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

Pull KVM fixes from Paolo Bonzini:
 "Bugfixes and strengthening the validity checks on inputs from new
  userspace APIs.

  Now I know why I shouldn't prepare pull requests on the weekend, it's
  hard to concentrate if your son is shouting about his latest Minecraft
  builds in your ear. Fortunately all the patches were ready and I just
  had to check the test results..."

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: SVM: Fix disable pause loop exit/pause filtering capability on SVM
  KVM: LAPIC: Prevent setting the tscdeadline timer if the lapic is hw disabled
  KVM: arm64: Don't inherit exec permission across page-table levels
  KVM: arm64: Prevent vcpu_has_ptrauth from generating OOL functions
  KVM: nVMX: check for invalid hdr.vmx.flags
  KVM: nVMX: check for required but missing VMCS12 in KVM_SET_NESTED_STATE
  selftests: kvm: do not set guest mode flag

Author: torvalds

arch/arm64/include/asm/kvm_host.h

@@ -380,9 +380,14 @@ struct kvm_vcpu_arch {
 #define vcpu_has_sve(vcpu) (system_supports_sve() && \
 			    ((vcpu)->arch.flags & KVM_ARM64_GUEST_HAS_SVE))
 
-#define vcpu_has_ptrauth(vcpu)	((system_supports_address_auth() || \
-				  system_supports_generic_auth()) && \
-				 ((vcpu)->arch.flags & KVM_ARM64_GUEST_HAS_PTRAUTH))
+#ifdef CONFIG_ARM64_PTR_AUTH
+#define vcpu_has_ptrauth(vcpu)						\
+	((cpus_have_final_cap(ARM64_HAS_ADDRESS_AUTH) ||		\
+	  cpus_have_final_cap(ARM64_HAS_GENERIC_AUTH)) &&		\
+	 (vcpu)->arch.flags & KVM_ARM64_GUEST_HAS_PTRAUTH)
+#else
+#define vcpu_has_ptrauth(vcpu)		false
+#endif
 
 #define vcpu_gp_regs(v)		(&(v)->arch.ctxt.gp_regs)
 

arch/arm64/kvm/mmu.c

@@ -1326,7 +1326,7 @@ static bool stage2_get_leaf_entry(struct kvm *kvm, phys_addr_t addr,
 	return true;
 }
 
-static bool stage2_is_exec(struct kvm *kvm, phys_addr_t addr)
+static bool stage2_is_exec(struct kvm *kvm, phys_addr_t addr, unsigned long sz)
 {
 	pud_t *pudp;
 	pmd_t *pmdp;
@@ -1338,11 +1338,11 @@ static bool stage2_is_exec(struct kvm *kvm, phys_addr_t addr)
 		return false;
 
 	if (pudp)
-		return kvm_s2pud_exec(pudp);
+		return sz <= PUD_SIZE && kvm_s2pud_exec(pudp);
 	else if (pmdp)
-		return kvm_s2pmd_exec(pmdp);
+		return sz <= PMD_SIZE && kvm_s2pmd_exec(pmdp);
 	else
-		return kvm_s2pte_exec(ptep);
+		return sz == PAGE_SIZE && kvm_s2pte_exec(ptep);
 }
 
 static int stage2_set_pte(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
@@ -1958,7 +1958,8 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 	 * execute permissions, and we preserve whatever we have.
 	 */
 	needs_exec = exec_fault ||
-		(fault_status == FSC_PERM && stage2_is_exec(kvm, fault_ipa));
+		(fault_status == FSC_PERM &&
+		 stage2_is_exec(kvm, fault_ipa, vma_pagesize));
 
 	if (vma_pagesize == PUD_SIZE) {
 		pud_t new_pud = kvm_pfn_pud(pfn, mem_type);

arch/x86/kvm/lapic.c

@@ -2195,7 +2195,7 @@ void kvm_set_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu, u64 data)
 {
 	struct kvm_lapic *apic = vcpu->arch.apic;
 
-	if (!lapic_in_kernel(vcpu) || apic_lvtt_oneshot(apic) ||
+	if (!kvm_apic_present(vcpu) || apic_lvtt_oneshot(apic) ||
 			apic_lvtt_period(apic))
 		return;
 

arch/x86/kvm/svm/svm.c

@@ -1090,7 +1090,7 @@ static void init_vmcb(struct vcpu_svm *svm)
 	svm->nested.vmcb = 0;
 	svm->vcpu.arch.hflags = 0;
 
-	if (pause_filter_count) {
+	if (!kvm_pause_in_guest(svm->vcpu.kvm)) {
 		control->pause_filter_count = pause_filter_count;
 		if (pause_filter_thresh)
 			control->pause_filter_thresh = pause_filter_thresh;
@@ -2693,7 +2693,7 @@ static int pause_interception(struct vcpu_svm *svm)
 	struct kvm_vcpu *vcpu = &svm->vcpu;
 	bool in_kernel = (svm_get_cpl(vcpu) == 0);
 
-	if (pause_filter_thresh)
+	if (!kvm_pause_in_guest(vcpu->kvm))
 		grow_ple_window(vcpu);
 
 	kvm_vcpu_on_spin(vcpu, in_kernel);
@@ -3780,7 +3780,7 @@ static void svm_handle_exit_irqoff(struct kvm_vcpu *vcpu)
 
 static void svm_sched_in(struct kvm_vcpu *vcpu, int cpu)
 {
-	if (pause_filter_thresh)
+	if (!kvm_pause_in_guest(vcpu->kvm))
 		shrink_ple_window(vcpu);
 }
 
@@ -3958,6 +3958,9 @@ static void svm_vm_destroy(struct kvm *kvm)
 
 static int svm_vm_init(struct kvm *kvm)
 {
+	if (!pause_filter_count || !pause_filter_thresh)
+		kvm->arch.pause_in_guest = true;
+
 	if (avic) {
 		int ret = avic_vm_init(kvm);
 		if (ret)

arch/x86/kvm/vmx/nested.c

@@ -6079,6 +6079,9 @@ static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
 	    ~(KVM_STATE_NESTED_SMM_GUEST_MODE | KVM_STATE_NESTED_SMM_VMXON))
 		return -EINVAL;
 
+	if (kvm_state->hdr.vmx.flags & ~KVM_STATE_VMX_PREEMPTION_TIMER_DEADLINE)
+		return -EINVAL;
+
 	/*
 	 * SMM temporarily disables VMX, so we cannot be in guest mode,
 	 * nor can VMLAUNCH/VMRESUME be pending.  Outside SMM, SMM flags
@@ -6108,9 +6111,16 @@ static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
 	if (ret)
 		return ret;
 
-	/* Empty 'VMXON' state is permitted */
-	if (kvm_state->size < sizeof(*kvm_state) + sizeof(*vmcs12))
-		return 0;
+	/* Empty 'VMXON' state is permitted if no VMCS loaded */
+	if (kvm_state->size < sizeof(*kvm_state) + sizeof(*vmcs12)) {
+		/* See vmx_has_valid_vmcs12.  */
+		if ((kvm_state->flags & KVM_STATE_NESTED_GUEST_MODE) ||
+		    (kvm_state->flags & KVM_STATE_NESTED_EVMCS) ||
+		    (kvm_state->hdr.vmx.vmcs12_pa != -1ull))
+			return -EINVAL;
+		else
+			return 0;
+	}
 
 	if (kvm_state->hdr.vmx.vmcs12_pa != -1ull) {
 		if (kvm_state->hdr.vmx.vmcs12_pa == kvm_state->hdr.vmx.vmxon_pa ||

arch/x86/kvm/vmx/nested.h

@@ -47,6 +47,11 @@ static inline struct vmcs12 *get_shadow_vmcs12(struct kvm_vcpu *vcpu)
 	return to_vmx(vcpu)->nested.cached_shadow_vmcs12;
 }
 
+/*
+ * Note: the same condition is checked against the state provided by userspace
+ * in vmx_set_nested_state; if it is satisfied, the nested state must include
+ * the VMCS12.
+ */
 static inline int vmx_has_valid_vmcs12(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);

tools/testing/selftests/kvm/x86_64/vmx_set_nested_state_test.c

@@ -76,10 +76,8 @@ void set_default_state(struct kvm_nested_state *state)
 void set_default_vmx_state(struct kvm_nested_state *state, int size)
 {
 	memset(state, 0, size);
-	state->flags = KVM_STATE_NESTED_GUEST_MODE  |
-			KVM_STATE_NESTED_RUN_PENDING;
 	if (have_evmcs)
-		state->flags |= KVM_STATE_NESTED_EVMCS;
+		state->flags = KVM_STATE_NESTED_EVMCS;
 	state->format = 0;
 	state->size = size;
 	state->hdr.vmx.vmxon_pa = 0x1000;
@@ -148,6 +146,11 @@ void test_vmx_nested_state(struct kvm_vm *vm)
 	state->hdr.vmx.smm.flags = 1;
 	test_nested_state_expect_einval(vm, state);
 
+	/* Invalid flags are rejected. */
+	set_default_vmx_state(state, state_sz);
+	state->hdr.vmx.flags = ~0;
+	test_nested_state_expect_einval(vm, state);
+
 	/* It is invalid to have vmxon_pa == -1ull and vmcs_pa != -1ull. */
 	set_default_vmx_state(state, state_sz);
 	state->hdr.vmx.vmxon_pa = -1ull;
@@ -185,20 +188,41 @@ void test_vmx_nested_state(struct kvm_vm *vm)
 	state->hdr.vmx.smm.flags = KVM_STATE_NESTED_SMM_GUEST_MODE;
 	test_nested_state_expect_einval(vm, state);
 
-	/* Size must be large enough to fit kvm_nested_state and vmcs12. */
+	/*
+	 * Size must be large enough to fit kvm_nested_state and vmcs12
+	 * if VMCS12 physical address is set
+	 */
 	set_default_vmx_state(state, state_sz);
 	state->size = sizeof(*state);
+	state->flags = 0;
+	test_nested_state_expect_einval(vm, state);
+
+	set_default_vmx_state(state, state_sz);
+	state->size = sizeof(*state);
+	state->flags = 0;
+	state->hdr.vmx.vmcs12_pa = -1;
 	test_nested_state(vm, state);
 
-	/* vmxon_pa cannot be the same address as vmcs_pa. */
+	/*
+	 * KVM_SET_NESTED_STATE succeeds with invalid VMCS
+	 * contents but L2 not running.
+	 */
 	set_default_vmx_state(state, state_sz);
-	state->hdr.vmx.vmxon_pa = 0;
-	state->hdr.vmx.vmcs12_pa = 0;
+	state->flags = 0;
+	test_nested_state(vm, state);
+
+	/* Invalid flags are rejected, even if no VMCS loaded. */
+	set_default_vmx_state(state, state_sz);
+	state->size = sizeof(*state);
+	state->flags = 0;
+	state->hdr.vmx.vmcs12_pa = -1;
+	state->hdr.vmx.flags = ~0;
 	test_nested_state_expect_einval(vm, state);
 
-	/* The revision id for vmcs12 must be VMCS12_REVISION. */
+	/* vmxon_pa cannot be the same address as vmcs_pa. */
 	set_default_vmx_state(state, state_sz);
-	set_revision_id_for_vmcs12(state, 0);
+	state->hdr.vmx.vmxon_pa = 0;
+	state->hdr.vmx.vmcs12_pa = 0;
 	test_nested_state_expect_einval(vm, state);
 
 	/*