1 files changed, 33 insertions, 34 deletions

diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 0aa158657f20..3a3a6854dcca 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -65,7 +65,9 @@
 #define APIC_BROADCAST			0xFF
 #define X2APIC_BROADCAST		0xFFFFFFFFul
 
-#define LAPIC_TIMER_ADVANCE_ADJUST_DONE 100
+static bool lapic_timer_advance_dynamic __read_mostly;
+#define LAPIC_TIMER_ADVANCE_ADJUST_MIN 100
+#define LAPIC_TIMER_ADVANCE_ADJUST_MAX 5000
 #define LAPIC_TIMER_ADVANCE_ADJUST_INIT 1000
 /* step-by-step approximation to mitigate fluctuation */
 #define LAPIC_TIMER_ADVANCE_ADJUST_STEP 8
@@ -216,6 +218,9 @@ static void recalculate_apic_map(struct kvm *kvm)
 		if (!apic_x2apic_mode(apic) && !new->phys_map[xapic_id])
 			new->phys_map[xapic_id] = apic;
 
+		if (!kvm_apic_sw_enabled(apic))
+			continue;
+
 		ldr = kvm_lapic_get_reg(apic, APIC_LDR);
 
 		if (apic_x2apic_mode(apic)) {
@@ -258,6 +263,8 @@ static inline void apic_set_spiv(struct kvm_lapic *apic, u32 val)
 			static_key_slow_dec_deferred(&apic_sw_disabled);
 		else
 			static_key_slow_inc(&apic_sw_disabled.key);
+
+		recalculate_apic_map(apic->vcpu->kvm);
 	}
 }
 
@@ -1193,10 +1200,8 @@ void kvm_apic_set_eoi_accelerated(struct kvm_vcpu *vcpu, int vector)
 }
 EXPORT_SYMBOL_GPL(kvm_apic_set_eoi_accelerated);
 
-static void apic_send_ipi(struct kvm_lapic *apic)
+static void apic_send_ipi(struct kvm_lapic *apic, u32 icr_low, u32 icr_high)
 {
-	u32 icr_low = kvm_lapic_get_reg(apic, APIC_ICR);
-	u32 icr_high = kvm_lapic_get_reg(apic, APIC_ICR2);
 	struct kvm_lapic_irq irq;
 
 	irq.vector = icr_low & APIC_VECTOR_MASK;
@@ -1482,26 +1487,25 @@ static inline void adjust_lapic_timer_advance(struct kvm_vcpu *vcpu,
 	u32 timer_advance_ns = apic->lapic_timer.timer_advance_ns;
 	u64 ns;
 
+	/* Do not adjust for tiny fluctuations or large random spikes. */
+	if (abs(advance_expire_delta) > LAPIC_TIMER_ADVANCE_ADJUST_MAX ||
+	    abs(advance_expire_delta) < LAPIC_TIMER_ADVANCE_ADJUST_MIN)
+		return;
+
 	/* too early */
 	if (advance_expire_delta < 0) {
 		ns = -advance_expire_delta * 1000000ULL;
 		do_div(ns, vcpu->arch.virtual_tsc_khz);
-		timer_advance_ns -= min((u32)ns,
-			timer_advance_ns / LAPIC_TIMER_ADVANCE_ADJUST_STEP);
+		timer_advance_ns -= ns/LAPIC_TIMER_ADVANCE_ADJUST_STEP;
 	} else {
 	/* too late */
 		ns = advance_expire_delta * 1000000ULL;
 		do_div(ns, vcpu->arch.virtual_tsc_khz);
-		timer_advance_ns += min((u32)ns,
-			timer_advance_ns / LAPIC_TIMER_ADVANCE_ADJUST_STEP);
+		timer_advance_ns += ns/LAPIC_TIMER_ADVANCE_ADJUST_STEP;
 	}
 
-	if (abs(advance_expire_delta) < LAPIC_TIMER_ADVANCE_ADJUST_DONE)
-		apic->lapic_timer.timer_advance_adjust_done = true;
-	if (unlikely(timer_advance_ns > 5000)) {
+	if (unlikely(timer_advance_ns > LAPIC_TIMER_ADVANCE_ADJUST_MAX))
 		timer_advance_ns = LAPIC_TIMER_ADVANCE_ADJUST_INIT;
-		apic->lapic_timer.timer_advance_adjust_done = false;
-	}
 	apic->lapic_timer.timer_advance_ns = timer_advance_ns;
 }
 
@@ -1521,7 +1525,7 @@ static void __kvm_wait_lapic_expire(struct kvm_vcpu *vcpu)
 	if (guest_tsc < tsc_deadline)
 		__wait_lapic_expire(vcpu, tsc_deadline - guest_tsc);
 
-	if (unlikely(!apic->lapic_timer.timer_advance_adjust_done))
+	if (lapic_timer_advance_dynamic)
 		adjust_lapic_timer_advance(vcpu, apic->lapic_timer.advance_expire_delta);
 }
 
@@ -1548,7 +1552,6 @@ static void kvm_apic_inject_pending_timer_irqs(struct kvm_lapic *apic)
 static void apic_timer_expired(struct kvm_lapic *apic)
 {
 	struct kvm_vcpu *vcpu = apic->vcpu;
-	struct swait_queue_head *q = &vcpu->wq;
 	struct kvm_timer *ktimer = &apic->lapic_timer;
 
 	if (atomic_read(&apic->lapic_timer.pending))
@@ -1566,13 +1569,6 @@ static void apic_timer_expired(struct kvm_lapic *apic)
 
 	atomic_inc(&apic->lapic_timer.pending);
 	kvm_set_pending_timer(vcpu);
-
-	/*
-	 * For x86, the atomic_inc() is serialized, thus
-	 * using swait_active() is safe.
-	 */
-	if (swait_active(q))
-		swake_up_one(q);
 }
 
 static void start_sw_tscdeadline(struct kvm_lapic *apic)
@@ -1601,7 +1597,7 @@ static void start_sw_tscdeadline(struct kvm_lapic *apic)
 	    likely(ns > apic->lapic_timer.timer_advance_ns)) {
 		expire = ktime_add_ns(now, ns);
 		expire = ktime_sub_ns(expire, ktimer->timer_advance_ns);
-		hrtimer_start(&ktimer->timer, expire, HRTIMER_MODE_ABS);
+		hrtimer_start(&ktimer->timer, expire, HRTIMER_MODE_ABS_HARD);
 	} else
 		apic_timer_expired(apic);
 
@@ -1917,8 +1913,9 @@ int kvm_lapic_reg_write(struct kvm_lapic *apic, u32 reg, u32 val)
 	}
 	case APIC_ICR:
 		/* No delay here, so we always clear the pending bit */
-		kvm_lapic_set_reg(apic, APIC_ICR, val & ~(1 << 12));
-		apic_send_ipi(apic);
+		val &= ~(1 << 12);
+		apic_send_ipi(apic, val, kvm_lapic_get_reg(apic, APIC_ICR2));
+		kvm_lapic_set_reg(apic, APIC_ICR, val);
 		break;
 
 	case APIC_ICR2:
@@ -2302,17 +2299,16 @@ int kvm_create_lapic(struct kvm_vcpu *vcpu, int timer_advance_ns)
 	apic->vcpu = vcpu;
 
 	hrtimer_init(&apic->lapic_timer.timer, CLOCK_MONOTONIC,
-		     HRTIMER_MODE_ABS);
+		     HRTIMER_MODE_ABS_HARD);
 	apic->lapic_timer.timer.function = apic_timer_fn;
 	if (timer_advance_ns == -1) {
 		apic->lapic_timer.timer_advance_ns = LAPIC_TIMER_ADVANCE_ADJUST_INIT;
-		apic->lapic_timer.timer_advance_adjust_done = false;
+		lapic_timer_advance_dynamic = true;
 	} else {
 		apic->lapic_timer.timer_advance_ns = timer_advance_ns;
-		apic->lapic_timer.timer_advance_adjust_done = true;
+		lapic_timer_advance_dynamic = false;
 	}
 
-
 	/*
 	 * APIC is created enabled. This will prevent kvm_lapic_set_base from
 	 * thinking that APIC state has changed.
@@ -2487,7 +2483,7 @@ void __kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
 
 	timer = &vcpu->arch.apic->lapic_timer.timer;
 	if (hrtimer_cancel(timer))
-		hrtimer_start_expires(timer, HRTIMER_MODE_ABS);
+		hrtimer_start_expires(timer, HRTIMER_MODE_ABS_HARD);
 }
 
 /*
@@ -2710,11 +2706,14 @@ void kvm_apic_accept_events(struct kvm_vcpu *vcpu)
 		return;
 
 	/*
-	 * INITs are latched while in SMM.  Because an SMM CPU cannot
-	 * be in KVM_MP_STATE_INIT_RECEIVED state, just eat SIPIs
-	 * and delay processing of INIT until the next RSM.
+	 * INITs are latched while CPU is in specific states
+	 * (SMM, VMX non-root mode, SVM with GIF=0).
+	 * Because a CPU cannot be in these states immediately
+	 * after it has processed an INIT signal (and thus in
+	 * KVM_MP_STATE_INIT_RECEIVED state), just eat SIPIs
+	 * and leave the INIT pending.
 	 */
-	if (is_smm(vcpu)) {
+	if (is_smm(vcpu) || kvm_x86_ops->apic_init_signal_blocked(vcpu)) {
 		WARN_ON_ONCE(vcpu->arch.mp_state == KVM_MP_STATE_INIT_RECEIVED);
 		if (test_bit(KVM_APIC_SIPI, &apic->pending_events))
 			clear_bit(KVM_APIC_SIPI, &apic->pending_events);