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

Pull kvm fixes from Paolo Bonzini:
 "Miscellaneous bugfixes.

  The main interesting one is a NULL pointer dereference reported by
  syzkaller ("KVM: x86: Immediately reset the MMU context when the SMM
  flag is cleared")"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: selftests: Fix kvm_check_cap() assertion
  KVM: x86/mmu: Calculate and check "full" mmu_role for nested MMU
  KVM: X86: Fix x86_emulator slab cache leak
  KVM: SVM: Call SEV Guest Decommission if ASID binding fails
  KVM: x86: Immediately reset the MMU context when the SMM flag is cleared
  KVM: x86: Fix fall-through warnings for Clang
  KVM: SVM: fix doc warnings
  KVM: selftests: Fix compiling errors when initializing the static structure
  kvm: LAPIC: Restore guard to prevent illegal APIC register access

arch/x86/kvm/cpuid.c

@@ -655,6 +655,7 @@ static int __do_cpuid_func_emulated(struct kvm_cpuid_array *array, u32 func)
 		if (kvm_cpu_cap_has(X86_FEATURE_RDTSCP))
 			entry->ecx = F(RDPID);
 		++array->nent;
+		break;
 	default:
 		break;
 	}

arch/x86/kvm/lapic.c

@@ -1410,6 +1410,9 @@ int kvm_lapic_reg_read(struct kvm_lapic *apic, u32 offset, int len,
 	if (!apic_x2apic_mode(apic))
 		valid_reg_mask |= APIC_REG_MASK(APIC_ARBPRI);
 
+	if (alignment + len > 4)
+		return 1;
+
 	if (offset > 0x3f0 || !(valid_reg_mask & APIC_REG_MASK(offset)))
 		return 1;
 

arch/x86/kvm/mmu/mmu.c

@@ -4739,9 +4739,33 @@ static void init_kvm_softmmu(struct kvm_vcpu *vcpu)
 	context->inject_page_fault = kvm_inject_page_fault;
 }
 
+static union kvm_mmu_role kvm_calc_nested_mmu_role(struct kvm_vcpu *vcpu)
+{
+	union kvm_mmu_role role = kvm_calc_shadow_root_page_role_common(vcpu, false);
+
+	/*
+	 * Nested MMUs are used only for walking L2's gva->gpa, they never have
+	 * shadow pages of their own and so "direct" has no meaning.   Set it
+	 * to "true" to try to detect bogus usage of the nested MMU.
+	 */
+	role.base.direct = true;
+
+	if (!is_paging(vcpu))
+		role.base.level = 0;
+	else if (is_long_mode(vcpu))
+		role.base.level = is_la57_mode(vcpu) ? PT64_ROOT_5LEVEL :
+						       PT64_ROOT_4LEVEL;
+	else if (is_pae(vcpu))
+		role.base.level = PT32E_ROOT_LEVEL;
+	else
+		role.base.level = PT32_ROOT_LEVEL;
+
+	return role;
+}
+
 static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
 {
-	union kvm_mmu_role new_role = kvm_calc_mmu_role_common(vcpu, false);
+	union kvm_mmu_role new_role = kvm_calc_nested_mmu_role(vcpu);
 	struct kvm_mmu *g_context = &vcpu->arch.nested_mmu;
 
 	if (new_role.as_u64 == g_context->mmu_role.as_u64)

arch/x86/kvm/svm/avic.c

@@ -221,7 +221,7 @@ static u64 *avic_get_physical_id_entry(struct kvm_vcpu *vcpu,
 	return &avic_physical_id_table[index];
 }
 
-/**
+/*
  * Note:
  * AVIC hardware walks the nested page table to check permissions,
  * but does not use the SPA address specified in the leaf page
@@ -764,7 +764,7 @@ static int svm_ir_list_add(struct vcpu_svm *svm, struct amd_iommu_pi_data *pi)
 	return ret;
 }
 
-/**
+/*
  * Note:
  * The HW cannot support posting multicast/broadcast
  * interrupts to a vCPU. So, we still use legacy interrupt
@@ -1005,7 +1005,7 @@ void avic_vcpu_put(struct kvm_vcpu *vcpu)
 	WRITE_ONCE(*(svm->avic_physical_id_cache), entry);
 }
 
-/**
+/*
  * This function is called during VCPU halt/unhalt.
  */
 static void avic_set_running(struct kvm_vcpu *vcpu, bool is_run)

arch/x86/kvm/svm/sev.c

@@ -199,9 +199,19 @@ static void sev_asid_free(struct kvm_sev_info *sev)
 	sev->misc_cg = NULL;
 }
 
-static void sev_unbind_asid(struct kvm *kvm, unsigned int handle)
+static void sev_decommission(unsigned int handle)
 {
 	struct sev_data_decommission decommission;
+
+	if (!handle)
+		return;
+
+	decommission.handle = handle;
+	sev_guest_decommission(&decommission, NULL);
+}
+
+static void sev_unbind_asid(struct kvm *kvm, unsigned int handle)
+{
 	struct sev_data_deactivate deactivate;
 
 	if (!handle)
@@ -214,9 +224,7 @@ static void sev_unbind_asid(struct kvm *kvm, unsigned int handle)
 	sev_guest_deactivate(&deactivate, NULL);
 	up_read(&sev_deactivate_lock);
 
-	/* decommission handle */
-	decommission.handle = handle;
-	sev_guest_decommission(&decommission, NULL);
+	sev_decommission(handle);
 }
 
 static int sev_guest_init(struct kvm *kvm, struct kvm_sev_cmd *argp)
@@ -341,8 +349,10 @@ static int sev_launch_start(struct kvm *kvm, struct kvm_sev_cmd *argp)
 
 	/* Bind ASID to this guest */
 	ret = sev_bind_asid(kvm, start.handle, error);
-	if (ret)
+	if (ret) {
+		sev_decommission(start.handle);
 		goto e_free_session;
+	}
 
 	/* return handle to userspace */
 	params.handle = start.handle;

arch/x86/kvm/vmx/vmx.c

@@ -6247,6 +6247,7 @@ void vmx_set_virtual_apic_mode(struct kvm_vcpu *vcpu)
 	switch (kvm_get_apic_mode(vcpu)) {
 	case LAPIC_MODE_INVALID:
 		WARN_ONCE(true, "Invalid local APIC state");
+		break;
 	case LAPIC_MODE_DISABLED:
 		break;
 	case LAPIC_MODE_XAPIC:

arch/x86/kvm/x86.c

@@ -7106,7 +7106,10 @@ static unsigned emulator_get_hflags(struct x86_emulate_ctxt *ctxt)
 
 static void emulator_set_hflags(struct x86_emulate_ctxt *ctxt, unsigned emul_flags)
 {
-	emul_to_vcpu(ctxt)->arch.hflags = emul_flags;
+	struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt);
+
+	vcpu->arch.hflags = emul_flags;
+	kvm_mmu_reset_context(vcpu);
 }
 
 static int emulator_pre_leave_smm(struct x86_emulate_ctxt *ctxt,
@@ -8258,6 +8261,7 @@ void kvm_arch_exit(void)
 	kvm_x86_ops.hardware_enable = NULL;
 	kvm_mmu_module_exit();
 	free_percpu(user_return_msrs);
+	kmem_cache_destroy(x86_emulator_cache);
 	kmem_cache_destroy(x86_fpu_cache);
 #ifdef CONFIG_KVM_XEN
 	static_key_deferred_flush(&kvm_xen_enabled);

tools/testing/selftests/kvm/lib/kvm_util.c

@@ -82,7 +82,7 @@ int kvm_check_cap(long cap)
 
 	kvm_fd = open_kvm_dev_path_or_exit();
 	ret = ioctl(kvm_fd, KVM_CHECK_EXTENSION, cap);
-	TEST_ASSERT(ret != -1, "KVM_CHECK_EXTENSION IOCTL failed,\n"
+	TEST_ASSERT(ret >= 0, "KVM_CHECK_EXTENSION IOCTL failed,\n"
 		"  rc: %i errno: %i", ret, errno);
 
 	close(kvm_fd);

tools/testing/selftests/kvm/lib/test_util.c

@@ -166,75 +166,75 @@ size_t get_def_hugetlb_pagesz(void)
 	return 0;
 }
 
+#define ANON_FLAGS	(MAP_PRIVATE | MAP_ANONYMOUS)
+#define ANON_HUGE_FLAGS	(ANON_FLAGS | MAP_HUGETLB)
+
 const struct vm_mem_backing_src_alias *vm_mem_backing_src_alias(uint32_t i)
 {
-	static const int anon_flags = MAP_PRIVATE | MAP_ANONYMOUS;
-	static const int anon_huge_flags = anon_flags | MAP_HUGETLB;
-
 	static const struct vm_mem_backing_src_alias aliases[] = {
 		[VM_MEM_SRC_ANONYMOUS] = {
 			.name = "anonymous",
-			.flag = anon_flags,
+			.flag = ANON_FLAGS,
 		},
 		[VM_MEM_SRC_ANONYMOUS_THP] = {
 			.name = "anonymous_thp",
-			.flag = anon_flags,
+			.flag = ANON_FLAGS,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB] = {
 			.name = "anonymous_hugetlb",
-			.flag = anon_huge_flags,
+			.flag = ANON_HUGE_FLAGS,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_16KB] = {
 			.name = "anonymous_hugetlb_16kb",
-			.flag = anon_huge_flags | MAP_HUGE_16KB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_16KB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_64KB] = {
 			.name = "anonymous_hugetlb_64kb",
-			.flag = anon_huge_flags | MAP_HUGE_64KB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_64KB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_512KB] = {
 			.name = "anonymous_hugetlb_512kb",
-			.flag = anon_huge_flags | MAP_HUGE_512KB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_512KB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_1MB] = {
 			.name = "anonymous_hugetlb_1mb",
-			.flag = anon_huge_flags | MAP_HUGE_1MB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_1MB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_2MB] = {
 			.name = "anonymous_hugetlb_2mb",
-			.flag = anon_huge_flags | MAP_HUGE_2MB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_2MB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_8MB] = {
 			.name = "anonymous_hugetlb_8mb",
-			.flag = anon_huge_flags | MAP_HUGE_8MB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_8MB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_16MB] = {
 			.name = "anonymous_hugetlb_16mb",
-			.flag = anon_huge_flags | MAP_HUGE_16MB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_16MB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_32MB] = {
 			.name = "anonymous_hugetlb_32mb",
-			.flag = anon_huge_flags | MAP_HUGE_32MB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_32MB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_256MB] = {
 			.name = "anonymous_hugetlb_256mb",
-			.flag = anon_huge_flags | MAP_HUGE_256MB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_256MB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_512MB] = {
 			.name = "anonymous_hugetlb_512mb",
-			.flag = anon_huge_flags | MAP_HUGE_512MB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_512MB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_1GB] = {
 			.name = "anonymous_hugetlb_1gb",
-			.flag = anon_huge_flags | MAP_HUGE_1GB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_1GB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_2GB] = {
 			.name = "anonymous_hugetlb_2gb",
-			.flag = anon_huge_flags | MAP_HUGE_2GB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_2GB,
 		},
 		[VM_MEM_SRC_ANONYMOUS_HUGETLB_16GB] = {
 			.name = "anonymous_hugetlb_16gb",
-			.flag = anon_huge_flags | MAP_HUGE_16GB,
+			.flag = ANON_HUGE_FLAGS | MAP_HUGE_16GB,
 		},
 		[VM_MEM_SRC_SHMEM] = {
 			.name = "shmem",