Merge branch kvm-arm64/pmu-bl into kvmarm-master/next

* kvm-arm64/pmu-bl:
  : .
  : Improve PMU support on heterogeneous systems, courtesy of Alexandru Elisei
  : .
  KVM: arm64: Refuse to run VCPU if the PMU doesn't match the physical CPU
  KVM: arm64: Add KVM_ARM_VCPU_PMU_V3_SET_PMU attribute
  KVM: arm64: Keep a list of probed PMUs
  KVM: arm64: Keep a per-VM pointer to the default PMU
  perf: Fix wrong name in comment for struct perf_cpu_context
  KVM: arm64: Do not change the PMU event filter after a VCPU has run
Signed-off-by: Marc Zyngier <maz@kernel.org>

Documentation/virt/kvm/devices/vcpu.rst

@@ -70,7 +70,7 @@ 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
+	 -EBUSY   PMUv3 already initialized or a VCPU has already run
 	 -EINVAL  Invalid filter range
 	 =======  ======================================================
 
@@ -104,6 +104,38 @@ hardware event. Filtering event 0x1E (CHAIN) has no effect either, as it
 isn't strictly speaking an event. Filtering the cycle counter is possible
 using event 0x11 (CPU_CYCLES).
 
+1.4 ATTRIBUTE: KVM_ARM_VCPU_PMU_V3_SET_PMU
+------------------------------------------
+
+:Parameters: in kvm_device_attr.addr the address to an int representing the PMU
+             identifier.
+
+:Returns:
+
+	 =======  ====================================================
+	 -EBUSY   PMUv3 already initialized, a VCPU has already run or
+                  an event filter 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"
+file for the desired PMU instance under /sys/devices (or, equivalent,
+/sys/bus/even_source). This attribute is particularly useful on heterogeneous
+systems where there are at least two CPU PMUs on the system. The PMU that is set
+for one VCPU will be used by all the other VCPUs. It isn't possible to set a PMU
+if a PMU event filter is already present.
+
+Note that KVM will not make any attempts to run the VCPU on the physical CPUs
+associated with the PMU specified by this attribute. This is entirely left to
+userspace. However, attempting to run the VCPU on a physical CPU not supported
+by the PMU will fail and KVM_RUN will return with
+exit_reason = KVM_EXIT_FAIL_ENTRY and populate the fail_entry struct by setting
+hardare_entry_failure_reason field to KVM_EXIT_FAIL_ENTRY_CPU_UNSUPPORTED and
+the cpu field to the processor id.
 
 2. GROUP: KVM_ARM_VCPU_TIMER_CTRL
 =================================

arch/arm64/include/asm/kvm_host.h

@@ -129,13 +129,16 @@ struct kvm_arch {
 	 * up to 2^10 events (ARMv8.0) or 2^16 events (ARMv8.1+).
 	 */
 	unsigned long *pmu_filter;
-	unsigned int pmuver;
+	struct arm_pmu *arm_pmu;
+
+	cpumask_var_t supported_cpus;
 
 	u8 pfr0_csv2;
 	u8 pfr0_csv3;
 
 	/* Memory Tagging Extension enabled for the guest */
 	bool mte_enabled;
+	bool ran_once;
 };
 
 struct kvm_vcpu_fault_info {
@@ -436,6 +439,7 @@ struct kvm_vcpu_arch {
 #define KVM_ARM64_DEBUG_STATE_SAVE_SPE	(1 << 12) /* Save SPE context if active  */
 #define KVM_ARM64_DEBUG_STATE_SAVE_TRBE	(1 << 13) /* Save TRBE context if active  */
 #define KVM_ARM64_FP_FOREIGN_FPSTATE	(1 << 14)
+#define KVM_ARM64_ON_UNSUPPORTED_CPU	(1 << 15) /* Physical CPU not in supported_cpus */
 
 #define KVM_GUESTDBG_VALID_MASK (KVM_GUESTDBG_ENABLE | \
 				 KVM_GUESTDBG_USE_SW_BP | \
@@ -454,6 +458,15 @@ struct kvm_vcpu_arch {
 #define vcpu_has_ptrauth(vcpu)		false
 #endif
 
+#define vcpu_on_unsupported_cpu(vcpu)					\
+	((vcpu)->arch.flags & KVM_ARM64_ON_UNSUPPORTED_CPU)
+
+#define vcpu_set_on_unsupported_cpu(vcpu)				\
+	((vcpu)->arch.flags |= KVM_ARM64_ON_UNSUPPORTED_CPU)
+
+#define vcpu_clear_on_unsupported_cpu(vcpu)				\
+	((vcpu)->arch.flags &= ~KVM_ARM64_ON_UNSUPPORTED_CPU)
+
 #define vcpu_gp_regs(v)		(&(v)->arch.ctxt.regs)
 
 /*

arch/arm64/include/uapi/asm/kvm.h

@@ -362,6 +362,7 @@ struct kvm_arm_copy_mte_tags {
 #define   KVM_ARM_VCPU_PMU_V3_IRQ	0
 #define   KVM_ARM_VCPU_PMU_V3_INIT	1
 #define   KVM_ARM_VCPU_PMU_V3_FILTER	2
+#define   KVM_ARM_VCPU_PMU_V3_SET_PMU	3
 #define KVM_ARM_VCPU_TIMER_CTRL		1
 #define   KVM_ARM_VCPU_TIMER_IRQ_VTIMER		0
 #define   KVM_ARM_VCPU_TIMER_IRQ_PTIMER		1
@@ -413,6 +414,9 @@ struct kvm_arm_copy_mte_tags {
 #define KVM_PSCI_RET_INVAL		PSCI_RET_INVALID_PARAMS
 #define KVM_PSCI_RET_DENIED		PSCI_RET_DENIED
 
+/* run->fail_entry.hardware_entry_failure_reason codes. */
+#define KVM_EXIT_FAIL_ENTRY_CPU_UNSUPPORTED	(1ULL << 0)
+
 #endif
 
 #endif /* __ARM_KVM_H__ */

arch/arm64/kvm/arm.c

@@ -145,6 +145,10 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	if (ret)
 		goto out_free_stage2_pgd;
 
+	if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL))
+		goto out_free_stage2_pgd;
+	cpumask_copy(kvm->arch.supported_cpus, cpu_possible_mask);
+
 	kvm_vgic_early_init(kvm);
 
 	/* The maximum number of VCPUs is limited by the host's GIC model */
@@ -171,6 +175,7 @@ vm_fault_t kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
 	bitmap_free(kvm->arch.pmu_filter);
+	free_cpumask_var(kvm->arch.supported_cpus);
 
 	kvm_vgic_destroy(kvm);
 
@@ -406,6 +411,9 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	if (vcpu_has_ptrauth(vcpu))
 		vcpu_ptrauth_disable(vcpu);
 	kvm_arch_vcpu_load_debug_state_flags(vcpu);
+
+	if (!cpumask_test_cpu(smp_processor_id(), vcpu->kvm->arch.supported_cpus))
+		vcpu_set_on_unsupported_cpu(vcpu);
 }
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
@@ -419,6 +427,7 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 	kvm_vcpu_pmu_restore_host(vcpu);
 	kvm_arm_vmid_clear_active();
 
+	vcpu_clear_on_unsupported_cpu(vcpu);
 	vcpu->cpu = -1;
 }
 
@@ -549,6 +558,10 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu)
 	if (kvm_vm_is_protected(kvm))
 		kvm_call_hyp_nvhe(__pkvm_vcpu_init_traps, vcpu);
 
+	mutex_lock(&kvm->lock);
+	kvm->arch.ran_once = true;
+	mutex_unlock(&kvm->lock);
+
 	return ret;
 }
 
@@ -707,6 +720,14 @@ static bool kvm_vcpu_exit_request(struct kvm_vcpu *vcpu, int *ret)
 		}
 	}
 
+	if (unlikely(vcpu_on_unsupported_cpu(vcpu))) {
+		run->exit_reason = KVM_EXIT_FAIL_ENTRY;
+		run->fail_entry.hardware_entry_failure_reason = KVM_EXIT_FAIL_ENTRY_CPU_UNSUPPORTED;
+		run->fail_entry.cpu = smp_processor_id();
+		*ret = 0;
+		return true;
+	}
+
 	return kvm_request_pending(vcpu) ||
 			xfer_to_guest_mode_work_pending();
 }

arch/arm64/kvm/pmu-emul.c

@@ -7,6 +7,7 @@
 #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/uaccess.h>
@@ -16,6 +17,9 @@
 
 DEFINE_STATIC_KEY_FALSE(kvm_arm_pmu_available);
 
+static LIST_HEAD(arm_pmus);
+static DEFINE_MUTEX(arm_pmus_lock);
+
 static void kvm_pmu_create_perf_event(struct kvm_vcpu *vcpu, u64 select_idx);
 static void kvm_pmu_update_pmc_chained(struct kvm_vcpu *vcpu, u64 select_idx);
 static void kvm_pmu_stop_counter(struct kvm_vcpu *vcpu, struct kvm_pmc *pmc);
@@ -24,7 +28,11 @@ static void kvm_pmu_stop_counter(struct kvm_vcpu *vcpu, struct kvm_pmc *pmc);
 
 static u32 kvm_pmu_event_mask(struct kvm *kvm)
 {
-	switch (kvm->arch.pmuver) {
+	unsigned int pmuver;
+
+	pmuver = kvm->arch.arm_pmu->pmuver;
+
+	switch (pmuver) {
 	case ID_AA64DFR0_PMUVER_8_0:
 		return GENMASK(9, 0);
 	case ID_AA64DFR0_PMUVER_8_1:
@@ -33,7 +41,7 @@ static u32 kvm_pmu_event_mask(struct kvm *kvm)
 	case ID_AA64DFR0_PMUVER_8_7:
 		return GENMASK(15, 0);
 	default:		/* Shouldn't be here, just for sanity */
-		WARN_ONCE(1, "Unknown PMU version %d\n", kvm->arch.pmuver);
+		WARN_ONCE(1, "Unknown PMU version %d\n", pmuver);
 		return 0;
 	}
 }
@@ -600,6 +608,7 @@ static bool kvm_pmu_counter_is_enabled(struct kvm_vcpu *vcpu, u64 select_idx)
  */
 static void kvm_pmu_create_perf_event(struct kvm_vcpu *vcpu, u64 select_idx)
 {
+	struct arm_pmu *arm_pmu = vcpu->kvm->arch.arm_pmu;
 	struct kvm_pmu *pmu = &vcpu->arch.pmu;
 	struct kvm_pmc *pmc;
 	struct perf_event *event;
@@ -636,7 +645,7 @@ static void kvm_pmu_create_perf_event(struct kvm_vcpu *vcpu, u64 select_idx)
 		return;
 
 	memset(&attr, 0, sizeof(struct perf_event_attr));
-	attr.type = PERF_TYPE_RAW;
+	attr.type = arm_pmu->pmu.type;
 	attr.size = sizeof(attr);
 	attr.pinned = 1;
 	attr.disabled = !kvm_pmu_counter_is_enabled(vcpu, pmc->idx);
@@ -745,17 +754,33 @@ void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data,
 
 void kvm_host_pmu_init(struct arm_pmu *pmu)
 {
-	if (pmu->pmuver != 0 && pmu->pmuver != ID_AA64DFR0_PMUVER_IMP_DEF &&
-	    !kvm_arm_support_pmu_v3() && !is_protected_kvm_enabled())
+	struct arm_pmu_entry *entry;
+
+	if (pmu->pmuver == 0 || pmu->pmuver == ID_AA64DFR0_PMUVER_IMP_DEF ||
+	    is_protected_kvm_enabled())
+		return;
+
+	mutex_lock(&arm_pmus_lock);
+
+	entry = kmalloc(sizeof(*entry), GFP_KERNEL);
+	if (!entry)
+		goto out_unlock;
+
+	entry->arm_pmu = pmu;
+	list_add_tail(&entry->entry, &arm_pmus);
+
+	if (list_is_singular(&arm_pmus))
 		static_branch_enable(&kvm_arm_pmu_available);
+
+out_unlock:
+	mutex_unlock(&arm_pmus_lock);
 }
 
-static int kvm_pmu_probe_pmuver(void)
+static struct arm_pmu *kvm_pmu_probe_armpmu(void)
 {
 	struct perf_event_attr attr = { };
 	struct perf_event *event;
-	struct arm_pmu *pmu;
-	int pmuver = ID_AA64DFR0_PMUVER_IMP_DEF;
+	struct arm_pmu *pmu = NULL;
 
 	/*
 	 * Create a dummy event that only counts user cycles. As we'll never
@@ -780,19 +805,20 @@ static int kvm_pmu_probe_pmuver(void)
 	if (IS_ERR(event)) {
 		pr_err_once("kvm: pmu event creation failed %ld\n",
 			    PTR_ERR(event));
-		return ID_AA64DFR0_PMUVER_IMP_DEF;
+		return NULL;
 	}
 
 	if (event->pmu) {
 		pmu = to_arm_pmu(event->pmu);
-		if (pmu->pmuver)
-			pmuver = pmu->pmuver;
+		if (pmu->pmuver == 0 ||
+		    pmu->pmuver == ID_AA64DFR0_PMUVER_IMP_DEF)
+			pmu = NULL;
 	}
 
 	perf_event_disable(event);
 	perf_event_release_kernel(event);
 
-	return pmuver;
+	return pmu;
 }
 
 u64 kvm_pmu_get_pmceid(struct kvm_vcpu *vcpu, bool pmceid1)
@@ -810,7 +836,7 @@ u64 kvm_pmu_get_pmceid(struct kvm_vcpu *vcpu, bool pmceid1)
 		 * Don't advertise STALL_SLOT, as PMMIR_EL0 is handled
 		 * as RAZ
 		 */
-		if (vcpu->kvm->arch.pmuver >= ID_AA64DFR0_PMUVER_8_4)
+		if (vcpu->kvm->arch.arm_pmu->pmuver >= ID_AA64DFR0_PMUVER_8_4)
 			val &= ~BIT_ULL(ARMV8_PMUV3_PERFCTR_STALL_SLOT - 32);
 		base = 32;
 	}
@@ -922,26 +948,64 @@ static bool pmu_irq_is_valid(struct kvm *kvm, int irq)
 	return true;
 }
 
+static int kvm_arm_pmu_v3_set_pmu(struct kvm_vcpu *vcpu, int pmu_id)
+{
+	struct kvm *kvm = vcpu->kvm;
+	struct arm_pmu_entry *entry;
+	struct arm_pmu *arm_pmu;
+	int ret = -ENXIO;
+
+	mutex_lock(&kvm->lock);
+	mutex_lock(&arm_pmus_lock);
+
+	list_for_each_entry(entry, &arm_pmus, entry) {
+		arm_pmu = entry->arm_pmu;
+		if (arm_pmu->pmu.type == pmu_id) {
+			if (kvm->arch.ran_once ||
+			    (kvm->arch.pmu_filter && kvm->arch.arm_pmu != arm_pmu)) {
+				ret = -EBUSY;
+				break;
+			}
+
+			kvm->arch.arm_pmu = arm_pmu;
+			cpumask_copy(kvm->arch.supported_cpus, &arm_pmu->supported_cpus);
+			ret = 0;
+			break;
+		}
+	}
+
+	mutex_unlock(&arm_pmus_lock);
+	mutex_unlock(&kvm->lock);
+	return ret;
+}
+
 int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 {
+	struct kvm *kvm = vcpu->kvm;
+
 	if (!kvm_vcpu_has_pmu(vcpu))
 		return -ENODEV;
 
 	if (vcpu->arch.pmu.created)
 		return -EBUSY;
 
-	if (!vcpu->kvm->arch.pmuver)
-		vcpu->kvm->arch.pmuver = kvm_pmu_probe_pmuver();
-
-	if (vcpu->kvm->arch.pmuver == ID_AA64DFR0_PMUVER_IMP_DEF)
-		return -ENODEV;
+	mutex_lock(&kvm->lock);
+	if (!kvm->arch.arm_pmu) {
+		/* No PMU set, get the default one */
+		kvm->arch.arm_pmu = kvm_pmu_probe_armpmu();
+		if (!kvm->arch.arm_pmu) {
+			mutex_unlock(&kvm->lock);
+			return -ENODEV;
+		}
+	}
+	mutex_unlock(&kvm->lock);
 
 	switch (attr->attr) {
 	case KVM_ARM_VCPU_PMU_V3_IRQ: {
 		int __user *uaddr = (int __user *)(long)attr->addr;
 		int irq;
 
-		if (!irqchip_in_kernel(vcpu->kvm))
+		if (!irqchip_in_kernel(kvm))
 			return -EINVAL;
 
 		if (get_user(irq, uaddr))
@@ -951,7 +1015,7 @@ int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 		if (!(irq_is_ppi(irq) || irq_is_spi(irq)))
 			return -EINVAL;
 
-		if (!pmu_irq_is_valid(vcpu->kvm, irq))
+		if (!pmu_irq_is_valid(kvm, irq))
 			return -EINVAL;
 
 		if (kvm_arm_pmu_irq_initialized(vcpu))
@@ -966,7 +1030,7 @@ int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 		struct kvm_pmu_event_filter filter;
 		int nr_events;
 
-		nr_events = kvm_pmu_event_mask(vcpu->kvm) + 1;
+		nr_events = kvm_pmu_event_mask(kvm) + 1;
 
 		uaddr = (struct kvm_pmu_event_filter __user *)(long)attr->addr;
 
@@ -978,12 +1042,17 @@ int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 		     filter.action != KVM_PMU_EVENT_DENY))
 			return -EINVAL;
 
-		mutex_lock(&vcpu->kvm->lock);
+		mutex_lock(&kvm->lock);
 
-		if (!vcpu->kvm->arch.pmu_filter) {
-			vcpu->kvm->arch.pmu_filter = bitmap_alloc(nr_events, GFP_KERNEL_ACCOUNT);
-			if (!vcpu->kvm->arch.pmu_filter) {
-				mutex_unlock(&vcpu->kvm->lock);
+		if (kvm->arch.ran_once) {
+			mutex_unlock(&kvm->lock);
+			return -EBUSY;
+		}
+
+		if (!kvm->arch.pmu_filter) {
+			kvm->arch.pmu_filter = bitmap_alloc(nr_events, GFP_KERNEL_ACCOUNT);
+			if (!kvm->arch.pmu_filter) {
+				mutex_unlock(&kvm->lock);
 				return -ENOMEM;
 			}
 
@@ -994,20 +1063,29 @@ int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 			 * events, the default is to allow.
 			 */
 			if (filter.action == KVM_PMU_EVENT_ALLOW)
-				bitmap_zero(vcpu->kvm->arch.pmu_filter, nr_events);
+				bitmap_zero(kvm->arch.pmu_filter, nr_events);
 			else
-				bitmap_fill(vcpu->kvm->arch.pmu_filter, nr_events);
+				bitmap_fill(kvm->arch.pmu_filter, nr_events);
 		}
 
 		if (filter.action == KVM_PMU_EVENT_ALLOW)
-			bitmap_set(vcpu->kvm->arch.pmu_filter, filter.base_event, filter.nevents);
+			bitmap_set(kvm->arch.pmu_filter, filter.base_event, filter.nevents);
 		else
-			bitmap_clear(vcpu->kvm->arch.pmu_filter, filter.base_event, filter.nevents);
+			bitmap_clear(kvm->arch.pmu_filter, filter.base_event, filter.nevents);
 
-		mutex_unlock(&vcpu->kvm->lock);
+		mutex_unlock(&kvm->lock);
 
 		return 0;
 	}
+	case KVM_ARM_VCPU_PMU_V3_SET_PMU: {
+		int __user *uaddr = (int __user *)(long)attr->addr;
+		int pmu_id;
+
+		if (get_user(pmu_id, uaddr))
+			return -EFAULT;
+
+		return kvm_arm_pmu_v3_set_pmu(vcpu, pmu_id);
+	}
 	case KVM_ARM_VCPU_PMU_V3_INIT:
 		return kvm_arm_pmu_v3_init(vcpu);
 	}
@@ -1045,6 +1123,7 @@ int kvm_arm_pmu_v3_has_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 	case KVM_ARM_VCPU_PMU_V3_IRQ:
 	case KVM_ARM_VCPU_PMU_V3_INIT:
 	case KVM_ARM_VCPU_PMU_V3_FILTER:
+	case KVM_ARM_VCPU_PMU_V3_SET_PMU:
 		if (kvm_vcpu_has_pmu(vcpu))
 			return 0;
 	}

include/kvm/arm_pmu.h

@@ -29,6 +29,11 @@ struct kvm_pmu {
 	struct irq_work overflow_work;
 };
 
+struct arm_pmu_entry {
+	struct list_head entry;
+	struct arm_pmu *arm_pmu;
+};
+
 DECLARE_STATIC_KEY_FALSE(kvm_arm_pmu_available);
 
 static __always_inline bool kvm_arm_support_pmu_v3(void)

include/linux/perf_event.h

@@ -864,7 +864,7 @@ struct perf_event_context {
 #define PERF_NR_CONTEXTS	4
 
 /**
- * struct perf_event_cpu_context - per cpu event context structure
+ * struct perf_cpu_context - per cpu event context structure
  */
 struct perf_cpu_context {
 	struct perf_event_context	ctx;

tools/arch/arm64/include/uapi/asm/kvm.h

@@ -362,6 +362,7 @@ struct kvm_arm_copy_mte_tags {
 #define   KVM_ARM_VCPU_PMU_V3_IRQ	0
 #define   KVM_ARM_VCPU_PMU_V3_INIT	1
 #define   KVM_ARM_VCPU_PMU_V3_FILTER	2
+#define   KVM_ARM_VCPU_PMU_V3_SET_PMU	3
 #define KVM_ARM_VCPU_TIMER_CTRL		1
 #define   KVM_ARM_VCPU_TIMER_IRQ_VTIMER		0
 #define   KVM_ARM_VCPU_TIMER_IRQ_PTIMER		1