1 files changed, 229 insertions, 142 deletions

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 884e5b3838c7..efc7a82ab140 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -156,9 +156,9 @@ module_param(tsc_tolerance_ppm, uint, S_IRUGO | S_IWUSR);
 
 /*
  * lapic timer advance (tscdeadline mode only) in nanoseconds.  '-1' enables
- * adaptive tuning starting from default advancment of 1000ns.  '0' disables
+ * adaptive tuning starting from default advancement of 1000ns.  '0' disables
  * advancement entirely.  Any other value is used as-is and disables adaptive
- * tuning, i.e. allows priveleged userspace to set an exact advancement time.
+ * tuning, i.e. allows privileged userspace to set an exact advancement time.
  */
 static int __read_mostly lapic_timer_advance_ns = -1;
 module_param(lapic_timer_advance_ns, int, S_IRUGO | S_IWUSR);
@@ -271,8 +271,7 @@ static struct kmem_cache *x86_emulator_cache;
  * When called, it means the previous get/set msr reached an invalid msr.
  * Return true if we want to ignore/silent this failed msr access.
  */
-static bool kvm_msr_ignored_check(struct kvm_vcpu *vcpu, u32 msr,
-				  u64 data, bool write)
+static bool kvm_msr_ignored_check(u32 msr, u64 data, bool write)
 {
 	const char *op = write ? "wrmsr" : "rdmsr";
 
@@ -1288,7 +1287,7 @@ static const u32 emulated_msrs_all[] = {
 	MSR_KVM_PV_EOI_EN, MSR_KVM_ASYNC_PF_INT, MSR_KVM_ASYNC_PF_ACK,
 
 	MSR_IA32_TSC_ADJUST,
-	MSR_IA32_TSCDEADLINE,
+	MSR_IA32_TSC_DEADLINE,
 	MSR_IA32_ARCH_CAPABILITIES,
 	MSR_IA32_PERF_CAPABILITIES,
 	MSR_IA32_MISC_ENABLE,
@@ -1373,7 +1372,7 @@ static u64 kvm_get_arch_capabilities(void)
 	/*
 	 * If nx_huge_pages is enabled, KVM's shadow paging will ensure that
 	 * the nested hypervisor runs with NX huge pages.  If it is not,
-	 * L1 is anyway vulnerable to ITLB_MULTIHIT explots from other
+	 * L1 is anyway vulnerable to ITLB_MULTIHIT exploits from other
 	 * L1 guests, so it need not worry about its own (L2) guests.
 	 */
 	data |= ARCH_CAP_PSCHANGE_MC_NO;
@@ -1445,7 +1444,7 @@ static int do_get_msr_feature(struct kvm_vcpu *vcpu, unsigned index, u64 *data)
 	if (r == KVM_MSR_RET_INVALID) {
 		/* Unconditionally clear the output for simplicity */
 		*data = 0;
-		if (kvm_msr_ignored_check(vcpu, index, 0, false))
+		if (kvm_msr_ignored_check(index, 0, false))
 			r = 0;
 	}
 
@@ -1526,35 +1525,44 @@ EXPORT_SYMBOL_GPL(kvm_enable_efer_bits);
 
 bool kvm_msr_allowed(struct kvm_vcpu *vcpu, u32 index, u32 type)
 {
+	struct kvm_x86_msr_filter *msr_filter;
+	struct msr_bitmap_range *ranges;
 	struct kvm *kvm = vcpu->kvm;
-	struct msr_bitmap_range *ranges = kvm->arch.msr_filter.ranges;
-	u32 count = kvm->arch.msr_filter.count;
-	u32 i;
-	bool r = kvm->arch.msr_filter.default_allow;
+	bool allowed;
 	int idx;
+	u32 i;
 
-	/* MSR filtering not set up or x2APIC enabled, allow everything */
-	if (!count || (index >= 0x800 && index <= 0x8ff))
+	/* x2APIC MSRs do not support filtering. */
+	if (index >= 0x800 && index <= 0x8ff)
 		return true;
 
-	/* Prevent collision with set_msr_filter */
 	idx = srcu_read_lock(&kvm->srcu);
 
-	for (i = 0; i < count; i++) {
+	msr_filter = srcu_dereference(kvm->arch.msr_filter, &kvm->srcu);
+	if (!msr_filter) {
+		allowed = true;
+		goto out;
+	}
+
+	allowed = msr_filter->default_allow;
+	ranges = msr_filter->ranges;
+
+	for (i = 0; i < msr_filter->count; i++) {
 		u32 start = ranges[i].base;
 		u32 end = start + ranges[i].nmsrs;
 		u32 flags = ranges[i].flags;
 		unsigned long *bitmap = ranges[i].bitmap;
 
 		if ((index >= start) && (index < end) && (flags & type)) {
-			r = !!test_bit(index - start, bitmap);
+			allowed = !!test_bit(index - start, bitmap);
 			break;
 		}
 	}
 
+out:
 	srcu_read_unlock(&kvm->srcu, idx);
 
-	return r;
+	return allowed;
 }
 EXPORT_SYMBOL_GPL(kvm_msr_allowed);
 
@@ -1611,7 +1619,7 @@ static int kvm_set_msr_ignored_check(struct kvm_vcpu *vcpu,
 	int ret = __kvm_set_msr(vcpu, index, data, host_initiated);
 
 	if (ret == KVM_MSR_RET_INVALID)
-		if (kvm_msr_ignored_check(vcpu, index, data, true))
+		if (kvm_msr_ignored_check(index, data, true))
 			ret = 0;
 
 	return ret;
@@ -1649,7 +1657,7 @@ static int kvm_get_msr_ignored_check(struct kvm_vcpu *vcpu,
 	if (ret == KVM_MSR_RET_INVALID) {
 		/* Unconditionally clear *data for simplicity */
 		*data = 0;
-		if (kvm_msr_ignored_check(vcpu, index, 0, false))
+		if (kvm_msr_ignored_check(index, 0, false))
 			ret = 0;
 	}
 
@@ -1841,7 +1849,7 @@ fastpath_t handle_fastpath_set_msr_irqoff(struct kvm_vcpu *vcpu)
 			ret = EXIT_FASTPATH_EXIT_HANDLED;
 		}
 		break;
-	case MSR_IA32_TSCDEADLINE:
+	case MSR_IA32_TSC_DEADLINE:
 		data = kvm_read_edx_eax(vcpu);
 		if (!handle_fastpath_set_tscdeadline(vcpu, data)) {
 			kvm_skip_emulated_instruction(vcpu);
@@ -2320,7 +2328,7 @@ static void kvm_synchronize_tsc(struct kvm_vcpu *vcpu, u64 data)
 	kvm_vcpu_write_tsc_offset(vcpu, offset);
 	raw_spin_unlock_irqrestore(&kvm->arch.tsc_write_lock, flags);
 
-	spin_lock(&kvm->arch.pvclock_gtod_sync_lock);
+	spin_lock_irqsave(&kvm->arch.pvclock_gtod_sync_lock, flags);
 	if (!matched) {
 		kvm->arch.nr_vcpus_matched_tsc = 0;
 	} else if (!already_matched) {
@@ -2328,7 +2336,7 @@ static void kvm_synchronize_tsc(struct kvm_vcpu *vcpu, u64 data)
 	}
 
 	kvm_track_tsc_matching(vcpu);
-	spin_unlock(&kvm->arch.pvclock_gtod_sync_lock);
+	spin_unlock_irqrestore(&kvm->arch.pvclock_gtod_sync_lock, flags);
 }
 
 static inline void adjust_tsc_offset_guest(struct kvm_vcpu *vcpu,
@@ -2550,11 +2558,16 @@ static void kvm_gen_update_masterclock(struct kvm *kvm)
 	int i;
 	struct kvm_vcpu *vcpu;
 	struct kvm_arch *ka = &kvm->arch;
+	unsigned long flags;
+
+	kvm_hv_invalidate_tsc_page(kvm);
 
-	spin_lock(&ka->pvclock_gtod_sync_lock);
 	kvm_make_mclock_inprogress_request(kvm);
+
 	/* no guest entries from this point */
+	spin_lock_irqsave(&ka->pvclock_gtod_sync_lock, flags);
 	pvclock_update_vm_gtod_copy(kvm);
+	spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
 
 	kvm_for_each_vcpu(i, vcpu, kvm)
 		kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
@@ -2562,8 +2575,6 @@ static void kvm_gen_update_masterclock(struct kvm *kvm)
 	/* guest entries allowed */
 	kvm_for_each_vcpu(i, vcpu, kvm)
 		kvm_clear_request(KVM_REQ_MCLOCK_INPROGRESS, vcpu);
-
-	spin_unlock(&ka->pvclock_gtod_sync_lock);
 #endif
 }
 
@@ -2571,17 +2582,18 @@ u64 get_kvmclock_ns(struct kvm *kvm)
 {
 	struct kvm_arch *ka = &kvm->arch;
 	struct pvclock_vcpu_time_info hv_clock;
+	unsigned long flags;
 	u64 ret;
 
-	spin_lock(&ka->pvclock_gtod_sync_lock);
+	spin_lock_irqsave(&ka->pvclock_gtod_sync_lock, flags);
 	if (!ka->use_master_clock) {
-		spin_unlock(&ka->pvclock_gtod_sync_lock);
+		spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
 		return get_kvmclock_base_ns() + ka->kvmclock_offset;
 	}
 
 	hv_clock.tsc_timestamp = ka->master_cycle_now;
 	hv_clock.system_time = ka->master_kernel_ns + ka->kvmclock_offset;
-	spin_unlock(&ka->pvclock_gtod_sync_lock);
+	spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
 
 	/* both __this_cpu_read() and rdtsc() should be on the same cpu */
 	get_cpu();
@@ -2675,13 +2687,13 @@ static int kvm_guest_time_update(struct kvm_vcpu *v)
 	 * If the host uses TSC clock, then passthrough TSC as stable
 	 * to the guest.
 	 */
-	spin_lock(&ka->pvclock_gtod_sync_lock);
+	spin_lock_irqsave(&ka->pvclock_gtod_sync_lock, flags);
 	use_master_clock = ka->use_master_clock;
 	if (use_master_clock) {
 		host_tsc = ka->master_cycle_now;
 		kernel_ns = ka->master_kernel_ns;
 	}
-	spin_unlock(&ka->pvclock_gtod_sync_lock);
+	spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
 
 	/* Keep irq disabled to prevent changes to the clock */
 	local_irq_save(flags);
@@ -2957,6 +2969,11 @@ static void record_steal_time(struct kvm_vcpu *vcpu)
 	struct kvm_host_map map;
 	struct kvm_steal_time *st;
 
+	if (kvm_xen_msr_enabled(vcpu->kvm)) {
+		kvm_xen_runstate_set_running(vcpu);
+		return;
+	}
+
 	if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED))
 		return;
 
@@ -3070,7 +3087,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		return kvm_set_apic_base(vcpu, msr_info);
 	case APIC_BASE_MSR ... APIC_BASE_MSR + 0xff:
 		return kvm_x2apic_msr_write(vcpu, msr, data);
-	case MSR_IA32_TSCDEADLINE:
+	case MSR_IA32_TSC_DEADLINE:
 		kvm_set_lapic_tscdeadline_msr(vcpu, data);
 		break;
 	case MSR_IA32_TSC_ADJUST:
@@ -3432,7 +3449,7 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		break;
 	case APIC_BASE_MSR ... APIC_BASE_MSR + 0xff:
 		return kvm_x2apic_msr_read(vcpu, msr_info->index, &msr_info->data);
-	case MSR_IA32_TSCDEADLINE:
+	case MSR_IA32_TSC_DEADLINE:
 		msr_info->data = kvm_get_lapic_tscdeadline_msr(vcpu);
 		break;
 	case MSR_IA32_TSC_ADJUST:
@@ -3756,11 +3773,15 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_ENFORCE_PV_FEATURE_CPUID:
 		r = 1;
 		break;
+#ifdef CONFIG_KVM_XEN
 	case KVM_CAP_XEN_HVM:
 		r = KVM_XEN_HVM_CONFIG_HYPERCALL_MSR |
 		    KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL |
 		    KVM_XEN_HVM_CONFIG_SHARED_INFO;
+		if (sched_info_on())
+			r |= KVM_XEN_HVM_CONFIG_RUNSTATE;
 		break;
+#endif
 	case KVM_CAP_SYNC_REGS:
 		r = KVM_SYNC_X86_VALID_FIELDS;
 		break;
@@ -4004,7 +4025,6 @@ static void kvm_steal_time_set_preempted(struct kvm_vcpu *vcpu)
 {
 	struct kvm_host_map map;
 	struct kvm_steal_time *st;
-	int idx;
 
 	if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED))
 		return;
@@ -4012,15 +4032,9 @@ static void kvm_steal_time_set_preempted(struct kvm_vcpu *vcpu)
 	if (vcpu->arch.st.preempted)
 		return;
 
-	/*
-	 * Take the srcu lock as memslots will be accessed to check the gfn
-	 * cache generation against the memslots generation.
-	 */
-	idx = srcu_read_lock(&vcpu->kvm->srcu);
-
 	if (kvm_map_gfn(vcpu, vcpu->arch.st.msr_val >> PAGE_SHIFT, &map,
 			&vcpu->arch.st.cache, true))
-		goto out;
+		return;
 
 	st = map.hva +
 		offset_in_page(vcpu->arch.st.msr_val & KVM_STEAL_VALID_BITS);
@@ -4028,17 +4042,26 @@ static void kvm_steal_time_set_preempted(struct kvm_vcpu *vcpu)
 	st->preempted = vcpu->arch.st.preempted = KVM_VCPU_PREEMPTED;
 
 	kvm_unmap_gfn(vcpu, &map, &vcpu->arch.st.cache, true, true);
-
-out:
-	srcu_read_unlock(&vcpu->kvm->srcu, idx);
 }
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
+	int idx;
+
 	if (vcpu->preempted && !vcpu->arch.guest_state_protected)
 		vcpu->arch.preempted_in_kernel = !static_call(kvm_x86_get_cpl)(vcpu);
 
-	kvm_steal_time_set_preempted(vcpu);
+	/*
+	 * Take the srcu lock as memslots will be accessed to check the gfn
+	 * cache generation against the memslots generation.
+	 */
+	idx = srcu_read_lock(&vcpu->kvm->srcu);
+	if (kvm_xen_msr_enabled(vcpu->kvm))
+		kvm_xen_runstate_set_preempted(vcpu);
+	else
+		kvm_steal_time_set_preempted(vcpu);
+	srcu_read_unlock(&vcpu->kvm->srcu, idx);
+
 	static_call(kvm_x86_vcpu_put)(vcpu);
 	vcpu->arch.last_host_tsc = rdtsc();
 	/*
@@ -5013,6 +5036,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 	case KVM_GET_SUPPORTED_HV_CPUID:
 		r = kvm_ioctl_get_supported_hv_cpuid(vcpu, argp);
 		break;
+#ifdef CONFIG_KVM_XEN
 	case KVM_XEN_VCPU_GET_ATTR: {
 		struct kvm_xen_vcpu_attr xva;
 
@@ -5033,6 +5057,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 		r = kvm_xen_vcpu_set_attr(vcpu, &xva);
 		break;
 	}
+#endif
 	default:
 		r = -EINVAL;
 	}
@@ -5215,10 +5240,18 @@ static int kvm_vm_ioctl_reinject(struct kvm *kvm,
 
 void kvm_arch_sync_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot)
 {
+
 	/*
-	 * Flush potentially hardware-cached dirty pages to dirty_bitmap.
+	 * Flush all CPUs' dirty log buffers to the  dirty_bitmap.  Called
+	 * before reporting dirty_bitmap to userspace.  KVM flushes the buffers
+	 * on all VM-Exits, thus we only need to kick running vCPUs to force a
+	 * VM-Exit.
 	 */
-	static_call_cond(kvm_x86_flush_log_dirty)(kvm);
+	struct kvm_vcpu *vcpu;
+	int i;
+
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		kvm_vcpu_kick(vcpu);
 }
 
 int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_event,
@@ -5329,25 +5362,34 @@ split_irqchip_unlock:
 	return r;
 }
 
-static void kvm_clear_msr_filter(struct kvm *kvm)
+static struct kvm_x86_msr_filter *kvm_alloc_msr_filter(bool default_allow)
+{
+	struct kvm_x86_msr_filter *msr_filter;
+
+	msr_filter = kzalloc(sizeof(*msr_filter), GFP_KERNEL_ACCOUNT);
+	if (!msr_filter)
+		return NULL;
+
+	msr_filter->default_allow = default_allow;
+	return msr_filter;
+}
+
+static void kvm_free_msr_filter(struct kvm_x86_msr_filter *msr_filter)
 {
 	u32 i;
-	u32 count = kvm->arch.msr_filter.count;
-	struct msr_bitmap_range ranges[16];
 
-	mutex_lock(&kvm->lock);
-	kvm->arch.msr_filter.count = 0;
-	memcpy(ranges, kvm->arch.msr_filter.ranges, count * sizeof(ranges[0]));
-	mutex_unlock(&kvm->lock);
-	synchronize_srcu(&kvm->srcu);
+	if (!msr_filter)
+		return;
+
+	for (i = 0; i < msr_filter->count; i++)
+		kfree(msr_filter->ranges[i].bitmap);
 
-	for (i = 0; i < count; i++)
-		kfree(ranges[i].bitmap);
+	kfree(msr_filter);
 }
 
-static int kvm_add_msr_filter(struct kvm *kvm, struct kvm_msr_filter_range *user_range)
+static int kvm_add_msr_filter(struct kvm_x86_msr_filter *msr_filter,
+			      struct kvm_msr_filter_range *user_range)
 {
-	struct msr_bitmap_range *ranges = kvm->arch.msr_filter.ranges;
 	struct msr_bitmap_range range;
 	unsigned long *bitmap = NULL;
 	size_t bitmap_size;
@@ -5381,11 +5423,9 @@ static int kvm_add_msr_filter(struct kvm *kvm, struct kvm_msr_filter_range *user
 		goto err;
 	}
 
-	/* Everything ok, add this range identifier to our global pool */
-	ranges[kvm->arch.msr_filter.count] = range;
-	/* Make sure we filled the array before we tell anyone to walk it */
-	smp_wmb();
-	kvm->arch.msr_filter.count++;
+	/* Everything ok, add this range identifier. */
+	msr_filter->ranges[msr_filter->count] = range;
+	msr_filter->count++;
 
 	return 0;
 err:
@@ -5396,10 +5436,11 @@ err:
 static int kvm_vm_ioctl_set_msr_filter(struct kvm *kvm, void __user *argp)
 {
 	struct kvm_msr_filter __user *user_msr_filter = argp;
+	struct kvm_x86_msr_filter *new_filter, *old_filter;
 	struct kvm_msr_filter filter;
 	bool default_allow;
-	int r = 0;
 	bool empty = true;
+	int r = 0;
 	u32 i;
 
 	if (copy_from_user(&filter, user_msr_filter, sizeof(filter)))
@@ -5412,25 +5453,32 @@ static int kvm_vm_ioctl_set_msr_filter(struct kvm *kvm, void __user *argp)
 	if (empty && !default_allow)
 		return -EINVAL;
 
-	kvm_clear_msr_filter(kvm);
-
-	kvm->arch.msr_filter.default_allow = default_allow;
+	new_filter = kvm_alloc_msr_filter(default_allow);
+	if (!new_filter)
+		return -ENOMEM;
 
-	/*
-	 * Protect from concurrent calls to this function that could trigger
-	 * a TOCTOU violation on kvm->arch.msr_filter.count.
-	 */
-	mutex_lock(&kvm->lock);
 	for (i = 0; i < ARRAY_SIZE(filter.ranges); i++) {
-		r = kvm_add_msr_filter(kvm, &filter.ranges[i]);
-		if (r)
-			break;
+		r = kvm_add_msr_filter(new_filter, &filter.ranges[i]);
+		if (r) {
+			kvm_free_msr_filter(new_filter);
+			return r;
+		}
 	}
 
+	mutex_lock(&kvm->lock);
+
+	/* The per-VM filter is protected by kvm->lock... */
+	old_filter = srcu_dereference_check(kvm->arch.msr_filter, &kvm->srcu, 1);
+
+	rcu_assign_pointer(kvm->arch.msr_filter, new_filter);
+	synchronize_srcu(&kvm->srcu);
+
+	kvm_free_msr_filter(old_filter);
+
 	kvm_make_all_cpus_request(kvm, KVM_REQ_MSR_FILTER_CHANGED);
 	mutex_unlock(&kvm->lock);
 
-	return r;
+	return 0;
 }
 
 long kvm_arch_vm_ioctl(struct file *filp,
@@ -5646,6 +5694,7 @@ set_pit2_out:
 			kvm->arch.bsp_vcpu_id = arg;
 		mutex_unlock(&kvm->lock);
 		break;
+#ifdef CONFIG_KVM_XEN
 	case KVM_XEN_HVM_CONFIG: {
 		struct kvm_xen_hvm_config xhc;
 		r = -EFAULT;
@@ -5674,7 +5723,9 @@ set_pit2_out:
 		r = kvm_xen_hvm_set_attr(kvm, &xha);
 		break;
 	}
+#endif
 	case KVM_SET_CLOCK: {
+		struct kvm_arch *ka = &kvm->arch;
 		struct kvm_clock_data user_ns;
 		u64 now_ns;
 
@@ -5693,8 +5744,22 @@ set_pit2_out:
 		 * pvclock_update_vm_gtod_copy().
 		 */
 		kvm_gen_update_masterclock(kvm);
-		now_ns = get_kvmclock_ns(kvm);
-		kvm->arch.kvmclock_offset += user_ns.clock - now_ns;
+
+		/*
+		 * This pairs with kvm_guest_time_update(): when masterclock is
+		 * in use, we use master_kernel_ns + kvmclock_offset to set
+		 * unsigned 'system_time' so if we use get_kvmclock_ns() (which
+		 * is slightly ahead) here we risk going negative on unsigned
+		 * 'system_time' when 'user_ns.clock' is very small.
+		 */
+		spin_lock_irq(&ka->pvclock_gtod_sync_lock);
+		if (kvm->arch.use_master_clock)
+			now_ns = ka->master_kernel_ns;
+		else
+			now_ns = get_kvmclock_base_ns();
+		ka->kvmclock_offset = user_ns.clock - now_ns;
+		spin_unlock_irq(&ka->pvclock_gtod_sync_lock);
+
 		kvm_make_all_cpus_request(kvm, KVM_REQ_CLOCK_UPDATE);
 		break;
 	}
@@ -6578,7 +6643,7 @@ static int kvm_emulate_wbinvd_noskip(struct kvm_vcpu *vcpu)
 		int cpu = get_cpu();
 
 		cpumask_set_cpu(cpu, vcpu->arch.wbinvd_dirty_mask);
-		smp_call_function_many(vcpu->arch.wbinvd_dirty_mask,
+		on_each_cpu_mask(vcpu->arch.wbinvd_dirty_mask,
 				wbinvd_ipi, NULL, 1);
 		put_cpu();
 		cpumask_clear(vcpu->arch.wbinvd_dirty_mask);
@@ -7673,6 +7738,7 @@ static void kvm_hyperv_tsc_notifier(void)
 	struct kvm *kvm;
 	struct kvm_vcpu *vcpu;
 	int cpu;
+	unsigned long flags;
 
 	mutex_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
@@ -7688,17 +7754,15 @@ static void kvm_hyperv_tsc_notifier(void)
 	list_for_each_entry(kvm, &vm_list, vm_list) {
 		struct kvm_arch *ka = &kvm->arch;
 
-		spin_lock(&ka->pvclock_gtod_sync_lock);
-
+		spin_lock_irqsave(&ka->pvclock_gtod_sync_lock, flags);
 		pvclock_update_vm_gtod_copy(kvm);
+		spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
 
 		kvm_for_each_vcpu(cpu, vcpu, kvm)
 			kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
 
 		kvm_for_each_vcpu(cpu, vcpu, kvm)
 			kvm_clear_request(KVM_REQ_MCLOCK_INPROGRESS, vcpu);
-
-		spin_unlock(&ka->pvclock_gtod_sync_lock);
 	}
 	mutex_unlock(&kvm_lock);
 }
@@ -8032,7 +8096,10 @@ void kvm_arch_exit(void)
 	kvm_mmu_module_exit();
 	free_percpu(user_return_msrs);
 	kmem_cache_destroy(x86_fpu_cache);
+#ifdef CONFIG_KVM_XEN
+	static_key_deferred_flush(&kvm_xen_enabled);
 	WARN_ON(static_branch_unlikely(&kvm_xen_enabled.key));
+#endif
 }
 
 static int __kvm_vcpu_halt(struct kvm_vcpu *vcpu, int state, int reason)
@@ -8980,6 +9047,9 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 			kvm_check_async_pf_completion(vcpu);
 		if (kvm_check_request(KVM_REQ_MSR_FILTER_CHANGED, vcpu))
 			static_call(kvm_x86_msr_filter_changed)(vcpu);
+
+		if (kvm_check_request(KVM_REQ_UPDATE_CPU_DIRTY_LOGGING, vcpu))
+			static_call(kvm_x86_update_cpu_dirty_logging)(vcpu);
 	}
 
 	if (kvm_check_request(KVM_REQ_EVENT, vcpu) || req_int_win ||
@@ -10570,7 +10640,7 @@ void __user * __x86_set_memory_region(struct kvm *kvm, int id, gpa_t gpa,
 			return (void __user *)hva;
 	} else {
 		if (!slot || !slot->npages)
-			return 0;
+			return NULL;
 
 		old_npages = slot->npages;
 		hva = slot->userspace_addr;
@@ -10603,8 +10673,6 @@ void kvm_arch_pre_destroy_vm(struct kvm *kvm)
 
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
-	u32 i;
-
 	if (current->mm == kvm->mm) {
 		/*
 		 * Free memory regions allocated on behalf of userspace,
@@ -10620,8 +10688,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 		mutex_unlock(&kvm->slots_lock);
 	}
 	static_call_cond(kvm_x86_vm_destroy)(kvm);
-	for (i = 0; i < kvm->arch.msr_filter.count; i++)
-		kfree(kvm->arch.msr_filter.ranges[i].bitmap);
+	kvm_free_msr_filter(srcu_dereference_check(kvm->arch.msr_filter, &kvm->srcu, 1));
 	kvm_pic_destroy(kvm);
 	kvm_ioapic_destroy(kvm);
 	kvm_free_vcpus(kvm);
@@ -10748,76 +10815,97 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
 	return 0;
 }
 
+
+static void kvm_mmu_update_cpu_dirty_logging(struct kvm *kvm, bool enable)
+{
+	struct kvm_arch *ka = &kvm->arch;
+
+	if (!kvm_x86_ops.cpu_dirty_log_size)
+		return;
+
+	if ((enable && ++ka->cpu_dirty_logging_count == 1) ||
+	    (!enable && --ka->cpu_dirty_logging_count == 0))
+		kvm_make_all_cpus_request(kvm, KVM_REQ_UPDATE_CPU_DIRTY_LOGGING);
+
+	WARN_ON_ONCE(ka->cpu_dirty_logging_count < 0);
+}
+
 static void kvm_mmu_slot_apply_flags(struct kvm *kvm,
 				     struct kvm_memory_slot *old,
 				     struct kvm_memory_slot *new,
 				     enum kvm_mr_change change)
 {
+	bool log_dirty_pages = new->flags & KVM_MEM_LOG_DIRTY_PAGES;
+
 	/*
-	 * Nothing to do for RO slots or CREATE/MOVE/DELETE of a slot.
-	 * See comments below.
+	 * Update CPU dirty logging if dirty logging is being toggled.  This
+	 * applies to all operations.
 	 */
-	if ((change != KVM_MR_FLAGS_ONLY) || (new->flags & KVM_MEM_READONLY))
-		return;
+	if ((old->flags ^ new->flags) & KVM_MEM_LOG_DIRTY_PAGES)
+		kvm_mmu_update_cpu_dirty_logging(kvm, log_dirty_pages);
 
 	/*
-	 * Dirty logging tracks sptes in 4k granularity, meaning that large
-	 * sptes have to be split.  If live migration is successful, the guest
-	 * in the source machine will be destroyed and large sptes will be
-	 * created in the destination. However, if the guest continues to run
-	 * in the source machine (for example if live migration fails), small
-	 * sptes will remain around and cause bad performance.
+	 * Nothing more to do for RO slots (which can't be dirtied and can't be
+	 * made writable) or CREATE/MOVE/DELETE of a slot.
 	 *
-	 * Scan sptes if dirty logging has been stopped, dropping those
-	 * which can be collapsed into a single large-page spte.  Later
-	 * page faults will create the large-page sptes.
-	 *
-	 * There is no need to do this in any of the following cases:
+	 * For a memslot with dirty logging disabled:
 	 * CREATE:      No dirty mappings will already exist.
 	 * MOVE/DELETE: The old mappings will already have been cleaned up by
 	 *		kvm_arch_flush_shadow_memslot()
+	 *
+	 * For a memslot with dirty logging enabled:
+	 * CREATE:      No shadow pages exist, thus nothing to write-protect
+	 *		and no dirty bits to clear.
+	 * MOVE/DELETE: The old mappings will already have been cleaned up by
+	 *		kvm_arch_flush_shadow_memslot().
 	 */
-	if ((old->flags & KVM_MEM_LOG_DIRTY_PAGES) &&
-	    !(new->flags & KVM_MEM_LOG_DIRTY_PAGES))
-		kvm_mmu_zap_collapsible_sptes(kvm, new);
+	if ((change != KVM_MR_FLAGS_ONLY) || (new->flags & KVM_MEM_READONLY))
+		return;
 
 	/*
-	 * Enable or disable dirty logging for the slot.
-	 *
-	 * For KVM_MR_DELETE and KVM_MR_MOVE, the shadow pages of the old
-	 * slot have been zapped so no dirty logging updates are needed for
-	 * the old slot.
-	 * For KVM_MR_CREATE and KVM_MR_MOVE, once the new slot is visible
-	 * any mappings that might be created in it will consume the
-	 * properties of the new slot and do not need to be updated here.
-	 *
-	 * When PML is enabled, the kvm_x86_ops dirty logging hooks are
-	 * called to enable/disable dirty logging.
-	 *
-	 * When disabling dirty logging with PML enabled, the D-bit is set
-	 * for sptes in the slot in order to prevent unnecessary GPA
-	 * logging in the PML buffer (and potential PML buffer full VMEXIT).
-	 * This guarantees leaving PML enabled for the guest's lifetime
-	 * won't have any additional overhead from PML when the guest is
-	 * running with dirty logging disabled.
-	 *
-	 * When enabling dirty logging, large sptes are write-protected
-	 * so they can be split on first write.  New large sptes cannot
-	 * be created for this slot until the end of the logging.
-	 * See the comments in fast_page_fault().
-	 * For small sptes, nothing is done if the dirty log is in the
-	 * initial-all-set state.  Otherwise, depending on whether pml
-	 * is enabled the D-bit or the W-bit will be cleared.
+	 * READONLY and non-flags changes were filtered out above, and the only
+	 * other flag is LOG_DIRTY_PAGES, i.e. something is wrong if dirty
+	 * logging isn't being toggled on or off.
 	 */
-	if (new->flags & KVM_MEM_LOG_DIRTY_PAGES) {
-		if (kvm_x86_ops.slot_enable_log_dirty) {
-			static_call(kvm_x86_slot_enable_log_dirty)(kvm, new);
-		} else {
-			int level =
-				kvm_dirty_log_manual_protect_and_init_set(kvm) ?
-				PG_LEVEL_2M : PG_LEVEL_4K;
+	if (WARN_ON_ONCE(!((old->flags ^ new->flags) & KVM_MEM_LOG_DIRTY_PAGES)))
+		return;
+
+	if (!log_dirty_pages) {
+		/*
+		 * Dirty logging tracks sptes in 4k granularity, meaning that
+		 * large sptes have to be split.  If live migration succeeds,
+		 * the guest in the source machine will be destroyed and large
+		 * sptes will be created in the destination.  However, if the
+		 * guest continues to run in the source machine (for example if
+		 * live migration fails), small sptes will remain around and
+		 * cause bad performance.
+		 *
+		 * Scan sptes if dirty logging has been stopped, dropping those
+		 * which can be collapsed into a single large-page spte.  Later
+		 * page faults will create the large-page sptes.
+		 */
+		kvm_mmu_zap_collapsible_sptes(kvm, new);
+	} else {
+		/* By default, write-protect everything to log writes. */
+		int level = PG_LEVEL_4K;
+
+		if (kvm_x86_ops.cpu_dirty_log_size) {
+			/*
+			 * Clear all dirty bits, unless pages are treated as
+			 * dirty from the get-go.
+			 */
+			if (!kvm_dirty_log_manual_protect_and_init_set(kvm))
+				kvm_mmu_slot_leaf_clear_dirty(kvm, new);
 
 			/*
+			 * Write-protect large pages on write so that dirty
+			 * logging happens at 4k granularity.  No need to
+			 * write-protect small SPTEs since write accesses are
+			 * logged by the CPU via dirty bits.
+			 */
+			level = PG_LEVEL_2M;
+		} else if (kvm_dirty_log_manual_protect_and_init_set(kvm)) {
+			/*
 			 * If we're with initial-all-set, we don't need
 			 * to write protect any small page because
 			 * they're reported as dirty already.  However
@@ -10825,10 +10913,9 @@ static void kvm_mmu_slot_apply_flags(struct kvm *kvm,
 			 * so that the page split can happen lazily on
 			 * the first write to the huge page.
 			 */
-			kvm_mmu_slot_remove_write_access(kvm, new, level);
+			level = PG_LEVEL_2M;
 		}
-	} else {
-		static_call_cond(kvm_x86_slot_disable_log_dirty)(kvm, new);
+		kvm_mmu_slot_remove_write_access(kvm, new, level);
 	}
 }