Merge branch 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull timer changes from Ingo Molnar:
 "Main changes in this cycle were:

   - Updated full dynticks support.

   - Event stream support for architected (ARM) timers.

   - ARM clocksource driver updates.

   - Move arm64 to using the generic sched_clock framework & resulting
     cleanup in the generic sched_clock code.

   - Misc fixes and cleanups"

* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (50 commits)
  x86/time: Honor ACPI FADT flag indicating absence of a CMOS RTC
  clocksource: sun4i: remove IRQF_DISABLED
  clocksource: sun4i: Report the minimum tick that we can program
  clocksource: sun4i: Select CLKSRC_MMIO
  clocksource: Provide timekeeping for efm32 SoCs
  clocksource: em_sti: convert to clk_prepare/unprepare
  time: Fix signedness bug in sysfs_get_uname() and its callers
  timekeeping: Fix some trivial typos in comments
  alarmtimer: return EINVAL instead of ENOTSUPP if rtcdev doesn't exist
  clocksource: arch_timer: Do not register arch_sys_counter twice
  timer stats: Add a 'Collection: active/inactive' line to timer usage statistics
  sched_clock: Remove sched_clock_func() hook
  arch_timer: Move to generic sched_clock framework
  clocksource: tcb_clksrc: Remove IRQF_DISABLED
  clocksource: tcb_clksrc: Improve driver robustness
  clocksource: tcb_clksrc: Replace clk_enable/disable with clk_prepare_enable/disable_unprepare
  clocksource: arm_arch_timer: Use clocksource for suspend timekeeping
  clocksource: dw_apb_timer_of: Mark a few more functions as __init
  clocksource: Put nodes passed to CLOCKSOURCE_OF_DECLARE callbacks centrally
  arm: zynq: Enable arm_global_timer
  ...

10 files changed, 107 insertions, 84 deletions

diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig
index 2b62fe8..3ce6e8c 100644
--- a/kernel/time/Kconfig
+++ b/kernel/time/Kconfig
@@ -100,7 +100,7 @@ config NO_HZ_FULL
 	# RCU_USER_QS dependency
 	depends on HAVE_CONTEXT_TRACKING
 	# VIRT_CPU_ACCOUNTING_GEN dependency
-	depends on 64BIT
+	depends on HAVE_VIRT_CPU_ACCOUNTING_GEN
 	select NO_HZ_COMMON
 	select RCU_USER_QS
 	select RCU_NOCB_CPU
diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c
index eec50fc..88c9c65 100644
--- a/kernel/time/alarmtimer.c
+++ b/kernel/time/alarmtimer.c
@@ -490,7 +490,7 @@ static int alarm_clock_getres(const clockid_t which_clock, struct timespec *tp)
 	clockid_t baseid = alarm_bases[clock2alarm(which_clock)].base_clockid;
 
 	if (!alarmtimer_get_rtcdev())
-		return -ENOTSUPP;
+		return -EINVAL;
 
 	return hrtimer_get_res(baseid, tp);
 }
@@ -507,7 +507,7 @@ static int alarm_clock_get(clockid_t which_clock, struct timespec *tp)
 	struct alarm_base *base = &alarm_bases[clock2alarm(which_clock)];
 
 	if (!alarmtimer_get_rtcdev())
-		return -ENOTSUPP;
+		return -EINVAL;
 
 	*tp = ktime_to_timespec(base->gettime());
 	return 0;
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 662c579..086ad60 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -619,7 +619,7 @@ static ssize_t sysfs_unbind_tick_dev(struct device *dev,
 				     const char *buf, size_t count)
 {
 	char name[CS_NAME_LEN];
-	size_t ret = sysfs_get_uname(buf, name, count);
+	ssize_t ret = sysfs_get_uname(buf, name, count);
 	struct clock_event_device *ce;
 
 	if (ret < 0)
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index 50a8736..ba3e502 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -479,6 +479,7 @@ static inline void clocksource_dequeue_watchdog(struct clocksource *cs) { }
 static inline void clocksource_resume_watchdog(void) { }
 static inline int __clocksource_watchdog_kthread(void) { return 0; }
 static bool clocksource_is_watchdog(struct clocksource *cs) { return false; }
+void clocksource_mark_unstable(struct clocksource *cs) { }
 
 #endif /* CONFIG_CLOCKSOURCE_WATCHDOG */
 
@@ -537,40 +538,55 @@ 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
@@ -893,7 +909,7 @@ sysfs_show_current_clocksources(struct device *dev,
 	return count;
 }
 
-size_t sysfs_get_uname(const char *buf, char *dst, size_t cnt)
+ssize_t sysfs_get_uname(const char *buf, char *dst, size_t cnt)
 {
 	size_t ret = cnt;
 
@@ -924,7 +940,7 @@ static ssize_t sysfs_override_clocksource(struct device *dev,
 					  struct device_attribute *attr,
 					  const char *buf, size_t count)
 {
-	size_t ret;
+	ssize_t ret;
 
 	mutex_lock(&clocksource_mutex);
 
@@ -952,7 +968,7 @@ static ssize_t sysfs_unbind_clocksource(struct device *dev,
 {
 	struct clocksource *cs;
 	char name[CS_NAME_LEN];
-	size_t ret;
+	ssize_t ret;
 
 	ret = sysfs_get_uname(buf, name, count);
 	if (ret < 0)
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index bb22151..af8d1d4 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -475,6 +475,7 @@ static void sync_cmos_clock(struct work_struct *work)
 	 * called as close as possible to 500 ms before the new second starts.
 	 * This code is run on a timer.  If the clock is set, that timer
 	 * may not expire at the correct time.  Thus, we adjust...
+	 * We want the clock to be within a couple of ticks from the target.
 	 */
 	if (!ntp_synced()) {
 		/*
@@ -485,7 +486,7 @@ static void sync_cmos_clock(struct work_struct *work)
 	}
 
 	getnstimeofday(&now);
-	if (abs(now.tv_nsec - (NSEC_PER_SEC / 2)) <= tick_nsec / 2) {
+	if (abs(now.tv_nsec - (NSEC_PER_SEC / 2)) <= tick_nsec * 5) {
 		struct timespec adjust = now;
 
 		fail = -ENODEV;
diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c
index 0b479a6..68b7993 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);
@@ -35,42 +38,46 @@ 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)
 {
 	return (cyc * mult) >> shift;
 }
 
-static unsigned long long notrace sched_clock_32(void)
+unsigned long long notrace sched_clock(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();
 
 	/*
@@ -166,11 +164,10 @@ void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
 	pr_debug("Registered %pF as sched_clock source\n", read);
 }
 
-unsigned long long __read_mostly (*sched_clock_func)(void) = sched_clock_32;
-
-unsigned long long notrace sched_clock(void)
+void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
 {
-	return sched_clock_func();
+	read_sched_clock_32 = read;
+	sched_clock_register(read_sched_clock_32_wrapper, bits, rate);
 }
 
 void __init sched_clock_postinit(void)
@@ -180,14 +177,22 @@ 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 +200,6 @@ 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;
 }
 
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 218bcb5..9532690 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -70,6 +70,7 @@ static bool tick_check_broadcast_device(struct clock_event_device *curdev,
 					struct clock_event_device *newdev)
 {
 	if ((newdev->features & CLOCK_EVT_FEAT_DUMMY) ||
+	    (newdev->features & CLOCK_EVT_FEAT_PERCPU) ||
 	    (newdev->features & CLOCK_EVT_FEAT_C3STOP))
 		return false;
 
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index bc906ca..18e71f7 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -31,7 +31,7 @@ extern void tick_install_replacement(struct clock_event_device *dev);
 
 extern void clockevents_shutdown(struct clock_event_device *dev);
 
-extern size_t sysfs_get_uname(const char *buf, char *dst, size_t cnt);
+extern ssize_t sysfs_get_uname(const char *buf, char *dst, size_t cnt);
 
 /*
  * NO_HZ / high resolution timer shared code
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 947ba25..3abf534 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -1613,9 +1613,10 @@ void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim,
  * ktime_get_update_offsets - hrtimer helper
  * @offs_real:	pointer to storage for monotonic -> realtime offset
  * @offs_boot:	pointer to storage for monotonic -> boottime offset
+ * @offs_tai:	pointer to storage for monotonic -> clock tai offset
  *
  * Returns current monotonic time and updates the offsets
- * Called from hrtimer_interupt() or retrigger_next_event()
+ * Called from hrtimer_interrupt() or retrigger_next_event()
  */
 ktime_t ktime_get_update_offsets(ktime_t *offs_real, ktime_t *offs_boot,
 							ktime_t *offs_tai)
diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c
index 0b537f2..1fb08f2 100644
--- a/kernel/time/timer_stats.c
+++ b/kernel/time/timer_stats.c
@@ -298,15 +298,15 @@ static int tstats_show(struct seq_file *m, void *v)
 	period = ktime_to_timespec(time);
 	ms = period.tv_nsec / 1000000;
 
-	seq_puts(m, "Timer Stats Version: v0.2\n");
+	seq_puts(m, "Timer Stats Version: v0.3\n");
 	seq_printf(m, "Sample period: %ld.%03ld s\n", period.tv_sec, ms);
 	if (atomic_read(&overflow_count))
-		seq_printf(m, "Overflow: %d entries\n",
-			atomic_read(&overflow_count));
+		seq_printf(m, "Overflow: %d entries\n", atomic_read(&overflow_count));
+	seq_printf(m, "Collection: %s\n", timer_stats_active ? "active" : "inactive");
 
 	for (i = 0; i < nr_entries; i++) {
 		entry = entries + i;
- 		if (entry->timer_flag & TIMER_STATS_FLAG_DEFERRABLE) {
+		if (entry->timer_flag & TIMER_STATS_FLAG_DEFERRABLE) {
 			seq_printf(m, "%4luD, %5d %-16s ",
 				entry->count, entry->pid, entry->comm);
 		} else {