1 files changed, 200 insertions, 100 deletions

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 1d26f3c4985b..e665aa7167cf 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -121,7 +121,6 @@ module_param_named(pml, enable_pml, bool, S_IRUGO);
 
 #define MSR_BITMAP_MODE_X2APIC		1
 #define MSR_BITMAP_MODE_X2APIC_APICV	2
-#define MSR_BITMAP_MODE_LM		4
 
 #define KVM_VMX_TSC_MULTIPLIER_MAX     0xffffffffffffffffULL
 
@@ -397,6 +396,7 @@ struct loaded_vmcs {
 	int cpu;
 	bool launched;
 	bool nmi_known_unmasked;
+	bool hv_timer_armed;
 	/* Support for vnmi-less CPUs */
 	int soft_vnmi_blocked;
 	ktime_t entry_time;
@@ -856,6 +856,7 @@ struct nested_vmx {
 
 	/* to migrate it to L2 if VM_ENTRY_LOAD_DEBUG_CONTROLS is off */
 	u64 vmcs01_debugctl;
+	u64 vmcs01_guest_bndcfgs;
 
 	u16 vpid02;
 	u16 last_vpid;
@@ -1019,6 +1020,8 @@ struct vcpu_vmx {
 	int ple_window;
 	bool ple_window_dirty;
 
+	bool req_immediate_exit;
+
 	/* Support for PML */
 #define PML_ENTITY_NUM		512
 	struct page *pml_pg;
@@ -1569,8 +1572,12 @@ static int vmx_hv_remote_flush_tlb(struct kvm *kvm)
 		goto out;
 	}
 
+	/*
+	 * FLUSH_GUEST_PHYSICAL_ADDRESS_SPACE hypercall needs the address of the
+	 * base of EPT PML4 table, strip off EPT configuration information.
+	 */
 	ret = hyperv_flush_guest_mapping(
-			to_vmx(kvm_get_vcpu(kvm, 0))->ept_pointer);
+			to_vmx(kvm_get_vcpu(kvm, 0))->ept_pointer & PAGE_MASK);
 
 out:
 	spin_unlock(&to_kvm_vmx(kvm)->ept_pointer_lock);
@@ -2864,6 +2871,8 @@ static void vmx_prepare_switch_to_guest(struct kvm_vcpu *vcpu)
 	u16 fs_sel, gs_sel;
 	int i;
 
+	vmx->req_immediate_exit = false;
+
 	if (vmx->loaded_cpu_state)
 		return;
 
@@ -2894,8 +2903,7 @@ static void vmx_prepare_switch_to_guest(struct kvm_vcpu *vcpu)
 		vmx->msr_host_kernel_gs_base = read_msr(MSR_KERNEL_GS_BASE);
 	}
 
-	if (is_long_mode(&vmx->vcpu))
-		wrmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base);
+	wrmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base);
 #else
 	savesegment(fs, fs_sel);
 	savesegment(gs, gs_sel);
@@ -2946,8 +2954,7 @@ static void vmx_prepare_switch_to_host(struct vcpu_vmx *vmx)
 	vmx->loaded_cpu_state = NULL;
 
 #ifdef CONFIG_X86_64
-	if (is_long_mode(&vmx->vcpu))
-		rdmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base);
+	rdmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base);
 #endif
 	if (host_state->ldt_sel || (host_state->gs_sel & 7)) {
 		kvm_load_ldt(host_state->ldt_sel);
@@ -2975,24 +2982,19 @@ static void vmx_prepare_switch_to_host(struct vcpu_vmx *vmx)
 #ifdef CONFIG_X86_64
 static u64 vmx_read_guest_kernel_gs_base(struct vcpu_vmx *vmx)
 {
-	if (is_long_mode(&vmx->vcpu)) {
-		preempt_disable();
-		if (vmx->loaded_cpu_state)
-			rdmsrl(MSR_KERNEL_GS_BASE,
-			       vmx->msr_guest_kernel_gs_base);
-		preempt_enable();
-	}
+	preempt_disable();
+	if (vmx->loaded_cpu_state)
+		rdmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base);
+	preempt_enable();
 	return vmx->msr_guest_kernel_gs_base;
 }
 
 static void vmx_write_guest_kernel_gs_base(struct vcpu_vmx *vmx, u64 data)
 {
-	if (is_long_mode(&vmx->vcpu)) {
-		preempt_disable();
-		if (vmx->loaded_cpu_state)
-			wrmsrl(MSR_KERNEL_GS_BASE, data);
-		preempt_enable();
-	}
+	preempt_disable();
+	if (vmx->loaded_cpu_state)
+		wrmsrl(MSR_KERNEL_GS_BASE, data);
+	preempt_enable();
 	vmx->msr_guest_kernel_gs_base = data;
 }
 #endif
@@ -3528,9 +3530,6 @@ static void nested_vmx_setup_ctls_msrs(struct nested_vmx_msrs *msrs, bool apicv)
 		VM_EXIT_LOAD_IA32_EFER | VM_EXIT_SAVE_IA32_EFER |
 		VM_EXIT_SAVE_VMX_PREEMPTION_TIMER | VM_EXIT_ACK_INTR_ON_EXIT;
 
-	if (kvm_mpx_supported())
-		msrs->exit_ctls_high |= VM_EXIT_CLEAR_BNDCFGS;
-
 	/* We support free control of debug control saving. */
 	msrs->exit_ctls_low &= ~VM_EXIT_SAVE_DEBUG_CONTROLS;
 
@@ -3547,8 +3546,6 @@ static void nested_vmx_setup_ctls_msrs(struct nested_vmx_msrs *msrs, bool apicv)
 		VM_ENTRY_LOAD_IA32_PAT;
 	msrs->entry_ctls_high |=
 		(VM_ENTRY_ALWAYSON_WITHOUT_TRUE_MSR | VM_ENTRY_LOAD_IA32_EFER);
-	if (kvm_mpx_supported())
-		msrs->entry_ctls_high |= VM_ENTRY_LOAD_BNDCFGS;
 
 	/* We support free control of debug control loading. */
 	msrs->entry_ctls_low &= ~VM_ENTRY_LOAD_DEBUG_CONTROLS;
@@ -3596,12 +3593,12 @@ static void nested_vmx_setup_ctls_msrs(struct nested_vmx_msrs *msrs, bool apicv)
 		msrs->secondary_ctls_high);
 	msrs->secondary_ctls_low = 0;
 	msrs->secondary_ctls_high &=
-		SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES |
 		SECONDARY_EXEC_DESC |
 		SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE |
 		SECONDARY_EXEC_APIC_REGISTER_VIRT |
 		SECONDARY_EXEC_VIRTUAL_INTR_DELIVERY |
 		SECONDARY_EXEC_WBINVD_EXITING;
+
 	/*
 	 * We can emulate "VMCS shadowing," even if the hardware
 	 * doesn't support it.
@@ -3658,6 +3655,10 @@ static void nested_vmx_setup_ctls_msrs(struct nested_vmx_msrs *msrs, bool apicv)
 		msrs->secondary_ctls_high |=
 			SECONDARY_EXEC_UNRESTRICTED_GUEST;
 
+	if (flexpriority_enabled)
+		msrs->secondary_ctls_high |=
+			SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES;
+
 	/* miscellaneous data */
 	rdmsr(MSR_IA32_VMX_MISC,
 		msrs->misc_low,
@@ -5068,19 +5069,6 @@ static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer)
 	if (!msr)
 		return;
 
-	/*
-	 * MSR_KERNEL_GS_BASE is not intercepted when the guest is in
-	 * 64-bit mode as a 64-bit kernel may frequently access the
-	 * MSR.  This means we need to manually save/restore the MSR
-	 * when switching between guest and host state, but only if
-	 * the guest is in 64-bit mode.  Sync our cached value if the
-	 * guest is transitioning to 32-bit mode and the CPU contains
-	 * guest state, i.e. the cache is stale.
-	 */
-#ifdef CONFIG_X86_64
-	if (!(efer & EFER_LMA))
-		(void)vmx_read_guest_kernel_gs_base(vmx);
-#endif
 	vcpu->arch.efer = efer;
 	if (efer & EFER_LMA) {
 		vm_entry_controls_setbit(to_vmx(vcpu), VM_ENTRY_IA32E_MODE);
@@ -5393,9 +5381,10 @@ static int vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
 		 * To use VMXON (and later other VMX instructions), a guest
 		 * must first be able to turn on cr4.VMXE (see handle_vmon()).
 		 * So basically the check on whether to allow nested VMX
-		 * is here.
+		 * is here.  We operate under the default treatment of SMM,
+		 * so VMX cannot be enabled under SMM.
 		 */
-		if (!nested_vmx_allowed(vcpu))
+		if (!nested_vmx_allowed(vcpu) || is_smm(vcpu))
 			return 1;
 	}
 
@@ -6072,9 +6061,6 @@ static u8 vmx_msr_bitmap_mode(struct kvm_vcpu *vcpu)
 			mode |= MSR_BITMAP_MODE_X2APIC_APICV;
 	}
 
-	if (is_long_mode(vcpu))
-		mode |= MSR_BITMAP_MODE_LM;
-
 	return mode;
 }
 
@@ -6115,9 +6101,6 @@ static void vmx_update_msr_bitmap(struct kvm_vcpu *vcpu)
 	if (!changed)
 		return;
 
-	vmx_set_intercept_for_msr(msr_bitmap, MSR_KERNEL_GS_BASE, MSR_TYPE_RW,
-				  !(mode & MSR_BITMAP_MODE_LM));
-
 	if (changed & (MSR_BITMAP_MODE_X2APIC | MSR_BITMAP_MODE_X2APIC_APICV))
 		vmx_update_msr_bitmap_x2apic(msr_bitmap, mode);
 
@@ -6183,6 +6166,32 @@ static void vmx_complete_nested_posted_interrupt(struct kvm_vcpu *vcpu)
 	nested_mark_vmcs12_pages_dirty(vcpu);
 }
 
+static u8 vmx_get_rvi(void)
+{
+	return vmcs_read16(GUEST_INTR_STATUS) & 0xff;
+}
+
+static bool vmx_guest_apic_has_interrupt(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	void *vapic_page;
+	u32 vppr;
+	int rvi;
+
+	if (WARN_ON_ONCE(!is_guest_mode(vcpu)) ||
+		!nested_cpu_has_vid(get_vmcs12(vcpu)) ||
+		WARN_ON_ONCE(!vmx->nested.virtual_apic_page))
+		return false;
+
+	rvi = vmx_get_rvi();
+
+	vapic_page = kmap(vmx->nested.virtual_apic_page);
+	vppr = *((u32 *)(vapic_page + APIC_PROCPRI));
+	kunmap(vmx->nested.virtual_apic_page);
+
+	return ((rvi & 0xf0) > (vppr & 0xf0));
+}
+
 static inline bool kvm_vcpu_trigger_posted_interrupt(struct kvm_vcpu *vcpu,
 						     bool nested)
 {
@@ -6983,7 +6992,7 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu,
 	 * Cause the #SS fault with 0 error code in VM86 mode.
 	 */
 	if (((vec == GP_VECTOR) || (vec == SS_VECTOR)) && err_code == 0) {
-		if (emulate_instruction(vcpu, 0) == EMULATE_DONE) {
+		if (kvm_emulate_instruction(vcpu, 0) == EMULATE_DONE) {
 			if (vcpu->arch.halt_request) {
 				vcpu->arch.halt_request = 0;
 				return kvm_vcpu_halt(vcpu);
@@ -7054,7 +7063,7 @@ static int handle_exception(struct kvm_vcpu *vcpu)
 
 	if (!vmx->rmode.vm86_active && is_gp_fault(intr_info)) {
 		WARN_ON_ONCE(!enable_vmware_backdoor);
-		er = emulate_instruction(vcpu,
+		er = kvm_emulate_instruction(vcpu,
 			EMULTYPE_VMWARE | EMULTYPE_NO_UD_ON_FAIL);
 		if (er == EMULATE_USER_EXIT)
 			return 0;
@@ -7157,7 +7166,7 @@ static int handle_io(struct kvm_vcpu *vcpu)
 	++vcpu->stat.io_exits;
 
 	if (string)
-		return emulate_instruction(vcpu, 0) == EMULATE_DONE;
+		return kvm_emulate_instruction(vcpu, 0) == EMULATE_DONE;
 
 	port = exit_qualification >> 16;
 	size = (exit_qualification & 7) + 1;
@@ -7231,7 +7240,7 @@ static int handle_set_cr4(struct kvm_vcpu *vcpu, unsigned long val)
 static int handle_desc(struct kvm_vcpu *vcpu)
 {
 	WARN_ON(!(vcpu->arch.cr4 & X86_CR4_UMIP));
-	return emulate_instruction(vcpu, 0) == EMULATE_DONE;
+	return kvm_emulate_instruction(vcpu, 0) == EMULATE_DONE;
 }
 
 static int handle_cr(struct kvm_vcpu *vcpu)
@@ -7480,7 +7489,7 @@ static int handle_vmcall(struct kvm_vcpu *vcpu)
 
 static int handle_invd(struct kvm_vcpu *vcpu)
 {
-	return emulate_instruction(vcpu, 0) == EMULATE_DONE;
+	return kvm_emulate_instruction(vcpu, 0) == EMULATE_DONE;
 }
 
 static int handle_invlpg(struct kvm_vcpu *vcpu)
@@ -7547,7 +7556,7 @@ static int handle_apic_access(struct kvm_vcpu *vcpu)
 			return kvm_skip_emulated_instruction(vcpu);
 		}
 	}
-	return emulate_instruction(vcpu, 0) == EMULATE_DONE;
+	return kvm_emulate_instruction(vcpu, 0) == EMULATE_DONE;
 }
 
 static int handle_apic_eoi_induced(struct kvm_vcpu *vcpu)
@@ -7704,8 +7713,8 @@ static int handle_ept_misconfig(struct kvm_vcpu *vcpu)
 		if (!static_cpu_has(X86_FEATURE_HYPERVISOR))
 			return kvm_skip_emulated_instruction(vcpu);
 		else
-			return x86_emulate_instruction(vcpu, gpa, EMULTYPE_SKIP,
-						       NULL, 0) == EMULATE_DONE;
+			return kvm_emulate_instruction(vcpu, EMULTYPE_SKIP) ==
+								EMULATE_DONE;
 	}
 
 	return kvm_mmu_page_fault(vcpu, gpa, PFERR_RSVD_MASK, NULL, 0);
@@ -7748,7 +7757,7 @@ static int handle_invalid_guest_state(struct kvm_vcpu *vcpu)
 		if (kvm_test_request(KVM_REQ_EVENT, vcpu))
 			return 1;
 
-		err = emulate_instruction(vcpu, 0);
+		err = kvm_emulate_instruction(vcpu, 0);
 
 		if (err == EMULATE_USER_EXIT) {
 			++vcpu->stat.mmio_exits;
@@ -7966,6 +7975,9 @@ static __init int hardware_setup(void)
 		kvm_x86_ops->enable_log_dirty_pt_masked = NULL;
 	}
 
+	if (!cpu_has_vmx_preemption_timer())
+		kvm_x86_ops->request_immediate_exit = __kvm_request_immediate_exit;
+
 	if (cpu_has_vmx_preemption_timer() && enable_preemption_timer) {
 		u64 vmx_msr;
 
@@ -9208,7 +9220,8 @@ static int handle_pml_full(struct kvm_vcpu *vcpu)
 
 static int handle_preemption_timer(struct kvm_vcpu *vcpu)
 {
-	kvm_lapic_expired_hv_timer(vcpu);
+	if (!to_vmx(vcpu)->req_immediate_exit)
+		kvm_lapic_expired_hv_timer(vcpu);
 	return 1;
 }
 
@@ -10214,15 +10227,16 @@ static void vmx_set_virtual_apic_mode(struct kvm_vcpu *vcpu)
 	if (!lapic_in_kernel(vcpu))
 		return;
 
+	if (!flexpriority_enabled &&
+	    !cpu_has_vmx_virtualize_x2apic_mode())
+		return;
+
 	/* Postpone execution until vmcs01 is the current VMCS. */
 	if (is_guest_mode(vcpu)) {
 		to_vmx(vcpu)->nested.change_vmcs01_virtual_apic_mode = true;
 		return;
 	}
 
-	if (!cpu_need_tpr_shadow(vcpu))
-		return;
-
 	sec_exec_control = vmcs_read32(SECONDARY_VM_EXEC_CONTROL);
 	sec_exec_control &= ~(SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES |
 			      SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE);
@@ -10344,6 +10358,14 @@ static int vmx_sync_pir_to_irr(struct kvm_vcpu *vcpu)
 	return max_irr;
 }
 
+static u8 vmx_has_apicv_interrupt(struct kvm_vcpu *vcpu)
+{
+	u8 rvi = vmx_get_rvi();
+	u8 vppr = kvm_lapic_get_reg(vcpu->arch.apic, APIC_PROCPRI);
+
+	return ((rvi & 0xf0) > (vppr & 0xf0));
+}
+
 static void vmx_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap)
 {
 	if (!kvm_vcpu_apicv_active(vcpu))
@@ -10595,24 +10617,43 @@ static void atomic_switch_perf_msrs(struct vcpu_vmx *vmx)
 					msrs[i].host, false);
 }
 
-static void vmx_arm_hv_timer(struct kvm_vcpu *vcpu)
+static void vmx_arm_hv_timer(struct vcpu_vmx *vmx, u32 val)
+{
+	vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, val);
+	if (!vmx->loaded_vmcs->hv_timer_armed)
+		vmcs_set_bits(PIN_BASED_VM_EXEC_CONTROL,
+			      PIN_BASED_VMX_PREEMPTION_TIMER);
+	vmx->loaded_vmcs->hv_timer_armed = true;
+}
+
+static void vmx_update_hv_timer(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	u64 tscl;
 	u32 delta_tsc;
 
-	if (vmx->hv_deadline_tsc == -1)
+	if (vmx->req_immediate_exit) {
+		vmx_arm_hv_timer(vmx, 0);
 		return;
+	}
 
-	tscl = rdtsc();
-	if (vmx->hv_deadline_tsc > tscl)
-		/* sure to be 32 bit only because checked on set_hv_timer */
-		delta_tsc = (u32)((vmx->hv_deadline_tsc - tscl) >>
-			cpu_preemption_timer_multi);
-	else
-		delta_tsc = 0;
+	if (vmx->hv_deadline_tsc != -1) {
+		tscl = rdtsc();
+		if (vmx->hv_deadline_tsc > tscl)
+			/* set_hv_timer ensures the delta fits in 32-bits */
+			delta_tsc = (u32)((vmx->hv_deadline_tsc - tscl) >>
+				cpu_preemption_timer_multi);
+		else
+			delta_tsc = 0;
 
-	vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, delta_tsc);
+		vmx_arm_hv_timer(vmx, delta_tsc);
+		return;
+	}
+
+	if (vmx->loaded_vmcs->hv_timer_armed)
+		vmcs_clear_bits(PIN_BASED_VM_EXEC_CONTROL,
+				PIN_BASED_VMX_PREEMPTION_TIMER);
+	vmx->loaded_vmcs->hv_timer_armed = false;
 }
 
 static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
@@ -10672,7 +10713,7 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 
 	atomic_switch_perf_msrs(vmx);
 
-	vmx_arm_hv_timer(vcpu);
+	vmx_update_hv_timer(vcpu);
 
 	/*
 	 * If this vCPU has touched SPEC_CTRL, restore the guest's value if
@@ -11214,6 +11255,23 @@ static void nested_vmx_cr_fixed1_bits_update(struct kvm_vcpu *vcpu)
 #undef cr4_fixed1_update
 }
 
+static void nested_vmx_entry_exit_ctls_update(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (kvm_mpx_supported()) {
+		bool mpx_enabled = guest_cpuid_has(vcpu, X86_FEATURE_MPX);
+
+		if (mpx_enabled) {
+			vmx->nested.msrs.entry_ctls_high |= VM_ENTRY_LOAD_BNDCFGS;
+			vmx->nested.msrs.exit_ctls_high |= VM_EXIT_CLEAR_BNDCFGS;
+		} else {
+			vmx->nested.msrs.entry_ctls_high &= ~VM_ENTRY_LOAD_BNDCFGS;
+			vmx->nested.msrs.exit_ctls_high &= ~VM_EXIT_CLEAR_BNDCFGS;
+		}
+	}
+}
+
 static void vmx_cpuid_update(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
@@ -11230,8 +11288,10 @@ static void vmx_cpuid_update(struct kvm_vcpu *vcpu)
 		to_vmx(vcpu)->msr_ia32_feature_control_valid_bits &=
 			~FEATURE_CONTROL_VMXON_ENABLED_OUTSIDE_SMX;
 
-	if (nested_vmx_allowed(vcpu))
+	if (nested_vmx_allowed(vcpu)) {
 		nested_vmx_cr_fixed1_bits_update(vcpu);
+		nested_vmx_entry_exit_ctls_update(vcpu);
+	}
 }
 
 static void vmx_set_supported_cpuid(u32 func, struct kvm_cpuid_entry2 *entry)
@@ -11427,16 +11487,18 @@ static void vmx_start_preemption_timer(struct kvm_vcpu *vcpu)
 	u64 preemption_timeout = get_vmcs12(vcpu)->vmx_preemption_timer_value;
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 
-	if (vcpu->arch.virtual_tsc_khz == 0)
-		return;
-
-	/* Make sure short timeouts reliably trigger an immediate vmexit.
-	 * hrtimer_start does not guarantee this. */
-	if (preemption_timeout <= 1) {
+	/*
+	 * A timer value of zero is architecturally guaranteed to cause
+	 * a VMExit prior to executing any instructions in the guest.
+	 */
+	if (preemption_timeout == 0) {
 		vmx_preemption_timer_fn(&vmx->nested.preemption_timer);
 		return;
 	}
 
+	if (vcpu->arch.virtual_tsc_khz == 0)
+		return;
+
 	preemption_timeout <<= VMX_MISC_EMULATED_PREEMPTION_TIMER_RATE;
 	preemption_timeout *= 1000000;
 	do_div(preemption_timeout, vcpu->arch.virtual_tsc_khz);
@@ -11646,11 +11708,15 @@ static int nested_vmx_check_apicv_controls(struct kvm_vcpu *vcpu,
 	 * bits 15:8 should be zero in posted_intr_nv,
 	 * the descriptor address has been already checked
 	 * in nested_get_vmcs12_pages.
+	 *
+	 * bits 5:0 of posted_intr_desc_addr should be zero.
 	 */
 	if (nested_cpu_has_posted_intr(vmcs12) &&
 	   (!nested_cpu_has_vid(vmcs12) ||
 	    !nested_exit_intr_ack_set(vcpu) ||
-	    vmcs12->posted_intr_nv & 0xff00))
+	    (vmcs12->posted_intr_nv & 0xff00) ||
+	    (vmcs12->posted_intr_desc_addr & 0x3f) ||
+	    (!page_address_valid(vcpu, vmcs12->posted_intr_desc_addr))))
 		return -EINVAL;
 
 	/* tpr shadow is needed by all apicv features. */
@@ -11993,8 +12059,13 @@ static void prepare_vmcs02_full(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 
 	set_cr4_guest_host_mask(vmx);
 
-	if (vmx_mpx_supported())
-		vmcs_write64(GUEST_BNDCFGS, vmcs12->guest_bndcfgs);
+	if (kvm_mpx_supported()) {
+		if (vmx->nested.nested_run_pending &&
+			(vmcs12->vm_entry_controls & VM_ENTRY_LOAD_BNDCFGS))
+			vmcs_write64(GUEST_BNDCFGS, vmcs12->guest_bndcfgs);
+		else
+			vmcs_write64(GUEST_BNDCFGS, vmx->nested.vmcs01_guest_bndcfgs);
+	}
 
 	if (enable_vpid) {
 		if (nested_cpu_has_vpid(vmcs12) && vmx->nested.vpid02)
@@ -12076,11 +12147,10 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12,
 
 	exec_control = vmcs12->pin_based_vm_exec_control;
 
-	/* Preemption timer setting is only taken from vmcs01.  */
-	exec_control &= ~PIN_BASED_VMX_PREEMPTION_TIMER;
+	/* Preemption timer setting is computed directly in vmx_vcpu_run.  */
 	exec_control |= vmcs_config.pin_based_exec_ctrl;
-	if (vmx->hv_deadline_tsc == -1)
-		exec_control &= ~PIN_BASED_VMX_PREEMPTION_TIMER;
+	exec_control &= ~PIN_BASED_VMX_PREEMPTION_TIMER;
+	vmx->loaded_vmcs->hv_timer_armed = false;
 
 	/* Posted interrupts setting is only taken from vmcs12.  */
 	if (nested_cpu_has_posted_intr(vmcs12)) {
@@ -12318,6 +12388,9 @@ static int check_vmentry_prereqs(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 	    vmcs12->guest_activity_state != GUEST_ACTIVITY_HLT)
 		return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
 
+	if (nested_cpu_has_vpid(vmcs12) && !vmcs12->virtual_processor_id)
+		return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
+
 	if (nested_vmx_check_io_bitmap_controls(vcpu, vmcs12))
 		return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
 
@@ -12537,12 +12610,21 @@ static int enter_vmx_non_root_mode(struct kvm_vcpu *vcpu, u32 *exit_qual)
 	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
 	bool from_vmentry = !!exit_qual;
 	u32 dummy_exit_qual;
+	bool evaluate_pending_interrupts;
 	int r = 0;
 
+	evaluate_pending_interrupts = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL) &
+		(CPU_BASED_VIRTUAL_INTR_PENDING | CPU_BASED_VIRTUAL_NMI_PENDING);
+	if (likely(!evaluate_pending_interrupts) && kvm_vcpu_apicv_active(vcpu))
+		evaluate_pending_interrupts |= vmx_has_apicv_interrupt(vcpu);
+
 	enter_guest_mode(vcpu);
 
 	if (!(vmcs12->vm_entry_controls & VM_ENTRY_LOAD_DEBUG_CONTROLS))
 		vmx->nested.vmcs01_debugctl = vmcs_read64(GUEST_IA32_DEBUGCTL);
+	if (kvm_mpx_supported() &&
+		!(vmcs12->vm_entry_controls & VM_ENTRY_LOAD_BNDCFGS))
+		vmx->nested.vmcs01_guest_bndcfgs = vmcs_read64(GUEST_BNDCFGS);
 
 	vmx_switch_vmcs(vcpu, &vmx->nested.vmcs02);
 	vmx_segment_cache_clear(vmx);
@@ -12575,6 +12657,23 @@ static int enter_vmx_non_root_mode(struct kvm_vcpu *vcpu, u32 *exit_qual)
 	}
 
 	/*
+	 * If L1 had a pending IRQ/NMI until it executed
+	 * VMLAUNCH/VMRESUME which wasn't delivered because it was
+	 * disallowed (e.g. interrupts disabled), L0 needs to
+	 * evaluate if this pending event should cause an exit from L2
+	 * to L1 or delivered directly to L2 (e.g. In case L1 don't
+	 * intercept EXTERNAL_INTERRUPT).
+	 *
+	 * Usually this would be handled by the processor noticing an
+	 * IRQ/NMI window request, or checking RVI during evaluation of
+	 * pending virtual interrupts.  However, this setting was done
+	 * on VMCS01 and now VMCS02 is active instead. Thus, we force L0
+	 * to perform pending event evaluation by requesting a KVM_REQ_EVENT.
+	 */
+	if (unlikely(evaluate_pending_interrupts))
+		kvm_make_request(KVM_REQ_EVENT, vcpu);
+
+	/*
 	 * Note no nested_vmx_succeed or nested_vmx_fail here. At this point
 	 * we are no longer running L1, and VMLAUNCH/VMRESUME has not yet
 	 * returned as far as L1 is concerned. It will only return (and set
@@ -12841,6 +12940,11 @@ static int vmx_check_nested_events(struct kvm_vcpu *vcpu, bool external_intr)
 	return 0;
 }
 
+static void vmx_request_immediate_exit(struct kvm_vcpu *vcpu)
+{
+	to_vmx(vcpu)->req_immediate_exit = true;
+}
+
 static u32 vmx_get_preemption_timer_value(struct kvm_vcpu *vcpu)
 {
 	ktime_t remaining =
@@ -13231,12 +13335,7 @@ static void nested_vmx_vmexit(struct kvm_vcpu *vcpu, u32 exit_reason,
 	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, vmx->msr_autoload.host.nr);
 	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, vmx->msr_autoload.guest.nr);
 	vmcs_write64(TSC_OFFSET, vcpu->arch.tsc_offset);
-	if (vmx->hv_deadline_tsc == -1)
-		vmcs_clear_bits(PIN_BASED_VM_EXEC_CONTROL,
-				PIN_BASED_VMX_PREEMPTION_TIMER);
-	else
-		vmcs_set_bits(PIN_BASED_VM_EXEC_CONTROL,
-			      PIN_BASED_VMX_PREEMPTION_TIMER);
+
 	if (kvm_has_tsc_control)
 		decache_tsc_multiplier(vmx);
 
@@ -13440,18 +13539,12 @@ static int vmx_set_hv_timer(struct kvm_vcpu *vcpu, u64 guest_deadline_tsc)
 		return -ERANGE;
 
 	vmx->hv_deadline_tsc = tscl + delta_tsc;
-	vmcs_set_bits(PIN_BASED_VM_EXEC_CONTROL,
-			PIN_BASED_VMX_PREEMPTION_TIMER);
-
 	return delta_tsc == 0;
 }
 
 static void vmx_cancel_hv_timer(struct kvm_vcpu *vcpu)
 {
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	vmx->hv_deadline_tsc = -1;
-	vmcs_clear_bits(PIN_BASED_VM_EXEC_CONTROL,
-			PIN_BASED_VMX_PREEMPTION_TIMER);
+	to_vmx(vcpu)->hv_deadline_tsc = -1;
 }
 #endif
 
@@ -13932,6 +14025,14 @@ static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
 	    ~(KVM_STATE_NESTED_SMM_GUEST_MODE | KVM_STATE_NESTED_SMM_VMXON))
 		return -EINVAL;
 
+	/*
+	 * SMM temporarily disables VMX, so we cannot be in guest mode,
+	 * nor can VMLAUNCH/VMRESUME be pending.  Outside SMM, SMM flags
+	 * must be zero.
+	 */
+	if (is_smm(vcpu) ? kvm_state->flags : kvm_state->vmx.smm.flags)
+		return -EINVAL;
+
 	if ((kvm_state->vmx.smm.flags & KVM_STATE_NESTED_SMM_GUEST_MODE) &&
 	    !(kvm_state->vmx.smm.flags & KVM_STATE_NESTED_SMM_VMXON))
 		return -EINVAL;
@@ -13988,9 +14089,6 @@ static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
 	    check_vmentry_postreqs(vcpu, vmcs12, &exit_qual))
 		return -EINVAL;
 
-	if (kvm_state->flags & KVM_STATE_NESTED_RUN_PENDING)
-		vmx->nested.nested_run_pending = 1;
-
 	vmx->nested.dirty_vmcs12 = true;
 	ret = enter_vmx_non_root_mode(vcpu, NULL);
 	if (ret)
@@ -14078,6 +14176,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
 	.apicv_post_state_restore = vmx_apicv_post_state_restore,
 	.hwapic_irr_update = vmx_hwapic_irr_update,
 	.hwapic_isr_update = vmx_hwapic_isr_update,
+	.guest_apic_has_interrupt = vmx_guest_apic_has_interrupt,
 	.sync_pir_to_irr = vmx_sync_pir_to_irr,
 	.deliver_posted_interrupt = vmx_deliver_posted_interrupt,
 
@@ -14111,6 +14210,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
 	.umip_emulated = vmx_umip_emulated,
 
 	.check_nested_events = vmx_check_nested_events,
+	.request_immediate_exit = vmx_request_immediate_exit,
 
 	.sched_in = vmx_sched_in,