Merge branch 'timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  hrtimers: Reorder clock bases
  hrtimers: Avoid touching inactive timer bases
  hrtimers: Make struct hrtimer_cpu_base layout less stupid
  timerfd: Manage cancelable timers in timerfd
  clockevents: Move C3 stop test outside lock
  alarmtimer: Drop device refcount after rtc_open()
  alarmtimer: Check return value of class_find_device()
  timerfd: Allow timers to be cancelled when clock was set
  hrtimers: Prepare for cancel on clock was set timers

6 files changed, 125 insertions, 92 deletions

diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index dbbbf7d43080..c541ee527ecb 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -64,17 +64,20 @@ DEFINE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases) =
 	.clock_base =
 	{
 		{
-			.index = CLOCK_REALTIME,
-			.get_time = &ktime_get_real,
+			.index = HRTIMER_BASE_MONOTONIC,
+			.clockid = CLOCK_MONOTONIC,
+			.get_time = &ktime_get,
 			.resolution = KTIME_LOW_RES,
 		},
 		{
-			.index = CLOCK_MONOTONIC,
-			.get_time = &ktime_get,
+			.index = HRTIMER_BASE_REALTIME,
+			.clockid = CLOCK_REALTIME,
+			.get_time = &ktime_get_real,
 			.resolution = KTIME_LOW_RES,
 		},
 		{
-			.index = CLOCK_BOOTTIME,
+			.index = HRTIMER_BASE_BOOTTIME,
+			.clockid = CLOCK_BOOTTIME,
 			.get_time = &ktime_get_boottime,
 			.resolution = KTIME_LOW_RES,
 		},
@@ -196,7 +199,7 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base,
 	struct hrtimer_cpu_base *new_cpu_base;
 	int this_cpu = smp_processor_id();
 	int cpu = hrtimer_get_target(this_cpu, pinned);
-	int basenum = hrtimer_clockid_to_base(base->index);
+	int basenum = base->index;
 
 again:
 	new_cpu_base = &per_cpu(hrtimer_bases, cpu);
@@ -621,66 +624,6 @@ static int hrtimer_reprogram(struct hrtimer *timer,
 	return res;
 }
 
-
-/*
- * Retrigger next event is called after clock was set
- *
- * Called with interrupts disabled via on_each_cpu()
- */
-static void retrigger_next_event(void *arg)
-{
-	struct hrtimer_cpu_base *base;
-	struct timespec realtime_offset, wtm, sleep;
-
-	if (!hrtimer_hres_active())
-		return;
-
-	get_xtime_and_monotonic_and_sleep_offset(&realtime_offset, &wtm,
-							&sleep);
-	set_normalized_timespec(&realtime_offset, -wtm.tv_sec, -wtm.tv_nsec);
-
-	base = &__get_cpu_var(hrtimer_bases);
-
-	/* Adjust CLOCK_REALTIME offset */
-	raw_spin_lock(&base->lock);
-	base->clock_base[HRTIMER_BASE_REALTIME].offset =
-		timespec_to_ktime(realtime_offset);
-	base->clock_base[HRTIMER_BASE_BOOTTIME].offset =
-		timespec_to_ktime(sleep);
-
-	hrtimer_force_reprogram(base, 0);
-	raw_spin_unlock(&base->lock);
-}
-
-/*
- * Clock realtime was set
- *
- * Change the offset of the realtime clock vs. the monotonic
- * clock.
- *
- * We might have to reprogram the high resolution timer interrupt. On
- * SMP we call the architecture specific code to retrigger _all_ high
- * resolution timer interrupts. On UP we just disable interrupts and
- * call the high resolution interrupt code.
- */
-void clock_was_set(void)
-{
-	/* Retrigger the CPU local events everywhere */
-	on_each_cpu(retrigger_next_event, NULL, 1);
-}
-
-/*
- * During resume we might have to reprogram the high resolution timer
- * interrupt (on the local CPU):
- */
-void hres_timers_resume(void)
-{
-	WARN_ONCE(!irqs_disabled(),
-		  KERN_INFO "hres_timers_resume() called with IRQs enabled!");
-
-	retrigger_next_event(NULL);
-}
-
 /*
  * Initialize the high resolution related parts of cpu_base
  */
@@ -715,11 +658,39 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 }
 
 /*
+ * Retrigger next event is called after clock was set
+ *
+ * Called with interrupts disabled via on_each_cpu()
+ */
+static void retrigger_next_event(void *arg)
+{
+	struct hrtimer_cpu_base *base = &__get_cpu_var(hrtimer_bases);
+	struct timespec realtime_offset, xtim, wtm, sleep;
+
+	if (!hrtimer_hres_active())
+		return;
+
+	/* Optimized out for !HIGH_RES */
+	get_xtime_and_monotonic_and_sleep_offset(&xtim, &wtm, &sleep);
+	set_normalized_timespec(&realtime_offset, -wtm.tv_sec, -wtm.tv_nsec);
+
+	/* Adjust CLOCK_REALTIME offset */
+	raw_spin_lock(&base->lock);
+	base->clock_base[HRTIMER_BASE_REALTIME].offset =
+		timespec_to_ktime(realtime_offset);
+	base->clock_base[HRTIMER_BASE_BOOTTIME].offset =
+		timespec_to_ktime(sleep);
+
+	hrtimer_force_reprogram(base, 0);
+	raw_spin_unlock(&base->lock);
+}
+
+/*
  * Switch to high resolution mode
  */
 static int hrtimer_switch_to_hres(void)
 {
-	int cpu = smp_processor_id();
+	int i, cpu = smp_processor_id();
 	struct hrtimer_cpu_base *base = &per_cpu(hrtimer_bases, cpu);
 	unsigned long flags;
 
@@ -735,9 +706,8 @@ static int hrtimer_switch_to_hres(void)
 		return 0;
 	}
 	base->hres_active = 1;
-	base->clock_base[HRTIMER_BASE_REALTIME].resolution = KTIME_HIGH_RES;
-	base->clock_base[HRTIMER_BASE_MONOTONIC].resolution = KTIME_HIGH_RES;
-	base->clock_base[HRTIMER_BASE_BOOTTIME].resolution = KTIME_HIGH_RES;
+	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++)
+		base->clock_base[i].resolution = KTIME_HIGH_RES;
 
 	tick_setup_sched_timer();
 
@@ -761,9 +731,43 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 	return 0;
 }
 static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base) { }
+static inline void retrigger_next_event(void *arg) { }
 
 #endif /* CONFIG_HIGH_RES_TIMERS */
 
+/*
+ * Clock realtime was set
+ *
+ * Change the offset of the realtime clock vs. the monotonic
+ * clock.
+ *
+ * We might have to reprogram the high resolution timer interrupt. On
+ * SMP we call the architecture specific code to retrigger _all_ high
+ * resolution timer interrupts. On UP we just disable interrupts and
+ * call the high resolution interrupt code.
+ */
+void clock_was_set(void)
+{
+#ifdef CONFIG_HIGHRES_TIMERS
+	/* Retrigger the CPU local events everywhere */
+	on_each_cpu(retrigger_next_event, NULL, 1);
+#endif
+	timerfd_clock_was_set();
+}
+
+/*
+ * During resume we might have to reprogram the high resolution timer
+ * interrupt (on the local CPU):
+ */
+void hrtimers_resume(void)
+{
+	WARN_ONCE(!irqs_disabled(),
+		  KERN_INFO "hrtimers_resume() called with IRQs enabled!");
+
+	retrigger_next_event(NULL);
+	timerfd_clock_was_set();
+}
+
 static inline void timer_stats_hrtimer_set_start_info(struct hrtimer *timer)
 {
 #ifdef CONFIG_TIMER_STATS
@@ -856,6 +860,7 @@ static int enqueue_hrtimer(struct hrtimer *timer,
 	debug_activate(timer);
 
 	timerqueue_add(&base->active, &timer->node);
+	base->cpu_base->active_bases |= 1 << base->index;
 
 	/*
 	 * HRTIMER_STATE_ENQUEUED is or'ed to the current state to preserve the
@@ -897,6 +902,8 @@ static void __remove_hrtimer(struct hrtimer *timer,
 #endif
 	}
 	timerqueue_del(&base->active, &timer->node);
+	if (!timerqueue_getnext(&base->active))
+		base->cpu_base->active_bases &= ~(1 << base->index);
 out:
 	timer->state = newstate;
 }
@@ -1234,7 +1241,6 @@ static void __run_hrtimer(struct hrtimer *timer, ktime_t *now)
 void hrtimer_interrupt(struct clock_event_device *dev)
 {
 	struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
-	struct hrtimer_clock_base *base;
 	ktime_t expires_next, now, entry_time, delta;
 	int i, retries = 0;
 
@@ -1256,12 +1262,15 @@ retry:
 	 */
 	cpu_base->expires_next.tv64 = KTIME_MAX;
 
-	base = cpu_base->clock_base;
-
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) {
-		ktime_t basenow;
+		struct hrtimer_clock_base *base;
 		struct timerqueue_node *node;
+		ktime_t basenow;
+
+		if (!(cpu_base->active_bases & (1 << i)))
+			continue;
 
+		base = cpu_base->clock_base + i;
 		basenow = ktime_add(now, base->offset);
 
 		while ((node = timerqueue_getnext(&base->active))) {
@@ -1294,7 +1303,6 @@ retry:
 
 			__run_hrtimer(timer, &basenow);
 		}
-		base++;
 	}
 
 	/*
@@ -1525,7 +1533,7 @@ long __sched hrtimer_nanosleep_restart(struct restart_block *restart)
 	struct timespec __user  *rmtp;
 	int ret = 0;
 
-	hrtimer_init_on_stack(&t.timer, restart->nanosleep.index,
+	hrtimer_init_on_stack(&t.timer, restart->nanosleep.clockid,
 				HRTIMER_MODE_ABS);
 	hrtimer_set_expires_tv64(&t.timer, restart->nanosleep.expires);
 
@@ -1577,7 +1585,7 @@ long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp,
 
 	restart = &current_thread_info()->restart_block;
 	restart->fn = hrtimer_nanosleep_restart;
-	restart->nanosleep.index = t.timer.base->index;
+	restart->nanosleep.clockid = t.timer.base->clockid;
 	restart->nanosleep.rmtp = rmtp;
 	restart->nanosleep.expires = hrtimer_get_expires_tv64(&t.timer);
 
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 0791b13df7bf..58f405b581e7 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -1514,7 +1514,7 @@ static int posix_cpu_nsleep(const clockid_t which_clock, int flags,
 			return -EFAULT;
 
 		restart_block->fn = posix_cpu_nsleep_restart;
-		restart_block->nanosleep.index = which_clock;
+		restart_block->nanosleep.clockid = which_clock;
 		restart_block->nanosleep.rmtp = rmtp;
 		restart_block->nanosleep.expires = timespec_to_ns(rqtp);
 	}
@@ -1523,7 +1523,7 @@ static int posix_cpu_nsleep(const clockid_t which_clock, int flags,
 
 static long posix_cpu_nsleep_restart(struct restart_block *restart_block)
 {
-	clockid_t which_clock = restart_block->nanosleep.index;
+	clockid_t which_clock = restart_block->nanosleep.clockid;
 	struct timespec t;
 	struct itimerspec it;
 	int error;
diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index e5498d7405c3..a1b5edf1bf92 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -1056,7 +1056,7 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags,
  */
 long clock_nanosleep_restart(struct restart_block *restart_block)
 {
-	clockid_t which_clock = restart_block->nanosleep.index;
+	clockid_t which_clock = restart_block->nanosleep.clockid;
 	struct k_clock *kc = clockid_to_kclock(which_clock);
 
 	if (WARN_ON_ONCE(!kc || !kc->nsleep_restart))
diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c
index 9265014cb4db..2d966244ea60 100644
--- a/kernel/time/alarmtimer.c
+++ b/kernel/time/alarmtimer.c
@@ -494,7 +494,7 @@ static int update_rmtp(ktime_t exp, enum  alarmtimer_type type,
  */
 static long __sched alarm_timer_nsleep_restart(struct restart_block *restart)
 {
-	enum  alarmtimer_type type = restart->nanosleep.index;
+	enum  alarmtimer_type type = restart->nanosleep.clockid;
 	ktime_t exp;
 	struct timespec __user  *rmtp;
 	struct alarm alarm;
@@ -573,7 +573,7 @@ static int alarm_timer_nsleep(const clockid_t which_clock, int flags,
 
 	restart = &current_thread_info()->restart_block;
 	restart->fn = alarm_timer_nsleep_restart;
-	restart->nanosleep.index = type;
+	restart->nanosleep.clockid = type;
 	restart->nanosleep.expires = exp.tv64;
 	restart->nanosleep.rmtp = rmtp;
 	ret = -ERESTART_RESTARTBLOCK;
@@ -669,12 +669,20 @@ static int __init has_wakealarm(struct device *dev, void *name_ptr)
  */
 static int __init alarmtimer_init_late(void)
 {
+	struct device *dev;
 	char *str;
 
 	/* Find an rtc device and init the rtc_timer */
-	class_find_device(rtc_class, NULL, &str, has_wakealarm);
-	if (str)
+	dev = class_find_device(rtc_class, NULL, &str, has_wakealarm);
+	/* If we have a device then str is valid. See has_wakealarm() */
+	if (dev) {
 		rtcdev = rtc_class_open(str);
+		/*
+		 * Drop the reference we got in class_find_device,
+		 * rtc_open takes its own.
+		 */
+		put_device(dev);
+	}
 	if (!rtcdev) {
 		printk(KERN_WARNING "No RTC device found, ALARM timers will"
 			" not wake from suspend");
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 723c7637e55a..c7218d132738 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -456,23 +456,27 @@ void tick_broadcast_oneshot_control(unsigned long reason)
 	unsigned long flags;
 	int cpu;
 
-	raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
-
 	/*
 	 * Periodic mode does not care about the enter/exit of power
 	 * states
 	 */
 	if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
-		goto out;
+		return;
 
-	bc = tick_broadcast_device.evtdev;
+	/*
+	 * We are called with preemtion disabled from the depth of the
+	 * idle code, so we can't be moved away.
+	 */
 	cpu = smp_processor_id();
 	td = &per_cpu(tick_cpu_device, cpu);
 	dev = td->evtdev;
 
 	if (!(dev->features & CLOCK_EVT_FEAT_C3STOP))
-		goto out;
+		return;
+
+	bc = tick_broadcast_device.evtdev;
 
+	raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
 	if (reason == CLOCK_EVT_NOTIFY_BROADCAST_ENTER) {
 		if (!cpumask_test_cpu(cpu, tick_get_broadcast_oneshot_mask())) {
 			cpumask_set_cpu(cpu, tick_get_broadcast_oneshot_mask());
@@ -489,8 +493,6 @@ void tick_broadcast_oneshot_control(unsigned long reason)
 				tick_program_event(dev->next_event, 1);
 		}
 	}
-
-out:
 	raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
 }
 
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 8e6a05a5915a..342408cf68dd 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -680,7 +680,7 @@ static void timekeeping_resume(void)
 	clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL);
 
 	/* Resume hrtimers */
-	hres_timers_resume();
+	hrtimers_resume();
 }
 
 static int timekeeping_suspend(void)
@@ -1099,6 +1099,21 @@ void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim,
 }
 
 /**
+ * ktime_get_monotonic_offset() - get wall_to_monotonic in ktime_t format
+ */
+ktime_t ktime_get_monotonic_offset(void)
+{
+	unsigned long seq;
+	struct timespec wtom;
+
+	do {
+		seq = read_seqbegin(&xtime_lock);
+		wtom = wall_to_monotonic;
+	} while (read_seqretry(&xtime_lock, seq));
+	return timespec_to_ktime(wtom);
+}
+
+/**
  * xtime_update() - advances the timekeeping infrastructure
  * @ticks:	number of ticks, that have elapsed since the last call.
  *