1 files changed, 250 insertions, 161 deletions

diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
index 09873f6488f7..de1fd7369736 100644
--- a/arch/x86/kvm/pmu.c
+++ b/arch/x86/kvm/pmu.c
@@ -13,7 +13,10 @@
 #include <linux/types.h>
 #include <linux/kvm_host.h>
 #include <linux/perf_event.h>
+#include <linux/bsearch.h>
+#include <linux/sort.h>
 #include <asm/perf_event.h>
+#include <asm/cpu_device_id.h>
 #include "x86.h"
 #include "cpuid.h"
 #include "lapic.h"
@@ -22,6 +25,15 @@
 /* This is enough to filter the vast majority of currently defined events. */
 #define KVM_PMU_EVENT_FILTER_MAX_EVENTS 300
 
+struct x86_pmu_capability __read_mostly kvm_pmu_cap;
+EXPORT_SYMBOL_GPL(kvm_pmu_cap);
+
+static const struct x86_cpu_id vmx_icl_pebs_cpu[] = {
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_D, NULL),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X, NULL),
+	{}
+};
+
 /* NOTE:
  * - Each perf counter is defined as "struct kvm_pmc";
  * - There are two types of perf counters: general purpose (gp) and fixed.
@@ -32,7 +44,9 @@
  *   However AMD doesn't support fixed-counters;
  * - There are three types of index to access perf counters (PMC):
  *     1. MSR (named msr): For example Intel has MSR_IA32_PERFCTRn and AMD
- *        has MSR_K7_PERFCTRn.
+ *        has MSR_K7_PERFCTRn and, for families 15H and later,
+ *        MSR_F15H_PERF_CTRn, where MSR_F15H_PERF_CTR[0-3] are
+ *        aliased to MSR_K7_PERFCTRn.
  *     2. MSR Index (named idx): This normally is used by RDPMC instruction.
  *        For instance AMD RDPMC instruction uses 0000_0003h in ECX to access
  *        C001_0007h (MSR_K7_PERCTR3). Intel has a similar mechanism, except
@@ -42,11 +56,38 @@
  *        code. Each pmc, stored in kvm_pmc.idx field, is unique across
  *        all perf counters (both gp and fixed). The mapping relationship
  *        between pmc and perf counters is as the following:
- *        * Intel: [0 .. INTEL_PMC_MAX_GENERIC-1] <=> gp counters
+ *        * Intel: [0 .. KVM_INTEL_PMC_MAX_GENERIC-1] <=> gp counters
  *                 [INTEL_PMC_IDX_FIXED .. INTEL_PMC_IDX_FIXED + 2] <=> fixed
- *        * AMD:   [0 .. AMD64_NUM_COUNTERS-1] <=> gp counters
+ *        * AMD:   [0 .. AMD64_NUM_COUNTERS-1] and, for families 15H
+ *          and later, [0 .. AMD64_NUM_COUNTERS_CORE-1] <=> gp counters
  */
 
+static struct kvm_pmu_ops kvm_pmu_ops __read_mostly;
+
+#define KVM_X86_PMU_OP(func)					     \
+	DEFINE_STATIC_CALL_NULL(kvm_x86_pmu_##func,			     \
+				*(((struct kvm_pmu_ops *)0)->func));
+#define KVM_X86_PMU_OP_OPTIONAL KVM_X86_PMU_OP
+#include <asm/kvm-x86-pmu-ops.h>
+
+void kvm_pmu_ops_update(const struct kvm_pmu_ops *pmu_ops)
+{
+	memcpy(&kvm_pmu_ops, pmu_ops, sizeof(kvm_pmu_ops));
+
+#define __KVM_X86_PMU_OP(func) \
+	static_call_update(kvm_x86_pmu_##func, kvm_pmu_ops.func);
+#define KVM_X86_PMU_OP(func) \
+	WARN_ON(!kvm_pmu_ops.func); __KVM_X86_PMU_OP(func)
+#define KVM_X86_PMU_OP_OPTIONAL __KVM_X86_PMU_OP
+#include <asm/kvm-x86-pmu-ops.h>
+#undef __KVM_X86_PMU_OP
+}
+
+static inline bool pmc_is_enabled(struct kvm_pmc *pmc)
+{
+	return static_call(kvm_x86_pmu_pmc_is_enabled)(pmc);
+}
+
 static void kvm_pmi_trigger_fn(struct irq_work *irq_work)
 {
 	struct kvm_pmu *pmu = container_of(irq_work, struct kvm_pmu, irq_work);
@@ -55,50 +96,65 @@ static void kvm_pmi_trigger_fn(struct irq_work *irq_work)
 	kvm_pmu_deliver_pmi(vcpu);
 }
 
-static void kvm_perf_overflow(struct perf_event *perf_event,
-			      struct perf_sample_data *data,
-			      struct pt_regs *regs)
+static inline void __kvm_perf_overflow(struct kvm_pmc *pmc, bool in_pmi)
 {
-	struct kvm_pmc *pmc = perf_event->overflow_handler_context;
 	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
+	bool skip_pmi = false;
+
+	/* Ignore counters that have been reprogrammed already. */
+	if (test_and_set_bit(pmc->idx, pmu->reprogram_pmi))
+		return;
 
-	if (!test_and_set_bit(pmc->idx, pmu->reprogram_pmi)) {
+	if (pmc->perf_event && pmc->perf_event->attr.precise_ip) {
+		if (!in_pmi) {
+			/*
+			 * TODO: KVM is currently _choosing_ to not generate records
+			 * for emulated instructions, avoiding BUFFER_OVF PMI when
+			 * there are no records. Strictly speaking, it should be done
+			 * as well in the right context to improve sampling accuracy.
+			 */
+			skip_pmi = true;
+		} else {
+			/* Indicate PEBS overflow PMI to guest. */
+			skip_pmi = __test_and_set_bit(GLOBAL_STATUS_BUFFER_OVF_BIT,
+						      (unsigned long *)&pmu->global_status);
+		}
+	} else {
 		__set_bit(pmc->idx, (unsigned long *)&pmu->global_status);
-		kvm_make_request(KVM_REQ_PMU, pmc->vcpu);
 	}
+	kvm_make_request(KVM_REQ_PMU, pmc->vcpu);
+
+	if (!pmc->intr || skip_pmi)
+		return;
+
+	/*
+	 * Inject PMI. If vcpu was in a guest mode during NMI PMI
+	 * can be ejected on a guest mode re-entry. Otherwise we can't
+	 * be sure that vcpu wasn't executing hlt instruction at the
+	 * time of vmexit and is not going to re-enter guest mode until
+	 * woken up. So we should wake it, but this is impossible from
+	 * NMI context. Do it from irq work instead.
+	 */
+	if (in_pmi && !kvm_handling_nmi_from_guest(pmc->vcpu))
+		irq_work_queue(&pmc_to_pmu(pmc)->irq_work);
+	else
+		kvm_make_request(KVM_REQ_PMI, pmc->vcpu);
 }
 
-static void kvm_perf_overflow_intr(struct perf_event *perf_event,
-				   struct perf_sample_data *data,
-				   struct pt_regs *regs)
+static void kvm_perf_overflow(struct perf_event *perf_event,
+			      struct perf_sample_data *data,
+			      struct pt_regs *regs)
 {
 	struct kvm_pmc *pmc = perf_event->overflow_handler_context;
-	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
-
-	if (!test_and_set_bit(pmc->idx, pmu->reprogram_pmi)) {
-		__set_bit(pmc->idx, (unsigned long *)&pmu->global_status);
-		kvm_make_request(KVM_REQ_PMU, pmc->vcpu);
 
-		/*
-		 * Inject PMI. If vcpu was in a guest mode during NMI PMI
-		 * can be ejected on a guest mode re-entry. Otherwise we can't
-		 * be sure that vcpu wasn't executing hlt instruction at the
-		 * time of vmexit and is not going to re-enter guest mode until
-		 * woken up. So we should wake it, but this is impossible from
-		 * NMI context. Do it from irq work instead.
-		 */
-		if (!kvm_is_in_guest())
-			irq_work_queue(&pmc_to_pmu(pmc)->irq_work);
-		else
-			kvm_make_request(KVM_REQ_PMI, pmc->vcpu);
-	}
+	__kvm_perf_overflow(pmc, true);
 }
 
 static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
-				  unsigned config, bool exclude_user,
-				  bool exclude_kernel, bool intr,
-				  bool in_tx, bool in_tx_cp)
+				  u64 config, bool exclude_user,
+				  bool exclude_kernel, bool intr)
 {
+	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
 	struct perf_event *event;
 	struct perf_event_attr attr = {
 		.type = type,
@@ -110,23 +166,40 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
 		.exclude_kernel = exclude_kernel,
 		.config = config,
 	};
+	bool pebs = test_bit(pmc->idx, (unsigned long *)&pmu->pebs_enable);
 
 	attr.sample_period = get_sample_period(pmc, pmc->counter);
 
-	if (in_tx)
-		attr.config |= HSW_IN_TX;
-	if (in_tx_cp) {
+	if ((attr.config & HSW_IN_TX_CHECKPOINTED) &&
+	    guest_cpuid_is_intel(pmc->vcpu)) {
 		/*
 		 * HSW_IN_TX_CHECKPOINTED is not supported with nonzero
 		 * period. Just clear the sample period so at least
 		 * allocating the counter doesn't fail.
 		 */
 		attr.sample_period = 0;
-		attr.config |= HSW_IN_TX_CHECKPOINTED;
+	}
+	if (pebs) {
+		/*
+		 * The non-zero precision level of guest event makes the ordinary
+		 * guest event becomes a guest PEBS event and triggers the host
+		 * PEBS PMI handler to determine whether the PEBS overflow PMI
+		 * comes from the host counters or the guest.
+		 *
+		 * For most PEBS hardware events, the difference in the software
+		 * precision levels of guest and host PEBS events will not affect
+		 * the accuracy of the PEBS profiling result, because the "event IP"
+		 * in the PEBS record is calibrated on the guest side.
+		 *
+		 * On Icelake everything is fine. Other hardware (GLC+, TNT+) that
+		 * could possibly care here is unsupported and needs changes.
+		 */
+		attr.precise_ip = 1;
+		if (x86_match_cpu(vmx_icl_pebs_cpu) && pmc->idx == 32)
+			attr.precise_ip = 3;
 	}
 
 	event = perf_event_create_kernel_counter(&attr, -1, current,
-						 intr ? kvm_perf_overflow_intr :
 						 kvm_perf_overflow, pmc);
 	if (IS_ERR(event)) {
 		pr_debug_ratelimited("kvm_pmu: event creation failed %ld for pmc->idx = %d\n",
@@ -138,6 +211,7 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
 	pmc_to_pmu(pmc)->event_count++;
 	clear_bit(pmc->idx, pmc_to_pmu(pmc)->reprogram_pmi);
 	pmc->is_paused = false;
+	pmc->intr = intr || pebs;
 }
 
 static void pmc_pause_counter(struct kvm_pmc *pmc)
@@ -163,6 +237,10 @@ static bool pmc_resume_counter(struct kvm_pmc *pmc)
 			      get_sample_period(pmc, pmc->counter)))
 		return false;
 
+	if (test_bit(pmc->idx, (unsigned long *)&pmc_to_pmu(pmc)->pebs_enable) !=
+	    (!!pmc->perf_event->attr.precise_ip))
+		return false;
+
 	/* reuse perf_event to serve as pmc_reprogram_counter() does*/
 	perf_event_enable(pmc->perf_event);
 	pmc->is_paused = false;
@@ -171,125 +249,91 @@ static bool pmc_resume_counter(struct kvm_pmc *pmc)
 	return true;
 }
 
-void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel)
+static int cmp_u64(const void *pa, const void *pb)
 {
-	unsigned config, type = PERF_TYPE_RAW;
-	u8 event_select, unit_mask;
-	struct kvm *kvm = pmc->vcpu->kvm;
-	struct kvm_pmu_event_filter *filter;
-	int i;
-	bool allow_event = true;
-
-	if (eventsel & ARCH_PERFMON_EVENTSEL_PIN_CONTROL)
-		printk_once("kvm pmu: pin control bit is ignored\n");
+	u64 a = *(u64 *)pa;
+	u64 b = *(u64 *)pb;
 
-	pmc->eventsel = eventsel;
+	return (a > b) - (a < b);
+}
 
-	pmc_pause_counter(pmc);
+static bool check_pmu_event_filter(struct kvm_pmc *pmc)
+{
+	struct kvm_pmu_event_filter *filter;
+	struct kvm *kvm = pmc->vcpu->kvm;
+	bool allow_event = true;
+	__u64 key;
+	int idx;
 
-	if (!(eventsel & ARCH_PERFMON_EVENTSEL_ENABLE) || !pmc_is_enabled(pmc))
-		return;
+	if (!static_call(kvm_x86_pmu_hw_event_available)(pmc))
+		return false;
 
 	filter = srcu_dereference(kvm->arch.pmu_event_filter, &kvm->srcu);
-	if (filter) {
-		for (i = 0; i < filter->nevents; i++)
-			if (filter->events[i] ==
-			    (eventsel & AMD64_RAW_EVENT_MASK_NB))
-				break;
-		if (filter->action == KVM_PMU_EVENT_ALLOW &&
-		    i == filter->nevents)
-			allow_event = false;
+	if (!filter)
+		goto out;
+
+	if (pmc_is_gp(pmc)) {
+		key = pmc->eventsel & AMD64_RAW_EVENT_MASK_NB;
+		if (bsearch(&key, filter->events, filter->nevents,
+			    sizeof(__u64), cmp_u64))
+			allow_event = filter->action == KVM_PMU_EVENT_ALLOW;
+		else
+			allow_event = filter->action == KVM_PMU_EVENT_DENY;
+	} else {
+		idx = pmc->idx - INTEL_PMC_IDX_FIXED;
 		if (filter->action == KVM_PMU_EVENT_DENY &&
-		    i < filter->nevents)
+		    test_bit(idx, (ulong *)&filter->fixed_counter_bitmap))
+			allow_event = false;
+		if (filter->action == KVM_PMU_EVENT_ALLOW &&
+		    !test_bit(idx, (ulong *)&filter->fixed_counter_bitmap))
 			allow_event = false;
 	}
-	if (!allow_event)
-		return;
-
-	event_select = eventsel & ARCH_PERFMON_EVENTSEL_EVENT;
-	unit_mask = (eventsel & ARCH_PERFMON_EVENTSEL_UMASK) >> 8;
-
-	if (!(eventsel & (ARCH_PERFMON_EVENTSEL_EDGE |
-			  ARCH_PERFMON_EVENTSEL_INV |
-			  ARCH_PERFMON_EVENTSEL_CMASK |
-			  HSW_IN_TX |
-			  HSW_IN_TX_CHECKPOINTED))) {
-		config = kvm_x86_ops.pmu_ops->find_arch_event(pmc_to_pmu(pmc),
-						      event_select,
-						      unit_mask);
-		if (config != PERF_COUNT_HW_MAX)
-			type = PERF_TYPE_HARDWARE;
-	}
-
-	if (type == PERF_TYPE_RAW)
-		config = eventsel & X86_RAW_EVENT_MASK;
-
-	if (pmc->current_config == eventsel && pmc_resume_counter(pmc))
-		return;
 
-	pmc_release_perf_event(pmc);
-
-	pmc->current_config = eventsel;
-	pmc_reprogram_counter(pmc, type, config,
-			      !(eventsel & ARCH_PERFMON_EVENTSEL_USR),
-			      !(eventsel & ARCH_PERFMON_EVENTSEL_OS),
-			      eventsel & ARCH_PERFMON_EVENTSEL_INT,
-			      (eventsel & HSW_IN_TX),
-			      (eventsel & HSW_IN_TX_CHECKPOINTED));
+out:
+	return allow_event;
 }
-EXPORT_SYMBOL_GPL(reprogram_gp_counter);
 
-void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 ctrl, int idx)
+void reprogram_counter(struct kvm_pmc *pmc)
 {
-	unsigned en_field = ctrl & 0x3;
-	bool pmi = ctrl & 0x8;
-	struct kvm_pmu_event_filter *filter;
-	struct kvm *kvm = pmc->vcpu->kvm;
+	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
+	u64 eventsel = pmc->eventsel;
+	u64 new_config = eventsel;
+	u8 fixed_ctr_ctrl;
 
 	pmc_pause_counter(pmc);
 
-	if (!en_field || !pmc_is_enabled(pmc))
+	if (!pmc_speculative_in_use(pmc) || !pmc_is_enabled(pmc))
 		return;
 
-	filter = srcu_dereference(kvm->arch.pmu_event_filter, &kvm->srcu);
-	if (filter) {
-		if (filter->action == KVM_PMU_EVENT_DENY &&
-		    test_bit(idx, (ulong *)&filter->fixed_counter_bitmap))
-			return;
-		if (filter->action == KVM_PMU_EVENT_ALLOW &&
-		    !test_bit(idx, (ulong *)&filter->fixed_counter_bitmap))
-			return;
-	}
-
-	if (pmc->current_config == (u64)ctrl && pmc_resume_counter(pmc))
+	if (!check_pmu_event_filter(pmc))
 		return;
 
-	pmc_release_perf_event(pmc);
-
-	pmc->current_config = (u64)ctrl;
-	pmc_reprogram_counter(pmc, PERF_TYPE_HARDWARE,
-			      kvm_x86_ops.pmu_ops->find_fixed_event(idx),
-			      !(en_field & 0x2), /* exclude user */
-			      !(en_field & 0x1), /* exclude kernel */
-			      pmi, false, false);
-}
-EXPORT_SYMBOL_GPL(reprogram_fixed_counter);
+	if (eventsel & ARCH_PERFMON_EVENTSEL_PIN_CONTROL)
+		printk_once("kvm pmu: pin control bit is ignored\n");
 
-void reprogram_counter(struct kvm_pmu *pmu, int pmc_idx)
-{
-	struct kvm_pmc *pmc = kvm_x86_ops.pmu_ops->pmc_idx_to_pmc(pmu, pmc_idx);
+	if (pmc_is_fixed(pmc)) {
+		fixed_ctr_ctrl = fixed_ctrl_field(pmu->fixed_ctr_ctrl,
+						  pmc->idx - INTEL_PMC_IDX_FIXED);
+		if (fixed_ctr_ctrl & 0x1)
+			eventsel |= ARCH_PERFMON_EVENTSEL_OS;
+		if (fixed_ctr_ctrl & 0x2)
+			eventsel |= ARCH_PERFMON_EVENTSEL_USR;
+		if (fixed_ctr_ctrl & 0x8)
+			eventsel |= ARCH_PERFMON_EVENTSEL_INT;
+		new_config = (u64)fixed_ctr_ctrl;
+	}
 
-	if (!pmc)
+	if (pmc->current_config == new_config && pmc_resume_counter(pmc))
 		return;
 
-	if (pmc_is_gp(pmc))
-		reprogram_gp_counter(pmc, pmc->eventsel);
-	else {
-		int idx = pmc_idx - INTEL_PMC_IDX_FIXED;
-		u8 ctrl = fixed_ctrl_field(pmu->fixed_ctr_ctrl, idx);
+	pmc_release_perf_event(pmc);
 
-		reprogram_fixed_counter(pmc, ctrl, idx);
-	}
+	pmc->current_config = new_config;
+	pmc_reprogram_counter(pmc, PERF_TYPE_RAW,
+			      (eventsel & pmu->raw_event_mask),
+			      !(eventsel & ARCH_PERFMON_EVENTSEL_USR),
+			      !(eventsel & ARCH_PERFMON_EVENTSEL_OS),
+			      eventsel & ARCH_PERFMON_EVENTSEL_INT);
 }
 EXPORT_SYMBOL_GPL(reprogram_counter);
 
@@ -299,14 +343,13 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
 	int bit;
 
 	for_each_set_bit(bit, pmu->reprogram_pmi, X86_PMC_IDX_MAX) {
-		struct kvm_pmc *pmc = kvm_x86_ops.pmu_ops->pmc_idx_to_pmc(pmu, bit);
+		struct kvm_pmc *pmc = static_call(kvm_x86_pmu_pmc_idx_to_pmc)(pmu, bit);
 
 		if (unlikely(!pmc || !pmc->perf_event)) {
 			clear_bit(bit, pmu->reprogram_pmi);
 			continue;
 		}
-
-		reprogram_counter(pmu, bit);
+		reprogram_counter(pmc);
 	}
 
 	/*
@@ -321,7 +364,7 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
 /* check if idx is a valid index to access PMU */
 bool kvm_pmu_is_valid_rdpmc_ecx(struct kvm_vcpu *vcpu, unsigned int idx)
 {
-	return kvm_x86_ops.pmu_ops->is_valid_rdpmc_ecx(vcpu, idx);
+	return static_call(kvm_x86_pmu_is_valid_rdpmc_ecx)(vcpu, idx);
 }
 
 bool is_vmware_backdoor_pmc(u32 pmc_idx)
@@ -371,7 +414,7 @@ int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
 	if (is_vmware_backdoor_pmc(idx))
 		return kvm_pmu_rdpmc_vmware(vcpu, idx, data);
 
-	pmc = kvm_x86_ops.pmu_ops->rdpmc_ecx_to_pmc(vcpu, idx, &mask);
+	pmc = static_call(kvm_x86_pmu_rdpmc_ecx_to_pmc)(vcpu, idx, &mask);
 	if (!pmc)
 		return 1;
 
@@ -387,22 +430,21 @@ int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
 void kvm_pmu_deliver_pmi(struct kvm_vcpu *vcpu)
 {
 	if (lapic_in_kernel(vcpu)) {
-		if (kvm_x86_ops.pmu_ops->deliver_pmi)
-			kvm_x86_ops.pmu_ops->deliver_pmi(vcpu);
+		static_call_cond(kvm_x86_pmu_deliver_pmi)(vcpu);
 		kvm_apic_local_deliver(vcpu->arch.apic, APIC_LVTPC);
 	}
 }
 
 bool kvm_pmu_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
 {
-	return kvm_x86_ops.pmu_ops->msr_idx_to_pmc(vcpu, msr) ||
-		kvm_x86_ops.pmu_ops->is_valid_msr(vcpu, msr);
+	return static_call(kvm_x86_pmu_msr_idx_to_pmc)(vcpu, msr) ||
+		static_call(kvm_x86_pmu_is_valid_msr)(vcpu, msr);
 }
 
 static void kvm_pmu_mark_pmc_in_use(struct kvm_vcpu *vcpu, u32 msr)
 {
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
-	struct kvm_pmc *pmc = kvm_x86_ops.pmu_ops->msr_idx_to_pmc(vcpu, msr);
+	struct kvm_pmc *pmc = static_call(kvm_x86_pmu_msr_idx_to_pmc)(vcpu, msr);
 
 	if (pmc)
 		__set_bit(pmc->idx, pmu->pmc_in_use);
@@ -410,13 +452,13 @@ static void kvm_pmu_mark_pmc_in_use(struct kvm_vcpu *vcpu, u32 msr)
 
 int kvm_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 {
-	return kvm_x86_ops.pmu_ops->get_msr(vcpu, msr_info);
+	return static_call(kvm_x86_pmu_get_msr)(vcpu, msr_info);
 }
 
 int kvm_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 {
 	kvm_pmu_mark_pmc_in_use(vcpu, msr_info->index);
-	return kvm_x86_ops.pmu_ops->set_msr(vcpu, msr_info);
+	return static_call(kvm_x86_pmu_set_msr)(vcpu, msr_info);
 }
 
 /* refresh PMU settings. This function generally is called when underlying
@@ -425,7 +467,7 @@ int kvm_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
  */
 void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
 {
-	kvm_x86_ops.pmu_ops->refresh(vcpu);
+	static_call(kvm_x86_pmu_refresh)(vcpu);
 }
 
 void kvm_pmu_reset(struct kvm_vcpu *vcpu)
@@ -433,7 +475,7 @@ void kvm_pmu_reset(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 
 	irq_work_sync(&pmu->irq_work);
-	kvm_x86_ops.pmu_ops->reset(vcpu);
+	static_call(kvm_x86_pmu_reset)(vcpu);
 }
 
 void kvm_pmu_init(struct kvm_vcpu *vcpu)
@@ -441,24 +483,13 @@ void kvm_pmu_init(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 
 	memset(pmu, 0, sizeof(*pmu));
-	kvm_x86_ops.pmu_ops->init(vcpu);
+	static_call(kvm_x86_pmu_init)(vcpu);
 	init_irq_work(&pmu->irq_work, kvm_pmi_trigger_fn);
 	pmu->event_count = 0;
 	pmu->need_cleanup = false;
 	kvm_pmu_refresh(vcpu);
 }
 
-static inline bool pmc_speculative_in_use(struct kvm_pmc *pmc)
-{
-	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
-
-	if (pmc_is_fixed(pmc))
-		return fixed_ctrl_field(pmu->fixed_ctr_ctrl,
-			pmc->idx - INTEL_PMC_IDX_FIXED) & 0x3;
-
-	return pmc->eventsel & ARCH_PERFMON_EVENTSEL_ENABLE;
-}
-
 /* Release perf_events for vPMCs that have been unused for a full time slice.  */
 void kvm_pmu_cleanup(struct kvm_vcpu *vcpu)
 {
@@ -473,14 +504,13 @@ void kvm_pmu_cleanup(struct kvm_vcpu *vcpu)
 		      pmu->pmc_in_use, X86_PMC_IDX_MAX);
 
 	for_each_set_bit(i, bitmask, X86_PMC_IDX_MAX) {
-		pmc = kvm_x86_ops.pmu_ops->pmc_idx_to_pmc(pmu, i);
+		pmc = static_call(kvm_x86_pmu_pmc_idx_to_pmc)(pmu, i);
 
 		if (pmc && pmc->perf_event && !pmc_speculative_in_use(pmc))
 			pmc_stop_counter(pmc);
 	}
 
-	if (kvm_x86_ops.pmu_ops->cleanup)
-		kvm_x86_ops.pmu_ops->cleanup(vcpu);
+	static_call_cond(kvm_x86_pmu_cleanup)(vcpu);
 
 	bitmap_zero(pmu->pmc_in_use, X86_PMC_IDX_MAX);
 }
@@ -490,6 +520,60 @@ void kvm_pmu_destroy(struct kvm_vcpu *vcpu)
 	kvm_pmu_reset(vcpu);
 }
 
+static void kvm_pmu_incr_counter(struct kvm_pmc *pmc)
+{
+	u64 prev_count;
+
+	prev_count = pmc->counter;
+	pmc->counter = (pmc->counter + 1) & pmc_bitmask(pmc);
+
+	reprogram_counter(pmc);
+	if (pmc->counter < prev_count)
+		__kvm_perf_overflow(pmc, false);
+}
+
+static inline bool eventsel_match_perf_hw_id(struct kvm_pmc *pmc,
+	unsigned int perf_hw_id)
+{
+	return !((pmc->eventsel ^ perf_get_hw_event_config(perf_hw_id)) &
+		AMD64_RAW_EVENT_MASK_NB);
+}
+
+static inline bool cpl_is_matched(struct kvm_pmc *pmc)
+{
+	bool select_os, select_user;
+	u64 config = pmc->current_config;
+
+	if (pmc_is_gp(pmc)) {
+		select_os = config & ARCH_PERFMON_EVENTSEL_OS;
+		select_user = config & ARCH_PERFMON_EVENTSEL_USR;
+	} else {
+		select_os = config & 0x1;
+		select_user = config & 0x2;
+	}
+
+	return (static_call(kvm_x86_get_cpl)(pmc->vcpu) == 0) ? select_os : select_user;
+}
+
+void kvm_pmu_trigger_event(struct kvm_vcpu *vcpu, u64 perf_hw_id)
+{
+	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
+	struct kvm_pmc *pmc;
+	int i;
+
+	for_each_set_bit(i, pmu->all_valid_pmc_idx, X86_PMC_IDX_MAX) {
+		pmc = static_call(kvm_x86_pmu_pmc_idx_to_pmc)(pmu, i);
+
+		if (!pmc || !pmc_is_enabled(pmc) || !pmc_speculative_in_use(pmc))
+			continue;
+
+		/* Ignore checks for edge detect, pin control, invert and CMASK bits */
+		if (eventsel_match_perf_hw_id(pmc, perf_hw_id) && cpl_is_matched(pmc))
+			kvm_pmu_incr_counter(pmc);
+	}
+}
+EXPORT_SYMBOL_GPL(kvm_pmu_trigger_event);
+
 int kvm_vm_ioctl_set_pmu_event_filter(struct kvm *kvm, void __user *argp)
 {
 	struct kvm_pmu_event_filter tmp, *filter;
@@ -521,6 +605,11 @@ int kvm_vm_ioctl_set_pmu_event_filter(struct kvm *kvm, void __user *argp)
 	/* Ensure nevents can't be changed between the user copies. */
 	*filter = tmp;
 
+	/*
+	 * Sort the in-kernel list so that we can search it with bsearch.
+	 */
+	sort(&filter->events, filter->nevents, sizeof(__u64), cmp_u64, NULL);
+
 	mutex_lock(&kvm->lock);
 	filter = rcu_replace_pointer(kvm->arch.pmu_event_filter, filter,
 				     mutex_is_locked(&kvm->lock));