Release 4.11 kernel/sched/clock.c

/*
 * sched_clock for unstable cpu clocks
 *
 *  Copyright (C) 2008 Red Hat, Inc., Peter Zijlstra
 *
 *  Updates and enhancements:
 *    Copyright (C) 2008 Red Hat, Inc. Steven Rostedt <srostedt@redhat.com>
 *
 * Based on code by:
 *   Ingo Molnar <mingo@redhat.com>
 *   Guillaume Chazarain <guichaz@gmail.com>
 *
 *
 * What:
 *
 * cpu_clock(i) provides a fast (execution time) high resolution
 * clock with bounded drift between CPUs. The value of cpu_clock(i)
 * is monotonic for constant i. The timestamp returned is in nanoseconds.
 *
 * ######################### BIG FAT WARNING ##########################
 * # when comparing cpu_clock(i) to cpu_clock(j) for i != j, time can #
 * # go backwards !!                                                  #
 * ####################################################################
 *
 * There is no strict promise about the base, although it tends to start
 * at 0 on boot (but people really shouldn't rely on that).
 *
 * cpu_clock(i)       -- can be used from any context, including NMI.
 * local_clock()      -- is cpu_clock() on the current cpu.
 *
 * sched_clock_cpu(i)
 *
 * How:
 *
 * The implementation either uses sched_clock() when
 * !CONFIG_HAVE_UNSTABLE_SCHED_CLOCK, which means in that case the
 * sched_clock() is assumed to provide these properties (mostly it means
 * the architecture provides a globally synchronized highres time source).
 *
 * Otherwise it tries to create a semi stable clock from a mixture of other
 * clocks, including:
 *
 *  - GTOD (clock monotomic)
 *  - sched_clock()
 *  - explicit idle events
 *
 * We use GTOD as base and use sched_clock() deltas to improve resolution. The
 * deltas are filtered to provide monotonicity and keeping it within an
 * expected window.
 *
 * Furthermore, explicit sleep and wakeup hooks allow us to account for time
 * that is otherwise invisible (TSC gets stopped).
 *
 */
#include <linux/spinlock.h>
#include <linux/hardirq.h>
#include <linux/export.h>
#include <linux/percpu.h>
#include <linux/ktime.h>
#include <linux/sched.h>
#include <linux/nmi.h>
#include <linux/sched/clock.h>
#include <linux/static_key.h>
#include <linux/workqueue.h>
#include <linux/compiler.h>
#include <linux/tick.h>

/*
 * Scheduler clock - returns current time in nanosec units.
 * This is default implementation.
 * Architectures and sub-architectures can override this.
 */


unsigned long long __weak sched_clock(void)
{
	return (unsigned long long)(jiffies - INITIAL_JIFFIES)
					* (NSEC_PER_SEC / HZ);
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Hugh Dickins
19
67.86%
1
33.33%
Ron
6
21.43%
1
33.33%
Gideon Israel Dsouza
3
10.71%
1
33.33%
Total
28
100.00%
3
100.00%


EXPORT_SYMBOL_GPL(sched_clock);


__read_mostly int sched_clock_running;



void sched_clock_init(void)
{
	sched_clock_running = 1;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
11
100.00%
1
100.00%
Total
11
100.00%
1
100.00%

#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
/*
 * We must start with !__sched_clock_stable because the unstable -> stable
 * transition is accurate, while the stable -> unstable transition is not.
 *
 * Similarly we start with __sched_clock_stable_early, thereby assuming we
 * will become stable, such that there's only a single 1 -> 0 transition.
 */
static DEFINE_STATIC_KEY_FALSE(__sched_clock_stable);

static int __sched_clock_stable_early = 1;

/*
 * We want: ktime_get_ns() + __gtod_offset == sched_clock() + __sched_clock_offset
 */

__read_mostly u64 __sched_clock_offset;

static __read_mostly u64 __gtod_offset;


struct sched_clock_data {
	
u64			tick_raw;
	
u64			tick_gtod;
	
u64			clock;
};

static DEFINE_PER_CPU_SHARED_ALIGNED(struct sched_clock_data, sched_clock_data);



static inline struct sched_clock_data *this_scd(void)
{
	return this_cpu_ptr(&sched_clock_data);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
18
100.00%
1
100.00%
Total
18
100.00%
1
100.00%



static inline struct sched_clock_data *cpu_sdc(int cpu)
{
	return &per_cpu(sched_clock_data, cpu);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
21
100.00%
1
100.00%
Total
21
100.00%
1
100.00%



int sched_clock_stable(void)
{
	return static_branch_likely(&__sched_clock_stable);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
14
100.00%
3
100.00%
Total
14
100.00%
3
100.00%



static void __set_sched_clock_stable(void)
{
	struct sched_clock_data *scd = this_scd();

	/*
         * Attempt to make the (initial) unstable->stable transition continuous.
         */
	__sched_clock_offset = (scd->tick_gtod + __gtod_offset) - (scd->tick_raw);

	printk(KERN_INFO "sched_clock: Marking stable (%lld, %lld)->(%lld, %lld)\n",
			scd->tick_gtod, __gtod_offset,
			scd->tick_raw,  __sched_clock_offset);

	static_branch_enable(&__sched_clock_stable);
	tick_dep_clear(TICK_DEP_BIT_CLOCK_UNSTABLE);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
57
91.94%
5
83.33%
Frédéric Weisbecker
5
8.06%
1
16.67%
Total
62
100.00%
6
100.00%



static void __sched_clock_work(struct work_struct *work)
{
	static_branch_disable(&__sched_clock_stable);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
17
100.00%
3
100.00%
Total
17
100.00%
3
100.00%

static DECLARE_WORK(sched_clock_work, __sched_clock_work);



static void __clear_sched_clock_stable(void)
{
	struct sched_clock_data *scd = this_scd();

	/*
         * Attempt to make the stable->unstable transition continuous.
         *
         * Trouble is, this is typically called from the TSC watchdog
         * timer, which is late per definition. This means the tick
         * values can already be screwy.
         *
         * Still do what we can.
         */
	__gtod_offset = (scd->tick_raw + __sched_clock_offset) - (scd->tick_gtod);

	printk(KERN_INFO "sched_clock: Marking unstable (%lld, %lld)<-(%lld, %lld)\n",
			scd->tick_gtod, __gtod_offset,
			scd->tick_raw,  __sched_clock_offset);

	tick_dep_set(TICK_DEP_BIT_CLOCK_UNSTABLE);

	if (sched_clock_stable())
		schedule_work(&sched_clock_work);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
62
92.54%
4
80.00%
Frédéric Weisbecker
5
7.46%
1
20.00%
Total
67
100.00%
5
100.00%



void clear_sched_clock_stable(void)
{
	__sched_clock_stable_early = 0;

	smp_mb(); /* matches sched_clock_init_late() */

	if (sched_clock_running == 2)
		__clear_sched_clock_stable();
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
24
100.00%
4
100.00%
Total
24
100.00%
4
100.00%



void sched_clock_init_late(void)
{
	sched_clock_running = 2;
	/*
         * Ensure that it is impossible to not do a static_key update.
         *
         * Either {set,clear}_sched_clock_stable() must see sched_clock_running
         * and do the update, or we must see their __sched_clock_stable_early
         * and do the update, or both.
         */
	smp_mb(); /* matches {set,clear}_sched_clock_stable() */

	if (__sched_clock_stable_early)
		__set_sched_clock_stable();
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
23
100.00%
4
100.00%
Total
23
100.00%
4
100.00%

/*
 * min, max except they take wrapping into account
 */



static inline u64 wrap_min(u64 x, u64 y)
{
	return (s64)(x - y) < 0 ? x : y;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
29
100.00%
1
100.00%
Total
29
100.00%
1
100.00%



static inline u64 wrap_max(u64 x, u64 y)
{
	return (s64)(x - y) > 0 ? x : y;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
29
100.00%
1
100.00%
Total
29
100.00%
1
100.00%

/*
 * update the percpu scd from the raw @now value
 *
 *  - filter out backward motion
 *  - use the GTOD tick value to create a window to filter crazy TSC values
 */


static u64 sched_clock_local(struct sched_clock_data *scd)
{
	u64 now, clock, old_clock, min_clock, max_clock, gtod;
	s64 delta;

again:
	now = sched_clock();
	delta = now - scd->tick_raw;
	if (unlikely(delta < 0))
		delta = 0;

	old_clock = scd->clock;

	/*
         * scd->clock = clamp(scd->tick_gtod + delta,
         *                    max(scd->tick_gtod, scd->clock),
         *                    scd->tick_gtod + TICK_NSEC);
         */

	gtod = scd->tick_gtod + __gtod_offset;
	clock = gtod + delta;
	min_clock = wrap_max(gtod, old_clock);
	max_clock = wrap_max(old_clock, gtod + TICK_NSEC);

	clock = wrap_max(clock, min_clock);
	clock = wrap_min(clock, max_clock);

	if (cmpxchg64(&scd->clock, old_clock, clock) != old_clock)
		goto again;

	return clock;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
113
83.09%
5
45.45%
Ingo Molnar
9
6.62%
3
27.27%
Pavel Tatashin
9
6.62%
1
9.09%
Thomas Gleixner
4
2.94%
1
9.09%
Eric Dumazet
1
0.74%
1
9.09%
Total
136
100.00%
11
100.00%



static u64 sched_clock_remote(struct sched_clock_data *scd)
{
	struct sched_clock_data *my_scd = this_scd();
	u64 this_clock, remote_clock;
	u64 *ptr, old_val, val;

#if BITS_PER_LONG != 64
again:
	/*
         * Careful here: The local and the remote clock values need to
         * be read out atomic as we need to compare the values and
         * then update either the local or the remote side. So the
         * cmpxchg64 below only protects one readout.
         *
         * We must reread via sched_clock_local() in the retry case on
         * 32bit as an NMI could use sched_clock_local() via the
         * tracer and hit between the readout of
         * the low32bit and the high 32bit portion.
         */
	this_clock = sched_clock_local(my_scd);
	/*
         * We must enforce atomic readout on 32bit, otherwise the
         * update on the remote cpu can hit inbetween the readout of
         * the low32bit and the high 32bit portion.
         */
	remote_clock = cmpxchg64(&scd->clock, 0, 0);
#else
	/*
         * On 64bit the read of [my]scd->clock is atomic versus the
         * update, so we can avoid the above 32bit dance.
         */
	sched_clock_local(my_scd);
again:
	this_clock = my_scd->clock;
	remote_clock = scd->clock;
#endif

	/*
         * Use the opportunity that we have both locks
         * taken to couple the two clocks: we take the
         * larger time as the latest time for both
         * runqueues. (this creates monotonic movement)
         */
	if (likely((s64)(remote_clock - this_clock) < 0)) {
		ptr = &scd->clock;
		old_val = remote_clock;
		val = this_clock;
	} else {
		/*
                 * Should be rare, but possible:
                 */
		ptr = &my_scd->clock;
		old_val = this_clock;
		val = remote_clock;
	}

	if (cmpxchg64(ptr, old_val, val) != old_val)
		goto again;

	return val;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
87
55.06%
3
37.50%
Thomas Gleixner
35
22.15%
1
12.50%
Ingo Molnar
35
22.15%
3
37.50%
Eric Dumazet
1
0.63%
1
12.50%
Total
158
100.00%
8
100.00%

/*
 * Similar to cpu_clock(), but requires local IRQs to be disabled.
 *
 * See cpu_clock().
 */


u64 sched_clock_cpu(int cpu)
{
	struct sched_clock_data *scd;
	u64 clock;

	if (sched_clock_stable())
		return sched_clock() + __sched_clock_offset;

	if (unlikely(!sched_clock_running))
		return 0ull;

	preempt_disable_notrace();
	scd = cpu_sdc(cpu);

	if (cpu != smp_processor_id())
		clock = sched_clock_remote(scd);
	else
		clock = sched_clock_local(scd);
	preempt_enable_notrace();

	return clock;
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
72
94.74%
6
75.00%
Ingo Molnar
2
2.63%
1
12.50%
Fernando Luis Vázquez Cao
2
2.63%
1
12.50%
Total
76
100.00%
8
100.00%


EXPORT_SYMBOL_GPL(sched_clock_cpu);



void sched_clock_tick(void)
{
	struct sched_clock_data *scd;

	WARN_ON_ONCE(!irqs_disabled());

	/*
         * Update these values even if sched_clock_stable(), because it can
         * become unstable at any point in time at which point we need some
         * values to fall back on.
         *
         * XXX arguably we can skip this if we expose tsc_clocksource_reliable
         */
	scd = this_scd();
	scd->tick_raw  = sched_clock();
	scd->tick_gtod = ktime_get_ns();

	if (!sched_clock_stable() && likely(sched_clock_running))
		sched_clock_local(scd);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
55
100.00%
5
100.00%
Total
55
100.00%
5
100.00%

/*
 * We are going deep-idle (irqs are disabled):
 */


void sched_clock_idle_sleep_event(void)
{
	sched_clock_cpu(smp_processor_id());
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
13
100.00%
1
100.00%
Total
13
100.00%
1
100.00%


EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event);

/*
 * We just idled delta nanoseconds (called with irqs disabled):
 */


void sched_clock_idle_wakeup_event(u64 delta_ns)
{
	if (timekeeping_suspended)
		return;

	sched_clock_tick();
	touch_softlockup_watchdog_sched();
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
13
68.42%
2
50.00%
Thomas Gleixner
5
26.32%
1
25.00%
Tejun Heo
1
5.26%
1
25.00%
Total
19
100.00%
4
100.00%


EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);

#else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */



u64 sched_clock_cpu(int cpu)
{
	if (unlikely(!sched_clock_running))
		return 0;

	return sched_clock();
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
23
100.00%
1
100.00%
Total
23
100.00%
1
100.00%

#endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */

/*
 * Running clock - returns the time that has elapsed while a guest has been
 * running.
 * On a guest this value should be local_clock minus the time the guest was
 * suspended by the hypervisor (for any reason).
 * On bare metal this function should return the same as local_clock.
 * Architectures and sub-architectures can override this.
 */


u64 __weak running_clock(void)
{
	return local_clock();
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Cyril Bur
12
100.00%
1
100.00%
Total
12
100.00%
1
100.00%

Overall Contributors
Person
Tokens
Prop
Commits
CommitProp
Peter Zijlstra
779
80.64%
18
43.90%
Ingo Molnar
57
5.90%
8
19.51%
Thomas Gleixner
44
4.55%
2
4.88%
Hugh Dickins
20
2.07%
1
2.44%
Frédéric Weisbecker
13
1.35%
1
2.44%
Cyril Bur
13
1.35%
1
2.44%
Pavel Tatashin
9
0.93%
1
2.44%
Ron
6
0.62%
1
2.44%
Gideon Israel Dsouza
6
0.62%
1
2.44%
Daniel Lezcano
5
0.52%
1
2.44%
Divyesh Shah
5
0.52%
1
2.44%
David S. Miller
3
0.31%
1
2.44%
Eric Dumazet
2
0.21%
1
2.44%
Fernando Luis Vázquez Cao
2
0.21%
1
2.44%
Paul Gortmaker
1
0.10%
1
2.44%
Tejun Heo
1
0.10%
1
2.44%
Total
966
100.00%
41
100.00%
Directory: kernel/sched

Information contained on this website is for historical information purposes only and does not indicate or represent copyright ownership.