hrtimers: prepare full preemption

Make cancellation of a running callback in softirq context safe
against preemption.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

include/linux/hrtimer.h

@@ -177,6 +177,9 @@ struct hrtimer_cpu_base {
 	int				hres_active;
 	unsigned long			nr_events;
 #endif
+#ifdef CONFIG_PREEMPT_SOFTIRQS
+	wait_queue_head_t		wait;
+#endif
 };
 
 static inline void hrtimer_set_expires(struct hrtimer *timer, ktime_t time)
@@ -364,6 +367,13 @@ static inline int hrtimer_restart(struct hrtimer *timer)
 	return hrtimer_start_expires(timer, HRTIMER_MODE_ABS);
 }
 
+/* Softirq preemption could deadlock timer removal */
+#ifdef CONFIG_PREEMPT_SOFTIRQS
+  extern void hrtimer_wait_for_timer(const struct hrtimer *timer);
+#else
+# define hrtimer_wait_for_timer(timer)	do { cpu_relax(); } while (0)
+#endif
+
 /* Query timers: */
 extern ktime_t hrtimer_get_remaining(const struct hrtimer *timer);
 extern int hrtimer_get_res(const clockid_t which_clock, struct timespec *tp);

kernel/hrtimer.c

@@ -837,6 +837,32 @@ static int enqueue_hrtimer(struct hrtimer *timer,
 	return leftmost;
 }
 
+#ifdef CONFIG_PREEMPT_SOFTIRQS
+# define wake_up_timer_waiters(b)	wake_up(&(b)->wait)
+
+/**
+ * hrtimer_wait_for_timer - Wait for a running timer
+ *
+ * @timer:	timer to wait for
+ *
+ * The function waits in case the timers callback function is
+ * currently executed on the waitqueue of the timer base. The
+ * waitqueue is woken up after the timer callback function has
+ * finished execution.
+ */
+void hrtimer_wait_for_timer(const struct hrtimer *timer)
+{
+	struct hrtimer_clock_base *base = timer->base;
+
+	if (base && base->cpu_base)
+		wait_event(base->cpu_base->wait,
+				!(timer->state & HRTIMER_STATE_CALLBACK));
+}
+
+#else
+# define wake_up_timer_waiters(b)	do { } while (0)
+#endif
+
 /*
  * __remove_hrtimer - internal function to remove a timer
  *
@@ -865,6 +891,8 @@ static void __remove_hrtimer(struct hrtimer *timer,
 		rb_erase(&timer->node, &base->active);
 	}
 	timer->state = newstate;
+
+	wake_up_timer_waiters(base->cpu_base);
 }
 
 /*
@@ -1023,7 +1051,7 @@ int hrtimer_cancel(struct hrtimer *timer)
 
 		if (ret >= 0)
 			return ret;
-		cpu_relax();
+		hrtimer_wait_for_timer(timer);
 	}
 }
 EXPORT_SYMBOL_GPL(hrtimer_cancel);
@@ -1558,6 +1586,9 @@ static void __cpuinit init_hrtimers_cpu(int cpu)
 		cpu_base->clock_base[i].cpu_base = cpu_base;
 
 	hrtimer_init_hres(cpu_base);
+#ifdef CONFIG_PREEMPT_SOFTIRQS
+	init_waitqueue_head(&cpu_base->wait);
+#endif
 }
 
 #ifdef CONFIG_HOTPLUG_CPU

kernel/itimer.c

@@ -161,6 +161,7 @@ again:
 		/* We are sharing ->siglock with it_real_fn() */
 		if (hrtimer_try_to_cancel(timer) < 0) {
 			spin_unlock_irq(&tsk->sighand->siglock);
+			hrtimer_wait_for_timer(&tsk->signal->real_timer);
 			goto again;
 		}
 		expires = timeval_to_ktime(value->it_value);

kernel/posix-timers.c

@@ -789,6 +789,7 @@ retry:
 
 	unlock_timer(timr, flag);
 	if (error == TIMER_RETRY) {
+		hrtimer_wait_for_timer(&timr->it.real.timer);
 		rtn = NULL;	// We already got the old time...
 		goto retry;
 	}
@@ -827,6 +828,7 @@ retry_delete:
 
 	if (timer_delete_hook(timer) == TIMER_RETRY) {
 		unlock_timer(timer, flags);
+		hrtimer_wait_for_timer(&timer->it.real.timer);
 		goto retry_delete;
 	}
 
@@ -856,6 +858,7 @@ retry_delete:
 
 	if (timer_delete_hook(timer) == TIMER_RETRY) {
 		unlock_timer(timer, flags);
+		hrtimer_wait_for_timer(&timer->it.real.timer);
 		goto retry_delete;
 	}
 	list_del(&timer->list);