Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull more kvm updates from Paolo Bonzini:
 "Generic:

   - selftest compilation fix for non-x86

   - KVM: avoid warning on s390 in mark_page_dirty

 x86:

   - fix page write-protection bug and improve comments

   - use binary search to lookup the PMU event filter, add test

   - enable_pmu module parameter support for Intel CPUs

   - switch blocked_vcpu_on_cpu_lock to raw spinlock

   - cleanups of blocked vCPU logic

   - partially allow KVM_SET_CPUID{,2} after KVM_RUN (5.16 regression)

   - various small fixes"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (46 commits)
  docs: kvm: fix WARNINGs from api.rst
  selftests: kvm/x86: Fix the warning in lib/x86_64/processor.c
  selftests: kvm/x86: Fix the warning in pmu_event_filter_test.c
  kvm: selftests: Do not indent with spaces
  kvm: selftests: sync uapi/linux/kvm.h with Linux header
  selftests: kvm: add amx_test to .gitignore
  KVM: SVM: Nullify vcpu_(un)blocking() hooks if AVIC is disabled
  KVM: SVM: Move svm_hardware_setup() and its helpers below svm_x86_ops
  KVM: SVM: Drop AVIC's intermediate avic_set_running() helper
  KVM: VMX: Don't do full kick when handling posted interrupt wakeup
  KVM: VMX: Fold fallback path into triggering posted IRQ helper
  KVM: VMX: Pass desired vector instead of bool for triggering posted IRQ
  KVM: VMX: Don't do full kick when triggering posted interrupt "fails"
  KVM: SVM: Skip AVIC and IRTE updates when loading blocking vCPU
  KVM: SVM: Use kvm_vcpu_is_blocking() in AVIC load to handle preemption
  KVM: SVM: Remove unnecessary APICv/AVIC update in vCPU unblocking path
  KVM: SVM: Don't bother checking for "running" AVIC when kicking for IPIs
  KVM: SVM: Signal AVIC doorbell iff vCPU is in guest mode
  KVM: x86: Remove defunct pre_block/post_block kvm_x86_ops hooks
  KVM: x86: Unexport LAPIC's switch_to_{hv,sw}_timer() helpers
  ...

12 files changed, 756 insertions, 88 deletions

diff --git a/tools/include/uapi/linux/kvm.h b/tools/include/uapi/linux/kvm.h
index f066637ee206..9563d294f181 100644
--- a/tools/include/uapi/linux/kvm.h
+++ b/tools/include/uapi/linux/kvm.h
@@ -1131,7 +1131,8 @@ struct kvm_ppc_resize_hpt {
 #define KVM_CAP_EXIT_ON_EMULATION_FAILURE 204
 #define KVM_CAP_ARM_MTE 205
 #define KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM 206
-#define KVM_CAP_XSAVE2 207
+#define KVM_CAP_VM_GPA_BITS 207
+#define KVM_CAP_XSAVE2 208
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
@@ -1163,11 +1164,20 @@ struct kvm_irq_routing_hv_sint {
 	__u32 sint;
 };
 
+struct kvm_irq_routing_xen_evtchn {
+	__u32 port;
+	__u32 vcpu;
+	__u32 priority;
+};
+
+#define KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL ((__u32)(-1))
+
 /* gsi routing entry types */
 #define KVM_IRQ_ROUTING_IRQCHIP 1
 #define KVM_IRQ_ROUTING_MSI 2
 #define KVM_IRQ_ROUTING_S390_ADAPTER 3
 #define KVM_IRQ_ROUTING_HV_SINT 4
+#define KVM_IRQ_ROUTING_XEN_EVTCHN 5
 
 struct kvm_irq_routing_entry {
 	__u32 gsi;
@@ -1179,6 +1189,7 @@ struct kvm_irq_routing_entry {
 		struct kvm_irq_routing_msi msi;
 		struct kvm_irq_routing_s390_adapter adapter;
 		struct kvm_irq_routing_hv_sint hv_sint;
+		struct kvm_irq_routing_xen_evtchn xen_evtchn;
 		__u32 pad[8];
 	} u;
 };
@@ -1209,6 +1220,7 @@ struct kvm_x86_mce {
 #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL	(1 << 1)
 #define KVM_XEN_HVM_CONFIG_SHARED_INFO		(1 << 2)
 #define KVM_XEN_HVM_CONFIG_RUNSTATE		(1 << 3)
+#define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL	(1 << 4)
 
 struct kvm_xen_hvm_config {
 	__u32 flags;
@@ -1552,8 +1564,6 @@ struct kvm_s390_ucas_mapping {
 /* Available with KVM_CAP_XSAVE */
 #define KVM_GET_XSAVE		  _IOR(KVMIO,  0xa4, struct kvm_xsave)
 #define KVM_SET_XSAVE		  _IOW(KVMIO,  0xa5, struct kvm_xsave)
-/* Available with KVM_CAP_XSAVE2 */
-#define KVM_GET_XSAVE2		  _IOR(KVMIO,  0xcf, struct kvm_xsave)
 /* Available with KVM_CAP_XCRS */
 #define KVM_GET_XCRS		  _IOR(KVMIO,  0xa6, struct kvm_xcrs)
 #define KVM_SET_XCRS		  _IOW(KVMIO,  0xa7, struct kvm_xcrs)
@@ -1613,6 +1623,9 @@ struct kvm_enc_region {
 #define KVM_S390_NORMAL_RESET	_IO(KVMIO,   0xc3)
 #define KVM_S390_CLEAR_RESET	_IO(KVMIO,   0xc4)
 
+/* Available with KVM_CAP_XSAVE2 */
+#define KVM_GET_XSAVE2		  _IOR(KVMIO,  0xcf, struct kvm_xsave)
+
 struct kvm_s390_pv_sec_parm {
 	__u64 origin;
 	__u64 length;
diff --git a/tools/testing/selftests/kvm/.gitignore b/tools/testing/selftests/kvm/.gitignore
index 8c129961accf..dce7de7755e6 100644
--- a/tools/testing/selftests/kvm/.gitignore
+++ b/tools/testing/selftests/kvm/.gitignore
@@ -8,11 +8,12 @@
 /s390x/memop
 /s390x/resets
 /s390x/sync_regs_test
+/x86_64/amx_test
+/x86_64/cpuid_test
 /x86_64/cr4_cpuid_sync_test
 /x86_64/debug_regs
 /x86_64/evmcs_test
 /x86_64/emulator_error_test
-/x86_64/get_cpuid_test
 /x86_64/get_msr_index_features
 /x86_64/kvm_clock_test
 /x86_64/kvm_pv_test
@@ -22,6 +23,7 @@
 /x86_64/mmio_warning_test
 /x86_64/mmu_role_test
 /x86_64/platform_info_test
+/x86_64/pmu_event_filter_test
 /x86_64/set_boot_cpu_id
 /x86_64/set_sregs_test
 /x86_64/sev_migrate_tests
@@ -36,6 +38,7 @@
 /x86_64/vmx_apic_access_test
 /x86_64/vmx_close_while_nested_test
 /x86_64/vmx_dirty_log_test
+/x86_64/vmx_exception_with_invalid_guest_state
 /x86_64/vmx_invalid_nested_guest_state
 /x86_64/vmx_preemption_timer_test
 /x86_64/vmx_set_nested_state_test
diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile
index ee8cf2149824..81ebf99d6ff0 100644
--- a/tools/testing/selftests/kvm/Makefile
+++ b/tools/testing/selftests/kvm/Makefile
@@ -43,11 +43,11 @@ LIBKVM_aarch64 = lib/aarch64/processor.c lib/aarch64/ucall.c lib/aarch64/handler
 LIBKVM_s390x = lib/s390x/processor.c lib/s390x/ucall.c lib/s390x/diag318_test_handler.c
 LIBKVM_riscv = lib/riscv/processor.c lib/riscv/ucall.c
 
-TEST_GEN_PROGS_x86_64 = x86_64/cr4_cpuid_sync_test
+TEST_GEN_PROGS_x86_64 = x86_64/cpuid_test
+TEST_GEN_PROGS_x86_64 += x86_64/cr4_cpuid_sync_test
 TEST_GEN_PROGS_x86_64 += x86_64/get_msr_index_features
 TEST_GEN_PROGS_x86_64 += x86_64/evmcs_test
 TEST_GEN_PROGS_x86_64 += x86_64/emulator_error_test
-TEST_GEN_PROGS_x86_64 += x86_64/get_cpuid_test
 TEST_GEN_PROGS_x86_64 += x86_64/hyperv_clock
 TEST_GEN_PROGS_x86_64 += x86_64/hyperv_cpuid
 TEST_GEN_PROGS_x86_64 += x86_64/hyperv_features
@@ -56,6 +56,7 @@ TEST_GEN_PROGS_x86_64 += x86_64/kvm_pv_test
 TEST_GEN_PROGS_x86_64 += x86_64/mmio_warning_test
 TEST_GEN_PROGS_x86_64 += x86_64/mmu_role_test
 TEST_GEN_PROGS_x86_64 += x86_64/platform_info_test
+TEST_GEN_PROGS_x86_64 += x86_64/pmu_event_filter_test
 TEST_GEN_PROGS_x86_64 += x86_64/set_boot_cpu_id
 TEST_GEN_PROGS_x86_64 += x86_64/set_sregs_test
 TEST_GEN_PROGS_x86_64 += x86_64/smm_test
@@ -69,6 +70,7 @@ TEST_GEN_PROGS_x86_64 += x86_64/userspace_msr_exit_test
 TEST_GEN_PROGS_x86_64 += x86_64/vmx_apic_access_test
 TEST_GEN_PROGS_x86_64 += x86_64/vmx_close_while_nested_test
 TEST_GEN_PROGS_x86_64 += x86_64/vmx_dirty_log_test
+TEST_GEN_PROGS_x86_64 += x86_64/vmx_exception_with_invalid_guest_state
 TEST_GEN_PROGS_x86_64 += x86_64/vmx_invalid_nested_guest_state
 TEST_GEN_PROGS_x86_64 += x86_64/vmx_set_nested_state_test
 TEST_GEN_PROGS_x86_64 += x86_64/vmx_tsc_adjust_test
diff --git a/tools/testing/selftests/kvm/include/x86_64/processor.h b/tools/testing/selftests/kvm/include/x86_64/processor.h
index e94ba0fc67d8..423d8a61bd2e 100644
--- a/tools/testing/selftests/kvm/include/x86_64/processor.h
+++ b/tools/testing/selftests/kvm/include/x86_64/processor.h
@@ -364,6 +364,24 @@ static inline unsigned long get_xmm(int n)
 }
 
 bool is_intel_cpu(void);
+bool is_amd_cpu(void);
+
+static inline unsigned int x86_family(unsigned int eax)
+{
+	unsigned int x86;
+
+	x86 = (eax >> 8) & 0xf;
+
+	if (x86 == 0xf)
+		x86 += (eax >> 20) & 0xff;
+
+	return x86;
+}
+
+static inline unsigned int x86_model(unsigned int eax)
+{
+	return ((eax >> 12) & 0xf0) | ((eax >> 4) & 0x0f);
+}
 
 struct kvm_x86_state *vcpu_save_state(struct kvm_vm *vm, uint32_t vcpuid);
 void vcpu_load_state(struct kvm_vm *vm, uint32_t vcpuid,
@@ -375,6 +393,8 @@ uint64_t kvm_get_feature_msr(uint64_t msr_index);
 struct kvm_cpuid2 *kvm_get_supported_cpuid(void);
 
 struct kvm_cpuid2 *vcpu_get_cpuid(struct kvm_vm *vm, uint32_t vcpuid);
+int __vcpu_set_cpuid(struct kvm_vm *vm, uint32_t vcpuid,
+		     struct kvm_cpuid2 *cpuid);
 void vcpu_set_cpuid(struct kvm_vm *vm, uint32_t vcpuid,
 		    struct kvm_cpuid2 *cpuid);
 
@@ -419,6 +439,11 @@ void vm_set_page_table_entry(struct kvm_vm *vm, int vcpuid, uint64_t vaddr,
 			     uint64_t pte);
 
 /*
+ * get_cpuid() - find matching CPUID entry and return pointer to it.
+ */
+struct kvm_cpuid_entry2 *get_cpuid(struct kvm_cpuid2 *cpuid, uint32_t function,
+				   uint32_t index);
+/*
  * set_cpuid() - overwrites a matching cpuid entry with the provided value.
  *		 matches based on ent->function && ent->index. returns true
  *		 if a match was found and successfully overwritten.
diff --git a/tools/testing/selftests/kvm/lib/kvm_util.c b/tools/testing/selftests/kvm/lib/kvm_util.c
index 4a645dc77f34..8c53f96ab7fe 100644
--- a/tools/testing/selftests/kvm/lib/kvm_util.c
+++ b/tools/testing/selftests/kvm/lib/kvm_util.c
@@ -393,10 +393,12 @@ struct kvm_vm *vm_create_with_vcpus(enum vm_guest_mode mode, uint32_t nr_vcpus,
 	struct kvm_vm *vm;
 	int i;
 
+#ifdef __x86_64__
 	/*
 	 * Permission needs to be requested before KVM_SET_CPUID2.
 	 */
 	vm_xsave_req_perm();
+#endif
 
 	/* Force slot0 memory size not small than DEFAULT_GUEST_PHY_PAGES */
 	if (slot0_mem_pages < DEFAULT_GUEST_PHY_PAGES)
@@ -497,9 +499,11 @@ void kvm_vm_get_dirty_log(struct kvm_vm *vm, int slot, void *log)
 void kvm_vm_clear_dirty_log(struct kvm_vm *vm, int slot, void *log,
 			    uint64_t first_page, uint32_t num_pages)
 {
-	struct kvm_clear_dirty_log args = { .dirty_bitmap = log, .slot = slot,
-		                            .first_page = first_page,
-	                                    .num_pages = num_pages };
+	struct kvm_clear_dirty_log args = {
+		.dirty_bitmap = log, .slot = slot,
+		.first_page = first_page,
+		.num_pages = num_pages
+	};
 	int ret;
 
 	ret = ioctl(vm->fd, KVM_CLEAR_DIRTY_LOG, &args);
diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index babb0f28575c..5f9d7e91dc69 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -886,6 +886,17 @@ kvm_get_supported_cpuid_index(uint32_t function, uint32_t index)
 	return entry;
 }
 
+
+int __vcpu_set_cpuid(struct kvm_vm *vm, uint32_t vcpuid,
+		     struct kvm_cpuid2 *cpuid)
+{
+	struct vcpu *vcpu = vcpu_find(vm, vcpuid);
+
+	TEST_ASSERT(vcpu != NULL, "vcpu not found, vcpuid: %u", vcpuid);
+
+	return ioctl(vcpu->fd, KVM_SET_CPUID2, cpuid);
+}
+
 /*
  * VM VCPU CPUID Set
  *
@@ -903,12 +914,9 @@ kvm_get_supported_cpuid_index(uint32_t function, uint32_t index)
 void vcpu_set_cpuid(struct kvm_vm *vm,
 		uint32_t vcpuid, struct kvm_cpuid2 *cpuid)
 {
-	struct vcpu *vcpu = vcpu_find(vm, vcpuid);
 	int rc;
 
-	TEST_ASSERT(vcpu != NULL, "vcpu not found, vcpuid: %u", vcpuid);
-
-	rc = ioctl(vcpu->fd, KVM_SET_CPUID2, cpuid);
+	rc = __vcpu_set_cpuid(vm, vcpuid, cpuid);
 	TEST_ASSERT(rc == 0, "KVM_SET_CPUID2 failed, rc: %i errno: %i",
 		    rc, errno);
 
@@ -1136,25 +1144,25 @@ struct kvm_x86_state *vcpu_save_state(struct kvm_vm *vm, uint32_t vcpuid)
 	list = malloc(sizeof(*list) + nmsrs * sizeof(list->indices[0]));
 	list->nmsrs = nmsrs;
 	r = ioctl(vm->kvm_fd, KVM_GET_MSR_INDEX_LIST, list);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_MSR_INDEX_LIST, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_MSR_INDEX_LIST, r: %i",
+		    r);
 
 	state = malloc(sizeof(*state) + nmsrs * sizeof(state->msrs.entries[0]));
 	r = ioctl(vcpu->fd, KVM_GET_VCPU_EVENTS, &state->events);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_VCPU_EVENTS, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_VCPU_EVENTS, r: %i",
+		    r);
 
 	r = ioctl(vcpu->fd, KVM_GET_MP_STATE, &state->mp_state);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_MP_STATE, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_MP_STATE, r: %i",
+		    r);
 
 	r = ioctl(vcpu->fd, KVM_GET_REGS, &state->regs);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_REGS, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_REGS, r: %i",
+		    r);
 
 	r = vcpu_save_xsave_state(vm, vcpu, state);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_XSAVE, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_XSAVE, r: %i",
+		    r);
 
 	if (kvm_check_cap(KVM_CAP_XCRS)) {
 		r = ioctl(vcpu->fd, KVM_GET_XCRS, &state->xcrs);
@@ -1163,17 +1171,17 @@ struct kvm_x86_state *vcpu_save_state(struct kvm_vm *vm, uint32_t vcpuid)
 	}
 
 	r = ioctl(vcpu->fd, KVM_GET_SREGS, &state->sregs);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_SREGS, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_SREGS, r: %i",
+		    r);
 
 	if (nested_size) {
 		state->nested.size = sizeof(state->nested_);
 		r = ioctl(vcpu->fd, KVM_GET_NESTED_STATE, &state->nested);
 		TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_NESTED_STATE, r: %i",
-			r);
+			    r);
 		TEST_ASSERT(state->nested.size <= nested_size,
-			"Nested state size too big, %i (KVM_CHECK_CAP gave %i)",
-			state->nested.size, nested_size);
+			    "Nested state size too big, %i (KVM_CHECK_CAP gave %i)",
+			    state->nested.size, nested_size);
 	} else
 		state->nested.size = 0;
 
@@ -1181,12 +1189,12 @@ struct kvm_x86_state *vcpu_save_state(struct kvm_vm *vm, uint32_t vcpuid)
 	for (i = 0; i < nmsrs; i++)
 		state->msrs.entries[i].index = list->indices[i];
 	r = ioctl(vcpu->fd, KVM_GET_MSRS, &state->msrs);
-        TEST_ASSERT(r == nmsrs, "Unexpected result from KVM_GET_MSRS, r: %i (failed MSR was 0x%x)",
-                r, r == nmsrs ? -1 : list->indices[r]);
+	TEST_ASSERT(r == nmsrs, "Unexpected result from KVM_GET_MSRS, r: %i (failed MSR was 0x%x)",
+		    r, r == nmsrs ? -1 : list->indices[r]);
 
 	r = ioctl(vcpu->fd, KVM_GET_DEBUGREGS, &state->debugregs);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_DEBUGREGS, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_GET_DEBUGREGS, r: %i",
+		    r);
 
 	free(list);
 	return state;
@@ -1199,7 +1207,7 @@ void vcpu_load_state(struct kvm_vm *vm, uint32_t vcpuid, struct kvm_x86_state *s
 
 	r = ioctl(vcpu->fd, KVM_SET_SREGS, &state->sregs);
 	TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_SREGS, r: %i",
-                r);
+		    r);
 
 	r = ioctl(vcpu->fd, KVM_SET_MSRS, &state->msrs);
 	TEST_ASSERT(r == state->msrs.nmsrs,
@@ -1214,28 +1222,28 @@ void vcpu_load_state(struct kvm_vm *vm, uint32_t vcpuid, struct kvm_x86_state *s
 
 	r = ioctl(vcpu->fd, KVM_SET_XSAVE, state->xsave);
 	TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_XSAVE, r: %i",
-                r);
+		    r);
 
 	r = ioctl(vcpu->fd, KVM_SET_VCPU_EVENTS, &state->events);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_VCPU_EVENTS, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_VCPU_EVENTS, r: %i",
+		    r);
 
 	r = ioctl(vcpu->fd, KVM_SET_MP_STATE, &state->mp_state);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_MP_STATE, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_MP_STATE, r: %i",
+		    r);
 
 	r = ioctl(vcpu->fd, KVM_SET_DEBUGREGS, &state->debugregs);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_DEBUGREGS, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_DEBUGREGS, r: %i",
+		    r);
 
 	r = ioctl(vcpu->fd, KVM_SET_REGS, &state->regs);
-        TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_REGS, r: %i",
-                r);
+	TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_REGS, r: %i",
+		    r);
 
 	if (state->nested.size) {
 		r = ioctl(vcpu->fd, KVM_SET_NESTED_STATE, &state->nested);
 		TEST_ASSERT(r == 0, "Unexpected result from KVM_SET_NESTED_STATE, r: %i",
-			r);
+			    r);
 	}
 }
 
@@ -1245,10 +1253,10 @@ void kvm_x86_state_cleanup(struct kvm_x86_state *state)
 	free(state);
 }
 
-bool is_intel_cpu(void)
+static bool cpu_vendor_string_is(const char *vendor)
 {
+	const uint32_t *chunk = (const uint32_t *)vendor;
 	int eax, ebx, ecx, edx;
-	const uint32_t *chunk;
 	const int leaf = 0;
 
 	__asm__ __volatile__(
@@ -1257,10 +1265,22 @@ bool is_intel_cpu(void)
 		  "=c"(ecx), "=d"(edx)
 		: /* input */ "0"(leaf), "2"(0));
 
-	chunk = (const uint32_t *)("GenuineIntel");
 	return (ebx == chunk[0] && edx == chunk[1] && ecx == chunk[2]);
 }
 
+bool is_intel_cpu(void)
+{
+	return cpu_vendor_string_is("GenuineIntel");
+}
+
+/*
+ * Exclude early K5 samples with a vendor string of "AMDisbetter!"
+ */
+bool is_amd_cpu(void)
+{
+	return cpu_vendor_string_is("AuthenticAMD");
+}
+
 uint32_t kvm_get_cpuid_max_basic(void)
 {
 	return kvm_get_supported_cpuid_entry(0)->eax;
@@ -1384,6 +1404,23 @@ void assert_on_unhandled_exception(struct kvm_vm *vm, uint32_t vcpuid)
 	}
 }
 
+struct kvm_cpuid_entry2 *get_cpuid(struct kvm_cpuid2 *cpuid, uint32_t function,
+				   uint32_t index)
+{
+	int i;
+
+	for (i = 0; i < cpuid->nent; i++) {
+		struct kvm_cpuid_entry2 *cur = &cpuid->entries[i];
+
+		if (cur->function == function && cur->index == index)
+			return cur;
+	}
+
+	TEST_FAIL("CPUID function 0x%x index 0x%x not found ", function, index);
+
+	return NULL;
+}
+
 bool set_cpuid(struct kvm_cpuid2 *cpuid,
 	       struct kvm_cpuid_entry2 *ent)
 {
@@ -1479,22 +1516,6 @@ struct kvm_cpuid2 *vcpu_get_supported_hv_cpuid(struct kvm_vm *vm, uint32_t vcpui
 	return cpuid;
 }
 
-#define X86EMUL_CPUID_VENDOR_AuthenticAMD_ebx 0x68747541
-#define X86EMUL_CPUID_VENDOR_AuthenticAMD_ecx 0x444d4163
-#define X86EMUL_CPUID_VENDOR_AuthenticAMD_edx 0x69746e65
-
-static inline unsigned x86_family(unsigned int eax)
-{
-        unsigned int x86;
-
-        x86 = (eax >> 8) & 0xf;
-
-        if (x86 == 0xf)
-                x86 += (eax >> 20) & 0xff;
-
-        return x86;
-}
-
 unsigned long vm_compute_max_gfn(struct kvm_vm *vm)
 {
 	const unsigned long num_ht_pages = 12 << (30 - vm->page_shift); /* 12 GiB */
@@ -1504,11 +1525,7 @@ unsigned long vm_compute_max_gfn(struct kvm_vm *vm)
 	max_gfn = (1ULL << (vm->pa_bits - vm->page_shift)) - 1;
 
 	/* Avoid reserved HyperTransport region on AMD processors.  */
-	eax = ecx = 0;
-	cpuid(&eax, &ebx, &ecx, &edx);
-	if (ebx != X86EMUL_CPUID_VENDOR_AuthenticAMD_ebx ||
-	    ecx != X86EMUL_CPUID_VENDOR_AuthenticAMD_ecx ||
-	    edx != X86EMUL_CPUID_VENDOR_AuthenticAMD_edx)
+	if (!is_amd_cpu())
 		return max_gfn;
 
 	/* On parts with <40 physical address bits, the area is fully hidden */
@@ -1518,6 +1535,7 @@ unsigned long vm_compute_max_gfn(struct kvm_vm *vm)
 	/* Before family 17h, the HyperTransport area is just below 1T.  */
 	ht_gfn = (1 << 28) - num_ht_pages;
 	eax = 1;
+	ecx = 0;
 	cpuid(&eax, &ebx, &ecx, &edx);
 	if (x86_family(eax) < 0x17)
 		goto done;
diff --git a/tools/testing/selftests/kvm/x86_64/get_cpuid_test.c b/tools/testing/selftests/kvm/x86_64/cpuid_test.c
index a711f83749ea..16d2465c5634 100644
--- a/tools/testing/selftests/kvm/x86_64/get_cpuid_test.c
+++ b/tools/testing/selftests/kvm/x86_64/cpuid_test.c
@@ -154,6 +154,34 @@ struct kvm_cpuid2 *vcpu_alloc_cpuid(struct kvm_vm *vm, vm_vaddr_t *p_gva, struct
 	return guest_cpuids;
 }
 
+static void set_cpuid_after_run(struct kvm_vm *vm, struct kvm_cpuid2 *cpuid)
+{
+	struct kvm_cpuid_entry2 *ent;
+	int rc;
+	u32 eax, ebx, x;
+
+	/* Setting unmodified CPUID is allowed */
+	rc = __vcpu_set_cpuid(vm, VCPU_ID, cpuid);
+	TEST_ASSERT(!rc, "Setting unmodified CPUID after KVM_RUN failed: %d", rc);
+
+	/* Changing CPU features is forbidden */
+	ent = get_cpuid(cpuid, 0x7, 0);
+	ebx = ent->ebx;
+	ent->ebx--;
+	rc = __vcpu_set_cpuid(vm, VCPU_ID, cpuid);
+	TEST_ASSERT(rc, "Changing CPU features should fail");
+	ent->ebx = ebx;
+
+	/* Changing MAXPHYADDR is forbidden */
+	ent = get_cpuid(cpuid, 0x80000008, 0);
+	eax = ent->eax;
+	x = eax & 0xff;
+	ent->eax = (eax & ~0xffu) | (x - 1);
+	rc = __vcpu_set_cpuid(vm, VCPU_ID, cpuid);
+	TEST_ASSERT(rc, "Changing MAXPHYADDR should fail");
+	ent->eax = eax;
+}
+
 int main(void)
 {
 	struct kvm_cpuid2 *supp_cpuid, *cpuid2;
@@ -175,5 +203,7 @@ int main(void)
 	for (stage = 0; stage < 3; stage++)
 		run_vcpu(vm, VCPU_ID, stage);
 
+	set_cpuid_after_run(vm, cpuid2);
+
 	kvm_vm_free(vm);
 }
diff --git a/tools/testing/selftests/kvm/x86_64/pmu_event_filter_test.c b/tools/testing/selftests/kvm/x86_64/pmu_event_filter_test.c
new file mode 100644
index 000000000000..c715adcbd487
--- /dev/null
+++ b/tools/testing/selftests/kvm/x86_64/pmu_event_filter_test.c
@@ -0,0 +1,434 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Test for x86 KVM_SET_PMU_EVENT_FILTER.
+ *
+ * Copyright (C) 2022, Google LLC.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ *
+ * Verifies the expected behavior of allow lists and deny lists for
+ * virtual PMU events.
+ */
+
+#define _GNU_SOURCE /* for program_invocation_short_name */
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+/*
+ * In lieu of copying perf_event.h into tools...
+ */
+#define ARCH_PERFMON_EVENTSEL_OS			(1ULL << 17)
+#define ARCH_PERFMON_EVENTSEL_ENABLE			(1ULL << 22)
+
+union cpuid10_eax {
+	struct {
+		unsigned int version_id:8;
+		unsigned int num_counters:8;
+		unsigned int bit_width:8;
+		unsigned int mask_length:8;
+	} split;
+	unsigned int full;
+};
+
+union cpuid10_ebx {
+	struct {
+		unsigned int no_unhalted_core_cycles:1;
+		unsigned int no_instructions_retired:1;
+		unsigned int no_unhalted_reference_cycles:1;
+		unsigned int no_llc_reference:1;
+		unsigned int no_llc_misses:1;
+		unsigned int no_branch_instruction_retired:1;
+		unsigned int no_branch_misses_retired:1;
+	} split;
+	unsigned int full;
+};
+
+/* End of stuff taken from perf_event.h. */
+
+/* Oddly, this isn't in perf_event.h. */
+#define ARCH_PERFMON_BRANCHES_RETIRED		5
+
+#define VCPU_ID 0
+#define NUM_BRANCHES 42
+
+/*
+ * This is how the event selector and unit mask are stored in an AMD
+ * core performance event-select register. Intel's format is similar,
+ * but the event selector is only 8 bits.
+ */
+#define EVENT(select, umask) ((select & 0xf00UL) << 24 | (select & 0xff) | \
+			      (umask & 0xff) << 8)
+
+/*
+ * "Branch instructions retired", from the Intel SDM, volume 3,
+ * "Pre-defined Architectural Performance Events."
+ */
+
+#define INTEL_BR_RETIRED EVENT(0xc4, 0)
+
+/*
+ * "Retired branch instructions", from Processor Programming Reference
+ * (PPR) for AMD Family 17h Model 01h, Revision B1 Processors,
+ * Preliminary Processor Programming Reference (PPR) for AMD Family
+ * 17h Model 31h, Revision B0 Processors, and Preliminary Processor
+ * Programming Reference (PPR) for AMD Family 19h Model 01h, Revision
+ * B1 Processors Volume 1 of 2.
+ */
+
+#define AMD_ZEN_BR_RETIRED EVENT(0xc2, 0)
+
+/*
+ * This event list comprises Intel's eight architectural events plus
+ * AMD's "retired branch instructions" for Zen[123] (and possibly
+ * other AMD CPUs).
+ */
+static const uint64_t event_list[] = {
+	EVENT(0x3c, 0),
+	EVENT(0xc0, 0),
+	EVENT(0x3c, 1),
+	EVENT(0x2e, 0x4f),
+	EVENT(0x2e, 0x41),
+	EVENT(0xc4, 0),
+	EVENT(0xc5, 0),
+	EVENT(0xa4, 1),
+	AMD_ZEN_BR_RETIRED,
+};
+
+/*
+ * If we encounter a #GP during the guest PMU sanity check, then the guest
+ * PMU is not functional. Inform the hypervisor via GUEST_SYNC(0).
+ */
+static void guest_gp_handler(struct ex_regs *regs)
+{
+	GUEST_SYNC(0);
+}
+
+/*
+ * Check that we can write a new value to the given MSR and read it back.
+ * The caller should provide a non-empty set of bits that are safe to flip.
+ *
+ * Return on success. GUEST_SYNC(0) on error.
+ */
+static void check_msr(uint32_t msr, uint64_t bits_to_flip)
+{
+	uint64_t v = rdmsr(msr) ^ bits_to_flip;
+
+	wrmsr(msr, v);
+	if (rdmsr(msr) != v)
+		GUEST_SYNC(0);
+
+	v ^= bits_to_flip;
+	wrmsr(msr, v);
+	if (rdmsr(msr) != v)
+		GUEST_SYNC(0);
+}
+
+static void intel_guest_code(void)
+{
+	check_msr(MSR_CORE_PERF_GLOBAL_CTRL, 1);
+	check_msr(MSR_P6_EVNTSEL0, 0xffff);
+	check_msr(MSR_IA32_PMC0, 0xffff);
+	GUEST_SYNC(1);
+
+	for (;;) {
+		uint64_t br0, br1;
+
+		wrmsr(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+		wrmsr(MSR_P6_EVNTSEL0, ARCH_PERFMON_EVENTSEL_ENABLE |
+		      ARCH_PERFMON_EVENTSEL_OS | INTEL_BR_RETIRED);
+		wrmsr(MSR_CORE_PERF_GLOBAL_CTRL, 1);
+		br0 = rdmsr(MSR_IA32_PMC0);
+		__asm__ __volatile__("loop ." : "+c"((int){NUM_BRANCHES}));
+		br1 = rdmsr(MSR_IA32_PMC0);
+		GUEST_SYNC(br1 - br0);
+	}
+}
+
+/*
+ * To avoid needing a check for CPUID.80000001:ECX.PerfCtrExtCore[bit 23],
+ * this code uses the always-available, legacy K7 PMU MSRs, which alias to
+ * the first four of the six extended core PMU MSRs.
+ */
+static void amd_guest_code(void)
+{
+	check_msr(MSR_K7_EVNTSEL0, 0xffff);
+	check_msr(MSR_K7_PERFCTR0, 0xffff);
+	GUEST_SYNC(1);
+
+	for (;;) {
+		uint64_t br0, br1;
+
+		wrmsr(MSR_K7_EVNTSEL0, 0);
+		wrmsr(MSR_K7_EVNTSEL0, ARCH_PERFMON_EVENTSEL_ENABLE |
+		      ARCH_PERFMON_EVENTSEL_OS | AMD_ZEN_BR_RETIRED);
+		br0 = rdmsr(MSR_K7_PERFCTR0);
+		__asm__ __volatile__("loop ." : "+c"((int){NUM_BRANCHES}));
+		br1 = rdmsr(MSR_K7_PERFCTR0);
+		GUEST_SYNC(br1 - br0);
+	}
+}
+
+/*
+ * Run the VM to the next GUEST_SYNC(value), and return the value passed
+ * to the sync. Any other exit from the guest is fatal.
+ */
+static uint64_t run_vm_to_sync(struct kvm_vm *vm)
+{
+	struct kvm_run *run = vcpu_state(vm, VCPU_ID);
+	struct ucall uc;
+
+	vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(run->exit_reason == KVM_EXIT_IO,
+		    "Exit_reason other than KVM_EXIT_IO: %u (%s)\n",
+		    run->exit_reason,
+		    exit_reason_str(run->exit_reason));
+	get_ucall(vm, VCPU_ID, &uc);
+	TEST_ASSERT(uc.cmd == UCALL_SYNC,
+		    "Received ucall other than UCALL_SYNC: %lu", uc.cmd);
+	return uc.args[1];
+}
+
+/*
+ * In a nested environment or if the vPMU is disabled, the guest PMU
+ * might not work as architected (accessing the PMU MSRs may raise
+ * #GP, or writes could simply be discarded). In those situations,
+ * there is no point in running these tests. The guest code will perform
+ * a sanity check and then GUEST_SYNC(success). In the case of failure,
+ * the behavior of the guest on resumption is undefined.
+ */
+static bool sanity_check_pmu(struct kvm_vm *vm)
+{
+	bool success;
+
+	vm_install_exception_handler(vm, GP_VECTOR, guest_gp_handler);
+	success = run_vm_to_sync(vm);
+	vm_install_exception_handler(vm, GP_VECTOR, NULL);
+
+	return success;
+}
+
+static struct kvm_pmu_event_filter *make_pmu_event_filter(uint32_t nevents)
+{
+	struct kvm_pmu_event_filter *f;
+	int size = sizeof(*f) + nevents * sizeof(f->events[0]);
+
+	f = malloc(size);
+	TEST_ASSERT(f, "Out of memory");
+	memset(f, 0, size);
+	f->nevents = nevents;
+	return f;
+}
+
+static struct kvm_pmu_event_filter *event_filter(uint32_t action)
+{
+	struct kvm_pmu_event_filter *f;
+	int i;
+
+	f = make_pmu_event_filter(ARRAY_SIZE(event_list));
+	f->action = action;
+	for (i = 0; i < ARRAY_SIZE(event_list); i++)
+		f->events[i] = event_list[i];
+
+	return f;
+}
+
+/*
+ * Remove the first occurrence of 'event' (if any) from the filter's
+ * event list.
+ */
+static struct kvm_pmu_event_filter *remove_event(struct kvm_pmu_event_filter *f,
+						 uint64_t event)
+{
+	bool found = false;
+	int i;
+
+	for (i = 0; i < f->nevents; i++) {
+		if (found)
+			f->events[i - 1] = f->events[i];
+		else
+			found = f->events[i] == event;
+	}
+	if (found)
+		f->nevents--;
+	return f;
+}
+
+static void test_without_filter(struct kvm_vm *vm)
+{
+	uint64_t count = run_vm_to_sync(vm);
+
+	if (count != NUM_BRANCHES)
+		pr_info("%s: Branch instructions retired = %lu (expected %u)\n",
+			__func__, count, NUM_BRANCHES);
+	TEST_ASSERT(count, "Allowed PMU event is not counting");
+}
+
+static uint64_t test_with_filter(struct kvm_vm *vm,
+				 struct kvm_pmu_event_filter *f)
+{
+	vm_ioctl(vm, KVM_SET_PMU_EVENT_FILTER, (void *)f);
+	return run_vm_to_sync(vm);
+}
+
+static void test_member_deny_list(struct kvm_vm *vm)
+{
+	struct kvm_pmu_event_filter *f = event_filter(KVM_PMU_EVENT_DENY);
+	uint64_t count = test_with_filter(vm, f);
+
+	free(f);
+	if (count)
+		pr_info("%s: Branch instructions retired = %lu (expected 0)\n",
+			__func__, count);
+	TEST_ASSERT(!count, "Disallowed PMU Event is counting");
+}
+
+static void test_member_allow_list(struct kvm_vm *vm)
+{
+	struct kvm_pmu_event_filter *f = event_filter(KVM_PMU_EVENT_ALLOW);
+	uint64_t count = test_with_filter(vm, f);
+
+	free(f);
+	if (count != NUM_BRANCHES)
+		pr_info("%s: Branch instructions retired = %lu (expected %u)\n",
+			__func__, count, NUM_BRANCHES);
+	TEST_ASSERT(count, "Allowed PMU event is not counting");
+}
+
+static void test_not_member_deny_list(struct kvm_vm *vm)
+{
+	struct kvm_pmu_event_filter *f = event_filter(KVM_PMU_EVENT_DENY);
+	uint64_t count;
+
+	remove_event(f, INTEL_BR_RETIRED);
+	remove_event(f, AMD_ZEN_BR_RETIRED);
+	count = test_with_filter(vm, f);
+	free(f);
+	if (count != NUM_BRANCHES)
+		pr_info("%s: Branch instructions retired = %lu (expected %u)\n",
+			__func__, count, NUM_BRANCHES);
+	TEST_ASSERT(count, "Allowed PMU event is not counting");
+}
+
+static void test_not_member_allow_list(struct kvm_vm *vm)
+{
+	struct kvm_pmu_event_filter *f = event_filter(KVM_PMU_EVENT_ALLOW);
+	uint64_t count;
+
+	remove_event(f, INTEL_BR_RETIRED);
+	remove_event(f, AMD_ZEN_BR_RETIRED);
+	count = test_with_filter(vm, f);
+	free(f);
+	if (count)
+		pr_info("%s: Branch instructions retired = %lu (expected 0)\n",
+			__func__, count);
+	TEST_ASSERT(!count, "Disallowed PMU Event is counting");
+}
+
+/*
+ * Check for a non-zero PMU version, at least one general-purpose
+ * counter per logical processor, an EBX bit vector of length greater
+ * than 5, and EBX[5] clear.
+ */
+static bool check_intel_pmu_leaf(struct kvm_cpuid_entry2 *entry)
+{
+	union cpuid10_eax eax = { .full = entry->eax };
+	union cpuid10_ebx ebx = { .full = entry->ebx };
+
+	return eax.split.version_id && eax.split.num_counters > 0 &&
+		eax.split.mask_length > ARCH_PERFMON_BRANCHES_RETIRED &&
+		!ebx.split.no_branch_instruction_retired;
+}
+
+/*
+ * Note that CPUID leaf 0xa is Intel-specific. This leaf should be
+ * clear on AMD hardware.
+ */
+static bool use_intel_pmu(void)
+{
+	struct kvm_cpuid_entry2 *entry;
+
+	entry = kvm_get_supported_cpuid_index(0xa, 0);
+	return is_intel_cpu() && entry && check_intel_pmu_leaf(entry);
+}
+
+static bool is_zen1(uint32_t eax)
+{
+	return x86_family(eax) == 0x17 && x86_model(eax) <= 0x0f;
+}
+
+static bool is_zen2(uint32_t eax)
+{
+	return x86_family(eax) == 0x17 &&
+		x86_model(eax) >= 0x30 && x86_model(eax) <= 0x3f;
+}
+
+static bool is_zen3(uint32_t eax)
+{
+	return x86_family(eax) == 0x19 && x86_model(eax) <= 0x0f;
+}
+
+/*
+ * Determining AMD support for a PMU event requires consulting the AMD
+ * PPR for the CPU or reference material derived therefrom. The AMD
+ * test code herein has been verified to work on Zen1, Zen2, and Zen3.
+ *
+ * Feel free to add more AMD CPUs that are documented to support event
+ * select 0xc2 umask 0 as "retired branch instructions."
+ */
+static bool use_amd_pmu(void)
+{
+	struct kvm_cpuid_entry2 *entry;
+
+	entry = kvm_get_supported_cpuid_index(1, 0);
+	return is_amd_cpu() && entry &&
+		(is_zen1(entry->eax) ||
+		 is_zen2(entry->eax) ||
+		 is_zen3(entry->eax));
+}
+
+int main(int argc, char *argv[])
+{
+	void (*guest_code)(void) = NULL;
+	struct kvm_vm *vm;
+	int r;
+
+	/* Tell stdout not to buffer its content */
+	setbuf(stdout, NULL);
+
+	r = kvm_check_cap(KVM_CAP_PMU_EVENT_FILTER);
+	if (!r) {
+		print_skip("KVM_CAP_PMU_EVENT_FILTER not supported");
+		exit(KSFT_SKIP);
+	}
+
+	if (use_intel_pmu())
+		guest_code = intel_guest_code;
+	else if (use_amd_pmu())
+		guest_code = amd_guest_code;
+
+	if (!guest_code) {
+		print_skip("Don't know how to test this guest PMU");
+		exit(KSFT_SKIP);
+	}
+
+	vm = vm_create_default(VCPU_ID, 0, guest_code);
+
+	vm_init_descriptor_tables(vm);
+	vcpu_init_descriptor_tables(vm, VCPU_ID);
+
+	if (!sanity_check_pmu(vm)) {
+		print_skip("Guest PMU is not functional");
+		exit(KSFT_SKIP);
+	}
+
+	test_without_filter(vm);
+	test_member_deny_list(vm);
+	test_member_allow_list(vm);
+	test_not_member_deny_list(vm);
+	test_not_member_allow_list(vm);
+
+	kvm_vm_free(vm);
+
+	return 0;
+}
diff --git a/tools/testing/selftests/kvm/x86_64/tsc_msrs_test.c b/tools/testing/selftests/kvm/x86_64/tsc_msrs_test.c
index 5a6a662f2e59..a426078b16a3 100644
--- a/tools/testing/selftests/kvm/x86_64/tsc_msrs_test.c
+++ b/tools/testing/selftests/kvm/x86_64/tsc_msrs_test.c
@@ -77,8 +77,8 @@ static void run_vcpu(struct kvm_vm *vm, uint32_t vcpuid, int stage)
 	switch (get_ucall(vm, vcpuid, &uc)) {
 	case UCALL_SYNC:
 		TEST_ASSERT(!strcmp((const char *)uc.args[0], "hello") &&
-                            uc.args[1] == stage + 1, "Stage %d: Unexpected register values vmexit, got %lx",
-                            stage + 1, (ulong)uc.args[1]);
+			    uc.args[1] == stage + 1, "Stage %d: Unexpected register values vmexit, got %lx",
+			    stage + 1, (ulong)uc.args[1]);
 		return;
 	case UCALL_DONE:
 		return;
diff --git a/tools/testing/selftests/kvm/x86_64/vmx_close_while_nested_test.c b/tools/testing/selftests/kvm/x86_64/vmx_close_while_nested_test.c
index 2835a17f1b7a..edac8839e717 100644
--- a/tools/testing/selftests/kvm/x86_64/vmx_close_while_nested_test.c
+++ b/tools/testing/selftests/kvm/x86_64/vmx_close_while_nested_test.c
@@ -30,8 +30,8 @@ static struct kvm_vm *vm;
 static void l2_guest_code(void)
 {
 	/* Exit to L0 */
-        asm volatile("inb %%dx, %%al"
-                     : : [port] "d" (PORT_L0_EXIT) : "rax");
+	asm volatile("inb %%dx, %%al"
+		     : : [port] "d" (PORT_L0_EXIT) : "rax");
 }
 
 static void l1_guest_code(struct vmx_pages *vmx_pages)
diff --git a/tools/testing/selftests/kvm/x86_64/vmx_exception_with_invalid_guest_state.c b/tools/testing/selftests/kvm/x86_64/vmx_exception_with_invalid_guest_state.c
new file mode 100644
index 000000000000..27a850f3d7ce
--- /dev/null
+++ b/tools/testing/selftests/kvm/x86_64/vmx_exception_with_invalid_guest_state.c
@@ -0,0 +1,139 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+#include <signal.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <sys/time.h>
+
+#include "kselftest.h"
+
+#define VCPU_ID	0
+
+static struct kvm_vm *vm;
+
+static void guest_ud_handler(struct ex_regs *regs)
+{
+	/* Loop on the ud2 until guest state is made invalid. */
+}
+
+static void guest_code(void)
+{
+	asm volatile("ud2");
+}
+
+static void __run_vcpu_with_invalid_state(void)
+{
+	struct kvm_run *run = vcpu_state(vm, VCPU_ID);
+
+	vcpu_run(vm, VCPU_ID);
+
+	TEST_ASSERT(run->exit_reason == KVM_EXIT_INTERNAL_ERROR,
+		    "Expected KVM_EXIT_INTERNAL_ERROR, got %d (%s)\n",
+		    run->exit_reason, exit_reason_str(run->exit_reason));
+	TEST_ASSERT(run->emulation_failure.suberror == KVM_INTERNAL_ERROR_EMULATION,
+		    "Expected emulation failure, got %d\n",
+		    run->emulation_failure.suberror);
+}
+
+static void run_vcpu_with_invalid_state(void)
+{
+	/*
+	 * Always run twice to verify KVM handles the case where _KVM_ queues
+	 * an exception with invalid state and then exits to userspace, i.e.
+	 * that KVM doesn't explode if userspace ignores the initial error.
+	 */
+	__run_vcpu_with_invalid_state();
+	__run_vcpu_with_invalid_state();
+}
+
+static void set_timer(void)
+{
+	struct itimerval timer;
+
+	timer.it_value.tv_sec  = 0;
+	timer.it_value.tv_usec = 200;
+	timer.it_interval = timer.it_value;
+	ASSERT_EQ(setitimer(ITIMER_REAL, &timer, NULL), 0);
+}
+
+static void set_or_clear_invalid_guest_state(bool set)
+{
+	static struct kvm_sregs sregs;
+
+	if (!sregs.cr0)
+		vcpu_sregs_get(vm, VCPU_ID, &sregs);
+	sregs.tr.unusable = !!set;
+	vcpu_sregs_set(vm, VCPU_ID, &sregs);
+}
+
+static void set_invalid_guest_state(void)
+{
+	set_or_clear_invalid_guest_state(true);
+}
+
+static void clear_invalid_guest_state(void)
+{
+	set_or_clear_invalid_guest_state(false);
+}
+
+static void sigalrm_handler(int sig)
+{
+	struct kvm_vcpu_events events;
+
+	TEST_ASSERT(sig == SIGALRM, "Unexpected signal = %d", sig);
+
+	vcpu_events_get(vm, VCPU_ID, &events);
+
+	/*
+	 * If an exception is pending, attempt KVM_RUN with invalid guest,
+	 * otherwise rearm the timer and keep doing so until the timer fires
+	 * between KVM queueing an exception and re-entering the guest.
+	 */
+	if (events.exception.pending) {
+		set_invalid_guest_state();
+		run_vcpu_with_invalid_state();
+	} else {
+		set_timer();
+	}
+}
+
+int main(int argc, char *argv[])
+{
+	if (!is_intel_cpu() || vm_is_unrestricted_guest(NULL)) {
+		print_skip("Must be run with kvm_intel.unrestricted_guest=0");
+		exit(KSFT_SKIP);
+	}
+
+	vm = vm_create_default(VCPU_ID, 0, (void *)guest_code);
+
+	vm_init_descriptor_tables(vm);
+	vcpu_init_descriptor_tables(vm, VCPU_ID);
+
+	vm_install_exception_handler(vm, UD_VECTOR, guest_ud_handler);
+
+	/*
+	 * Stuff invalid guest state for L2 by making TR unusuable.  The next
+	 * KVM_RUN should induce a TRIPLE_FAULT in L2 as KVM doesn't support
+	 * emulating invalid guest state for L2.
+	 */
+	set_invalid_guest_state();
+	run_vcpu_with_invalid_state();
+
+	/*
+	 * Verify KVM also handles the case where userspace gains control while
+	 * an exception is pending and stuffs invalid state.  Run with valid
+	 * guest state and a timer firing every 200us, and attempt to enter the
+	 * guest with invalid state when the handler interrupts KVM with an
+	 * exception pending.
+	 */
+	clear_invalid_guest_state();
+	TEST_ASSERT(signal(SIGALRM, sigalrm_handler) != SIG_ERR,
+		    "Failed to register SIGALRM handler, errno = %d (%s)",
+		    errno, strerror(errno));
+
+	set_timer();
+	run_vcpu_with_invalid_state();
+}
diff --git a/tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c b/tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c
index 478e0ae8b93e..865e17146815 100644
--- a/tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c
+++ b/tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c
@@ -46,20 +46,20 @@ static struct kvm_vm *vm;
 #define MIN_STEAL_TIME		50000
 
 struct pvclock_vcpu_time_info {
-        u32   version;
-        u32   pad0;
-        u64   tsc_timestamp;
-        u64   system_time;
-        u32   tsc_to_system_mul;
-        s8    tsc_shift;
-        u8    flags;
-        u8    pad[2];
+	u32   version;
+	u32   pad0;
+	u64   tsc_timestamp;
+	u64   system_time;
+	u32   tsc_to_system_mul;
+	s8    tsc_shift;
+	u8    flags;
+	u8    pad[2];
 } __attribute__((__packed__)); /* 32 bytes */
 
 struct pvclock_wall_clock {
-        u32   version;
-        u32   sec;
-        u32   nsec;
+	u32   version;
+	u32   sec;
+	u32   nsec;
 } __attribute__((__packed__));
 
 struct vcpu_runstate_info {
@@ -74,11 +74,11 @@ struct arch_vcpu_info {
 };
 
 struct vcpu_info {
-        uint8_t evtchn_upcall_pending;
-        uint8_t evtchn_upcall_mask;
-        unsigned long evtchn_pending_sel;
-        struct arch_vcpu_info arch;
-        struct pvclock_vcpu_time_info time;
+	uint8_t evtchn_upcall_pending;
+	uint8_t evtchn_upcall_mask;
+	unsigned long evtchn_pending_sel;
+	struct arch_vcpu_info arch;
+	struct pvclock_vcpu_time_info time;
 }; /* 64 bytes (x86) */
 
 struct shared_info {
@@ -493,7 +493,7 @@ int main(int argc, char *argv[])
 
 	vm_ts.tv_sec = wc->sec;
 	vm_ts.tv_nsec = wc->nsec;
-        TEST_ASSERT(wc->version && !(wc->version & 1),
+	TEST_ASSERT(wc->version && !(wc->version & 1),
 		    "Bad wallclock version %x", wc->version);
 	TEST_ASSERT(cmp_timespec(&min_ts, &vm_ts) <= 0, "VM time too old");
 	TEST_ASSERT(cmp_timespec(&max_ts, &vm_ts) >= 0, "VM time too new");