Merge branch 'fortglx/3.12/sched-clock64-base' into fortglx/3.13/time

Merge in 64bit sched_clock support that missed 3.12.

Conflicts:
	kernel/time/sched_clock.c

Signed-off-by: John.Stultz <john.stultz@linaro.org>

2 files changed, 91 insertions, 65 deletions

diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index 50a8736..637a14a 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -537,40 +537,55 @@ static u32 clocksource_max_adjustment(struct clocksource *cs)
 }
 
 /**
- * clocksource_max_deferment - Returns max time the clocksource can be deferred
- * @cs:         Pointer to clocksource
- *
+ * clocks_calc_max_nsecs - Returns maximum nanoseconds that can be converted
+ * @mult:	cycle to nanosecond multiplier
+ * @shift:	cycle to nanosecond divisor (power of two)
+ * @maxadj:	maximum adjustment value to mult (~11%)
+ * @mask:	bitmask for two's complement subtraction of non 64 bit counters
  */
-static u64 clocksource_max_deferment(struct clocksource *cs)
+u64 clocks_calc_max_nsecs(u32 mult, u32 shift, u32 maxadj, u64 mask)
 {
 	u64 max_nsecs, max_cycles;
 
 	/*
 	 * Calculate the maximum number of cycles that we can pass to the
 	 * cyc2ns function without overflowing a 64-bit signed result. The
-	 * maximum number of cycles is equal to ULLONG_MAX/(cs->mult+cs->maxadj)
+	 * maximum number of cycles is equal to ULLONG_MAX/(mult+maxadj)
 	 * which is equivalent to the below.
-	 * max_cycles < (2^63)/(cs->mult + cs->maxadj)
-	 * max_cycles < 2^(log2((2^63)/(cs->mult + cs->maxadj)))
-	 * max_cycles < 2^(log2(2^63) - log2(cs->mult + cs->maxadj))
-	 * max_cycles < 2^(63 - log2(cs->mult + cs->maxadj))
-	 * max_cycles < 1 << (63 - log2(cs->mult + cs->maxadj))
+	 * max_cycles < (2^63)/(mult + maxadj)
+	 * max_cycles < 2^(log2((2^63)/(mult + maxadj)))
+	 * max_cycles < 2^(log2(2^63) - log2(mult + maxadj))
+	 * max_cycles < 2^(63 - log2(mult + maxadj))
+	 * max_cycles < 1 << (63 - log2(mult + maxadj))
 	 * Please note that we add 1 to the result of the log2 to account for
 	 * any rounding errors, ensure the above inequality is satisfied and
 	 * no overflow will occur.
 	 */
-	max_cycles = 1ULL << (63 - (ilog2(cs->mult + cs->maxadj) + 1));
+	max_cycles = 1ULL << (63 - (ilog2(mult + maxadj) + 1));
 
 	/*
 	 * The actual maximum number of cycles we can defer the clocksource is
-	 * determined by the minimum of max_cycles and cs->mask.
+	 * determined by the minimum of max_cycles and mask.
 	 * Note: Here we subtract the maxadj to make sure we don't sleep for
 	 * too long if there's a large negative adjustment.
 	 */
-	max_cycles = min_t(u64, max_cycles, (u64) cs->mask);
-	max_nsecs = clocksource_cyc2ns(max_cycles, cs->mult - cs->maxadj,
-					cs->shift);
+	max_cycles = min(max_cycles, mask);
+	max_nsecs = clocksource_cyc2ns(max_cycles, mult - maxadj, shift);
+
+	return max_nsecs;
+}
+
+/**
+ * clocksource_max_deferment - Returns max time the clocksource can be deferred
+ * @cs:         Pointer to clocksource
+ *
+ */
+static u64 clocksource_max_deferment(struct clocksource *cs)
+{
+	u64 max_nsecs;
 
+	max_nsecs = clocks_calc_max_nsecs(cs->mult, cs->shift, cs->maxadj,
+					  cs->mask);
 	/*
 	 * To ensure that the clocksource does not wrap whilst we are idle,
 	 * limit the time the clocksource can be deferred by 12.5%. Please
diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c
index 0b479a6..f388bae 100644
--- a/kernel/time/sched_clock.c
+++ b/kernel/time/sched_clock.c
@@ -8,25 +8,28 @@
 #include <linux/clocksource.h>
 #include <linux/init.h>
 #include <linux/jiffies.h>
+#include <linux/ktime.h>
 #include <linux/kernel.h>
 #include <linux/moduleparam.h>
 #include <linux/sched.h>
 #include <linux/syscore_ops.h>
-#include <linux/timer.h>
+#include <linux/hrtimer.h>
 #include <linux/sched_clock.h>
+#include <linux/seqlock.h>
+#include <linux/bitops.h>
 
 struct clock_data {
+	ktime_t wrap_kt;
 	u64 epoch_ns;
-	u32 epoch_cyc;
-	u32 epoch_cyc_copy;
+	u64 epoch_cyc;
+	seqcount_t seq;
 	unsigned long rate;
 	u32 mult;
 	u32 shift;
 	bool suspended;
 };
 
-static void sched_clock_poll(unsigned long wrap_ticks);
-static DEFINE_TIMER(sched_clock_timer, sched_clock_poll, 0, 0);
+static struct hrtimer sched_clock_timer;
 static int irqtime = -1;
 
 core_param(irqtime, irqtime, int, 0400);
@@ -35,14 +38,25 @@ static struct clock_data cd = {
 	.mult	= NSEC_PER_SEC / HZ,
 };
 
-static u32 __read_mostly sched_clock_mask = 0xffffffff;
+static u64 __read_mostly sched_clock_mask;
 
-static u32 notrace jiffy_sched_clock_read(void)
+static u64 notrace jiffy_sched_clock_read(void)
 {
-	return (u32)(jiffies - INITIAL_JIFFIES);
+	/*
+	 * We don't need to use get_jiffies_64 on 32-bit arches here
+	 * because we register with BITS_PER_LONG
+	 */
+	return (u64)(jiffies - INITIAL_JIFFIES);
 }
 
-static u32 __read_mostly (*read_sched_clock)(void) = jiffy_sched_clock_read;
+static u32 __read_mostly (*read_sched_clock_32)(void);
+
+static u64 notrace read_sched_clock_32_wrapper(void)
+{
+	return read_sched_clock_32();
+}
+
+static u64 __read_mostly (*read_sched_clock)(void) = jiffy_sched_clock_read;
 
 static inline u64 notrace cyc_to_ns(u64 cyc, u32 mult, u32 shift)
 {
@@ -52,25 +66,18 @@ static inline u64 notrace cyc_to_ns(u64 cyc, u32 mult, u32 shift)
 static unsigned long long notrace sched_clock_32(void)
 {
 	u64 epoch_ns;
-	u32 epoch_cyc;
-	u32 cyc;
+	u64 epoch_cyc;
+	u64 cyc;
+	unsigned long seq;
 
 	if (cd.suspended)
 		return cd.epoch_ns;
 
-	/*
-	 * 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 {
+		seq = read_seqcount_begin(&cd.seq);
 		epoch_cyc = cd.epoch_cyc;
-		smp_rmb();
 		epoch_ns = cd.epoch_ns;
-		smp_rmb();
-	} while (epoch_cyc != cd.epoch_cyc_copy);
+	} while (read_seqcount_retry(&cd.seq, seq));
 
 	cyc = read_sched_clock();
 	cyc = (cyc - epoch_cyc) & sched_clock_mask;
@@ -83,49 +90,46 @@ static unsigned long long notrace sched_clock_32(void)
 static void notrace update_sched_clock(void)
 {
 	unsigned long flags;
-	u32 cyc;
+	u64 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_copy = cyc;
-	smp_wmb();
+	write_seqcount_begin(&cd.seq);
 	cd.epoch_ns = ns;
-	smp_wmb();
 	cd.epoch_cyc = cyc;
+	write_seqcount_end(&cd.seq);
 	raw_local_irq_restore(flags);
 }
 
-static void sched_clock_poll(unsigned long wrap_ticks)
+static enum hrtimer_restart sched_clock_poll(struct hrtimer *hrt)
 {
-	mod_timer(&sched_clock_timer, round_jiffies(jiffies + wrap_ticks));
 	update_sched_clock();
+	hrtimer_forward_now(hrt, cd.wrap_kt);
+	return HRTIMER_RESTART;
 }
 
-void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
+void __init sched_clock_register(u64 (*read)(void), int bits,
+				 unsigned long rate)
 {
-	unsigned long r, w;
+	unsigned long r;
 	u64 res, wrap;
 	char r_unit;
 
 	if (cd.rate > rate)
 		return;
 
-	BUG_ON(bits > 32);
 	WARN_ON(!irqs_disabled());
 	read_sched_clock = read;
-	sched_clock_mask = (1ULL << bits) - 1;
+	sched_clock_mask = CLOCKSOURCE_MASK(bits);
 	cd.rate = rate;
 
 	/* calculate the mult/shift to convert counter ticks to ns. */
-	clocks_calc_mult_shift(&cd.mult, &cd.shift, rate, NSEC_PER_SEC, 0);
+	clocks_calc_mult_shift(&cd.mult, &cd.shift, rate, NSEC_PER_SEC, 3600);
 
 	r = rate;
 	if (r >= 4000000) {
@@ -138,20 +142,14 @@ void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
 		r_unit = ' ';
 
 	/* calculate how many ns until we wrap */
-	wrap = cyc_to_ns((1ULL << bits) - 1, cd.mult, cd.shift);
-	do_div(wrap, NSEC_PER_MSEC);
-	w = wrap;
+	wrap = clocks_calc_max_nsecs(cd.mult, cd.shift, 0, sched_clock_mask);
+	cd.wrap_kt = ns_to_ktime(wrap - (wrap >> 3));
 
 	/* calculate the ns resolution of this counter */
 	res = cyc_to_ns(1ULL, cd.mult, cd.shift);
-	pr_info("sched_clock: %u bits at %lu%cHz, resolution %lluns, wraps every %lums\n",
-		bits, r, r_unit, res, w);
+	pr_info("sched_clock: %u bits at %lu%cHz, resolution %lluns, wraps every %lluns\n",
+		bits, r, r_unit, res, wrap);
 
-	/*
-	 * Start the timer to keep sched_clock() properly updated and
-	 * sets the initial epoch.
-	 */
-	sched_clock_timer.data = msecs_to_jiffies(w - (w / 10));
 	update_sched_clock();
 
 	/*
@@ -166,6 +164,12 @@ void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
 	pr_debug("Registered %pF as sched_clock source\n", read);
 }
 
+void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
+{
+	read_sched_clock_32 = read;
+	sched_clock_register(read_sched_clock_32_wrapper, bits, rate);
+}
+
 unsigned long long __read_mostly (*sched_clock_func)(void) = sched_clock_32;
 
 unsigned long long notrace sched_clock(void)
@@ -180,14 +184,22 @@ void __init sched_clock_postinit(void)
 	 * make it the final one one.
 	 */
 	if (read_sched_clock == jiffy_sched_clock_read)
-		setup_sched_clock(jiffy_sched_clock_read, 32, HZ);
+		sched_clock_register(jiffy_sched_clock_read, BITS_PER_LONG, HZ);
 
-	sched_clock_poll(sched_clock_timer.data);
+	update_sched_clock();
+
+	/*
+	 * Start the timer to keep sched_clock() properly updated and
+	 * sets the initial epoch.
+	 */
+	hrtimer_init(&sched_clock_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	sched_clock_timer.function = sched_clock_poll;
+	hrtimer_start(&sched_clock_timer, cd.wrap_kt, HRTIMER_MODE_REL);
 }
 
 static int sched_clock_suspend(void)
 {
-	sched_clock_poll(sched_clock_timer.data);
+	sched_clock_poll(&sched_clock_timer);
 	cd.suspended = true;
 	return 0;
 }
@@ -195,7 +207,6 @@ static int sched_clock_suspend(void)
 static void sched_clock_resume(void)
 {
 	cd.epoch_cyc = read_sched_clock();
-	cd.epoch_cyc_copy = cd.epoch_cyc;
 	cd.suspended = false;
 }