmu_v3_has_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 	case KVM_ARM_VCPU_PMU_V3_FILTER:
 	case KVM_ARM_VCPU_PMU_V3_SET_PMU:
 	case KVM_ARM_VCPU_PMU_V3_SET_NR_COUNTERS:
+	case KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY:
 		if (kvm_vcpu_has_pmu(vcpu))
 			return 0;
 	}
diff --git a/tools/arch/arm64/include/uapi/asm/kvm.h b/tools/arch/arm64/include/uapi/asm/kvm.h
index 1c13bfa2d38a..39a1a1e412e6 100644
--- a/tools/arch/arm64/include/uapi/asm/kvm.h
+++ b/tools/arch/arm64/include/uapi/asm/kvm.h
@@ -437,6 +437,7 @@ enum {
 #define   KVM_ARM_VCPU_PMU_V3_FILTER		2
 #define   KVM_ARM_VCPU_PMU_V3_SET_PMU		3
 #define   KVM_ARM_VCPU_PMU_V3_SET_NR_COUNTERS	4
+#define   KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY	5
 #define KVM_ARM_VCPU_TIMER_CTRL		1
 #define   KVM_ARM_VCPU_TIMER_IRQ_VTIMER		0
 #define   KVM_ARM_VCPU_TIMER_IRQ_PTIMER		1

-- 
2.55.0

[PATCH v9 16/16] KVM: arm64: selftests: Test PMU_V3_FIXED_COUNTERS_ONLY

[Akihiko Odaki]

Assert the following:
- FIXED_COUNTERS_ONLY is unset at initialization.
- FIXED_COUNTERS_ONLY can be set.
- Setting an event filter when FIXED_COUNTERS_ONLY has already been set
  results in EBUSY.
- Setting FIXED_COUNTERS_ONLY when an event filter has already been set
  results in EBUSY.
- Setting FIXED_COUNTERS_ONLY when a VCPU has already run results in
  EBUSY.

Run the existing PMU and MDCR_EL2 tests in fixed-only mode. Verify that
enabling the mode preserves the userspace MDCR_EL2 value until a
subsequent KVM_ARM_VCPU_INIT resets HPMN to zero.

Assisted-by: Codex:gpt-5.5
Signed-off-by: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
---
 .../selftests/kvm/arm64/vpmu_counter_access.c      | 166 +++++++++++++++++----
 1 file changed, 137 insertions(+), 29 deletions(-)

diff --git a/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c b/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
index 73f04e4c6ca3..b5f8ce8d3451 100644
--- a/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
+++ b/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
@@ -411,12 +411,7 @@ static void create_vpmu_vm(void *guest_code)
 {
 	struct kvm_vcpu_init init;
 	u8 pmuver, ec;
-	u64 dfr0, irq = 23;
-	struct kvm_device_attr irq_attr = {
-		.group = KVM_ARM_VCPU_PMU_V3_CTRL,
-		.attr = KVM_ARM_VCPU_PMU_V3_IRQ,
-		.addr = (u64)&irq,
-	};
+	u64 dfr0;
 
 	/* The test creates the vpmu_vm multiple times. Ensure a clean state */
 	memset(&vpmu_vm, 0, sizeof(vpmu_vm));
@@ -442,8 +437,6 @@ static void create_vpmu_vm(void *guest_code)
 	TEST_ASSERT(pmuver != ID_AA64DFR0_EL1_PMUVer_IMP_DEF &&
 		    pmuver >= ID_AA64DFR0_EL1_PMUVer_IMP,
 		    "Unexpected PMUVER (0x%x) on the vCPU with PMUv3", pmuver);
-
-	vcpu_ioctl(vpmu_vm.vcpu, KVM_SET_DEVICE_ATTR, &irq_attr);
 }
 
 static void destroy_vpmu_vm(void)
@@ -494,13 +487,22 @@ static void set_nr_counters(struct kvm_vcpu *vcpu,
 }
 
 static void test_create_vpmu_vm_with_nr_counters(unsigned int nr_counters,
+						 bool fixed_counters_only,
 						 bool expect_fail)
 {
 	struct kvm_vcpu *vcpu;
+	u64 irq = 23;
 
 	create_vpmu_vm(guest_code);
 	vcpu = vpmu_vm.vcpu;
 
+	if (fixed_counters_only)
+		vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+				     KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY, NULL);
+
+	vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_IRQ, &irq);
+
 	set_nr_counters(vcpu, nr_counters, expect_fail);
 
 	vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL, KVM_ARM_VCPU_PMU_V3_INIT, NULL);
@@ -510,15 +512,15 @@ static void test_create_vpmu_vm_with_nr_counters(unsigned int nr_counters,
  * Create a guest with one vCPU, set the PMCR_EL0.N for the vCPU to @pmcr_n,
  * and run the test.
  */
-static void run_access_test(u64 pmcr_n)
+static void run_access_test(u64 pmcr_n, bool fixed_counters_only)
 {
 	u64 sp;
 	struct kvm_vcpu *vcpu;
 	struct kvm_vcpu_init init;
 
-	pr_debug("Test with pmcr_n %lu\n", pmcr_n);
+	pr_debug("Test with pmcr_n %lu, fixed_counters_only %d\n", pmcr_n, fixed_counters_only);
 
-	test_create_vpmu_vm_with_nr_counters(pmcr_n, false);
+	test_create_vpmu_vm_with_nr_counters(pmcr_n, fixed_counters_only, false);
 	vcpu = vpmu_vm.vcpu;
 
 	/* Save the initial sp to restore them later to run the guest again */
@@ -552,14 +554,14 @@ static struct pmreg_sets validity_check_reg_sets[] = {
  * Create a VM, and check if KVM handles the userspace accesses of
  * the PMU register sets in @validity_check_reg_sets[] correctly.
  */
-static void run_pmregs_validity_test(u64 pmcr_n)
+static void run_pmregs_validity_test(u64 pmcr_n, bool fixed_counters_only)
 {
 	int i;
 	struct kvm_vcpu *vcpu;
 	u64 set_reg_id, clr_reg_id, reg_val;
 	u64 valid_counters_mask, max_counters_mask;
 
-	test_create_vpmu_vm_with_nr_counters(pmcr_n, false);
+	test_create_vpmu_vm_with_nr_counters(pmcr_n, fixed_counters_only, false);
 	vcpu = vpmu_vm.vcpu;
 
 	valid_counters_mask = get_counters_mask(pmcr_n);
@@ -604,13 +606,15 @@ static void run_pmregs_validity_test(u64 pmcr_n)
 	destroy_vpmu_vm();
 }
 
-static void run_mdcr_el2_validity_test(u64 pmcr_n)
+static void run_mdcr_el2_validity_test(u64 pmcr_n, bool fixed_counters_only)
 {
 	struct kvm_vcpu_init init;
 	struct kvm_vcpu *vcpu;
 	u64 expected_mdcr, mdcr;
+	u64 irq = 23;
 
-	pr_debug("MDCR_EL2 test with pmcr_n %lu\n", pmcr_n);
+	pr_debug("MDCR_EL2 test with pmcr_n %lu, fixed_counters_only %d\n",
+		 pmcr_n, fixed_counters_only);
 
 	create_vpmu_vm(guest_code);
 	if (!vm_supports_el2(vpmu_vm.vm)) {
@@ -631,6 +635,16 @@ static void run_mdcr_el2_validity_test(u64 pmcr_n)
 		    "MDCR_EL2 was not properly updated after HPMN write (expected 0x%lx, got 0x%lx)",
 		    expected_mdcr, mdcr);
 
+	if (fixed_counters_only) {
+		vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+				     KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY, NULL);
+
+		mdcr = vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2));
+		TEST_ASSERT(mdcr == expected_mdcr,
+			    "MDCR_EL2 changed after PMU_V3_FIXED_COUNTERS_ONLY (expected 0x%lx, got 0x%lx)",
+			    expected_mdcr, mdcr);
+	}
+
 	set_nr_counters(vcpu, pmcr_n, false);
 
 	mdcr = vcpu_get_reg(vcpu, KVM_ARM64_SYS_REG(SYS_MDCR_EL2));
@@ -638,6 +652,9 @@ static void run_mdcr_el2_validity_test(u64 pmcr_n)
 		    "MDCR_EL2 changed after PMU_V3_SET_NR_COUNTERS (expected 0x%lx, got 0x%lx)",
 		    expected_mdcr, mdcr);
 
+	vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_IRQ, &irq);
+
 	vcpu_device_attr_set(vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
 			     KVM_ARM_VCPU_PMU_V3_INIT, NULL);
 
@@ -672,11 +689,11 @@ static void run_mdcr_el2_validity_test(u64 pmcr_n)
  * the vCPU to @pmcr_n, which is larger than the host value.
  * The attempt should fail as @pmcr_n is too big to set for the vCPU.
  */
-static void run_error_test(u64 pmcr_n)
+static void run_error_test(u64 pmcr_n, bool fixed_counters_only)
 {
 	pr_debug("Error test with pmcr_n %lu (larger than the host)\n", pmcr_n);
 
-	test_create_vpmu_vm_with_nr_counters(pmcr_n, true);
+	test_create_vpmu_vm_with_nr_counters(pmcr_n, fixed_counters_only, true);
 	destroy_vpmu_vm();
 }
 
@@ -737,25 +754,116 @@ static void test_set_nr_counters_after_vcpu_run(void)
 	kvm_vm_free(vm);
 }
 
-int main(void)
+static void test_config(u64 pmcr_n, bool fixed_counters_only)
+{
+	u64 i;
+
+	for (i = 0; i <= pmcr_n; i++) {
+		run_access_test(i, fixed_counters_only);
+		run_pmregs_validity_test(i, fixed_counters_only);
+		run_mdcr_el2_validity_test(i, fixed_counters_only);
+	}
+
+	for (i = pmcr_n + 1; i < ARMV8_PMU_MAX_COUNTERS; i++)
+		run_error_test(i, fixed_counters_only);
+}
+
+static void test_fixed_counters_only(void)
 {
-	u64 i, pmcr_n;
+	struct kvm_pmu_event_filter filter = { .nevents = 0 };
+	struct kvm_vm *vm;
+	struct kvm_vcpu *running_vcpu;
+	struct kvm_vcpu *stopped_vcpu;
+	struct kvm_vcpu_init init;
+	int ret;
+	u64 irq = 23;
+
+	create_vpmu_vm(guest_code);
+	ret = __vcpu_has_device_attr(vpmu_vm.vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+				     KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY);
+	if (ret) {
+		TEST_ASSERT(ret == -1 && errno == ENXIO,
+			    KVM_IOCTL_ERROR(KVM_HAS_DEVICE_ATTR, ret));
+		destroy_vpmu_vm();
+		return;
+	}
+
+	/* Assert that FIXED_COUNTERS_ONLY is unset at initialization. */
+	ret = __vcpu_device_attr_get(vpmu_vm.vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+				     KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY, NULL);
+	TEST_ASSERT(ret == -1 && errno == ENXIO,
+		    KVM_IOCTL_ERROR(KVM_GET_DEVICE_ATTR, ret));
+
+	/* Assert that setting FIXED_COUNTERS_ONLY succeeds. */
+	vcpu_device_attr_set(vpmu_vm.vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY, NULL);
+
+	/* Assert that FIXED_COUNTERS_ONLY is set. */
+	vcpu_device_attr_get(vpmu_vm.vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY, NULL);
+
+	/*
+	 * Setting an event filter when FIXED_COUNTERS_ONLY has already been set
+	 * results in EBUSY.
+	 */
+	ret = __vcpu_device_attr_set(vpmu_vm.vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+				     KVM_ARM_VCPU_PMU_V3_FILTER, &filter);
+	TEST_ASSERT(ret == -1 && errno == EBUSY,
+		    KVM_IOCTL_ERROR(KVM_SET_DEVICE_ATTR, ret));
+
+	destroy_vpmu_vm();
+
+	create_vpmu_vm(guest_code);
+
+	/*
+	 * Assert that setting FIXED_COUNTERS_ONLY when an event filter has
+	 * already been set results in EBUSY.
+	 */
+	vcpu_device_attr_set(vpmu_vm.vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_FILTER, &filter);
+
+	ret = __vcpu_device_attr_set(vpmu_vm.vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+				     KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY, NULL);
+	TEST_ASSERT(ret == -1 && errno == EBUSY,
+		    KVM_IOCTL_ERROR(KVM_SET_DEVICE_ATTR, ret));
+
+	destroy_vpmu_vm();
+
+	/*
+	 * Assert that setting FIXED_COUNTERS_ONLY when a VCPU has already run
+	 * results in EBUSY.
+	 */
+	vm = vm_create(2);
+	vm_ioctl(vm, KVM_ARM_PREFERRED_TARGET, &init);
+	init.features[0] |= (1 << KVM_ARM_VCPU_PMU_V3);
+	running_vcpu = aarch64_vcpu_add(vm, 0, &init, guest_code_done);
+	stopped_vcpu = aarch64_vcpu_add(vm, 1, &init, guest_code_done);
+	kvm_arch_vm_finalize_vcpus(vm);
+	vcpu_device_attr_set(running_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_IRQ, &irq);
+	vcpu_device_attr_set(running_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+			     KVM_ARM_VCPU_PMU_V3_INIT, NULL);
+	vcpu_run(running_vcpu);
+
+	ret = __vcpu_device_attr_set(stopped_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
+				     KVM_ARM_VCPU_PMU_V3_FIXED_COUNTERS_ONLY, NULL);
+	TEST_ASSERT(ret == -1 && errno == EBUSY,
+		    KVM_IOCTL_ERROR(KVM_SET_DEVICE_ATTR, ret));
+
+	kvm_vm_free(vm);
+
+	test_config(0, true);
+}
 
+int main(void)
+{
 	TEST_REQUIRE(kvm_has_cap(KVM_CAP_ARM_PMU_V3));
 	TEST_REQUIRE(kvm_supports_vgic_v3());
 	TEST_REQUIRE(kvm_supports_nr_counters_attr());
 
 	test_set_nr_counters_after_vcpu_run();
-
-	pmcr_n = get_pmcr_n_limit();
-	for (i = 0; i <= pmcr_n; i++) {
-		run_access_test(i);
-		run_pmregs_validity_test(i);
-		run_mdcr_el2_validity_test(i);
-	}
-
-	for (i = pmcr_n + 1; i < ARMV8_PMU_MAX_COUNTERS; i++)
-		run_error_test(i);
+	test_config(get_pmcr_n_limit(), false);
+	test_fixed_counters_only();
 
 	return 0;
 }

-- 
2.55.0

Re: [PATCH v9 05/16] KVM: arm64: PMU: Keep implemented counter mask EL-independent

[sashiko-bot]

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

Pre-existing issues:
- [High] State loss of EL2-reserved PMU counters during live migration.
--

--- Patch [5]: [PATCH v9 05/16] KVM: arm64: PMU: Keep implemented counter mask EL-independent ---
commit 0e1ed9d5217011f901071c9f1da4dcee65204dd7
Author: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>

KVM: arm64: PMU: Keep implemented counter mask EL-independent
    
This commit changes kvm_pmu_implemented_counter_mask() to build the
implemented counter mask from nr_pmu_counters instead of PMCR_EL0.N. This
prevents EL2 counter state from being lost during PMU reloads when the
saved context is outside EL2.

> diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
> index b4b877f31097f..5281d37634a06 100644
> --- a/arch/arm64/kvm/pmu-emul.c
> +++ b/arch/arm64/kvm/pmu-emul.c
> @@ -317,12 +317,12 @@ u64 kvm_pmu_accessible_counter_mask(struct kvm_vcpu *vcpu)
>  
>  u64 kvm_pmu_implemented_counter_mask(struct kvm_vcpu *vcpu)
>  {
> -	u64 val = FIELD_GET(ARMV8_PMU_PMCR_N, kvm_vcpu_read_pmcr(vcpu));
> +	u64 n = vcpu->kvm->arch.nr_pmu_counters;
>  
> -	if (val == 0)
> +	if (n == 0)
>  		return BIT(ARMV8_PMU_CYCLE_IDX);
>  	else
> -		return GENMASK(val - 1, 0) | BIT(ARMV8_PMU_CYCLE_IDX);
> +		return GENMASK(n - 1, 0) | BIT(ARMV8_PMU_CYCLE_IDX);
>  }

[Severity: High]
This is a pre-existing issue, but does restricting guest access via
kvm_pmu_accessible_counter_mask() also inadvertently drop EL2 state during
live migration?

Looking at get_pmreg() in arch/arm64/kvm/sys_regs.c, when userspace invokes
KVM_GET_ONE_REG to save PMU registers like PMCNTENSET_EL0:

arch/arm64/kvm/sys_regs.c:get_pmreg() {
    ...
    u64 mask = kvm_pmu_accessible_counter_mask(vcpu);

    *val = __vcpu_sys_reg(vcpu, r->reg) & mask;
    ...
}

And similarly when restoring in set_pmreg():

arch/arm64/kvm/sys_regs.c:set_pmreg() {
    ...
    u64 mask = kvm_pmu_accessible_counter_mask(vcpu);

    __vcpu_assign_sys_reg(vcpu, r->reg, val & mask);
    ...
}

If a nested VM is migrated while its vCPU is executing in EL1/EL0,
kvm_pmu_accessible_counter_mask() strips out the EL2 counters because
vcpu_is_el2() is false. Could this cause KVM to silently discard the
EL2-reserved counter state on the source and fail to restore it on the
target?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260720-hybrid-v9-0-2b713ca1b5dc@rsg.ci.i.u-tokyo.ac.jp?part=5

Re: [PATCH v9 06/16] KVM: arm64: PMU: Recreate events after MDCR_EL2 changes

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

Pre-existing issues:
- [Medium] Modifying PMCR_EL0.LC or PMCR_EL0.LP does not recreate perf events, leading to stale sample periods for active counters.
- [Medium] Userspace modification of PMEVTYPERn_EL0 or PMCCFILTR_EL0 via KVM_SET_ONE_REG does not trigger a PMU reload or recreate the backing perf_event.
--

--- Patch [6]: [PATCH v9 06/16] KVM: arm64: PMU: Recreate events after MDCR_EL2 changes ---
commit 55bb4c8121d4871eb781b7b2feafd0970dbee364
Author: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>

KVM: arm64: PMU: Recreate events after MDCR_EL2 changes

MDCR_EL2.HPMN changes which counters are reserved for EL2 and thus which
enable control, event filter, and overflow width apply. Route guest writes,
userspace writes, and reset through a common helper to mark the vCPU's
perf events for recreation when HPMN, HPMD, or HLP changes.

> diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
> --- a/arch/arm64/kvm/pmu-emul.c
> +++ b/arch/arm64/kvm/pmu-emul.c

[ ... ]

> @@ -606,6 +609,21 @@ void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val)
>  	}
>  }
>  
> +void kvm_pmu_handle_mdcr(struct kvm_vcpu *vcpu, u64 old, u64 val)
> +{
> +	u64 changed = old ^ val;
> +
> +	/*
> +	 * HPMN determines which counters HPMD and HLP apply to. Changes to
> +	 * these fields require new perf event filters and sample periods.
> +	 */
> +	if (changed & MDCR_EL2_PMU_RECREATE_MASK)
> +		vcpu->arch.pmu.events_need_recreate = true;
> +
> +	if (changed & (MDCR_EL2_HPME | MDCR_EL2_PMU_RECREATE_MASK))
> +		kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu);
> +}

[Severity: Medium]
This is a pre-existing issue, but modifying PMCR_EL0.LC or PMCR_EL0.LP
doesn't seem to recreate perf events.

Guest writes to PMCR_EL0 in kvm_pmu_handle_pmcr() only trigger a PMU reload
if the E bit changes, and never mark events for recreation when LC or LP
change. This also applies to userspace KVM_SET_ONE_REG writes to PMCR_EL0 via
set_pmcr(), which request a reload but fail to set
events_need_recreate = true.

Because PMCR_EL0.LC and LP determine whether the PMU counters overflow at 32
or 64 bits, does this leave the events using stale sample periods, leading
to incorrect performance statistics and mistimed overflow interrupts?

> diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
> --- a/arch/arm64/kvm/sys_regs.c
> +++ b/arch/arm64/kvm/sys_regs.c

[ ... ]

> @@ -3741,7 +3751,8 @@ static const struct sys_reg_desc sys_reg_descs[] = {
>  	EL2_REG_FILTERED(SCTLR2_EL2, access_vm_reg, reset_val, 0,
>  			 sctlr2_el2_visibility),
>  	EL2_REG_VNCR(HCR_EL2, reset_hcr, 0),
> -	EL2_REG(MDCR_EL2, access_mdcr, reset_mdcr, 0),
> +	SYS_REG_USER_FILTER(MDCR_EL2, access_mdcr, reset_mdcr, 0,
> +			    NULL, set_mdcr, el2_visibility),

[Severity: Medium]
This is also a pre-existing issue, but userspace modifications of
PMEVTYPERn_EL0 or PMCCFILTR_EL0 via KVM_SET_ONE_REG don't appear to trigger a
PMU reload or recreate the backing perf_event.

Because the macros defining PMEVTYPERn_EL0 and PMCCFILTR_EL0 in
sys_reg_descs lack a .set_user accessor, kvm_sys_reg_set_user() falls back to
__vcpu_assign_sys_reg() without requesting a PMU reload or event recreation.

If userspace modifies a vCPU's PMU event type registers after the vCPU has
already run, does the PMU continue counting the old events, leading to
incorrect statistics?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260720-hybrid-v9-0-2b713ca1b5dc@rsg.ci.i.u-tokyo.ac.jp?part=6

Re: [PATCH v9 02/16] KVM: arm64: PMU: Stop updating MDCR_EL2.HPMN

[sashiko-bot]

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

Pre-existing issues:
- [High] The KVM_ARM_VCPU_PMU_V3_SET_NR_COUNTERS attribute handler allows post-run reconfiguration of VM-wide PMU state, violating finalization requirements.
--

--- Patch [2]: [PATCH v9 02/16] KVM: arm64: PMU: Stop updating MDCR_EL2.HPMN ---
commit 3727863e040affba175d6ddaaa4d17f6dbc0e769
Author: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>

KVM: arm64: PMU: Stop updating MDCR_EL2.HPMN

This commit stops KVM from rewriting MDCR_EL2.HPMN for every vCPU
whenever userspace changes the PMU or its counter count. Instead, it
updates nr_pmu_counters directly and relies on reset_mdcr() to initialize
HPMN at VCPU init time.

Closes: https://sashiko.dev/#/patchset/20260706-hybrid-v8-0-de459617b59d%40rsg.ci.i.u-tokyo.ac.jp?part=6

> diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
> index 98305bbfc095a..b5df6843dbcd5 100644
> --- a/arch/arm64/kvm/pmu-emul.c
> +++ b/arch/arm64/kvm/pmu-emul.c

[ ... ]

> @@ -1111,7 +1093,7 @@ static int kvm_arm_pmu_v3_set_nr_counters(struct kvm_vcpu *vcpu, unsigned int n)
>  	if (n > kvm_arm_pmu_get_max_counters(kvm))
>  		return -EINVAL;
>  
> -	kvm_arm_set_nr_counters(kvm, n);
> +	kvm->arch.nr_pmu_counters = n;

[Severity: High]
This is a pre-existing issue, but does this assignment allow post-run
reconfiguration of VM-wide PMU state?

The caller kvm_arm_pmu_v3_set_attr() gates this update using only the
local VCPU state:

	if (vcpu->arch.pmu.created)
		return -EBUSY;

If VCPU 0 has already run, and userspace configures a new VCPU 1 (where
vcpu->arch.pmu.created is false), couldn't a call to KVM_SET_DEVICE_ATTR
reduce the number of counters and overwrite the VM-wide
kvm->arch.nr_pmu_counters here?

When the running VCPU 0 accesses a previously valid counter, KVM's
pmu_counter_idx_valid() would read the newly reduced nr_pmu_counters and
could inject an Undefined Exception. Does this need to be gated by
checking if any VCPU has run?

>  	return 0;
>  }

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260720-hybrid-v9-0-2b713ca1b5dc@rsg.ci.i.u-tokyo.ac.jp?part=2

Re: [PATCH v9 09/16] arm64: errata: Require Apple IMPDEF PMUv3 traps on all CPUs

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

New issues:
- [Critical] Permitting late CPUs with IMPDEF PMUv3 to boot without enabling hypervisor traps allows untrapped guest access to host PMU registers.

Pre-existing issues:
- [High] Writes to control-plane system registers in CPU capability enable callbacks lack a mandatory `isb()` instruction.
--

--- Patch [9]: [PATCH v9 09/16] arm64: errata: Require Apple IMPDEF PMUv3 traps on all CPUs ---
commit 3e98f085c1f1b418fb0ec6e336b981ba1df07b03
Author: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>

arm64: errata: Require Apple IMPDEF PMUv3 traps on all CPUs

This commit changes the ARM64_WORKAROUND_PMUV3_IMPDEF_TRAPS erratum to an
all-early-CPU capability. It ensures the capability is only advertised if all
early CPUs match, aiming to reject late CPUs that miss the capability while
permitting late CPUs to have it when the system does not.

> diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
> index 1995e1198648e..a5ae628204a12 100644
> --- a/arch/arm64/kernel/cpu_errata.c
> +++ b/arch/arm64/kernel/cpu_errata.c

[ ... ]

> @@ -1005,7 +1005,7 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
>  	{
>  		.desc = "Apple IMPDEF PMUv3 Traps",
>  		.capability = ARM64_WORKAROUND_PMUV3_IMPDEF_TRAPS,
> -		.type = ARM64_CPUCAP_LOCAL_CPU_ERRATUM,
> +		.type = ARM64_CPUCAP_EARLY_LOCAL_CPU_FEATURE,

[Severity: Critical]
By changing the capability type to ARM64_CPUCAP_EARLY_LOCAL_CPU_FEATURE,
does this allow a late CPU with IMPDEF PMUv3 to boot without enabling the
hypervisor trap if the system lacks the capability system-wide?

If an asymmetric system boots without the PMUv3 IMPDEF capability system-wide
and a late Apple CPU with the feature is onlined, verify_local_cpu_caps()
will permit the CPU. However, because system_has_cap is false, it skips
caps->cpu_enable() which leaves HACR_EL2 bit 56 clear.

Since KVM relies on this hardware trap to intercept these instructions and
inject an UNDEF exception, could this missing trap allow a malicious guest
scheduled on the late CPU to execute IMPDEF PMU instructions natively and
manipulate host physical PMU state?

>  		.matches = has_impdef_pmuv3,
>  		.cpu_enable = cpu_enable_impdef_pmuv3_traps,

[Severity: High]
This is a pre-existing issue, but do the writes to control-plane system
registers in capability enable callbacks require an isb() synchronization
barrier?

Currently, callbacks like cpu_enable_impdef_pmuv3_traps update SYS_HACR_EL2
using sysreg_clear_set_s, and then return immediately without a trailing
isb(). Other callbacks like cpu_enable_trap_ctr_access and
cpu_enable_cache_maint_trap update sctlr_el1 in a similar way.

Under the ARM64 subsystem rules, every write to a control-plane system
register must be followed by an isb() as the very next instruction. Since
sysreg_clear_set_s does not issue this barrier, returning immediately places
the return branch instruction against an architecturally undefined pipeline
state. Should these callbacks issue an explicit isb() before returning?

>  	},

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260720-hybrid-v9-0-2b713ca1b5dc@rsg.ci.i.u-tokyo.ac.jp?part=9

Re: [PATCH v9 12/16] KVM: arm64: PMU: Pass the pPMU to kvm_map_pmu_event()

[sashiko-bot]

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

Pre-existing issues:
- [Critical] Host kernel panic via NULL pointer dereference in PMU event creation when a guest without a PMU toggles `MDCR_EL2.HPME`.
--

--- Patch [12]: [PATCH v9 12/16] KVM: arm64: PMU: Pass the pPMU to kvm_map_pmu_event() ---
commit 61c39cb093b2170b92839ae7097044db9d1474cc
Author: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>

KVM: arm64: PMU: Pass the pPMU to kvm_map_pmu_event()

This patch updates kvm_map_pmu_event() to accept the PMU structure directly
rather than implicitly looking it up via the VM argument. This prepares the
code to map events against a per-vCPU selected pPMU in the future.

> diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
> index c5c505b6fc61c..27db8877f96ea 100644
> --- a/arch/arm64/kvm/pmu-emul.c
> +++ b/arch/arm64/kvm/pmu-emul.c
[ ... ]
> @@ -725,7 +723,7 @@ static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
>  	 * Don't create an event if we're running on hardware that requires
>  	 * PMUv3 event translation and we couldn't find a valid mapping.
>  	 */
> -	eventsel = kvm_map_pmu_event(vcpu->kvm, eventsel);
> +	eventsel = kvm_map_pmu_event(vcpu->kvm->arch.arm_pmu, eventsel);

[Severity: Critical]
This is a pre-existing issue, but this change makes it a bit more explicit.
Could this pass a NULL pointer to kvm_map_pmu_event() and cause a host kernel
panic?

If a guest with nested virtualization enabled but no PMU configured toggles
the MDCR_EL2.HPME bit, access_mdcr() will unconditionally trigger a
KVM_REQ_RELOAD_PMU request without checking if the VM actually has a PMU.

When kvm_vcpu_reload_pmu() handles this request on the next vCPU entry,
it tries to reprogram the cycle counter (which defaults to enabled). This
eventually reaches kvm_pmu_create_perf_event():

arch/arm64/kvm/sys_regs.c:access_mdcr() {
    ...
    kvm_pmu_handle_mdcr(vcpu, old, val); /* Triggers KVM_REQ_RELOAD_PMU */
}

arch/arm64/kvm/pmu-emul.c:kvm_pmu_create_perf_event() {
    ...
    eventsel = kvm_map_pmu_event(vcpu->kvm->arch.arm_pmu, eventsel);
}

arch/arm64/kvm/pmu-emul.c:kvm_map_pmu_event() {
    ...
    if (unlikely(pmu->map_pmuv3_event)) /* NULL pointer dereference */
}

Because no PMU was initialized for this guest, vcpu->kvm->arch.arm_pmu is NULL.

Could access_mdcr() or kvm_pmu_create_perf_event() check if the PMU is
actually initialized before attempting to create events?

>  	if (eventsel < 0)
>  		return;
>

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260720-hybrid-v9-0-2b713ca1b5dc@rsg.ci.i.u-tokyo.ac.jp?part=12

Re: [PATCH v9 04/16] KVM: arm64: selftests: Test SET_NR_COUNTERS after first run

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The test ignores the guest's exit state and `ucall` status after executing `vcpu_run()`.
--

--- Patch [4]: [PATCH v9 04/16] KVM: arm64: selftests: Test SET_NR_COUNTERS after first run ---
commit 0382d3a32df9349f988ca69c7b11c92e61c3d384
Author: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>

KVM: arm64: selftests: Test SET_NR_COUNTERS after first run

This commit adds a regression test that runs one vCPU and then tries to change
the VM-scoped PMU counter count through an idle sibling to verify that
SET_NR_COUNTERS fails with EBUSY once any vCPU has run.

> diff --git a/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c b/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
> index 22223395969e0..3b5bb421f401b 100644
> --- a/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c
> +++ b/tools/testing/selftests/kvm/arm64/vpmu_counter_access.c

[ ... ]

> @@ -622,6 +627,37 @@ static bool kvm_supports_nr_counters_attr(void)
>  	return supported;
>  }
>  
> +static void test_set_nr_counters_after_vcpu_run(void)
> +{
> +	struct kvm_vcpu *running_vcpu, *stopped_vcpu;
> +	unsigned int nr_counters = 0;
> +	struct kvm_vcpu_init init;
> +	struct kvm_vm *vm;
> +	int ret;
> +	u64 irq = 23;
> +
> +	vm = vm_create(2);
> +	vm_ioctl(vm, KVM_ARM_PREFERRED_TARGET, &init);
> +	init.features[0] |= BIT(KVM_ARM_VCPU_PMU_V3);
> +	running_vcpu = aarch64_vcpu_add(vm, 0, &init, guest_code_done);
> +	stopped_vcpu = aarch64_vcpu_add(vm, 1, &init, guest_code_done);
> +	kvm_arch_vm_finalize_vcpus(vm);
> +
> +	vcpu_device_attr_set(running_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
> +			     KVM_ARM_VCPU_PMU_V3_IRQ, &irq);
> +	vcpu_device_attr_set(running_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
> +			     KVM_ARM_VCPU_PMU_V3_INIT, NULL);
> +	vcpu_run(running_vcpu);

[Severity: Medium]
Does this code need to check the guest's exit state and ucall payload?

If the guest payload crashes or hits an unexpected exception, this will silently
ignore the failure and proceed. Since the test only checks for an EBUSY error on
the subsequent ioctl (which will still be returned because the vCPU has been
run), the test will pass despite the guest crash, potentially masking 
virtualization bugs.

> +
> +	ret = __vcpu_device_attr_set(stopped_vcpu, KVM_ARM_VCPU_PMU_V3_CTRL,
> +				     KVM_ARM_VCPU_PMU_V3_SET_NR_COUNTERS,
> +				     &nr_counters);
> +	TEST_ASSERT(ret == -1 && errno == EBUSY,
> +		    KVM_IOCTL_ERROR(KVM_SET_DEVICE_ATTR, ret));
> +
> +	kvm_vm_free(vm);
> +}
> +

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260720-hybrid-v9-0-2b713ca1b5dc@rsg.ci.i.u-tokyo.ac.jp?part=4

Re: [PATCH v9 14/16] KVM: arm64: PMU: Implement fixed-counters-only emulation

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] State desynchronization in PMU virtualization due to retained stale perf_event.
--

--- Patch [14]: [PATCH v9 14/16] KVM: arm64: PMU: Implement fixed-counters-only emulation ---
commit 8eabffaec38175a063281e21d7253117967c6ca4
Author: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>

KVM: arm64: PMU: Implement fixed-counters-only emulation

This patch adds internal state for emulating PMUv3 without programmable
event counters. In fixed-counters-only mode, KVM tracks physical CPU
migrations and requests PMU reloads so the host perf event for the cycle
counter can be recreated on the new physical CPU's PMU. Uncovered CPUs are
treated as unsupported and taint the kernel.

> diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
> index 78611eaecca25..1ec09ef28467a 100644
> --- a/arch/arm64/kvm/pmu-emul.c
> +++ b/arch/arm64/kvm/pmu-emul.c
[ ... ]
> @@ -785,6 +791,32 @@ static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
>  	pmc->perf_event = event;
>  }
>  
> +/**
> + * kvm_pmu_create_perf_event - create a perf event for a counter
> + * @pmc: Counter context
> + */
> +static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
> +{
> +	struct kvm_vcpu *vcpu = kvm_pmc_to_vcpu(pmc);
> +	struct arm_pmu *arm_pmu = vcpu->kvm->arch.arm_pmu;
> +
> +	if (kvm_pmu_fixed_counters_only(vcpu->kvm)) {
> +		do {
> +			arm_pmu = kvm_pmu_probe_armpmu(READ_ONCE(vcpu->cpu));
> +
> +			if (!arm_pmu) {
> +				pr_warn_once("kvm: Unsupported PMU variation detected.\n");
> +				add_taint(TAINT_CPU_OUT_OF_SPEC, LOCKDEP_STILL_OK);
> +				return;

[Severity: Medium]
Does this early return leave a stale perf_event attached to the PMC?

If a vCPU in fixed-counters-only mode migrates to a physical CPU without a
supported PMU, and the guest attempts to reprogram the cycle counter's
filter, kvm_pmu_create_perf_event() will exit here without stopping the old
counter or replacing it.

If the vCPU later migrates back to a supported CPU, kvm_vcpu_load_pmu() will
trigger a KVM_REQ_RELOAD_PMU. When kvm_pmc_enable_perf_event() checks the
PMC, it will find the old perf_event. Since that old event is valid for the
now-current supported CPU, it will be re-enabled without recreating it.

Can this lead to KVM silently discarding the configuration updates the guest
made while on the unsupported CPU, resulting in incorrect performance
monitoring statistics?

> +			}
> +
> +			kvm_pmu_create_perf_event_with_pmu(pmc, arm_pmu);
> +		} while (!cpumask_test_cpu(READ_ONCE(vcpu->cpu), &arm_pmu->supported_cpus));
> +	} else {
> +		kvm_pmu_create_perf_event_with_pmu(pmc, arm_pmu);
> +	}
> +}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260720-hybrid-v9-0-2b713ca1b5dc@rsg.ci.i.u-tokyo.ac.jp?part=14