X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=kernel%2Fsched_rt.c;h=1bbd99014011894766334eee96da13d6fbc4a8de;hb=45ab6b0c76d0e4cce5bd608ccf97b0f6b20f18df;hp=1144bf55669d01c571d3d332df086efa39cc5d99;hpb=614ee1f61f667b02165c1ae0c1357048dc6d94a0;p=safe%2Fjmp%2Flinux-2.6

diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 1144bf5..1bbd990 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -12,7 +12,10 @@ static inline int rt_overloaded(struct rq *rq)
 
 static inline void rt_set_overload(struct rq *rq)
 {
-	cpu_set(rq->cpu, rq->rd->rto_mask);
+	if (!rq->online)
+		return;
+
+	cpumask_set_cpu(rq->cpu, rq->rd->rto_mask);
 	/*
 	 * Make sure the mask is visible before we set
 	 * the overload count. That is checked to determine
@@ -26,9 +29,12 @@ static inline void rt_set_overload(struct rq *rq)
 
 static inline void rt_clear_overload(struct rq *rq)
 {
+	if (!rq->online)
+		return;
+
 	/* the order here really doesn't matter */
 	atomic_dec(&rq->rd->rto_count);
-	cpu_clear(rq->cpu, rq->rd->rto_mask);
+	cpumask_clear_cpu(rq->cpu, rq->rd->rto_mask);
 }
 
 static void update_rt_migration(struct rq *rq)
@@ -55,14 +61,19 @@ static inline int on_rt_rq(struct sched_rt_entity *rt_se)
 	return !list_empty(&rt_se->run_list);
 }
 
-#ifdef CONFIG_FAIR_GROUP_SCHED
+#ifdef CONFIG_RT_GROUP_SCHED
 
-static inline unsigned int sched_rt_ratio(struct rt_rq *rt_rq)
+static inline u64 sched_rt_runtime(struct rt_rq *rt_rq)
 {
 	if (!rt_rq->tg)
-		return SCHED_RT_FRAC;
+		return RUNTIME_INF;
+
+	return rt_rq->rt_runtime;
+}
 
-	return rt_rq->tg->rt_ratio;
+static inline u64 sched_rt_period(struct rt_rq *rt_rq)
+{
+	return ktime_to_ns(rt_rq->tg->rt_bandwidth.rt_period);
 }
 
 #define for_each_leaf_rt_rq(rt_rq, rq) \
@@ -89,17 +100,20 @@ static inline struct rt_rq *group_rt_rq(struct sched_rt_entity *rt_se)
 static void enqueue_rt_entity(struct sched_rt_entity *rt_se);
 static void dequeue_rt_entity(struct sched_rt_entity *rt_se);
 
-static void sched_rt_ratio_enqueue(struct rt_rq *rt_rq)
+static void sched_rt_rq_enqueue(struct rt_rq *rt_rq)
 {
+	struct task_struct *curr = rq_of_rt_rq(rt_rq)->curr;
 	struct sched_rt_entity *rt_se = rt_rq->rt_se;
 
-	if (rt_se && !on_rt_rq(rt_se) && rt_rq->rt_nr_running) {
-		enqueue_rt_entity(rt_se);
-		resched_task(rq_of_rt_rq(rt_rq)->curr);
+	if (rt_rq->rt_nr_running) {
+		if (rt_se && !on_rt_rq(rt_se))
+			enqueue_rt_entity(rt_se);
+		if (rt_rq->highest_prio < curr->prio)
+			resched_task(curr);
 	}
 }
 
-static void sched_rt_ratio_dequeue(struct rt_rq *rt_rq)
+static void sched_rt_rq_dequeue(struct rt_rq *rt_rq)
 {
 	struct sched_rt_entity *rt_se = rt_rq->rt_se;
 
@@ -107,11 +121,56 @@ static void sched_rt_ratio_dequeue(struct rt_rq *rt_rq)
 		dequeue_rt_entity(rt_se);
 }
 
+static inline int rt_rq_throttled(struct rt_rq *rt_rq)
+{
+	return rt_rq->rt_throttled && !rt_rq->rt_nr_boosted;
+}
+
+static int rt_se_boosted(struct sched_rt_entity *rt_se)
+{
+	struct rt_rq *rt_rq = group_rt_rq(rt_se);
+	struct task_struct *p;
+
+	if (rt_rq)
+		return !!rt_rq->rt_nr_boosted;
+
+	p = rt_task_of(rt_se);
+	return p->prio != p->normal_prio;
+}
+
+#ifdef CONFIG_SMP
+static inline const struct cpumask *sched_rt_period_mask(void)
+{
+	return cpu_rq(smp_processor_id())->rd->span;
+}
 #else
+static inline const struct cpumask *sched_rt_period_mask(void)
+{
+	return cpu_online_mask;
+}
+#endif
+
+static inline
+struct rt_rq *sched_rt_period_rt_rq(struct rt_bandwidth *rt_b, int cpu)
+{
+	return container_of(rt_b, struct task_group, rt_bandwidth)->rt_rq[cpu];
+}
+
+static inline struct rt_bandwidth *sched_rt_bandwidth(struct rt_rq *rt_rq)
+{
+	return &rt_rq->tg->rt_bandwidth;
+}
 
-static inline unsigned int sched_rt_ratio(struct rt_rq *rt_rq)
+#else /* !CONFIG_RT_GROUP_SCHED */
+
+static inline u64 sched_rt_runtime(struct rt_rq *rt_rq)
 {
-	return sysctl_sched_rt_ratio;
+	return rt_rq->rt_runtime;
+}
+
+static inline u64 sched_rt_period(struct rt_rq *rt_rq)
+{
+	return ktime_to_ns(def_rt_bandwidth.rt_period);
 }
 
 #define for_each_leaf_rt_rq(rt_rq, rq) \
@@ -138,75 +197,312 @@ static inline struct rt_rq *group_rt_rq(struct sched_rt_entity *rt_se)
 	return NULL;
 }
 
-static inline void sched_rt_ratio_enqueue(struct rt_rq *rt_rq)
+static inline void sched_rt_rq_enqueue(struct rt_rq *rt_rq)
 {
+	if (rt_rq->rt_nr_running)
+		resched_task(rq_of_rt_rq(rt_rq)->curr);
 }
 
-static inline void sched_rt_ratio_dequeue(struct rt_rq *rt_rq)
+static inline void sched_rt_rq_dequeue(struct rt_rq *rt_rq)
 {
 }
 
-#endif
+static inline int rt_rq_throttled(struct rt_rq *rt_rq)
+{
+	return rt_rq->rt_throttled;
+}
 
-static inline int rt_se_prio(struct sched_rt_entity *rt_se)
+static inline const struct cpumask *sched_rt_period_mask(void)
 {
-#ifdef CONFIG_FAIR_GROUP_SCHED
-	struct rt_rq *rt_rq = group_rt_rq(rt_se);
+	return cpu_online_mask;
+}
 
-	if (rt_rq)
-		return rt_rq->highest_prio;
-#endif
+static inline
+struct rt_rq *sched_rt_period_rt_rq(struct rt_bandwidth *rt_b, int cpu)
+{
+	return &cpu_rq(cpu)->rt;
+}
 
-	return rt_task_of(rt_se)->prio;
+static inline struct rt_bandwidth *sched_rt_bandwidth(struct rt_rq *rt_rq)
+{
+	return &def_rt_bandwidth;
 }
 
-static int sched_rt_ratio_exceeded(struct rt_rq *rt_rq)
+#endif /* CONFIG_RT_GROUP_SCHED */
+
+#ifdef CONFIG_SMP
+/*
+ * We ran out of runtime, see if we can borrow some from our neighbours.
+ */
+static int do_balance_runtime(struct rt_rq *rt_rq)
 {
-	unsigned int rt_ratio = sched_rt_ratio(rt_rq);
-	u64 period, ratio;
+	struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq);
+	struct root_domain *rd = cpu_rq(smp_processor_id())->rd;
+	int i, weight, more = 0;
+	u64 rt_period;
 
-	if (rt_ratio == SCHED_RT_FRAC)
-		return 0;
+	weight = cpumask_weight(rd->span);
 
-	if (rt_rq->rt_throttled)
-		return 1;
+	spin_lock(&rt_b->rt_runtime_lock);
+	rt_period = ktime_to_ns(rt_b->rt_period);
+	for_each_cpu(i, rd->span) {
+		struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i);
+		s64 diff;
 
-	period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
-	ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+		if (iter == rt_rq)
+			continue;
 
-	if (rt_rq->rt_time > ratio) {
-		rt_rq->rt_throttled = 1;
-		sched_rt_ratio_dequeue(rt_rq);
-		return 1;
+		spin_lock(&iter->rt_runtime_lock);
+		/*
+		 * Either all rqs have inf runtime and there's nothing to steal
+		 * or __disable_runtime() below sets a specific rq to inf to
+		 * indicate its been disabled and disalow stealing.
+		 */
+		if (iter->rt_runtime == RUNTIME_INF)
+			goto next;
+
+		/*
+		 * From runqueues with spare time, take 1/n part of their
+		 * spare time, but no more than our period.
+		 */
+		diff = iter->rt_runtime - iter->rt_time;
+		if (diff > 0) {
+			diff = div_u64((u64)diff, weight);
+			if (rt_rq->rt_runtime + diff > rt_period)
+				diff = rt_period - rt_rq->rt_runtime;
+			iter->rt_runtime -= diff;
+			rt_rq->rt_runtime += diff;
+			more = 1;
+			if (rt_rq->rt_runtime == rt_period) {
+				spin_unlock(&iter->rt_runtime_lock);
+				break;
+			}
+		}
+next:
+		spin_unlock(&iter->rt_runtime_lock);
 	}
+	spin_unlock(&rt_b->rt_runtime_lock);
 
-	return 0;
+	return more;
 }
 
-static void __update_sched_rt_period(struct rt_rq *rt_rq, u64 period)
+/*
+ * Ensure this RQ takes back all the runtime it lend to its neighbours.
+ */
+static void __disable_runtime(struct rq *rq)
 {
-	unsigned long rt_ratio = sched_rt_ratio(rt_rq);
-	u64 ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+	struct root_domain *rd = rq->rd;
+	struct rt_rq *rt_rq;
 
-	rt_rq->rt_time -= min(rt_rq->rt_time, ratio);
-	if (rt_rq->rt_throttled) {
-		rt_rq->rt_throttled = 0;
-		sched_rt_ratio_enqueue(rt_rq);
+	if (unlikely(!scheduler_running))
+		return;
+
+	for_each_leaf_rt_rq(rt_rq, rq) {
+		struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq);
+		s64 want;
+		int i;
+
+		spin_lock(&rt_b->rt_runtime_lock);
+		spin_lock(&rt_rq->rt_runtime_lock);
+		/*
+		 * Either we're all inf and nobody needs to borrow, or we're
+		 * already disabled and thus have nothing to do, or we have
+		 * exactly the right amount of runtime to take out.
+		 */
+		if (rt_rq->rt_runtime == RUNTIME_INF ||
+				rt_rq->rt_runtime == rt_b->rt_runtime)
+			goto balanced;
+		spin_unlock(&rt_rq->rt_runtime_lock);
+
+		/*
+		 * Calculate the difference between what we started out with
+		 * and what we current have, that's the amount of runtime
+		 * we lend and now have to reclaim.
+		 */
+		want = rt_b->rt_runtime - rt_rq->rt_runtime;
+
+		/*
+		 * Greedy reclaim, take back as much as we can.
+		 */
+		for_each_cpu(i, rd->span) {
+			struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i);
+			s64 diff;
+
+			/*
+			 * Can't reclaim from ourselves or disabled runqueues.
+			 */
+			if (iter == rt_rq || iter->rt_runtime == RUNTIME_INF)
+				continue;
+
+			spin_lock(&iter->rt_runtime_lock);
+			if (want > 0) {
+				diff = min_t(s64, iter->rt_runtime, want);
+				iter->rt_runtime -= diff;
+				want -= diff;
+			} else {
+				iter->rt_runtime -= want;
+				want -= want;
+			}
+			spin_unlock(&iter->rt_runtime_lock);
+
+			if (!want)
+				break;
+		}
+
+		spin_lock(&rt_rq->rt_runtime_lock);
+		/*
+		 * We cannot be left wanting - that would mean some runtime
+		 * leaked out of the system.
+		 */
+		BUG_ON(want);
+balanced:
+		/*
+		 * Disable all the borrow logic by pretending we have inf
+		 * runtime - in which case borrowing doesn't make sense.
+		 */
+		rt_rq->rt_runtime = RUNTIME_INF;
+		spin_unlock(&rt_rq->rt_runtime_lock);
+		spin_unlock(&rt_b->rt_runtime_lock);
 	}
 }
 
-static void update_sched_rt_period(struct rq *rq)
+static void disable_runtime(struct rq *rq)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rq->lock, flags);
+	__disable_runtime(rq);
+	spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+static void __enable_runtime(struct rq *rq)
 {
 	struct rt_rq *rt_rq;
-	u64 period;
 
-	while (rq->clock > rq->rt_period_expire) {
-		period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
-		rq->rt_period_expire += period;
+	if (unlikely(!scheduler_running))
+		return;
+
+	/*
+	 * Reset each runqueue's bandwidth settings
+	 */
+	for_each_leaf_rt_rq(rt_rq, rq) {
+		struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq);
+
+		spin_lock(&rt_b->rt_runtime_lock);
+		spin_lock(&rt_rq->rt_runtime_lock);
+		rt_rq->rt_runtime = rt_b->rt_runtime;
+		rt_rq->rt_time = 0;
+		rt_rq->rt_throttled = 0;
+		spin_unlock(&rt_rq->rt_runtime_lock);
+		spin_unlock(&rt_b->rt_runtime_lock);
+	}
+}
+
+static void enable_runtime(struct rq *rq)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rq->lock, flags);
+	__enable_runtime(rq);
+	spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+static int balance_runtime(struct rt_rq *rt_rq)
+{
+	int more = 0;
+
+	if (rt_rq->rt_time > rt_rq->rt_runtime) {
+		spin_unlock(&rt_rq->rt_runtime_lock);
+		more = do_balance_runtime(rt_rq);
+		spin_lock(&rt_rq->rt_runtime_lock);
+	}
+
+	return more;
+}
+#else /* !CONFIG_SMP */
+static inline int balance_runtime(struct rt_rq *rt_rq)
+{
+	return 0;
+}
+#endif /* CONFIG_SMP */
+
+static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
+{
+	int i, idle = 1;
+	const struct cpumask *span;
+
+	if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF)
+		return 1;
+
+	span = sched_rt_period_mask();
+	for_each_cpu(i, span) {
+		int enqueue = 0;
+		struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i);
+		struct rq *rq = rq_of_rt_rq(rt_rq);
+
+		spin_lock(&rq->lock);
+		if (rt_rq->rt_time) {
+			u64 runtime;
+
+			spin_lock(&rt_rq->rt_runtime_lock);
+			if (rt_rq->rt_throttled)
+				balance_runtime(rt_rq);
+			runtime = rt_rq->rt_runtime;
+			rt_rq->rt_time -= min(rt_rq->rt_time, overrun*runtime);
+			if (rt_rq->rt_throttled && rt_rq->rt_time < runtime) {
+				rt_rq->rt_throttled = 0;
+				enqueue = 1;
+			}
+			if (rt_rq->rt_time || rt_rq->rt_nr_running)
+				idle = 0;
+			spin_unlock(&rt_rq->rt_runtime_lock);
+		} else if (rt_rq->rt_nr_running)
+			idle = 0;
+
+		if (enqueue)
+			sched_rt_rq_enqueue(rt_rq);
+		spin_unlock(&rq->lock);
+	}
+
+	return idle;
+}
+
+static inline int rt_se_prio(struct sched_rt_entity *rt_se)
+{
+#ifdef CONFIG_RT_GROUP_SCHED
+	struct rt_rq *rt_rq = group_rt_rq(rt_se);
+
+	if (rt_rq)
+		return rt_rq->highest_prio;
+#endif
+
+	return rt_task_of(rt_se)->prio;
+}
+
+static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
+{
+	u64 runtime = sched_rt_runtime(rt_rq);
+
+	if (rt_rq->rt_throttled)
+		return rt_rq_throttled(rt_rq);
+
+	if (sched_rt_runtime(rt_rq) >= sched_rt_period(rt_rq))
+		return 0;
+
+	balance_runtime(rt_rq);
+	runtime = sched_rt_runtime(rt_rq);
+	if (runtime == RUNTIME_INF)
+		return 0;
 
-		for_each_leaf_rt_rq(rt_rq, rq)
-			__update_sched_rt_period(rt_rq, period);
+	if (rt_rq->rt_time > runtime) {
+		rt_rq->rt_throttled = 1;
+		if (rt_rq_throttled(rt_rq)) {
+			sched_rt_rq_dequeue(rt_rq);
+			return 1;
+		}
 	}
+
+	return 0;
 }
 
 /*
@@ -230,17 +526,25 @@ static void update_curr_rt(struct rq *rq)
 	schedstat_set(curr->se.exec_max, max(curr->se.exec_max, delta_exec));
 
 	curr->se.sum_exec_runtime += delta_exec;
+	account_group_exec_runtime(curr, delta_exec);
+
 	curr->se.exec_start = rq->clock;
 	cpuacct_charge(curr, delta_exec);
 
-	rt_rq->rt_time += delta_exec;
-	/*
-	 * might make it a tad more accurate:
-	 *
-	 * update_sched_rt_period(rq);
-	 */
-	if (sched_rt_ratio_exceeded(rt_rq))
-		resched_task(curr);
+	if (!rt_bandwidth_enabled())
+		return;
+
+	for_each_sched_rt_entity(rt_se) {
+		rt_rq = rt_rq_of_se(rt_se);
+
+		if (sched_rt_runtime(rt_rq) != RUNTIME_INF) {
+			spin_lock(&rt_rq->rt_runtime_lock);
+			rt_rq->rt_time += delta_exec;
+			if (sched_rt_runtime_exceeded(rt_rq))
+				resched_task(curr);
+			spin_unlock(&rt_rq->rt_runtime_lock);
+		}
+	}
 }
 
 static inline
@@ -248,27 +552,51 @@ void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 {
 	WARN_ON(!rt_prio(rt_se_prio(rt_se)));
 	rt_rq->rt_nr_running++;
-#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
-	if (rt_se_prio(rt_se) < rt_rq->highest_prio)
+#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED
+	if (rt_se_prio(rt_se) < rt_rq->highest_prio) {
+#ifdef CONFIG_SMP
+		struct rq *rq = rq_of_rt_rq(rt_rq);
+#endif
+
 		rt_rq->highest_prio = rt_se_prio(rt_se);
+#ifdef CONFIG_SMP
+		if (rq->online)
+			cpupri_set(&rq->rd->cpupri, rq->cpu,
+				   rt_se_prio(rt_se));
+#endif
+	}
 #endif
 #ifdef CONFIG_SMP
 	if (rt_se->nr_cpus_allowed > 1) {
 		struct rq *rq = rq_of_rt_rq(rt_rq);
+
 		rq->rt.rt_nr_migratory++;
 	}
 
 	update_rt_migration(rq_of_rt_rq(rt_rq));
 #endif
+#ifdef CONFIG_RT_GROUP_SCHED
+	if (rt_se_boosted(rt_se))
+		rt_rq->rt_nr_boosted++;
+
+	if (rt_rq->tg)
+		start_rt_bandwidth(&rt_rq->tg->rt_bandwidth);
+#else
+	start_rt_bandwidth(&def_rt_bandwidth);
+#endif
 }
 
 static inline
 void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 {
+#ifdef CONFIG_SMP
+	int highest_prio = rt_rq->highest_prio;
+#endif
+
 	WARN_ON(!rt_prio(rt_se_prio(rt_se)));
 	WARN_ON(!rt_rq->rt_nr_running);
 	rt_rq->rt_nr_running--;
-#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED
 	if (rt_rq->rt_nr_running) {
 		struct rt_prio_array *array;
 
@@ -288,26 +616,47 @@ void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 		rq->rt.rt_nr_migratory--;
 	}
 
+	if (rt_rq->highest_prio != highest_prio) {
+		struct rq *rq = rq_of_rt_rq(rt_rq);
+
+		if (rq->online)
+			cpupri_set(&rq->rd->cpupri, rq->cpu,
+				   rt_rq->highest_prio);
+	}
+
 	update_rt_migration(rq_of_rt_rq(rt_rq));
 #endif /* CONFIG_SMP */
+#ifdef CONFIG_RT_GROUP_SCHED
+	if (rt_se_boosted(rt_se))
+		rt_rq->rt_nr_boosted--;
+
+	WARN_ON(!rt_rq->rt_nr_running && rt_rq->rt_nr_boosted);
+#endif
 }
 
-static void enqueue_rt_entity(struct sched_rt_entity *rt_se)
+static void __enqueue_rt_entity(struct sched_rt_entity *rt_se)
 {
 	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
 	struct rt_prio_array *array = &rt_rq->active;
 	struct rt_rq *group_rq = group_rt_rq(rt_se);
+	struct list_head *queue = array->queue + rt_se_prio(rt_se);
 
-	if (group_rq && group_rq->rt_throttled)
+	/*
+	 * Don't enqueue the group if its throttled, or when empty.
+	 * The latter is a consequence of the former when a child group
+	 * get throttled and the current group doesn't have any other
+	 * active members.
+	 */
+	if (group_rq && (rt_rq_throttled(group_rq) || !group_rq->rt_nr_running))
 		return;
 
-	list_add_tail(&rt_se->run_list, array->queue + rt_se_prio(rt_se));
+	list_add_tail(&rt_se->run_list, queue);
 	__set_bit(rt_se_prio(rt_se), array->bitmap);
 
 	inc_rt_tasks(rt_se, rt_rq);
 }
 
-static void dequeue_rt_entity(struct sched_rt_entity *rt_se)
+static void __dequeue_rt_entity(struct sched_rt_entity *rt_se)
 {
 	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
 	struct rt_prio_array *array = &rt_rq->active;
@@ -322,27 +671,39 @@ static void dequeue_rt_entity(struct sched_rt_entity *rt_se)
 /*
  * Because the prio of an upper entry depends on the lower
  * entries, we must remove entries top - down.
- *
- * XXX: O(1/2 h^2) because we can only walk up, not down the chain.
- *      doesn't matter much for now, as h=2 for GROUP_SCHED.
  */
-static void dequeue_rt_stack(struct task_struct *p)
+static void dequeue_rt_stack(struct sched_rt_entity *rt_se)
 {
-	struct sched_rt_entity *rt_se, *top_se;
+	struct sched_rt_entity *back = NULL;
 
-	/*
-	 * dequeue all, top - down.
-	 */
-	do {
-		rt_se = &p->rt;
-		top_se = NULL;
-		for_each_sched_rt_entity(rt_se) {
-			if (on_rt_rq(rt_se))
-				top_se = rt_se;
-		}
-		if (top_se)
-			dequeue_rt_entity(top_se);
-	} while (top_se);
+	for_each_sched_rt_entity(rt_se) {
+		rt_se->back = back;
+		back = rt_se;
+	}
+
+	for (rt_se = back; rt_se; rt_se = rt_se->back) {
+		if (on_rt_rq(rt_se))
+			__dequeue_rt_entity(rt_se);
+	}
+}
+
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se)
+{
+	dequeue_rt_stack(rt_se);
+	for_each_sched_rt_entity(rt_se)
+		__enqueue_rt_entity(rt_se);
+}
+
+static void dequeue_rt_entity(struct sched_rt_entity *rt_se)
+{
+	dequeue_rt_stack(rt_se);
+
+	for_each_sched_rt_entity(rt_se) {
+		struct rt_rq *rt_rq = group_rt_rq(rt_se);
+
+		if (rt_rq && rt_rq->rt_nr_running)
+			__enqueue_rt_entity(rt_se);
+	}
 }
 
 /*
@@ -355,13 +716,7 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
 	if (wakeup)
 		rt_se->timeout = 0;
 
-	dequeue_rt_stack(p);
-
-	/*
-	 * enqueue everybody, bottom - up.
-	 */
-	for_each_sched_rt_entity(rt_se)
-		enqueue_rt_entity(rt_se);
+	enqueue_rt_entity(rt_se);
 
 	inc_cpu_load(rq, p->se.load.weight);
 }
@@ -369,20 +724,9 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
 static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
 {
 	struct sched_rt_entity *rt_se = &p->rt;
-	struct rt_rq *rt_rq;
 
 	update_curr_rt(rq);
-
-	dequeue_rt_stack(p);
-
-	/*
-	 * re-enqueue all non-empty rt_rq entities.
-	 */
-	for_each_sched_rt_entity(rt_se) {
-		rt_rq = group_rt_rq(rt_se);
-		if (rt_rq && rt_rq->rt_nr_running)
-			enqueue_rt_entity(rt_se);
-	}
+	dequeue_rt_entity(rt_se);
 
 	dec_cpu_load(rq, p->se.load.weight);
 }
@@ -391,28 +735,34 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
  * Put task to the end of the run list without the overhead of dequeue
  * followed by enqueue.
  */
-static
-void requeue_rt_entity(struct rt_rq *rt_rq, struct sched_rt_entity *rt_se)
+static void
+requeue_rt_entity(struct rt_rq *rt_rq, struct sched_rt_entity *rt_se, int head)
 {
-	struct rt_prio_array *array = &rt_rq->active;
-
-	list_move_tail(&rt_se->run_list, array->queue + rt_se_prio(rt_se));
+	if (on_rt_rq(rt_se)) {
+		struct rt_prio_array *array = &rt_rq->active;
+		struct list_head *queue = array->queue + rt_se_prio(rt_se);
+
+		if (head)
+			list_move(&rt_se->run_list, queue);
+		else
+			list_move_tail(&rt_se->run_list, queue);
+	}
 }
 
-static void requeue_task_rt(struct rq *rq, struct task_struct *p)
+static void requeue_task_rt(struct rq *rq, struct task_struct *p, int head)
 {
 	struct sched_rt_entity *rt_se = &p->rt;
 	struct rt_rq *rt_rq;
 
 	for_each_sched_rt_entity(rt_se) {
 		rt_rq = rt_rq_of_se(rt_se);
-		requeue_rt_entity(rt_rq, rt_se);
+		requeue_rt_entity(rt_rq, rt_se, head);
 	}
 }
 
 static void yield_task_rt(struct rq *rq)
 {
-	requeue_task_rt(rq, rq->curr);
+	requeue_task_rt(rq, rq->curr, 0);
 }
 
 #ifdef CONFIG_SMP
@@ -452,15 +802,63 @@ static int select_task_rq_rt(struct task_struct *p, int sync)
 	 */
 	return task_cpu(p);
 }
+
+static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p)
+{
+	cpumask_var_t mask;
+
+	if (rq->curr->rt.nr_cpus_allowed == 1)
+		return;
+
+	if (!alloc_cpumask_var(&mask, GFP_ATOMIC))
+		return;
+
+	if (p->rt.nr_cpus_allowed != 1
+	    && cpupri_find(&rq->rd->cpupri, p, mask))
+		goto free;
+
+	if (!cpupri_find(&rq->rd->cpupri, rq->curr, mask))
+		goto free;
+
+	/*
+	 * There appears to be other cpus that can accept
+	 * current and none to run 'p', so lets reschedule
+	 * to try and push current away:
+	 */
+	requeue_task_rt(rq, p, 1);
+	resched_task(rq->curr);
+free:
+	free_cpumask_var(mask);
+}
+
 #endif /* CONFIG_SMP */
 
 /*
  * Preempt the current task with a newly woken task if needed:
  */
-static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p)
+static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int sync)
 {
-	if (p->prio < rq->curr->prio)
+	if (p->prio < rq->curr->prio) {
 		resched_task(rq->curr);
+		return;
+	}
+
+#ifdef CONFIG_SMP
+	/*
+	 * If:
+	 *
+	 * - the newly woken task is of equal priority to the current task
+	 * - the newly woken task is non-migratable while current is migratable
+	 * - current will be preempted on the next reschedule
+	 *
+	 * we should check to see if current can readily move to a different
+	 * cpu.  If so, we will reschedule to allow the push logic to try
+	 * to move current somewhere else, making room for our non-migratable
+	 * task.
+	 */
+	if (p->prio == rq->curr->prio && !need_resched())
+		check_preempt_equal_prio(rq, p);
+#endif
 }
 
 static struct sched_rt_entity *pick_next_rt_entity(struct rq *rq,
@@ -471,15 +869,12 @@ static struct sched_rt_entity *pick_next_rt_entity(struct rq *rq,
 	struct list_head *queue;
 	int idx;
 
-	if (sched_rt_ratio_exceeded(rt_rq))
-		goto out;
-
 	idx = sched_find_first_bit(array->bitmap);
 	BUG_ON(idx >= MAX_RT_PRIO);
 
 	queue = array->queue + idx;
 	next = list_entry(queue->next, struct sched_rt_entity, run_list);
- out:
+
 	return next;
 }
 
@@ -489,19 +884,17 @@ static struct task_struct *pick_next_task_rt(struct rq *rq)
 	struct task_struct *p;
 	struct rt_rq *rt_rq;
 
- retry:
 	rt_rq = &rq->rt;
 
 	if (unlikely(!rt_rq->rt_nr_running))
 		return NULL;
 
-	if (sched_rt_ratio_exceeded(rt_rq))
+	if (rt_rq_throttled(rt_rq))
 		return NULL;
 
 	do {
 		rt_se = pick_next_rt_entity(rq, rt_rq);
-		if (unlikely(!rt_se))
-			goto retry;
+		BUG_ON(!rt_se);
 		rt_rq = group_rt_rq(rt_se);
 	} while (rt_rq);
 
@@ -521,13 +914,12 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
 /* Only try algorithms three times */
 #define RT_MAX_TRIES 3
 
-static int double_lock_balance(struct rq *this_rq, struct rq *busiest);
 static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep);
 
 static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 {
 	if (!task_running(rq, p) &&
-	    (cpu < 0 || cpu_isset(cpu, p->cpus_allowed)) &&
+	    (cpu < 0 || cpumask_test_cpu(cpu, &p->cpus_allowed)) &&
 	    (p->rt.nr_cpus_allowed > 1))
 		return 1;
 	return 0;
@@ -566,74 +958,7 @@ static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu)
 	return next;
 }
 
-static DEFINE_PER_CPU(cpumask_t, local_cpu_mask);
-
-static int find_lowest_cpus(struct task_struct *task, cpumask_t *lowest_mask)
-{
-	int       lowest_prio = -1;
-	int       lowest_cpu  = -1;
-	int       count       = 0;
-	int       cpu;
-
-	cpus_and(*lowest_mask, task_rq(task)->rd->online, task->cpus_allowed);
-
-	/*
-	 * Scan each rq for the lowest prio.
-	 */
-	for_each_cpu_mask(cpu, *lowest_mask) {
-		struct rq *rq = cpu_rq(cpu);
-
-		/* We look for lowest RT prio or non-rt CPU */
-		if (rq->rt.highest_prio >= MAX_RT_PRIO) {
-			/*
-			 * if we already found a low RT queue
-			 * and now we found this non-rt queue
-			 * clear the mask and set our bit.
-			 * Otherwise just return the queue as is
-			 * and the count==1 will cause the algorithm
-			 * to use the first bit found.
-			 */
-			if (lowest_cpu != -1) {
-				cpus_clear(*lowest_mask);
-				cpu_set(rq->cpu, *lowest_mask);
-			}
-			return 1;
-		}
-
-		/* no locking for now */
-		if ((rq->rt.highest_prio > task->prio)
-		    && (rq->rt.highest_prio >= lowest_prio)) {
-			if (rq->rt.highest_prio > lowest_prio) {
-				/* new low - clear old data */
-				lowest_prio = rq->rt.highest_prio;
-				lowest_cpu = cpu;
-				count = 0;
-			}
-			count++;
-		} else
-			cpu_clear(cpu, *lowest_mask);
-	}
-
-	/*
-	 * Clear out all the set bits that represent
-	 * runqueues that were of higher prio than
-	 * the lowest_prio.
-	 */
-	if (lowest_cpu > 0) {
-		/*
-		 * Perhaps we could add another cpumask op to
-		 * zero out bits. Like cpu_zero_bits(cpumask, nrbits);
-		 * Then that could be optimized to use memset and such.
-		 */
-		for_each_cpu_mask(cpu, *lowest_mask) {
-			if (cpu >= lowest_cpu)
-				break;
-			cpu_clear(cpu, *lowest_mask);
-		}
-	}
-
-	return count;
-}
+static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask);
 
 static inline int pick_optimal_cpu(int this_cpu, cpumask_t *mask)
 {
@@ -653,20 +978,22 @@ static inline int pick_optimal_cpu(int this_cpu, cpumask_t *mask)
 static int find_lowest_rq(struct task_struct *task)
 {
 	struct sched_domain *sd;
-	cpumask_t *lowest_mask = &__get_cpu_var(local_cpu_mask);
+	struct cpumask *lowest_mask = __get_cpu_var(local_cpu_mask);
 	int this_cpu = smp_processor_id();
 	int cpu      = task_cpu(task);
-	int count    = find_lowest_cpus(task, lowest_mask);
 
-	if (!count)
+	if (task->rt.nr_cpus_allowed == 1)
+		return -1; /* No other targets possible */
+
+	if (!cpupri_find(&task_rq(task)->rd->cpupri, task, lowest_mask))
 		return -1; /* No targets found */
 
 	/*
-	 * There is no sense in performing an optimal search if only one
-	 * target is found.
+	 * Only consider CPUs that are usable for migration.
+	 * I guess we might want to change cpupri_find() to ignore those
+	 * in the first place.
 	 */
-	if (count == 1)
-		return first_cpu(*lowest_mask);
+	cpumask_and(lowest_mask, lowest_mask, cpu_active_mask);
 
 	/*
 	 * At this point we have built a mask of cpus representing the
@@ -676,7 +1003,7 @@ static int find_lowest_rq(struct task_struct *task)
 	 * We prioritize the last cpu that the task executed on since
 	 * it is most likely cache-hot in that location.
 	 */
-	if (cpu_isset(cpu, *lowest_mask))
+	if (cpumask_test_cpu(cpu, lowest_mask))
 		return cpu;
 
 	/*
@@ -691,7 +1018,8 @@ static int find_lowest_rq(struct task_struct *task)
 			cpumask_t domain_mask;
 			int       best_cpu;
 
-			cpus_and(domain_mask, sd->span, *lowest_mask);
+			cpumask_and(&domain_mask, sched_domain_span(sd),
+				    lowest_mask);
 
 			best_cpu = pick_optimal_cpu(this_cpu,
 						    &domain_mask);
@@ -732,8 +1060,8 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 			 * Also make sure that it wasn't scheduled on its rq.
 			 */
 			if (unlikely(task_rq(task) != rq ||
-				     !cpu_isset(lowest_rq->cpu,
-						task->cpus_allowed) ||
+				     !cpumask_test_cpu(lowest_rq->cpu,
+						       &task->cpus_allowed) ||
 				     task_running(rq, task) ||
 				     !task->se.on_rq)) {
 
@@ -748,7 +1076,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 			break;
 
 		/* try again */
-		spin_unlock(&lowest_rq->lock);
+		double_unlock_balance(rq, lowest_rq);
 		lowest_rq = NULL;
 	}
 
@@ -817,7 +1145,7 @@ static int push_rt_task(struct rq *rq)
 
 	resched_task(lowest_rq->curr);
 
-	spin_unlock(&lowest_rq->lock);
+	double_unlock_balance(rq, lowest_rq);
 
 	ret = 1;
 out:
@@ -854,7 +1182,7 @@ static int pull_rt_task(struct rq *this_rq)
 
 	next = pick_next_task_rt(this_rq);
 
-	for_each_cpu_mask(cpu, this_rq->rd->rto_mask) {
+	for_each_cpu(cpu, this_rq->rd->rto_mask) {
 		if (this_cpu == cpu)
 			continue;
 
@@ -923,7 +1251,7 @@ static int pull_rt_task(struct rq *this_rq)
 
 		}
  skip:
-		spin_unlock(&src_rq->lock);
+		double_unlock_balance(this_rq, src_rq);
 	}
 
 	return ret;
@@ -952,11 +1280,14 @@ static void post_schedule_rt(struct rq *rq)
 	}
 }
 
-
+/*
+ * If we are not running and we are not going to reschedule soon, we should
+ * try to push tasks away now
+ */
 static void task_wake_up_rt(struct rq *rq, struct task_struct *p)
 {
 	if (!task_running(rq, p) &&
-	    (p->prio >= rq->rt.highest_prio) &&
+	    !test_tsk_need_resched(rq->curr) &&
 	    rq->rt.overloaded)
 		push_rt_tasks(rq);
 }
@@ -979,9 +1310,10 @@ move_one_task_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
 	return 0;
 }
 
-static void set_cpus_allowed_rt(struct task_struct *p, cpumask_t *new_mask)
+static void set_cpus_allowed_rt(struct task_struct *p,
+				const struct cpumask *new_mask)
 {
-	int weight = cpus_weight(*new_mask);
+	int weight = cpumask_weight(new_mask);
 
 	BUG_ON(!rt_task(p));
 
@@ -1002,22 +1334,30 @@ static void set_cpus_allowed_rt(struct task_struct *p, cpumask_t *new_mask)
 		update_rt_migration(rq);
 	}
 
-	p->cpus_allowed    = *new_mask;
+	cpumask_copy(&p->cpus_allowed, new_mask);
 	p->rt.nr_cpus_allowed = weight;
 }
 
 /* Assumes rq->lock is held */
-static void join_domain_rt(struct rq *rq)
+static void rq_online_rt(struct rq *rq)
 {
 	if (rq->rt.overloaded)
 		rt_set_overload(rq);
+
+	__enable_runtime(rq);
+
+	cpupri_set(&rq->rd->cpupri, rq->cpu, rq->rt.highest_prio);
 }
 
 /* Assumes rq->lock is held */
-static void leave_domain_rt(struct rq *rq)
+static void rq_offline_rt(struct rq *rq)
 {
 	if (rq->rt.overloaded)
 		rt_clear_overload(rq);
+
+	__disable_runtime(rq);
+
+	cpupri_set(&rq->rd->cpupri, rq->cpu, CPUPRI_INVALID);
 }
 
 /*
@@ -1037,6 +1377,14 @@ static void switched_from_rt(struct rq *rq, struct task_struct *p,
 	if (!rq->rt.rt_nr_running)
 		pull_rt_task(rq);
 }
+
+static inline void init_sched_rt_class(void)
+{
+	unsigned int i;
+
+	for_each_possible_cpu(i)
+		alloc_cpumask_var(&per_cpu(local_cpu_mask, i), GFP_KERNEL);
+}
 #endif /* CONFIG_SMP */
 
 /*
@@ -1085,9 +1433,11 @@ static void prio_changed_rt(struct rq *rq, struct task_struct *p,
 			pull_rt_task(rq);
 		/*
 		 * If there's a higher priority task waiting to run
-		 * then reschedule.
+		 * then reschedule. Note, the above pull_rt_task
+		 * can release the rq lock and p could migrate.
+		 * Only reschedule if p is still on the same runqueue.
 		 */
-		if (p->prio > rq->rt.highest_prio)
+		if (p->prio > rq->rt.highest_prio && rq->curr == p)
 			resched_task(p);
 #else
 		/* For UP simply resched on drop of prio */
@@ -1120,14 +1470,8 @@ static void watchdog(struct rq *rq, struct task_struct *p)
 
 		p->rt.timeout++;
 		next = DIV_ROUND_UP(min(soft, hard), USEC_PER_SEC/HZ);
-		if (next > p->rt.timeout) {
-			u64 next_time = p->se.sum_exec_runtime;
-
-			next_time += next * (NSEC_PER_SEC/HZ);
-			if (p->it_sched_expires > next_time)
-				p->it_sched_expires = next_time;
-		} else
-			p->it_sched_expires = p->se.sum_exec_runtime;
+		if (p->rt.timeout > next)
+			p->cputime_expires.sched_exp = p->se.sum_exec_runtime;
 	}
 }
 
@@ -1154,7 +1498,7 @@ static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued)
 	 * on the queue:
 	 */
 	if (p->rt.run_list.prev != p->rt.run_list.next) {
-		requeue_task_rt(rq, p);
+		requeue_task_rt(rq, p, 0);
 		set_tsk_need_resched(p);
 	}
 }
@@ -1166,14 +1510,11 @@ static void set_curr_task_rt(struct rq *rq)
 	p->se.exec_start = rq->clock;
 }
 
-const struct sched_class rt_sched_class = {
+static const struct sched_class rt_sched_class = {
 	.next			= &fair_sched_class,
 	.enqueue_task		= enqueue_task_rt,
 	.dequeue_task		= dequeue_task_rt,
 	.yield_task		= yield_task_rt,
-#ifdef CONFIG_SMP
-	.select_task_rq		= select_task_rq_rt,
-#endif /* CONFIG_SMP */
 
 	.check_preempt_curr	= check_preempt_curr_rt,
 
@@ -1181,11 +1522,13 @@ const struct sched_class rt_sched_class = {
 	.put_prev_task		= put_prev_task_rt,
 
 #ifdef CONFIG_SMP
+	.select_task_rq		= select_task_rq_rt,
+
 	.load_balance		= load_balance_rt,
 	.move_one_task		= move_one_task_rt,
 	.set_cpus_allowed       = set_cpus_allowed_rt,
-	.join_domain            = join_domain_rt,
-	.leave_domain           = leave_domain_rt,
+	.rq_online              = rq_online_rt,
+	.rq_offline             = rq_offline_rt,
 	.pre_schedule		= pre_schedule_rt,
 	.post_schedule		= post_schedule_rt,
 	.task_wake_up		= task_wake_up_rt,
@@ -1198,3 +1541,18 @@ const struct sched_class rt_sched_class = {
 	.prio_changed		= prio_changed_rt,
 	.switched_to		= switched_to_rt,
 };
+
+#ifdef CONFIG_SCHED_DEBUG
+extern void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq);
+
+static void print_rt_stats(struct seq_file *m, int cpu)
+{
+	struct rt_rq *rt_rq;
+
+	rcu_read_lock();
+	for_each_leaf_rt_rq(rt_rq, cpu_rq(cpu))
+		print_rt_rq(m, cpu, rt_rq);
+	rcu_read_unlock();
+}
+#endif /* CONFIG_SCHED_DEBUG */
+