hrtimer: removing all ur callback modes

Impact: cleanup, move all hrtimer processing into hardirq context

This is an attempt at removing some of the hrtimer complexity by
reducing the number of callback modes to 1.

This means that all hrtimer callback functions will be ran from HARD-irq
context.

I went through all the 30 odd hrtimer callback functions in the kernel
and saw only one that I'm not quite sure of, which is the one in
net/can/bcm.c - hence I'm CC-ing the folks responsible for that code.

Furthermore, the hrtimer core now calls callbacks directly with IRQs
disabled in case you try to enqueue an expired timer. If this timer is a
periodic timer (which should use hrtimer_forward() to advance its time)
then it might be possible to end up in an inf. recursive loop due to the
fact that hrtimer_forward() doesn't round up to the next timer
granularity, and therefore keeps on calling the callback - obviously
this needs a fix.

Aside from that, this seems to compile and actually boot on my dual core
test box - although I'm sure there are some bugs in, me not hitting any
makes me certain :-)
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

drivers/input/touchscreen/ads7846.c

@@ -697,7 +697,7 @@ static enum hrtimer_restart ads7846_timer(struct hrtimer *handle)
 	struct ads7846	*ts = container_of(handle, struct ads7846, timer);
 	int		status = 0;
 
-	spin_lock_irq(&ts->lock);
+	spin_lock(&ts->lock);
 
 	if (unlikely(!get_pendown_state(ts) ||
 		     device_suspended(&ts->spi->dev))) {
@@ -728,7 +728,7 @@ static enum hrtimer_restart ads7846_timer(struct hrtimer *handle)
 			dev_err(&ts->spi->dev, "spi_async --> %d\n", status);
 	}
 
-	spin_unlock_irq(&ts->lock);
+	spin_unlock(&ts->lock);
 	return HRTIMER_NORESTART;
 }
 

include/linux/hrtimer.h

@@ -42,26 +42,6 @@ enum hrtimer_restart {
 	HRTIMER_RESTART,	/* Timer must be restarted */
 };
 
-/*
- * hrtimer callback modes:
- *
- *	HRTIMER_CB_SOFTIRQ:		Callback must run in softirq context
- *	HRTIMER_CB_IRQSAFE_PERCPU:	Callback must run in hardirq context
- *					Special mode for tick emulation and
- *					scheduler timer. Such timers are per
- *					cpu and not allowed to be migrated on
- *					cpu unplug.
- *	HRTIMER_CB_IRQSAFE_UNLOCKED:	Callback should run in hardirq context
- *					with timer->base lock unlocked
- *					used for timers which call wakeup to
- *					avoid lock order problems with rq->lock
- */
-enum hrtimer_cb_mode {
-	HRTIMER_CB_SOFTIRQ,
-	HRTIMER_CB_IRQSAFE_PERCPU,
-	HRTIMER_CB_IRQSAFE_UNLOCKED,
-};
-
 /*
  * Values to track state of the timer
  *
@@ -70,7 +50,6 @@ enum hrtimer_cb_mode {
  * 0x00		inactive
  * 0x01		enqueued into rbtree
  * 0x02		callback function running
- * 0x04		callback pending (high resolution mode)
  *
  * Special cases:
  * 0x03		callback function running and enqueued
@@ -92,8 +71,7 @@ enum hrtimer_cb_mode {
 #define HRTIMER_STATE_INACTIVE	0x00
 #define HRTIMER_STATE_ENQUEUED	0x01
 #define HRTIMER_STATE_CALLBACK	0x02
-#define HRTIMER_STATE_PENDING	0x04
-#define HRTIMER_STATE_MIGRATE	0x08
+#define HRTIMER_STATE_MIGRATE	0x04
 
 /**
  * struct hrtimer - the basic hrtimer structure
@@ -109,8 +87,6 @@ enum hrtimer_cb_mode {
  * @function:	timer expiry callback function
  * @base:	pointer to the timer base (per cpu and per clock)
  * @state:	state information (See bit values above)
- * @cb_mode:	high resolution timer feature to select the callback execution
- *		 mode
  * @cb_entry:	list head to enqueue an expired timer into the callback list
  * @start_site:	timer statistics field to store the site where the timer
  *		was started
@@ -129,7 +105,6 @@ struct hrtimer {
 	struct hrtimer_clock_base	*base;
 	unsigned long			state;
 	struct list_head		cb_entry;
-	enum hrtimer_cb_mode		cb_mode;
 #ifdef CONFIG_TIMER_STATS
 	int				start_pid;
 	void				*start_site;
@@ -188,15 +163,11 @@ struct hrtimer_clock_base {
  * @check_clocks:	Indictator, when set evaluate time source and clock
  *			event devices whether high resolution mode can be
  *			activated.
- * @cb_pending:		Expired timers are moved from the rbtree to this
- *			list in the timer interrupt. The list is processed
- *			in the softirq.
  * @nr_events:		Total number of timer interrupt events
  */
 struct hrtimer_cpu_base {
 	spinlock_t			lock;
 	struct hrtimer_clock_base	clock_base[HRTIMER_MAX_CLOCK_BASES];
-	struct list_head		cb_pending;
 #ifdef CONFIG_HIGH_RES_TIMERS
 	ktime_t				expires_next;
 	int				hres_active;
@@ -404,8 +375,7 @@ static inline int hrtimer_active(const struct hrtimer *timer)
  */
 static inline int hrtimer_is_queued(struct hrtimer *timer)
 {
-	return timer->state &
-		(HRTIMER_STATE_ENQUEUED | HRTIMER_STATE_PENDING);
+	return timer->state & HRTIMER_STATE_ENQUEUED;
 }
 
 /*

include/linux/interrupt.h

@@ -251,9 +251,6 @@ enum
 	BLOCK_SOFTIRQ,
 	TASKLET_SOFTIRQ,
 	SCHED_SOFTIRQ,
-#ifdef CONFIG_HIGH_RES_TIMERS
-	HRTIMER_SOFTIRQ,
-#endif
 	RCU_SOFTIRQ, 	/* Preferable RCU should always be the last softirq */
 
 	NR_SOFTIRQS

kernel/sched.c

@@ -203,7 +203,6 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime)
 	hrtimer_init(&rt_b->rt_period_timer,
 			CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	rt_b->rt_period_timer.function = sched_rt_period_timer;
-	rt_b->rt_period_timer.cb_mode = HRTIMER_CB_IRQSAFE_UNLOCKED;
 }
 
 static inline int rt_bandwidth_enabled(void)
@@ -1139,7 +1138,6 @@ static void init_rq_hrtick(struct rq *rq)
 
 	hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	rq->hrtick_timer.function = hrtick;
-	rq->hrtick_timer.cb_mode = HRTIMER_CB_IRQSAFE_PERCPU;
 }
 #else	/* CONFIG_SCHED_HRTICK */
 static inline void hrtick_clear(struct rq *rq)

kernel/time/ntp.c

@@ -131,7 +131,7 @@ static enum hrtimer_restart ntp_leap_second(struct hrtimer *timer)
 {
 	enum hrtimer_restart res = HRTIMER_NORESTART;
 
-	write_seqlock_irq(&xtime_lock);
+	write_seqlock(&xtime_lock);
 
 	switch (time_state) {
 	case TIME_OK:
@@ -164,7 +164,7 @@ static enum hrtimer_restart ntp_leap_second(struct hrtimer *timer)
 	}
 	update_vsyscall(&xtime, clock);
 
-	write_sequnlock_irq(&xtime_lock);
+	write_sequnlock(&xtime_lock);
 
 	return res;
 }

kernel/time/tick-sched.c

@@ -681,7 +681,6 @@ void tick_setup_sched_timer(void)
 	 */
 	hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
 	ts->sched_timer.function = tick_sched_timer;
-	ts->sched_timer.cb_mode = HRTIMER_CB_IRQSAFE_PERCPU;
 
 	/* Get the next period (per cpu) */
 	hrtimer_set_expires(&ts->sched_timer, tick_init_jiffy_update());

kernel/trace/trace_sysprof.c

@@ -202,7 +202,6 @@ static void start_stack_timer(int cpu)
 
 	hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	hrtimer->function = stack_trace_timer_fn;
-	hrtimer->cb_mode = HRTIMER_CB_IRQSAFE_PERCPU;
 
 	hrtimer_start(hrtimer, ns_to_ktime(sample_period), HRTIMER_MODE_REL);
 }

sound/drivers/pcsp/pcsp.c

@@ -96,7 +96,6 @@ static int __devinit snd_card_pcsp_probe(int devnum, struct device *dev)
 		return -EINVAL;
 
 	hrtimer_init(&pcsp_chip.timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-	pcsp_chip.timer.cb_mode = HRTIMER_CB_SOFTIRQ;
 	pcsp_chip.timer.function = pcsp_do_timer;
 
 	card = snd_card_new(index, id, THIS_MODULE, 0);