1 files changed, 177 insertions, 73 deletions

diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index f3ab08caa2c3..646645e981f9 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -122,8 +122,7 @@ static ktime_t tick_init_jiffy_update(void)
 	return period;
 }
 
-
-static void tick_sched_do_timer(ktime_t now)
+static void tick_sched_do_timer(struct tick_sched *ts, ktime_t now)
 {
 	int cpu = smp_processor_id();
 
@@ -143,6 +142,9 @@ static void tick_sched_do_timer(ktime_t now)
 	/* Check, if the jiffies need an update */
 	if (tick_do_timer_cpu == cpu)
 		tick_do_update_jiffies64(now);
+
+	if (ts->inidle)
+		ts->got_idle_tick = 1;
 }
 
 static void tick_sched_handle(struct tick_sched *ts, struct pt_regs *regs)
@@ -474,7 +476,9 @@ __setup("nohz=", setup_tick_nohz);
 
 bool tick_nohz_tick_stopped(void)
 {
-	return __this_cpu_read(tick_cpu_sched.tick_stopped);
+	struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched);
+
+	return ts->tick_stopped;
 }
 
 bool tick_nohz_tick_stopped_cpu(int cpu)
@@ -537,14 +541,11 @@ static void tick_nohz_stop_idle(struct tick_sched *ts, ktime_t now)
 	sched_clock_idle_wakeup_event();
 }
 
-static ktime_t tick_nohz_start_idle(struct tick_sched *ts)
+static void tick_nohz_start_idle(struct tick_sched *ts)
 {
-	ktime_t now = ktime_get();
-
-	ts->idle_entrytime = now;
+	ts->idle_entrytime = ktime_get();
 	ts->idle_active = 1;
 	sched_clock_idle_sleep_event();
-	return now;
 }
 
 /**
@@ -653,13 +654,10 @@ static inline bool local_timer_softirq_pending(void)
 	return local_softirq_pending() & TIMER_SOFTIRQ;
 }
 
-static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
-					 ktime_t now, int cpu)
+static ktime_t tick_nohz_next_event(struct tick_sched *ts, int cpu)
 {
-	struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev);
 	u64 basemono, next_tick, next_tmr, next_rcu, delta, expires;
 	unsigned long seq, basejiff;
-	ktime_t	tick;
 
 	/* Read jiffies and the time when jiffies were updated last */
 	do {
@@ -668,6 +666,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
 		basejiff = jiffies;
 	} while (read_seqretry(&jiffies_lock, seq));
 	ts->last_jiffies = basejiff;
+	ts->timer_expires_base = basemono;
 
 	/*
 	 * Keep the periodic tick, when RCU, architecture or irq_work
@@ -712,47 +711,63 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
 		 * next period, so no point in stopping it either, bail.
 		 */
 		if (!ts->tick_stopped) {
-			tick = 0;
+			ts->timer_expires = 0;
 			goto out;
 		}
 	}
 
 	/*
+	 * If this CPU is the one which had the do_timer() duty last, we limit
+	 * the sleep time to the timekeeping max_deferment value.
+	 * Otherwise we can sleep as long as we want.
+	 */
+	delta = timekeeping_max_deferment();
+	if (cpu != tick_do_timer_cpu &&
+	    (tick_do_timer_cpu != TICK_DO_TIMER_NONE || !ts->do_timer_last))
+		delta = KTIME_MAX;
+
+	/* Calculate the next expiry time */
+	if (delta < (KTIME_MAX - basemono))
+		expires = basemono + delta;
+	else
+		expires = KTIME_MAX;
+
+	ts->timer_expires = min_t(u64, expires, next_tick);
+
+out:
+	return ts->timer_expires;
+}
+
+static void tick_nohz_stop_tick(struct tick_sched *ts, int cpu)
+{
+	struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev);
+	u64 basemono = ts->timer_expires_base;
+	u64 expires = ts->timer_expires;
+	ktime_t tick = expires;
+
+	/* Make sure we won't be trying to stop it twice in a row. */
+	ts->timer_expires_base = 0;
+
+	/*
 	 * If this CPU is the one which updates jiffies, then give up
 	 * the assignment and let it be taken by the CPU which runs
 	 * the tick timer next, which might be this CPU as well. If we
 	 * don't drop this here the jiffies might be stale and
 	 * do_timer() never invoked. Keep track of the fact that it
-	 * was the one which had the do_timer() duty last. If this CPU
-	 * is the one which had the do_timer() duty last, we limit the
-	 * sleep time to the timekeeping max_deferment value.
-	 * Otherwise we can sleep as long as we want.
+	 * was the one which had the do_timer() duty last.
 	 */
-	delta = timekeeping_max_deferment();
 	if (cpu == tick_do_timer_cpu) {
 		tick_do_timer_cpu = TICK_DO_TIMER_NONE;
 		ts->do_timer_last = 1;
 	} else if (tick_do_timer_cpu != TICK_DO_TIMER_NONE) {
-		delta = KTIME_MAX;
 		ts->do_timer_last = 0;
-	} else if (!ts->do_timer_last) {
-		delta = KTIME_MAX;
 	}
 
-	/* Calculate the next expiry time */
-	if (delta < (KTIME_MAX - basemono))
-		expires = basemono + delta;
-	else
-		expires = KTIME_MAX;
-
-	expires = min_t(u64, expires, next_tick);
-	tick = expires;
-
 	/* Skip reprogram of event if its not changed */
 	if (ts->tick_stopped && (expires == ts->next_tick)) {
 		/* Sanity check: make sure clockevent is actually programmed */
 		if (tick == KTIME_MAX || ts->next_tick == hrtimer_get_expires(&ts->sched_timer))
-			goto out;
+			return;
 
 		WARN_ON_ONCE(1);
 		printk_once("basemono: %llu ts->next_tick: %llu dev->next_event: %llu timer->active: %d timer->expires: %llu\n",
@@ -786,7 +801,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
 	if (unlikely(expires == KTIME_MAX)) {
 		if (ts->nohz_mode == NOHZ_MODE_HIGHRES)
 			hrtimer_cancel(&ts->sched_timer);
-		goto out;
+		return;
 	}
 
 	hrtimer_set_expires(&ts->sched_timer, tick);
@@ -795,15 +810,23 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
 		hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED);
 	else
 		tick_program_event(tick, 1);
-out:
-	/*
-	 * Update the estimated sleep length until the next timer
-	 * (not only the tick).
-	 */
-	ts->sleep_length = ktime_sub(dev->next_event, now);
-	return tick;
 }
 
+static void tick_nohz_retain_tick(struct tick_sched *ts)
+{
+	ts->timer_expires_base = 0;
+}
+
+#ifdef CONFIG_NO_HZ_FULL
+static void tick_nohz_stop_sched_tick(struct tick_sched *ts, int cpu)
+{
+	if (tick_nohz_next_event(ts, cpu))
+		tick_nohz_stop_tick(ts, cpu);
+	else
+		tick_nohz_retain_tick(ts);
+}
+#endif /* CONFIG_NO_HZ_FULL */
+
 static void tick_nohz_restart_sched_tick(struct tick_sched *ts, ktime_t now)
 {
 	/* Update jiffies first */
@@ -839,7 +862,7 @@ static void tick_nohz_full_update_tick(struct tick_sched *ts)
 		return;
 
 	if (can_stop_full_tick(cpu, ts))
-		tick_nohz_stop_sched_tick(ts, ktime_get(), cpu);
+		tick_nohz_stop_sched_tick(ts, cpu);
 	else if (ts->tick_stopped)
 		tick_nohz_restart_sched_tick(ts, ktime_get());
 #endif
@@ -865,10 +888,8 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts)
 		return false;
 	}
 
-	if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE)) {
-		ts->sleep_length = NSEC_PER_SEC / HZ;
+	if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE))
 		return false;
-	}
 
 	if (need_resched())
 		return false;
@@ -903,42 +924,65 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts)
 	return true;
 }
 
-static void __tick_nohz_idle_enter(struct tick_sched *ts)
+static void __tick_nohz_idle_stop_tick(struct tick_sched *ts)
 {
-	ktime_t now, expires;
+	ktime_t expires;
 	int cpu = smp_processor_id();
 
-	now = tick_nohz_start_idle(ts);
+	/*
+	 * If tick_nohz_get_sleep_length() ran tick_nohz_next_event(), the
+	 * tick timer expiration time is known already.
+	 */
+	if (ts->timer_expires_base)
+		expires = ts->timer_expires;
+	else if (can_stop_idle_tick(cpu, ts))
+		expires = tick_nohz_next_event(ts, cpu);
+	else
+		return;
+
+	ts->idle_calls++;
 
-	if (can_stop_idle_tick(cpu, ts)) {
+	if (expires > 0LL) {
 		int was_stopped = ts->tick_stopped;
 
-		ts->idle_calls++;
+		tick_nohz_stop_tick(ts, cpu);
 
-		expires = tick_nohz_stop_sched_tick(ts, now, cpu);
-		if (expires > 0LL) {
-			ts->idle_sleeps++;
-			ts->idle_expires = expires;
-		}
+		ts->idle_sleeps++;
+		ts->idle_expires = expires;
 
 		if (!was_stopped && ts->tick_stopped) {
 			ts->idle_jiffies = ts->last_jiffies;
 			nohz_balance_enter_idle(cpu);
 		}
+	} else {
+		tick_nohz_retain_tick(ts);
 	}
 }
 
 /**
- * tick_nohz_idle_enter - stop the idle tick from the idle task
+ * tick_nohz_idle_stop_tick - stop the idle tick from the idle task
  *
  * When the next event is more than a tick into the future, stop the idle tick
- * Called when we start the idle loop.
- *
- * The arch is responsible of calling:
+ */
+void tick_nohz_idle_stop_tick(void)
+{
+	__tick_nohz_idle_stop_tick(this_cpu_ptr(&tick_cpu_sched));
+}
+
+void tick_nohz_idle_retain_tick(void)
+{
+	tick_nohz_retain_tick(this_cpu_ptr(&tick_cpu_sched));
+	/*
+	 * Undo the effect of get_next_timer_interrupt() called from
+	 * tick_nohz_next_event().
+	 */
+	timer_clear_idle();
+}
+
+/**
+ * tick_nohz_idle_enter - prepare for entering idle on the current CPU
  *
- * - rcu_idle_enter() after its last use of RCU before the CPU is put
- *  to sleep.
- * - rcu_idle_exit() before the first use of RCU after the CPU is woken up.
+ * Called when we start the idle loop.
  */
 void tick_nohz_idle_enter(void)
 {
@@ -949,8 +993,11 @@ void tick_nohz_idle_enter(void)
 	local_irq_disable();
 
 	ts = this_cpu_ptr(&tick_cpu_sched);
+
+	WARN_ON_ONCE(ts->timer_expires_base);
+
 	ts->inidle = 1;
-	__tick_nohz_idle_enter(ts);
+	tick_nohz_start_idle(ts);
 
 	local_irq_enable();
 }
@@ -968,21 +1015,62 @@ void tick_nohz_irq_exit(void)
 	struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched);
 
 	if (ts->inidle)
-		__tick_nohz_idle_enter(ts);
+		tick_nohz_start_idle(ts);
 	else
 		tick_nohz_full_update_tick(ts);
 }
 
 /**
- * tick_nohz_get_sleep_length - return the length of the current sleep
+ * tick_nohz_idle_got_tick - Check whether or not the tick handler has run
+ */
+bool tick_nohz_idle_got_tick(void)
+{
+	struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched);
+
+	if (ts->got_idle_tick) {
+		ts->got_idle_tick = 0;
+		return true;
+	}
+	return false;
+}
+
+/**
+ * tick_nohz_get_sleep_length - return the expected length of the current sleep
+ * @delta_next: duration until the next event if the tick cannot be stopped
  *
  * Called from power state control code with interrupts disabled
  */
-ktime_t tick_nohz_get_sleep_length(void)
+ktime_t tick_nohz_get_sleep_length(ktime_t *delta_next)
 {
+	struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev);
 	struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched);
+	int cpu = smp_processor_id();
+	/*
+	 * The idle entry time is expected to be a sufficient approximation of
+	 * the current time at this point.
+	 */
+	ktime_t now = ts->idle_entrytime;
+	ktime_t next_event;
+
+	WARN_ON_ONCE(!ts->inidle);
+
+	*delta_next = ktime_sub(dev->next_event, now);
+
+	if (!can_stop_idle_tick(cpu, ts))
+		return *delta_next;
+
+	next_event = tick_nohz_next_event(ts, cpu);
+	if (!next_event)
+		return *delta_next;
+
+	/*
+	 * If the next highres timer to expire is earlier than next_event, the
+	 * idle governor needs to know that.
+	 */
+	next_event = min_t(u64, next_event,
+			   hrtimer_next_event_without(&ts->sched_timer));
 
-	return ts->sleep_length;
+	return ktime_sub(next_event, now);
 }
 
 /**
@@ -1031,6 +1119,20 @@ static void tick_nohz_account_idle_ticks(struct tick_sched *ts)
 #endif
 }
 
+static void __tick_nohz_idle_restart_tick(struct tick_sched *ts, ktime_t now)
+{
+	tick_nohz_restart_sched_tick(ts, now);
+	tick_nohz_account_idle_ticks(ts);
+}
+
+void tick_nohz_idle_restart_tick(void)
+{
+	struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched);
+
+	if (ts->tick_stopped)
+		__tick_nohz_idle_restart_tick(ts, ktime_get());
+}
+
 /**
  * tick_nohz_idle_exit - restart the idle tick from the idle task
  *
@@ -1041,24 +1143,26 @@ static void tick_nohz_account_idle_ticks(struct tick_sched *ts)
 void tick_nohz_idle_exit(void)
 {
 	struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched);
+	bool idle_active, tick_stopped;
 	ktime_t now;
 
 	local_irq_disable();
 
 	WARN_ON_ONCE(!ts->inidle);
+	WARN_ON_ONCE(ts->timer_expires_base);
 
 	ts->inidle = 0;
+	idle_active = ts->idle_active;
+	tick_stopped = ts->tick_stopped;
 
-	if (ts->idle_active || ts->tick_stopped)
+	if (idle_active || tick_stopped)
 		now = ktime_get();
 
-	if (ts->idle_active)
+	if (idle_active)
 		tick_nohz_stop_idle(ts, now);
 
-	if (ts->tick_stopped) {
-		tick_nohz_restart_sched_tick(ts, now);
-		tick_nohz_account_idle_ticks(ts);
-	}
+	if (tick_stopped)
+		__tick_nohz_idle_restart_tick(ts, now);
 
 	local_irq_enable();
 }
@@ -1074,7 +1178,7 @@ static void tick_nohz_handler(struct clock_event_device *dev)
 
 	dev->next_event = KTIME_MAX;
 
-	tick_sched_do_timer(now);
+	tick_sched_do_timer(ts, now);
 	tick_sched_handle(ts, regs);
 
 	/* No need to reprogram if we are running tickless  */
@@ -1169,7 +1273,7 @@ static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer)
 	struct pt_regs *regs = get_irq_regs();
 	ktime_t now = ktime_get();
 
-	tick_sched_do_timer(now);
+	tick_sched_do_timer(ts, now);
 
 	/*
 	 * Do not call, when we are not in irq context and have