5 files changed, 100 insertions, 1 deletions

diff --git a/kernel/arch/x86/include/asm/kvm_host.h b/kernel/arch/x86/include/asm/kvm_host.h
index 30cfd6429..21637c9e9 100644
--- a/kernel/arch/x86/include/asm/kvm_host.h
+++ b/kernel/arch/x86/include/asm/kvm_host.h
@@ -911,6 +911,9 @@ struct kvm_x86_ops {
 	void (*post_block)(struct kvm_vcpu *vcpu);
 	int (*update_pi_irte)(struct kvm *kvm, unsigned int host_irq,
 			      uint32_t guest_irq, bool set);
+
+	int (*set_hv_timer)(struct kvm_vcpu *vcpu, u64 guest_deadline_tsc);
+	void (*cancel_hv_timer)(struct kvm_vcpu *vcpu);
 };
 
 struct kvm_arch_async_pf {
diff --git a/kernel/arch/x86/kvm/lapic.c b/kernel/arch/x86/kvm/lapic.c
index 1e1e7eb8e..701f91234 100644
--- a/kernel/arch/x86/kvm/lapic.c
+++ b/kernel/arch/x86/kvm/lapic.c
@@ -1288,6 +1288,68 @@ static void start_sw_tscdeadline(struct kvm_lapic *apic)
 	local_irq_restore(flags);
 }
 
+bool kvm_lapic_hv_timer_in_use(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.apic->lapic_timer.hv_timer_in_use;
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_hv_timer_in_use);
+
+void kvm_lapic_expired_hv_timer(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	WARN_ON(!apic->lapic_timer.hv_timer_in_use);
+	WARN_ON(swait_active(&vcpu->wq));
+	kvm_x86_ops->cancel_hv_timer(vcpu);
+	apic->lapic_timer.hv_timer_in_use = false;
+	apic_timer_expired(apic);
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_expired_hv_timer);
+
+void kvm_lapic_switch_to_hv_timer(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	WARN_ON(apic->lapic_timer.hv_timer_in_use);
+
+	if (apic_lvtt_tscdeadline(apic) &&
+	    !atomic_read(&apic->lapic_timer.pending)) {
+		u64 tscdeadline = apic->lapic_timer.tscdeadline;
+
+		if (!kvm_x86_ops->set_hv_timer(vcpu, tscdeadline)) {
+			apic->lapic_timer.hv_timer_in_use = true;
+			hrtimer_cancel(&apic->lapic_timer.timer);
+
+			/* In case the sw timer triggered in the window */
+			if (atomic_read(&apic->lapic_timer.pending)) {
+				apic->lapic_timer.hv_timer_in_use = false;
+				kvm_x86_ops->cancel_hv_timer(apic->vcpu);
+			}
+		}
+		trace_kvm_hv_timer_state(vcpu->vcpu_id,
+				apic->lapic_timer.hv_timer_in_use);
+	}
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_switch_to_hv_timer);
+
+void kvm_lapic_switch_to_sw_timer(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	/* Possibly the TSC deadline timer is not enabled yet */
+	if (!apic->lapic_timer.hv_timer_in_use)
+		return;
+
+	kvm_x86_ops->cancel_hv_timer(vcpu);
+	apic->lapic_timer.hv_timer_in_use = false;
+
+	if (atomic_read(&apic->lapic_timer.pending))
+		return;
+
+	start_sw_tscdeadline(apic);
+}
+EXPORT_SYMBOL_GPL(kvm_lapic_switch_to_sw_timer);
+
 static void start_apic_timer(struct kvm_lapic *apic)
 {
 	ktime_t now;
@@ -1334,7 +1396,16 @@ static void start_apic_timer(struct kvm_lapic *apic)
 			   ktime_to_ns(ktime_add_ns(now,
 					apic->lapic_timer.period)));
 	} else if (apic_lvtt_tscdeadline(apic)) {
-		start_sw_tscdeadline(apic);
+		/* lapic timer in tsc deadline mode */
+		u64 tscdeadline = apic->lapic_timer.tscdeadline;
+
+		if (kvm_x86_ops->set_hv_timer &&
+		    !kvm_x86_ops->set_hv_timer(apic->vcpu, tscdeadline)) {
+			apic->lapic_timer.hv_timer_in_use = true;
+			trace_kvm_hv_timer_state(apic->vcpu->vcpu_id,
+					apic->lapic_timer.hv_timer_in_use);
+		} else
+			start_sw_tscdeadline(apic);
 	}
 }
 
diff --git a/kernel/arch/x86/kvm/lapic.h b/kernel/arch/x86/kvm/lapic.h
index fde8e35d5..640ad275b 100644
--- a/kernel/arch/x86/kvm/lapic.h
+++ b/kernel/arch/x86/kvm/lapic.h
@@ -16,6 +16,7 @@ struct kvm_timer {
 	u64 tscdeadline;
 	u64 expired_tscdeadline;
 	atomic_t pending;			/* accumulated triggered timers */
+	bool hv_timer_in_use;
 };
 
 struct kvm_lapic {
@@ -170,4 +171,8 @@ void wait_lapic_expire(struct kvm_vcpu *vcpu);
 
 bool kvm_intr_is_single_vcpu_fast(struct kvm *kvm, struct kvm_lapic_irq *irq,
 			struct kvm_vcpu **dest_vcpu);
+void kvm_lapic_switch_to_sw_timer(struct kvm_vcpu *vcpu);
+void kvm_lapic_switch_to_hv_timer(struct kvm_vcpu *vcpu);
+void kvm_lapic_expired_hv_timer(struct kvm_vcpu *vcpu);
+bool kvm_lapic_hv_timer_in_use(struct kvm_vcpu *vcpu);
 #endif
diff --git a/kernel/arch/x86/kvm/trace.h b/kernel/arch/x86/kvm/trace.h
index ab9ae67a8..b41f7a013 100644
--- a/kernel/arch/x86/kvm/trace.h
+++ b/kernel/arch/x86/kvm/trace.h
@@ -1025,6 +1025,21 @@ TRACE_EVENT(kvm_pi_irte_update,
 		  __entry->pi_desc_addr)
 );
 
+TRACE_EVENT(kvm_hv_timer_state,
+		TP_PROTO(unsigned int vcpu_id, unsigned int hv_timer_in_use),
+		TP_ARGS(vcpu_id, hv_timer_in_use),
+		TP_STRUCT__entry(
+			__field(unsigned int, vcpu_id)
+			__field(unsigned int, hv_timer_in_use)
+			),
+		TP_fast_assign(
+			__entry->vcpu_id = vcpu_id;
+			__entry->hv_timer_in_use = hv_timer_in_use;
+			),
+		TP_printk("vcpu_id %x hv_timer %x\n",
+			__entry->vcpu_id,
+			__entry->hv_timer_in_use)
+);
 #endif /* _TRACE_KVM_H */
 
 #undef TRACE_INCLUDE_PATH
diff --git a/kernel/arch/x86/kvm/x86.c b/kernel/arch/x86/kvm/x86.c
index 27419ba5a..e3e1a8c1b 100644
--- a/kernel/arch/x86/kvm/x86.c
+++ b/kernel/arch/x86/kvm/x86.c
@@ -2718,6 +2718,11 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 				rdtsc() - vcpu->arch.last_host_tsc;
 		if (tsc_delta < 0)
 			mark_tsc_unstable("KVM discovered backwards TSC");
+
+		if (kvm_lapic_hv_timer_in_use(vcpu) &&
+				kvm_x86_ops->set_hv_timer(vcpu,
+					kvm_get_lapic_tscdeadline_msr(vcpu)))
+			kvm_lapic_switch_to_sw_timer(vcpu);
 		if (check_tsc_unstable()) {
 			u64 offset = kvm_compute_tsc_offset(vcpu,
 						vcpu->arch.last_guest_tsc);