sched_clock.c 4.3 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12
/*
 * sched_clock.c: support for extending counters to full 64-bit ns counter
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */
#include <linux/clocksource.h>
#include <linux/init.h>
#include <linux/jiffies.h>
#include <linux/kernel.h>
#include <linux/sched.h>
13
#include <linux/syscore_ops.h>
14 15 16 17
#include <linux/timer.h>

#include <asm/sched_clock.h>

18 19 20 21 22 23 24 25
struct clock_data {
	u64 epoch_ns;
	u32 epoch_cyc;
	u32 epoch_cyc_copy;
	u32 mult;
	u32 shift;
};

26 27
static void sched_clock_poll(unsigned long wrap_ticks);
static DEFINE_TIMER(sched_clock_timer, sched_clock_poll, 0, 0);
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93

static struct clock_data cd = {
	.mult	= NSEC_PER_SEC / HZ,
};

static u32 __read_mostly sched_clock_mask = 0xffffffff;

static u32 notrace jiffy_sched_clock_read(void)
{
	return (u32)(jiffies - INITIAL_JIFFIES);
}

static u32 __read_mostly (*read_sched_clock)(void) = jiffy_sched_clock_read;

static inline u64 cyc_to_ns(u64 cyc, u32 mult, u32 shift)
{
	return (cyc * mult) >> shift;
}

static unsigned long long cyc_to_sched_clock(u32 cyc, u32 mask)
{
	u64 epoch_ns;
	u32 epoch_cyc;

	/*
	 * Load the epoch_cyc and epoch_ns atomically.  We do this by
	 * ensuring that we always write epoch_cyc, epoch_ns and
	 * epoch_cyc_copy in strict order, and read them in strict order.
	 * If epoch_cyc and epoch_cyc_copy are not equal, then we're in
	 * the middle of an update, and we should repeat the load.
	 */
	do {
		epoch_cyc = cd.epoch_cyc;
		smp_rmb();
		epoch_ns = cd.epoch_ns;
		smp_rmb();
	} while (epoch_cyc != cd.epoch_cyc_copy);

	return epoch_ns + cyc_to_ns((cyc - epoch_cyc) & mask, cd.mult, cd.shift);
}

/*
 * Atomically update the sched_clock epoch.
 */
static void notrace update_sched_clock(void)
{
	unsigned long flags;
	u32 cyc;
	u64 ns;

	cyc = read_sched_clock();
	ns = cd.epoch_ns +
		cyc_to_ns((cyc - cd.epoch_cyc) & sched_clock_mask,
			  cd.mult, cd.shift);
	/*
	 * Write epoch_cyc and epoch_ns in a way that the update is
	 * detectable in cyc_to_fixed_sched_clock().
	 */
	raw_local_irq_save(flags);
	cd.epoch_cyc = cyc;
	smp_wmb();
	cd.epoch_ns = ns;
	smp_wmb();
	cd.epoch_cyc_copy = cyc;
	raw_local_irq_restore(flags);
}
94 95 96 97

static void sched_clock_poll(unsigned long wrap_ticks)
{
	mod_timer(&sched_clock_timer, round_jiffies(jiffies + wrap_ticks));
98
	update_sched_clock();
99 100
}

101
void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
102 103 104 105 106
{
	unsigned long r, w;
	u64 res, wrap;
	char r_unit;

107 108 109 110 111
	BUG_ON(bits > 32);
	WARN_ON(!irqs_disabled());
	WARN_ON(read_sched_clock != jiffy_sched_clock_read);
	read_sched_clock = read;
	sched_clock_mask = (1 << bits) - 1;
112 113

	/* calculate the mult/shift to convert counter ticks to ns. */
114
	clocks_calc_mult_shift(&cd.mult, &cd.shift, rate, NSEC_PER_SEC, 0);
115 116 117 118 119

	r = rate;
	if (r >= 4000000) {
		r /= 1000000;
		r_unit = 'M';
120
	} else if (r >= 1000) {
121 122
		r /= 1000;
		r_unit = 'k';
123 124
	} else
		r_unit = ' ';
125 126

	/* calculate how many ns until we wrap */
127
	wrap = cyc_to_ns((1ULL << bits) - 1, cd.mult, cd.shift);
128 129 130 131
	do_div(wrap, NSEC_PER_MSEC);
	w = wrap;

	/* calculate the ns resolution of this counter */
132
	res = cyc_to_ns(1ULL, cd.mult, cd.shift);
133
	pr_info("sched_clock: %u bits at %lu%cHz, resolution %lluns, wraps every %lums\n",
134
		bits, r, r_unit, res, w);
135 136 137 138 139 140

	/*
	 * Start the timer to keep sched_clock() properly updated and
	 * sets the initial epoch.
	 */
	sched_clock_timer.data = msecs_to_jiffies(w - (w / 10));
141
	update_sched_clock();
142 143 144 145

	/*
	 * Ensure that sched_clock() starts off at 0ns
	 */
146 147 148 149 150 151 152 153 154
	cd.epoch_ns = 0;

	pr_debug("Registered %pF as sched_clock source\n", read);
}

unsigned long long notrace sched_clock(void)
{
	u32 cyc = read_sched_clock();
	return cyc_to_sched_clock(cyc, sched_clock_mask);
155
}
156 157 158

void __init sched_clock_postinit(void)
{
159 160 161 162 163 164 165
	/*
	 * If no sched_clock function has been provided at that point,
	 * make it the final one one.
	 */
	if (read_sched_clock == jiffy_sched_clock_read)
		setup_sched_clock(jiffy_sched_clock_read, 32, HZ);

166 167
	sched_clock_poll(sched_clock_timer.data);
}
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184

static int sched_clock_suspend(void)
{
	sched_clock_poll(sched_clock_timer.data);
	return 0;
}

static struct syscore_ops sched_clock_ops = {
	.suspend = sched_clock_suspend,
};

static int __init sched_clock_syscore_init(void)
{
	register_syscore_ops(&sched_clock_ops);
	return 0;
}
device_initcall(sched_clock_syscore_init);