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

Pull timer fixes from Thomas Gleixner:
 "This update from the timer departement contains:

   - A series of patches which address a shortcoming in the tick
     broadcast code.

     If the broadcast device is not available or an hrtimer emulated
     broadcast device, some of the original assumptions lead to boot
     failures.  I rather plugged all of the corner cases instead of only
     addressing the issue reported, so the change got a little larger.

     Has been extensivly tested on x86 and arm.

   - Get rid of the last holdouts using do_posix_clock_monotonic_gettime()

   - A regression fix for the imx clocksource driver

   - An update to the new state callbacks mechanism for clockevents.
     This is required to simplify the conversion, which will take place
     in 4.3"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  tick/broadcast: Prevent NULL pointer dereference
  time: Get rid of do_posix_clock_monotonic_gettime
  cris: Replace do_posix_clock_monotonic_gettime()
  tick/broadcast: Unbreak CONFIG_GENERIC_CLOCKEVENTS=n build
  tick/broadcast: Handle spurious interrupts gracefully
  tick/broadcast: Check for hrtimer broadcast active early
  tick/broadcast: Return busy when IPI is pending
  tick/broadcast: Return busy if periodic mode and hrtimer broadcast
  tick/broadcast: Move the check for periodic mode inside state handling
  tick/broadcast: Prevent deep idle if no broadcast device available
  tick/broadcast: Make idle check independent from mode and config
  tick/broadcast: Sanity check the shutdown of the local clock_event
  tick/broadcast: Prevent hrtimer recursion
  clockevents: Allow set-state callbacks to be optional
  clocksource/imx: Define clocksource for mx27

arch/cris/arch-v32/drivers/sync_serial.c

@@ -1464,7 +1464,7 @@ static inline void handle_rx_packet(struct sync_port *port)
 		if (port->write_ts_idx == NBR_IN_DESCR)
 			port->write_ts_idx = 0;
 		idx = port->write_ts_idx++;
-		do_posix_clock_monotonic_gettime(&port->timestamp[idx]);
+		ktime_get_ts(&port->timestamp[idx]);
 		port->in_buffer_len += port->inbufchunk;
 	}
 	spin_unlock_irqrestore(&port->lock, flags);

drivers/clocksource/timer-imx-gpt.c

@@ -529,6 +529,7 @@ static void __init imx6dl_timer_init_dt(struct device_node *np)
 
 CLOCKSOURCE_OF_DECLARE(imx1_timer, "fsl,imx1-gpt", imx1_timer_init_dt);
 CLOCKSOURCE_OF_DECLARE(imx21_timer, "fsl,imx21-gpt", imx21_timer_init_dt);
+CLOCKSOURCE_OF_DECLARE(imx27_timer, "fsl,imx27-gpt", imx21_timer_init_dt);
 CLOCKSOURCE_OF_DECLARE(imx31_timer, "fsl,imx31-gpt", imx31_timer_init_dt);
 CLOCKSOURCE_OF_DECLARE(imx25_timer, "fsl,imx25-gpt", imx31_timer_init_dt);
 CLOCKSOURCE_OF_DECLARE(imx50_timer, "fsl,imx50-gpt", imx31_timer_init_dt);

include/linux/tick.h

@@ -67,10 +67,13 @@ extern void tick_broadcast_control(enum tick_broadcast_mode mode);
 static inline void tick_broadcast_control(enum tick_broadcast_mode mode) { }
 #endif /* BROADCAST */
 
-#if defined(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST) && defined(CONFIG_TICK_ONESHOT)
+#ifdef CONFIG_GENERIC_CLOCKEVENTS
 extern int tick_broadcast_oneshot_control(enum tick_broadcast_state state);
 #else
-static inline int tick_broadcast_oneshot_control(enum tick_broadcast_state state) { return 0; }
+static inline int tick_broadcast_oneshot_control(enum tick_broadcast_state state)
+{
+	return 0;
+}
 #endif
 
 static inline void tick_broadcast_enable(void)

include/linux/timekeeping.h

@@ -145,7 +145,6 @@ static inline void getboottime(struct timespec *ts)
 }
 #endif
 
-#define do_posix_clock_monotonic_gettime(ts) ktime_get_ts(ts)
 #define ktime_get_real_ts64(ts)	getnstimeofday64(ts)
 
 /*

kernel/time/clockevents.c

@@ -120,19 +120,25 @@ static int __clockevents_switch_state(struct clock_event_device *dev,
 		/* The clockevent device is getting replaced. Shut it down. */
 
 	case CLOCK_EVT_STATE_SHUTDOWN:
+		if (dev->set_state_shutdown)
 			return dev->set_state_shutdown(dev);
+		return 0;
 
 	case CLOCK_EVT_STATE_PERIODIC:
 		/* Core internal bug */
 		if (!(dev->features & CLOCK_EVT_FEAT_PERIODIC))
 			return -ENOSYS;
+		if (dev->set_state_periodic)
 			return dev->set_state_periodic(dev);
+		return 0;
 
 	case CLOCK_EVT_STATE_ONESHOT:
 		/* Core internal bug */
 		if (!(dev->features & CLOCK_EVT_FEAT_ONESHOT))
 			return -ENOSYS;
+		if (dev->set_state_oneshot)
 			return dev->set_state_oneshot(dev);
+		return 0;
 
 	case CLOCK_EVT_STATE_ONESHOT_STOPPED:
 		/* Core internal bug */
@@ -471,18 +477,6 @@ static int clockevents_sanity_check(struct clock_event_device *dev)
 	if (dev->features & CLOCK_EVT_FEAT_DUMMY)
 		return 0;
 
-	/* New state-specific callbacks */
-	if (!dev->set_state_shutdown)
-		return -EINVAL;
-
-	if ((dev->features & CLOCK_EVT_FEAT_PERIODIC) &&
-	    !dev->set_state_periodic)
-		return -EINVAL;
-
-	if ((dev->features & CLOCK_EVT_FEAT_ONESHOT) &&
-	    !dev->set_state_oneshot)
-		return -EINVAL;
-
 	return 0;
 }
 

kernel/time/tick-broadcast.c

@@ -159,7 +159,7 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu)
 {
 	struct clock_event_device *bc = tick_broadcast_device.evtdev;
 	unsigned long flags;
-	int ret;
+	int ret = 0;
 
 	raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
 
@@ -221,13 +221,14 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu)
 			 * If we kept the cpu in the broadcast mask,
 			 * tell the caller to leave the per cpu device
 			 * in shutdown state. The periodic interrupt
-			 * is delivered by the broadcast device.
+			 * is delivered by the broadcast device, if
+			 * the broadcast device exists and is not
+			 * hrtimer based.
 			 */
+			if (bc && !(bc->features & CLOCK_EVT_FEAT_HRTIMER))
 				ret = cpumask_test_cpu(cpu, tick_broadcast_mask);
 			break;
 		default:
-			/* Nothing to do */
-			ret = 0;
 			break;
 		}
 	}
@@ -265,8 +266,22 @@ static bool tick_do_broadcast(struct cpumask *mask)
 	 * Check, if the current cpu is in the mask
 	 */
 	if (cpumask_test_cpu(cpu, mask)) {
+		struct clock_event_device *bc = tick_broadcast_device.evtdev;
+
 		cpumask_clear_cpu(cpu, mask);
-		local = true;
+		/*
+		 * We only run the local handler, if the broadcast
+		 * device is not hrtimer based. Otherwise we run into
+		 * a hrtimer recursion.
+		 *
+		 * local timer_interrupt()
+		 *   local_handler()
+		 *     expire_hrtimers()
+		 *       bc_handler()
+		 *         local_handler()
+		 *	     expire_hrtimers()
+		 */
+		local = !(bc->features & CLOCK_EVT_FEAT_HRTIMER);
 	}
 
 	if (!cpumask_empty(mask)) {
@@ -301,6 +316,13 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev)
 	bool bc_local;
 
 	raw_spin_lock(&tick_broadcast_lock);
+
+	/* Handle spurious interrupts gracefully */
+	if (clockevent_state_shutdown(tick_broadcast_device.evtdev)) {
+		raw_spin_unlock(&tick_broadcast_lock);
+		return;
+	}
+
 	bc_local = tick_do_periodic_broadcast();
 
 	if (clockevent_state_oneshot(dev)) {
@@ -359,8 +381,16 @@ void tick_broadcast_control(enum tick_broadcast_mode mode)
 	case TICK_BROADCAST_ON:
 		cpumask_set_cpu(cpu, tick_broadcast_on);
 		if (!cpumask_test_and_set_cpu(cpu, tick_broadcast_mask)) {
-			if (tick_broadcast_device.mode ==
+			/*
-			    TICKDEV_MODE_PERIODIC)
+			 * Only shutdown the cpu local device, if:
+			 *
+			 * - the broadcast device exists
+			 * - the broadcast device is not a hrtimer based one
+			 * - the broadcast device is in periodic mode to
+			 *   avoid a hickup during switch to oneshot mode
+			 */
+			if (bc && !(bc->features & CLOCK_EVT_FEAT_HRTIMER) &&
+			    tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
 				clockevents_shutdown(dev);
 		}
 		break;
@@ -379,6 +409,7 @@ void tick_broadcast_control(enum tick_broadcast_mode mode)
 		break;
 	}
 
+	if (bc) {
 		if (cpumask_empty(tick_broadcast_mask)) {
 			if (!bc_stopped)
 				clockevents_shutdown(bc);
@@ -388,6 +419,7 @@ void tick_broadcast_control(enum tick_broadcast_mode mode)
 			else
 				tick_broadcast_setup_oneshot(bc);
 		}
+	}
 	raw_spin_unlock(&tick_broadcast_lock);
 }
 EXPORT_SYMBOL_GPL(tick_broadcast_control);
@@ -662,71 +694,82 @@ static void broadcast_shutdown_local(struct clock_event_device *bc,
 	clockevents_switch_state(dev, CLOCK_EVT_STATE_SHUTDOWN);
 }
 
-/**
+int __tick_broadcast_oneshot_control(enum tick_broadcast_state state)
- * tick_broadcast_oneshot_control - Enter/exit broadcast oneshot mode
- * @state:	The target state (enter/exit)
- *
- * The system enters/leaves a state, where affected devices might stop
- * Returns 0 on success, -EBUSY if the cpu is used to broadcast wakeups.
- *
- * Called with interrupts disabled, so clockevents_lock is not
- * required here because the local clock event device cannot go away
- * under us.
- */
-int tick_broadcast_oneshot_control(enum tick_broadcast_state state)
 {
 	struct clock_event_device *bc, *dev;
-	struct tick_device *td;
 	int cpu, ret = 0;
 	ktime_t now;
 
 	/*
-	 * Periodic mode does not care about the enter/exit of power
+	 * If there is no broadcast device, tell the caller not to go
-	 * states
+	 * into deep idle.
 	 */
-	if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
+	if (!tick_broadcast_device.evtdev)
-		return 0;
+		return -EBUSY;
 
-	/*
+	dev = this_cpu_ptr(&tick_cpu_device)->evtdev;
-	 * We are called with preemtion disabled from the depth of the
-	 * idle code, so we can't be moved away.
-	 */
-	td = this_cpu_ptr(&tick_cpu_device);
-	dev = td->evtdev;
-
-	if (!(dev->features & CLOCK_EVT_FEAT_C3STOP))
-		return 0;
 
 	raw_spin_lock(&tick_broadcast_lock);
 	bc = tick_broadcast_device.evtdev;
 	cpu = smp_processor_id();
 
 	if (state == TICK_BROADCAST_ENTER) {
+		/*
+		 * If the current CPU owns the hrtimer broadcast
+		 * mechanism, it cannot go deep idle and we do not add
+		 * the CPU to the broadcast mask. We don't have to go
+		 * through the EXIT path as the local timer is not
+		 * shutdown.
+		 */
+		ret = broadcast_needs_cpu(bc, cpu);
+		if (ret)
+			goto out;
+
+		/*
+		 * If the broadcast device is in periodic mode, we
+		 * return.
+		 */
+		if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) {
+			/* If it is a hrtimer based broadcast, return busy */
+			if (bc->features & CLOCK_EVT_FEAT_HRTIMER)
+				ret = -EBUSY;
+			goto out;
+		}
+
 		if (!cpumask_test_and_set_cpu(cpu, tick_broadcast_oneshot_mask)) {
 			WARN_ON_ONCE(cpumask_test_cpu(cpu, tick_broadcast_pending_mask));
+
+			/* Conditionally shut down the local timer. */
 			broadcast_shutdown_local(bc, dev);
+
 			/*
 			 * We only reprogram the broadcast timer if we
 			 * did not mark ourself in the force mask and
 			 * if the cpu local event is earlier than the
 			 * broadcast event. If the current CPU is in
 			 * the force mask, then we are going to be
-			 * woken by the IPI right away.
+			 * woken by the IPI right away; we return
+			 * busy, so the CPU does not try to go deep
+			 * idle.
 			 */
-			if (!cpumask_test_cpu(cpu, tick_broadcast_force_mask) &&
+			if (cpumask_test_cpu(cpu, tick_broadcast_force_mask)) {
-			    dev->next_event.tv64 < bc->next_event.tv64)
+				ret = -EBUSY;
+			} else if (dev->next_event.tv64 < bc->next_event.tv64) {
 				tick_broadcast_set_event(bc, cpu, dev->next_event);
-		}
 				/*
-		 * If the current CPU owns the hrtimer broadcast
+				 * In case of hrtimer broadcasts the
-		 * mechanism, it cannot go deep idle and we remove the
+				 * programming might have moved the
-		 * CPU from the broadcast mask. We don't have to go
+				 * timer to this cpu. If yes, remove
-		 * through the EXIT path as the local timer is not
+				 * us from the broadcast mask and
-		 * shutdown.
+				 * return busy.
 				 */
 				ret = broadcast_needs_cpu(bc, cpu);
-		if (ret)
+				if (ret) {
-			cpumask_clear_cpu(cpu, tick_broadcast_oneshot_mask);
+					cpumask_clear_cpu(cpu,
+						tick_broadcast_oneshot_mask);
+				}
+			}
+		}
 	} else {
 		if (cpumask_test_and_clear_cpu(cpu, tick_broadcast_oneshot_mask)) {
 			clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT);
@@ -938,6 +981,16 @@ bool tick_broadcast_oneshot_available(void)
 	return bc ? bc->features & CLOCK_EVT_FEAT_ONESHOT : false;
 }
 
+#else
+int __tick_broadcast_oneshot_control(enum tick_broadcast_state state)
+{
+	struct clock_event_device *bc = tick_broadcast_device.evtdev;
+
+	if (!bc || (bc->features & CLOCK_EVT_FEAT_HRTIMER))
+		return -EBUSY;
+
+	return 0;
+}
 #endif
 
 void __init tick_broadcast_init(void)

kernel/time/tick-common.c

@@ -343,6 +343,27 @@ out_bc:
 	tick_install_broadcast_device(newdev);
 }
 
+/**
+ * tick_broadcast_oneshot_control - Enter/exit broadcast oneshot mode
+ * @state:	The target state (enter/exit)
+ *
+ * The system enters/leaves a state, where affected devices might stop
+ * Returns 0 on success, -EBUSY if the cpu is used to broadcast wakeups.
+ *
+ * Called with interrupts disabled, so clockevents_lock is not
+ * required here because the local clock event device cannot go away
+ * under us.
+ */
+int tick_broadcast_oneshot_control(enum tick_broadcast_state state)
+{
+	struct tick_device *td = this_cpu_ptr(&tick_cpu_device);
+
+	if (!(td->evtdev->features & CLOCK_EVT_FEAT_C3STOP))
+		return 0;
+
+	return __tick_broadcast_oneshot_control(state);
+}
+
 #ifdef CONFIG_HOTPLUG_CPU
 /*
  * Transfer the do_timer job away from a dying cpu.

kernel/time/tick-sched.h

@@ -71,4 +71,14 @@ extern void tick_cancel_sched_timer(int cpu);
 static inline void tick_cancel_sched_timer(int cpu) { }
 #endif
 
+#ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST
+extern int __tick_broadcast_oneshot_control(enum tick_broadcast_state state);
+#else
+static inline int
+__tick_broadcast_oneshot_control(enum tick_broadcast_state state)
+{
+	return -EBUSY;
+}
+#endif
+
 #endif