Release 4.15 kernel/time/tick-sched.c

/*
 *  linux/kernel/time/tick-sched.c
 *
 *  Copyright(C) 2005-2006, Thomas Gleixner <tglx@linutronix.de>
 *  Copyright(C) 2005-2007, Red Hat, Inc., Ingo Molnar
 *  Copyright(C) 2006-2007  Timesys Corp., Thomas Gleixner
 *
 *  No idle tick implementation for low and high resolution timers
 *
 *  Started by: Thomas Gleixner and Ingo Molnar
 *
 *  Distribute under GPLv2.
 */
#include <linux/cpu.h>
#include <linux/err.h>
#include <linux/hrtimer.h>
#include <linux/interrupt.h>
#include <linux/kernel_stat.h>
#include <linux/percpu.h>
#include <linux/nmi.h>
#include <linux/profile.h>
#include <linux/sched/signal.h>
#include <linux/sched/clock.h>
#include <linux/sched/stat.h>
#include <linux/sched/nohz.h>
#include <linux/module.h>
#include <linux/irq_work.h>
#include <linux/posix-timers.h>
#include <linux/context_tracking.h>
#include <linux/mm.h>

#include <asm/irq_regs.h>

#include "tick-internal.h"

#include <trace/events/timer.h>

/*
 * Per-CPU nohz control structure
 */
static DEFINE_PER_CPU(struct tick_sched, tick_cpu_sched);



struct tick_sched *tick_get_tick_sched(int cpu)
{
	return &per_cpu(tick_cpu_sched, cpu);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Ingo Molnar
19
100.00%
1
100.00%
Total
19
100.00%
1
100.00%

#if defined(CONFIG_NO_HZ_COMMON) || defined(CONFIG_HIGH_RES_TIMERS)
/*
 * The time, when the last jiffy update happened. Protected by jiffies_lock.
 */

static ktime_t last_jiffies_update;

/*
 * Must be called with interrupts disabled !
 */


static void tick_do_update_jiffies64(ktime_t now)
{
	unsigned long ticks = 0;
	ktime_t delta;

	/*
         * Do a quick check without holding jiffies_lock:
         */
	delta = ktime_sub(now, last_jiffies_update);
	if (delta < tick_period)
		return;

	/* Reevaluate with jiffies_lock held */
	write_seqlock(&jiffies_lock);

	delta = ktime_sub(now, last_jiffies_update);
	if (delta >= tick_period) {

		delta = ktime_sub(delta, tick_period);
		last_jiffies_update = ktime_add(last_jiffies_update,
						tick_period);

		/* Slow path for long timeouts */
		if (unlikely(delta >= tick_period)) {
			s64 incr = ktime_to_ns(tick_period);

			ticks = ktime_divns(delta, incr);

			last_jiffies_update = ktime_add_ns(last_jiffies_update,
							   incr * ticks);
		}
		do_timer(++ticks);

		/* Keep the tick_next_period variable up to date */
		tick_next_period = ktime_add(last_jiffies_update, tick_period);
	} else {
		write_sequnlock(&jiffies_lock);
		return;
	}
	write_sequnlock(&jiffies_lock);
	update_wall_time();
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Thomas Gleixner
119
78.29%
2
28.57%
Ingo Molnar
16
10.53%
1
14.29%
Viresh Kumar
10
6.58%
1
14.29%
John Stultz
6
3.95%
2
28.57%
Wei Jiangang
1
0.66%
1
14.29%
Total
152
100.00%
7
100.00%

/*
 * Initialize and return retrieve the jiffies update.
 */


static ktime_t tick_init_jiffy_update(void)
{
	ktime_t period;

	write_seqlock(&jiffies_lock);
	/* Did we start the jiffies update yet ? */
	if (last_jiffies_update == 0)
		last_jiffies_update = tick_next_period;
	period = last_jiffies_update;
	write_sequnlock(&jiffies_lock);
	return period;
}


Contributors
Person
Tokens
Prop
Commits
CommitProp
Thomas Gleixner
39
95.12%
1
50.00%
John Stultz
2
4.88%
1
50.00%
Total
41
100.00%
2
100.00%



static void tick_sched_do_timer(ktime_t now)
{
	int cpu = smp_processor_id();

#ifdef CONFIG_NO_HZ_COMMON
	/*
         * Check if the do_timer duty was dropped. We don't care about
         * concurrency: This happens only when the CPU in charge went
         * into a long sleep. If two CPUs happen to assign themselves to
         * this duty, then the jiffies update is still serialized by
         * jiffies_lock.
         */
	if (unlikely(tick_do_timer_cpu == TICK_DO_TIMER_NONE)
	    && !tick_nohz_full_cpu(cpu))
		tick_do_timer_cpu = cpu;
#endif

	/* Check, if the jiffies need an update */
	if (tick_do_timer_cpu == cpu)
		tick_do_update_jiffies64(now);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
51
98.08%
4
80.00%
Ingo Molnar
1
1.92%
1
20.00%
Total
52
100.00%
5
100.00%



static void tick_sched_handle(struct tick_sched *ts, struct pt_regs *regs)
{
#ifdef CONFIG_NO_HZ_COMMON
	/*
         * When we are idle and the tick is stopped, we have to touch
         * the watchdog as we might not schedule for a really long
         * time. This happens on complete idle SMP systems while
         * waiting on the login prompt. We also increment the "start of
         * idle" jiffy stamp so the idle accounting adjustment we do
         * when we go busy again does not account too much ticks.
         */
	if (ts->tick_stopped) {
		touch_softlockup_watchdog_sched();
		if (is_idle_task(current))
			ts->idle_jiffies++;
		/*
                 * In case the current tick fired too early past its expected
                 * expiration, make sure we don't bypass the next clock reprogramming
                 * to the same deadline.
                 */
		ts->next_tick = 0;
	}
#endif
	update_process_times(user_mode(regs));
	profile_tick(CPU_PROFILING);
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
64
98.46%
4
80.00%
Tejun Heo
1
1.54%
1
20.00%
Total
65
100.00%
5
100.00%

#endif

#ifdef CONFIG_NO_HZ_FULL

cpumask_var_t tick_nohz_full_mask;

bool tick_nohz_full_running;

static atomic_t tick_dep_mask;



static bool check_tick_dependency(atomic_t *dep)
{
	int val = atomic_read(dep);

	if (val & TICK_DEP_MASK_POSIX_TIMER) {
		trace_tick_stop(0, TICK_DEP_MASK_POSIX_TIMER);
		return true;
	}

	if (val & TICK_DEP_MASK_PERF_EVENTS) {
		trace_tick_stop(0, TICK_DEP_MASK_PERF_EVENTS);
		return true;
	}

	if (val & TICK_DEP_MASK_SCHED) {
		trace_tick_stop(0, TICK_DEP_MASK_SCHED);
		return true;
	}

	if (val & TICK_DEP_MASK_CLOCK_UNSTABLE) {
		trace_tick_stop(0, TICK_DEP_MASK_CLOCK_UNSTABLE);
		return true;
	}

	return false;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
93
100.00%
3
100.00%
Total
93
100.00%
3
100.00%



static bool can_stop_full_tick(int cpu, struct tick_sched *ts)
{
	lockdep_assert_irqs_disabled();

	if (unlikely(!cpu_online(cpu)))
		return false;

	if (check_tick_dependency(&tick_dep_mask))
		return false;

	if (check_tick_dependency(&ts->tick_dep_mask))
		return false;

	if (check_tick_dependency(&current->tick_dep_mask))
		return false;

	if (check_tick_dependency(&current->signal->tick_dep_mask))
		return false;

	return true;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
69
80.23%
4
80.00%
Wanpeng Li
17
19.77%
1
20.00%
Total
86
100.00%
5
100.00%



static void nohz_full_kick_func(struct irq_work *work)
{
	/* Empty, the tick restart happens on tick_nohz_irq_exit() */
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
12
100.00%
3
100.00%
Total
12
100.00%
3
100.00%

static DEFINE_PER_CPU(struct irq_work, nohz_full_kick_work) = {
	.func = nohz_full_kick_func,
};

/*
 * Kick this CPU if it's full dynticks in order to force it to
 * re-evaluate its dependency on the tick and restart it if necessary.
 * This kick, unlike tick_nohz_full_kick_cpu() and tick_nohz_full_kick_all(),
 * is NMI safe.
 */


static void tick_nohz_full_kick(void)
{
	if (!tick_nohz_full_cpu(smp_processor_id()))
		return;

	irq_work_queue(this_cpu_ptr(&nohz_full_kick_work));
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
25
92.59%
2
66.67%
Christoph Lameter
2
7.41%
1
33.33%
Total
27
100.00%
3
100.00%

/*
 * Kick the CPU if it's full dynticks in order to force it to
 * re-evaluate its dependency on the tick and restart it if necessary.
 */


void tick_nohz_full_kick_cpu(int cpu)
{
	if (!tick_nohz_full_cpu(cpu))
		return;

	irq_work_queue_on(&per_cpu(nohz_full_kick_work, cpu), cpu);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
30
100.00%
2
100.00%
Total
30
100.00%
2
100.00%

/*
 * Kick all full dynticks CPUs in order to force these to re-evaluate
 * their dependency on the tick and restart it if necessary.
 */


static void tick_nohz_full_kick_all(void)
{
	int cpu;

	if (!tick_nohz_full_running)
		return;

	preempt_disable();
	for_each_cpu_and(cpu, tick_nohz_full_mask, cpu_online_mask)
		tick_nohz_full_kick_cpu(cpu);
	preempt_enable();
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
36
100.00%
5
100.00%
Total
36
100.00%
5
100.00%



static void tick_nohz_dep_set_all(atomic_t *dep,
				  enum tick_dep_bits bit)
{
	int prev;

	prev = atomic_fetch_or(BIT(bit), dep);
	if (!prev)
		tick_nohz_full_kick_all();
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
35
94.59%
2
66.67%
Peter Zijlstra
2
5.41%
1
33.33%
Total
37
100.00%
3
100.00%

/*
 * Set a global tick dependency. Used by perf events that rely on freq and
 * by unstable clock.
 */


void tick_nohz_dep_set(enum tick_dep_bits bit)
{
	tick_nohz_dep_set_all(&tick_dep_mask, bit);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
17
100.00%
1
100.00%
Total
17
100.00%
1
100.00%



void tick_nohz_dep_clear(enum tick_dep_bits bit)
{
	atomic_andnot(BIT(bit), &tick_dep_mask);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
20
100.00%
2
100.00%
Total
20
100.00%
2
100.00%

/*
 * Set per-CPU tick dependency. Used by scheduler and perf events in order to
 * manage events throttling.
 */


void tick_nohz_dep_set_cpu(int cpu, enum tick_dep_bits bit)
{
	int prev;
	struct tick_sched *ts;

	ts = per_cpu_ptr(&tick_cpu_sched, cpu);

	prev = atomic_fetch_or(BIT(bit), &ts->tick_dep_mask);
	if (!prev) {
		preempt_disable();
		/* Perf needs local kick that is NMI safe */
		if (cpu == smp_processor_id()) {
			tick_nohz_full_kick();
		} else {
			/* Remote irq work not NMI-safe */
			if (!WARN_ON_ONCE(in_nmi()))
				tick_nohz_full_kick_cpu(cpu);
		}
		preempt_enable();
	}
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
84
94.38%
2
66.67%
Peter Zijlstra
5
5.62%
1
33.33%
Total
89
100.00%
3
100.00%



void tick_nohz_dep_clear_cpu(int cpu, enum tick_dep_bits bit)
{
	struct tick_sched *ts = per_cpu_ptr(&tick_cpu_sched, cpu);

	atomic_andnot(BIT(bit), &ts->tick_dep_mask);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
38
100.00%
2
100.00%
Total
38
100.00%
2
100.00%

/*
 * Set a per-task tick dependency. Posix CPU timers need this in order to elapse
 * per task timers.
 */


void tick_nohz_dep_set_task(struct task_struct *tsk, enum tick_dep_bits bit)
{
	/*
         * We could optimize this with just kicking the target running the task
         * if that noise matters for nohz full users.
         */
	tick_nohz_dep_set_all(&tsk->tick_dep_mask, bit);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
25
100.00%
1
100.00%
Total
25
100.00%
1
100.00%



void tick_nohz_dep_clear_task(struct task_struct *tsk, enum tick_dep_bits bit)
{
	atomic_andnot(BIT(bit), &tsk->tick_dep_mask);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
27
100.00%
2
100.00%
Total
27
100.00%
2
100.00%

/*
 * Set a per-taskgroup tick dependency. Posix CPU timers need this in order to elapse
 * per process timers.
 */


void tick_nohz_dep_set_signal(struct signal_struct *sig, enum tick_dep_bits bit)
{
	tick_nohz_dep_set_all(&sig->tick_dep_mask, bit);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
24
100.00%
1
100.00%
Total
24
100.00%
1
100.00%



void tick_nohz_dep_clear_signal(struct signal_struct *sig, enum tick_dep_bits bit)
{
	atomic_andnot(BIT(bit), &sig->tick_dep_mask);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
27
100.00%
2
100.00%
Total
27
100.00%
2
100.00%

/*
 * Re-evaluate the need for the tick as we switch the current task.
 * It might need the tick due to per task/process properties:
 * perf events, posix CPU timers, ...
 */


void __tick_nohz_task_switch(void)
{
	unsigned long flags;
	struct tick_sched *ts;

	local_irq_save(flags);

	if (!tick_nohz_full_cpu(smp_processor_id()))
		goto out;

	ts = this_cpu_ptr(&tick_cpu_sched);

	if (ts->tick_stopped) {
		if (atomic_read(&current->tick_dep_mask) ||
		    atomic_read(&current->signal->tick_dep_mask))
			tick_nohz_full_kick();
	}
out:
	local_irq_restore(flags);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
69
87.34%
5
83.33%
Li Zhong
10
12.66%
1
16.67%
Total
79
100.00%
6
100.00%

/* Get the boot-time nohz CPU list from the kernel parameters. */


void __init tick_nohz_full_setup(cpumask_var_t cpumask)
{
	alloc_bootmem_cpumask_var(&tick_nohz_full_mask);
	cpumask_copy(tick_nohz_full_mask, cpumask);
	tick_nohz_full_running = true;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
26
100.00%
5
100.00%
Total
26
100.00%
5
100.00%



static int tick_nohz_cpu_down(unsigned int cpu)
{
	/*
         * The boot CPU handles housekeeping duty (unbound timers,
         * workqueues, timekeeping, ...) on behalf of full dynticks
         * CPUs. It must remain online when nohz full is enabled.
         */
	if (tick_nohz_full_running && tick_do_timer_cpu == cpu)
		return -EBUSY;
	return 0;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
20
76.92%
2
66.67%
Sebastian Andrzej Siewior
6
23.08%
1
33.33%
Total
26
100.00%
3
100.00%



static int tick_nohz_init_all(void)
{
	int err = -1;

#ifdef CONFIG_NO_HZ_FULL_ALL
	if (!alloc_cpumask_var(&tick_nohz_full_mask, GFP_KERNEL)) {
		WARN(1, "NO_HZ: Can't allocate full dynticks cpumask\n");
		return err;
	}
	err = 0;
	cpumask_setall(tick_nohz_full_mask);
	tick_nohz_full_running = true;
#endif
	return err;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
50
86.21%
4
80.00%
Paul E. McKenney
8
13.79%
1
20.00%
Total
58
100.00%
5
100.00%



void __init tick_nohz_init(void)
{
	int cpu, ret;

	if (!tick_nohz_full_running) {
		if (tick_nohz_init_all() < 0)
			return;
	}

	/*
         * Full dynticks uses irq work to drive the tick rescheduling on safe
         * locking contexts. But then we need irq work to raise its own
         * interrupts to avoid circular dependency on the tick
         */
	if (!arch_irq_work_has_interrupt()) {
		pr_warn("NO_HZ: Can't run full dynticks because arch doesn't support irq work self-IPIs\n");
		cpumask_clear(tick_nohz_full_mask);
		tick_nohz_full_running = false;
		return;
	}

	cpu = smp_processor_id();

	if (cpumask_test_cpu(cpu, tick_nohz_full_mask)) {
		pr_warn("NO_HZ: Clearing %d from nohz_full range for timekeeping\n",
			cpu);
		cpumask_clear_cpu(cpu, tick_nohz_full_mask);
	}

	for_each_cpu(cpu, tick_nohz_full_mask)
		context_tracking_cpu_set(cpu);

	ret = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
					"kernel/nohz:predown", NULL,
					tick_nohz_cpu_down);
	WARN_ON(ret < 0);
	pr_info("NO_HZ: Full dynticks CPUs: %*pbl.\n",
		cpumask_pr_args(tick_nohz_full_mask));
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
100
81.30%
10
76.92%
Sebastian Andrzej Siewior
17
13.82%
1
7.69%
Joe Perches
3
2.44%
1
7.69%
Tejun Heo
3
2.44%
1
7.69%
Total
123
100.00%
13
100.00%

#endif

/*
 * NOHZ - aka dynamic tick functionality
 */
#ifdef CONFIG_NO_HZ_COMMON
/*
 * NO HZ enabled ?
 */

bool tick_nohz_enabled __read_mostly  = true;

unsigned long tick_nohz_active  __read_mostly;
/*
 * Enable / Disable tickless mode
 */


static int __init setup_tick_nohz(char *str)
{
	return (kstrtobool(str, &tick_nohz_enabled) == 0);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Thomas Gleixner
18
75.00%
1
50.00%
Kees Cook
6
25.00%
1
50.00%
Total
24
100.00%
2
100.00%

__setup("nohz=", setup_tick_nohz);



int tick_nohz_tick_stopped(void)
{
	return __this_cpu_read(tick_cpu_sched.tick_stopped);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Thomas Gleixner
15
100.00%
1
100.00%
Total
15
100.00%
1
100.00%

/**
 * tick_nohz_update_jiffies - update jiffies when idle was interrupted
 *
 * Called from interrupt entry when the CPU was idle
 *
 * In case the sched_tick was stopped on this CPU, we have to check if jiffies
 * must be updated. Otherwise an interrupt handler could use a stale jiffy
 * value. We do this unconditionally on any CPU, as we don't know whether the
 * CPU, which has the update task assigned is in a long sleep.
 */


static void tick_nohz_update_jiffies(ktime_t now)
{
	unsigned long flags;

	__this_cpu_write(tick_cpu_sched.idle_waketime, now);

	local_irq_save(flags);
	tick_do_update_jiffies64(now);
	local_irq_restore(flags);

	touch_softlockup_watchdog_sched();
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Thomas Gleixner
28
70.00%
2
28.57%
Frédéric Weisbecker
6
15.00%
1
14.29%
Ingo Molnar
2
5.00%
1
14.29%
Martin Schwidefsky
2
5.00%
1
14.29%
Tejun Heo
1
2.50%
1
14.29%
Jaswinder Singh Rajput
1
2.50%
1
14.29%
Total
40
100.00%
7
100.00%

/*
 * Updates the per-CPU time idle statistics counters
 */

static void

update_ts_time_stats(int cpu, struct tick_sched *ts, ktime_t now, u64 *last_update_time)
{
	ktime_t delta;

	if (ts->idle_active) {
		delta = ktime_sub(now, ts->idle_entrytime);
		if (nr_iowait_cpu(cpu) > 0)
			ts->iowait_sleeptime = ktime_add(ts->iowait_sleeptime, delta);
		else
			ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
		ts->idle_entrytime = now;
	}

	if (last_update_time)
		*last_update_time = ktime_to_us(now);

}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Arjan van de Ven
55
56.70%
4
44.44%
Venkatesh Pallipadi
18
18.56%
1
11.11%
Michal Hocko
14
14.43%
1
11.11%
Peter Zijlstra
6
6.19%
1
11.11%
Martin Schwidefsky
3
3.09%
1
11.11%
Thomas Gleixner
1
1.03%
1
11.11%
Total
97
100.00%
9
100.00%



static void tick_nohz_stop_idle(struct tick_sched *ts, ktime_t now)
{
	update_ts_time_stats(smp_processor_id(), ts, now, NULL);
	ts->idle_active = 0;

	sched_clock_idle_wakeup_event();
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Arjan van de Ven
18
51.43%
2
28.57%
Venkatesh Pallipadi
7
20.00%
1
14.29%
Frédéric Weisbecker
6
17.14%
1
14.29%
Peter Zijlstra
4
11.43%
3
42.86%
Total
35
100.00%
7
100.00%



static ktime_t tick_nohz_start_idle(struct tick_sched *ts)
{
	ktime_t now = ktime_get();

	ts->idle_entrytime = now;
	ts->idle_active = 1;
	sched_clock_idle_sleep_event();
	return now;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Venkatesh Pallipadi
28
80.00%
1
25.00%
Peter Zijlstra
3
8.57%
1
25.00%
Karsten Wiese
2
5.71%
1
25.00%
Michal Hocko
2
5.71%
1
25.00%
Total
35
100.00%
4
100.00%

/**
 * get_cpu_idle_time_us - get the total idle time of a CPU
 * @cpu: CPU number to query
 * @last_update_time: variable to store update time in. Do not update
 * counters if NULL.
 *
 * Return the cumulative idle time (since boot) for a given
 * CPU, in microseconds.
 *
 * This time is measured via accounting rather than sampling,
 * and is as accurate as ktime_get() is.
 *
 * This function returns -1 if NOHZ is not enabled.
 */


u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time)
{
	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
	ktime_t now, idle;

	if (!tick_nohz_active)
		return -1;

	now = ktime_get();
	if (last_update_time) {
		update_ts_time_stats(cpu, ts, now, last_update_time);
		idle = ts->idle_sleeptime;
	} else {
		if (ts->idle_active && !nr_iowait_cpu(cpu)) {
			ktime_t delta = ktime_sub(now, ts->idle_entrytime);

			idle = ktime_add(ts->idle_sleeptime, delta);
		} else {
			idle = ts->idle_sleeptime;
		}
	}

	return ktime_to_us(idle);

}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Michal Hocko
72
59.02%
1
14.29%
Venkatesh Pallipadi
42
34.43%
2
28.57%
Arjan van de Ven
5
4.10%
2
28.57%
Peter Zijlstra
2
1.64%
1
14.29%
Thomas Gleixner
1
0.82%
1
14.29%
Total
122
100.00%
7
100.00%


EXPORT_SYMBOL_GPL(get_cpu_idle_time_us);

/**
 * get_cpu_iowait_time_us - get the total iowait time of a CPU
 * @cpu: CPU number to query
 * @last_update_time: variable to store update time in. Do not update
 * counters if NULL.
 *
 * Return the cumulative iowait time (since boot) for a given
 * CPU, in microseconds.
 *
 * This time is measured via accounting rather than sampling,
 * and is as accurate as ktime_get() is.
 *
 * This function returns -1 if NOHZ is not enabled.
 */


u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time)
{
	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
	ktime_t now, iowait;

	if (!tick_nohz_active)
		return -1;

	now = ktime_get();
	if (last_update_time) {
		update_ts_time_stats(cpu, ts, now, last_update_time);
		iowait = ts->iowait_sleeptime;
	} else {
		if (ts->idle_active && nr_iowait_cpu(cpu) > 0) {
			ktime_t delta = ktime_sub(now, ts->idle_entrytime);

			iowait = ktime_add(ts->iowait_sleeptime, delta);
		} else {
			iowait = ts->iowait_sleeptime;
		}
	}

	return ktime_to_us(iowait);
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Michal Hocko
69
56.10%
1
25.00%
Arjan van de Ven
51
41.46%
1
25.00%
Peter Zijlstra
2
1.63%
1
25.00%
Thomas Gleixner
1
0.81%
1
25.00%
Total
123
100.00%
4
100.00%


EXPORT_SYMBOL_GPL(get_cpu_iowait_time_us);



static void tick_nohz_restart(struct tick_sched *ts, ktime_t now)
{
	hrtimer_cancel(&ts->sched_timer);
	hrtimer_set_expires(&ts->sched_timer, ts->last_tick);

	/* Forward the time to expire in the future */
	hrtimer_forward(&ts->sched_timer, now, tick_period);

	if (ts->nohz_mode == NOHZ_MODE_HIGHRES)
		hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED);
	else
		tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1);

	/*
         * Reset to make sure next tick stop doesn't get fooled by past
         * cached clock deadline.
         */
	ts->next_tick = 0;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Thomas Gleixner
79
91.86%
1
50.00%
Frédéric Weisbecker
7
8.14%
1
50.00%
Total
86
100.00%
2
100.00%



static inline bool local_timer_softirq_pending(void)
{
	return local_softirq_pending() & TIMER_SOFTIRQ;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Thomas Gleixner
15
100.00%
1
100.00%
Total
15
100.00%
1
100.00%



static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
					 ktime_t now, 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 {
		seq = read_seqbegin(&jiffies_lock);
		basemono = last_jiffies_update;
		basejiff = jiffies;
	} while (read_seqretry(&jiffies_lock, seq));
	ts->last_jiffies = basejiff;

	/*
         * Keep the periodic tick, when RCU, architecture or irq_work
         * requests it.
         * Aside of that check whether the local timer softirq is
         * pending. If so its a bad idea to call get_next_timer_interrupt()
         * because there is an already expired timer, so it will request
         * immeditate expiry, which rearms the hardware timer with a
         * minimal delta which brings us back to this place
         * immediately. Lather, rinse and repeat...
         */
	if (rcu_needs_cpu(basemono, &next_rcu) || arch_needs_cpu() ||
	    irq_work_needs_cpu() || local_timer_softirq_pending()) {
		next_tick = basemono + TICK_NSEC;
	} else {
		/*
                 * Get the next pending timer. If high resolution
                 * timers are enabled this only takes the timer wheel
                 * timers into account. If high resolution timers are
                 * disabled this also looks at the next expiring
                 * hrtimer.
                 */
		next_tmr = get_next_timer_interrupt(basejiff, basemono);
		ts->next_timer = next_tmr;
		/* Take the next rcu event into account */
		next_tick = next_rcu < next_tmr ? next_rcu : next_tmr;
	}

	/*
         * If the tick is due in the next period, keep it ticking or
         * force prod the timer.
         */
	delta = next_tick - basemono;
	if (delta <= (u64)TICK_NSEC) {
		/*
                 * Tell the timer code that the base is not idle, i.e. undo
                 * the effect of get_next_timer_interrupt():
                 */
		timer_clear_idle();
		/*
                 * We've not stopped the tick yet, and there's a timer in the
                 * next period, so no point in stopping it either, bail.
                 */
		if (!ts->tick_stopped) {
			tick = 0;
			goto out;
		}
	}

	/*
         * 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.
         */
	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;
	}

#ifdef CONFIG_NO_HZ_FULL
	/* Limit the tick delta to the maximum scheduler deferment */
	if (!ts->inidle)
		delta = min(delta, scheduler_tick_max_deferment());
#endif

	/* 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;

		WARN_ON_ONCE(1);
		printk_once("basemono: %llu ts->next_tick: %llu dev->next_event: %llu timer->active: %d timer->expires: %llu\n",
			    basemono, ts->next_tick, dev->next_event,
			    hrtimer_active(&ts->sched_timer), hrtimer_get_expires(&ts->sched_timer));
	}

	/*
         * nohz_stop_sched_tick can be called several times before
         * the nohz_restart_sched_tick is called. This happens when
         * interrupts arrive which do not cause a reschedule. In the
         * first call we save the current tick time, so we can restart
         * the scheduler tick in nohz_restart_sched_tick.
         */
	if (!ts->tick_stopped) {
		calc_load_nohz_start();
		cpu_load_update_nohz_start();
		quiet_vmstat();

		ts->last_tick = hrtimer_get_expires(&ts->sched_timer);
		ts->tick_stopped = 1;
		trace_tick_stop(1, TICK_DEP_MASK_NONE);
	}

	ts->next_tick = tick;

	/*
         * If the expiration time == KTIME_MAX, then we simply stop
         * the tick timer.
         */
	if (unlikely(expires == KTIME_MAX)) {
		if (ts->nohz_mode == NOHZ_MODE_HIGHRES)
			hrtimer_cancel(&ts->sched_timer);
		goto out;
	}

	hrtimer_set_expires(&ts->sched_timer, tick);

	if (ts->nohz_mode == NOHZ_MODE_HIGHRES)
		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;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Thomas Gleixner
280
54.58%
8
20.51%
Frédéric Weisbecker
130
25.34%
15
38.46%
Len Brown
23
4.48%
1
2.56%
Richard Woodruff
18
3.51%
1
2.56%
Jon Hunter
17
3.31%
1
2.56%
Martin Schwidefsky
15
2.92%
2
5.13%
Ingo Molnar
8
1.56%
2
5.13%
Peter Zijlstra
8
1.56%
4
10.26%
Paul E. McKenney
5
0.97%
1
2.56%
Arjan van de Ven
4
0.78%
1
2.56%
John Stultz
2
0.39%
1
2.56%
Christoph Lameter
2
0.39%
1
2.56%
Arun R Bharadwaj
1
0.19%
1
2.56%
Total
513
100.00%
39
100.00%



static void tick_nohz_restart_sched_tick(struct tick_sched *ts, ktime_t now)
{
	/* Update jiffies first */
	tick_do_update_jiffies64(now);
	cpu_load_update_nohz_stop();

	/*
         * Clear the timer idle flag, so we avoid IPIs on remote queueing and
         * the clock forward checks in the enqueue path:
         */
	timer_clear_idle();

	calc_load_nohz_stop();
	touch_softlockup_watchdog_sched();
	/*
         * Cancel the scheduled timer and restore the tick
         */
	ts->tick_stopped  = 0;
	ts->idle_exittime = now;

	tick_nohz_restart(ts, now);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
48
90.57%
4
66.67%
Thomas Gleixner
4
7.55%
1
16.67%
Tejun Heo
1
1.89%
1
16.67%
Total
53
100.00%
6
100.00%



static void tick_nohz_full_update_tick(struct tick_sched *ts)
{
#ifdef CONFIG_NO_HZ_FULL
	int cpu = smp_processor_id();

	if (!tick_nohz_full_cpu(cpu))
		return;

	if (!ts->tick_stopped && ts->nohz_mode == NOHZ_MODE_INACTIVE)
		return;

	if (can_stop_full_tick(cpu, ts))
		tick_nohz_stop_sched_tick(ts, ktime_get(), cpu);
	else if (ts->tick_stopped)
		tick_nohz_restart_sched_tick(ts, ktime_get());
#endif
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Frédéric Weisbecker
77
97.47%
3
75.00%
Wanpeng Li
2
2.53%
1
25.00%
Total
79
100.00%
4
100.00%



static bool can_stop_idle_tick(int cpu, struct tick_sched *ts)
{
	/*
         * If this CPU is offline and it 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. If we don't drop
         * this here the jiffies might be stale and do_timer() never
         * invoked.
         */
	if (unlikely(!cpu_online(cpu))) {
		if (cpu == tick_do_timer_cpu