X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=kernel%2Fsched_rt.c;h=8afb953e31c6c1ba9a263a78929a04a0d82ff2c7;hb=8fade6aff706b2ae3f02864b4023d34b002cd226;hp=4eda5f795f04323f994bcc44399f03cf28d50aaa;hpb=b07430ac37103218b5c1e542490a1b98e6deb3d6;p=safe%2Fjmp%2Flinux-2.6

diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 4eda5f7..8afb953 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -3,6 +3,52 @@
  * policies)
  */
 
+#ifdef CONFIG_RT_GROUP_SCHED
+
+#define rt_entity_is_task(rt_se) (!(rt_se)->my_q)
+
+static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se)
+{
+#ifdef CONFIG_SCHED_DEBUG
+	WARN_ON_ONCE(!rt_entity_is_task(rt_se));
+#endif
+	return container_of(rt_se, struct task_struct, rt);
+}
+
+static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq)
+{
+	return rt_rq->rq;
+}
+
+static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se)
+{
+	return rt_se->rt_rq;
+}
+
+#else /* CONFIG_RT_GROUP_SCHED */
+
+#define rt_entity_is_task(rt_se) (1)
+
+static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se)
+{
+	return container_of(rt_se, struct task_struct, rt);
+}
+
+static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq)
+{
+	return container_of(rt_rq, struct rq, rt);
+}
+
+static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se)
+{
+	struct task_struct *p = rt_task_of(rt_se);
+	struct rq *rq = task_rq(p);
+
+	return &rq->rt;
+}
+
+#endif /* CONFIG_RT_GROUP_SCHED */
+
 #ifdef CONFIG_SMP
 
 static inline int rt_overloaded(struct rq *rq)
@@ -37,19 +83,47 @@ static inline void rt_clear_overload(struct rq *rq)
 	cpumask_clear_cpu(rq->cpu, rq->rd->rto_mask);
 }
 
-static void update_rt_migration(struct rq *rq)
+static void update_rt_migration(struct rt_rq *rt_rq)
 {
-	if (rq->rt.rt_nr_migratory && (rq->rt.rt_nr_running > 1)) {
-		if (!rq->rt.overloaded) {
-			rt_set_overload(rq);
-			rq->rt.overloaded = 1;
+	if (rt_rq->rt_nr_migratory && rt_rq->rt_nr_total > 1) {
+		if (!rt_rq->overloaded) {
+			rt_set_overload(rq_of_rt_rq(rt_rq));
+			rt_rq->overloaded = 1;
 		}
-	} else if (rq->rt.overloaded) {
-		rt_clear_overload(rq);
-		rq->rt.overloaded = 0;
+	} else if (rt_rq->overloaded) {
+		rt_clear_overload(rq_of_rt_rq(rt_rq));
+		rt_rq->overloaded = 0;
 	}
 }
 
+static void inc_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+	if (!rt_entity_is_task(rt_se))
+		return;
+
+	rt_rq = &rq_of_rt_rq(rt_rq)->rt;
+
+	rt_rq->rt_nr_total++;
+	if (rt_se->nr_cpus_allowed > 1)
+		rt_rq->rt_nr_migratory++;
+
+	update_rt_migration(rt_rq);
+}
+
+static void dec_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+	if (!rt_entity_is_task(rt_se))
+		return;
+
+	rt_rq = &rq_of_rt_rq(rt_rq)->rt;
+
+	rt_rq->rt_nr_total--;
+	if (rt_se->nr_cpus_allowed > 1)
+		rt_rq->rt_nr_migratory--;
+
+	update_rt_migration(rt_rq);
+}
+
 static void enqueue_pushable_task(struct rq *rq, struct task_struct *p)
 {
 	plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks);
@@ -62,20 +136,33 @@ static void dequeue_pushable_task(struct rq *rq, struct task_struct *p)
 	plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks);
 }
 
+static inline int has_pushable_tasks(struct rq *rq)
+{
+	return !plist_head_empty(&rq->rt.pushable_tasks);
+}
+
 #else
 
-static inline
-void enqueue_pushable_task(struct rq *rq, struct task_struct *p) {}
-static inline
-void dequeue_pushable_task(struct rq *rq, struct task_struct *p) {}
+static inline void enqueue_pushable_task(struct rq *rq, struct task_struct *p)
+{
+}
 
-#endif /* CONFIG_SMP */
+static inline void dequeue_pushable_task(struct rq *rq, struct task_struct *p)
+{
+}
 
-static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se)
+static inline
+void inc_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 {
-	return container_of(rt_se, struct task_struct, rt);
 }
 
+static inline
+void dec_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+}
+
+#endif /* CONFIG_SMP */
+
 static inline int on_rt_rq(struct sched_rt_entity *rt_se)
 {
 	return !list_empty(&rt_se->run_list);
@@ -99,16 +186,6 @@ static inline u64 sched_rt_period(struct rt_rq *rt_rq)
 #define for_each_leaf_rt_rq(rt_rq, rq) \
 	list_for_each_entry_rcu(rt_rq, &rq->leaf_rt_rq_list, leaf_rt_rq_list)
 
-static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq)
-{
-	return rt_rq->rq;
-}
-
-static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se)
-{
-	return rt_se->rt_rq;
-}
-
 #define for_each_sched_rt_entity(rt_se) \
 	for (; rt_se; rt_se = rt_se->parent)
 
@@ -117,17 +194,20 @@ static inline struct rt_rq *group_rt_rq(struct sched_rt_entity *rt_se)
 	return rt_se->my_q;
 }
 
-static void enqueue_rt_entity(struct sched_rt_entity *rt_se);
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se, bool head);
 static void dequeue_rt_entity(struct sched_rt_entity *rt_se);
 
 static void sched_rt_rq_enqueue(struct rt_rq *rt_rq)
 {
+	int this_cpu = smp_processor_id();
 	struct task_struct *curr = rq_of_rt_rq(rt_rq)->curr;
-	struct sched_rt_entity *rt_se = rt_rq->rt_se;
+	struct sched_rt_entity *rt_se;
+
+	rt_se = rt_rq->tg->rt_se[this_cpu];
 
 	if (rt_rq->rt_nr_running) {
 		if (rt_se && !on_rt_rq(rt_se))
-			enqueue_rt_entity(rt_se);
+			enqueue_rt_entity(rt_se, false);
 		if (rt_rq->highest_prio.curr < curr->prio)
 			resched_task(curr);
 	}
@@ -135,7 +215,10 @@ static void sched_rt_rq_enqueue(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;
+	int this_cpu = smp_processor_id();
+	struct sched_rt_entity *rt_se;
+
+	rt_se = rt_rq->tg->rt_se[this_cpu];
 
 	if (rt_se && on_rt_rq(rt_se))
 		dequeue_rt_entity(rt_se);
@@ -196,19 +279,6 @@ static inline u64 sched_rt_period(struct rt_rq *rt_rq)
 #define for_each_leaf_rt_rq(rt_rq, rq) \
 	for (rt_rq = &rq->rt; rt_rq; rt_rq = NULL)
 
-static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq)
-{
-	return container_of(rt_rq, struct rq, rt);
-}
-
-static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se)
-{
-	struct task_struct *p = rt_task_of(rt_se);
-	struct rq *rq = task_rq(p);
-
-	return &rq->rt;
-}
-
 #define for_each_sched_rt_entity(rt_se) \
 	for (; rt_se; rt_se = NULL)
 
@@ -263,7 +333,7 @@ static int do_balance_runtime(struct rt_rq *rt_rq)
 
 	weight = cpumask_weight(rd->span);
 
-	spin_lock(&rt_b->rt_runtime_lock);
+	raw_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);
@@ -272,7 +342,7 @@ static int do_balance_runtime(struct rt_rq *rt_rq)
 		if (iter == rt_rq)
 			continue;
 
-		spin_lock(&iter->rt_runtime_lock);
+		raw_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
@@ -294,14 +364,14 @@ static int do_balance_runtime(struct rt_rq *rt_rq)
 			rt_rq->rt_runtime += diff;
 			more = 1;
 			if (rt_rq->rt_runtime == rt_period) {
-				spin_unlock(&iter->rt_runtime_lock);
+				raw_spin_unlock(&iter->rt_runtime_lock);
 				break;
 			}
 		}
 next:
-		spin_unlock(&iter->rt_runtime_lock);
+		raw_spin_unlock(&iter->rt_runtime_lock);
 	}
-	spin_unlock(&rt_b->rt_runtime_lock);
+	raw_spin_unlock(&rt_b->rt_runtime_lock);
 
 	return more;
 }
@@ -322,8 +392,8 @@ static void __disable_runtime(struct rq *rq)
 		s64 want;
 		int i;
 
-		spin_lock(&rt_b->rt_runtime_lock);
-		spin_lock(&rt_rq->rt_runtime_lock);
+		raw_spin_lock(&rt_b->rt_runtime_lock);
+		raw_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
@@ -332,7 +402,7 @@ static void __disable_runtime(struct rq *rq)
 		if (rt_rq->rt_runtime == RUNTIME_INF ||
 				rt_rq->rt_runtime == rt_b->rt_runtime)
 			goto balanced;
-		spin_unlock(&rt_rq->rt_runtime_lock);
+		raw_spin_unlock(&rt_rq->rt_runtime_lock);
 
 		/*
 		 * Calculate the difference between what we started out with
@@ -354,7 +424,7 @@ static void __disable_runtime(struct rq *rq)
 			if (iter == rt_rq || iter->rt_runtime == RUNTIME_INF)
 				continue;
 
-			spin_lock(&iter->rt_runtime_lock);
+			raw_spin_lock(&iter->rt_runtime_lock);
 			if (want > 0) {
 				diff = min_t(s64, iter->rt_runtime, want);
 				iter->rt_runtime -= diff;
@@ -363,13 +433,13 @@ static void __disable_runtime(struct rq *rq)
 				iter->rt_runtime -= want;
 				want -= want;
 			}
-			spin_unlock(&iter->rt_runtime_lock);
+			raw_spin_unlock(&iter->rt_runtime_lock);
 
 			if (!want)
 				break;
 		}
 
-		spin_lock(&rt_rq->rt_runtime_lock);
+		raw_spin_lock(&rt_rq->rt_runtime_lock);
 		/*
 		 * We cannot be left wanting - that would mean some runtime
 		 * leaked out of the system.
@@ -381,8 +451,8 @@ balanced:
 		 * 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);
+		raw_spin_unlock(&rt_rq->rt_runtime_lock);
+		raw_spin_unlock(&rt_b->rt_runtime_lock);
 	}
 }
 
@@ -390,9 +460,9 @@ static void disable_runtime(struct rq *rq)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&rq->lock, flags);
+	raw_spin_lock_irqsave(&rq->lock, flags);
 	__disable_runtime(rq);
-	spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_unlock_irqrestore(&rq->lock, flags);
 }
 
 static void __enable_runtime(struct rq *rq)
@@ -408,13 +478,13 @@ static void __enable_runtime(struct rq *rq)
 	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);
+		raw_spin_lock(&rt_b->rt_runtime_lock);
+		raw_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);
+		raw_spin_unlock(&rt_rq->rt_runtime_lock);
+		raw_spin_unlock(&rt_b->rt_runtime_lock);
 	}
 }
 
@@ -422,9 +492,9 @@ static void enable_runtime(struct rq *rq)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&rq->lock, flags);
+	raw_spin_lock_irqsave(&rq->lock, flags);
 	__enable_runtime(rq);
-	spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_unlock_irqrestore(&rq->lock, flags);
 }
 
 static int balance_runtime(struct rt_rq *rt_rq)
@@ -432,9 +502,9 @@ 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);
+		raw_spin_unlock(&rt_rq->rt_runtime_lock);
 		more = do_balance_runtime(rt_rq);
-		spin_lock(&rt_rq->rt_runtime_lock);
+		raw_spin_lock(&rt_rq->rt_runtime_lock);
 	}
 
 	return more;
@@ -460,11 +530,11 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
 		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);
+		raw_spin_lock(&rq->lock);
 		if (rt_rq->rt_time) {
 			u64 runtime;
 
-			spin_lock(&rt_rq->rt_runtime_lock);
+			raw_spin_lock(&rt_rq->rt_runtime_lock);
 			if (rt_rq->rt_throttled)
 				balance_runtime(rt_rq);
 			runtime = rt_rq->rt_runtime;
@@ -475,13 +545,13 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
 			}
 			if (rt_rq->rt_time || rt_rq->rt_nr_running)
 				idle = 0;
-			spin_unlock(&rt_rq->rt_runtime_lock);
+			raw_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);
+		raw_spin_unlock(&rq->lock);
 	}
 
 	return idle;
@@ -543,7 +613,7 @@ static void update_curr_rt(struct rq *rq)
 	if (unlikely((s64)delta_exec < 0))
 		delta_exec = 0;
 
-	schedstat_set(curr->se.exec_max, max(curr->se.exec_max, delta_exec));
+	schedstat_set(curr->se.statistics.exec_max, max(curr->se.statistics.exec_max, delta_exec));
 
 	curr->se.sum_exec_runtime += delta_exec;
 	account_group_exec_runtime(curr, delta_exec);
@@ -551,6 +621,8 @@ static void update_curr_rt(struct rq *rq)
 	curr->se.exec_start = rq->clock;
 	cpuacct_charge(curr, delta_exec);
 
+	sched_rt_avg_update(rq, delta_exec);
+
 	if (!rt_bandwidth_enabled())
 		return;
 
@@ -558,16 +630,16 @@ static void update_curr_rt(struct rq *rq)
 		rt_rq = rt_rq_of_se(rt_se);
 
 		if (sched_rt_runtime(rt_rq) != RUNTIME_INF) {
-			spin_lock(&rt_rq->rt_runtime_lock);
+			raw_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);
+			raw_spin_unlock(&rt_rq->rt_runtime_lock);
 		}
 	}
 }
 
-#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED
+#if defined CONFIG_SMP
 
 static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu);
 
@@ -580,33 +652,24 @@ static inline int next_prio(struct rq *rq)
 	else
 		return MAX_RT_PRIO;
 }
-#endif
 
-static inline
-void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+static void
+inc_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio)
 {
-	int prio = rt_se_prio(rt_se);
-#ifdef CONFIG_SMP
 	struct rq *rq = rq_of_rt_rq(rt_rq);
-#endif
 
-	WARN_ON(!rt_prio(prio));
-	rt_rq->rt_nr_running++;
-#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED
-	if (prio < rt_rq->highest_prio.curr) {
+	if (prio < prev_prio) {
 
 		/*
 		 * If the new task is higher in priority than anything on the
-		 * run-queue, we have a new high that must be published to
-		 * the world.  We also know that the previous high becomes
-		 * our next-highest.
+		 * run-queue, we know that the previous high becomes our
+		 * next-highest.
 		 */
-		rt_rq->highest_prio.next = rt_rq->highest_prio.curr;
-		rt_rq->highest_prio.curr = prio;
-#ifdef CONFIG_SMP
+		rt_rq->highest_prio.next = prev_prio;
+
 		if (rq->online)
 			cpupri_set(&rq->rd->cpupri, rq->cpu, prio);
-#endif
+
 	} else if (prio == rt_rq->highest_prio.curr)
 		/*
 		 * If the next task is equal in priority to the highest on
@@ -619,75 +682,134 @@ void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 		 * Otherwise, we need to recompute next-highest
 		 */
 		rt_rq->highest_prio.next = next_prio(rq);
-#endif
-#ifdef CONFIG_SMP
-	if (rt_se->nr_cpus_allowed > 1)
-		rq->rt.rt_nr_migratory++;
+}
 
-	update_rt_migration(rq);
-#endif
-#ifdef CONFIG_RT_GROUP_SCHED
-	if (rt_se_boosted(rt_se))
-		rt_rq->rt_nr_boosted++;
+static void
+dec_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio)
+{
+	struct rq *rq = rq_of_rt_rq(rt_rq);
 
-	if (rt_rq->tg)
-		start_rt_bandwidth(&rt_rq->tg->rt_bandwidth);
-#else
-	start_rt_bandwidth(&def_rt_bandwidth);
-#endif
+	if (rt_rq->rt_nr_running && (prio <= rt_rq->highest_prio.next))
+		rt_rq->highest_prio.next = next_prio(rq);
+
+	if (rq->online && rt_rq->highest_prio.curr != prev_prio)
+		cpupri_set(&rq->rd->cpupri, rq->cpu, rt_rq->highest_prio.curr);
 }
 
+#else /* CONFIG_SMP */
+
 static inline
-void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
-{
-#ifdef CONFIG_SMP
-	struct rq *rq = rq_of_rt_rq(rt_rq);
-	int highest_prio = rt_rq->highest_prio.curr;
-#endif
+void inc_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) {}
+static inline
+void dec_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) {}
+
+#endif /* CONFIG_SMP */
 
-	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_RT_GROUP_SCHED
+static void
+inc_rt_prio(struct rt_rq *rt_rq, int prio)
+{
+	int prev_prio = rt_rq->highest_prio.curr;
+
+	if (prio < prev_prio)
+		rt_rq->highest_prio.curr = prio;
+
+	inc_rt_prio_smp(rt_rq, prio, prev_prio);
+}
+
+static void
+dec_rt_prio(struct rt_rq *rt_rq, int prio)
+{
+	int prev_prio = rt_rq->highest_prio.curr;
+
 	if (rt_rq->rt_nr_running) {
-		int prio = rt_se_prio(rt_se);
 
-		WARN_ON(prio < rt_rq->highest_prio.curr);
+		WARN_ON(prio < prev_prio);
 
 		/*
-		 * This may have been our highest or next-highest priority
-		 * task and therefore we may have some recomputation to do
+		 * This may have been our highest task, and therefore
+		 * we may have some recomputation to do
 		 */
-		if (prio == rt_rq->highest_prio.curr) {
+		if (prio == prev_prio) {
 			struct rt_prio_array *array = &rt_rq->active;
 
 			rt_rq->highest_prio.curr =
 				sched_find_first_bit(array->bitmap);
 		}
 
-		if (prio <= rt_rq->highest_prio.next)
-			rt_rq->highest_prio.next = next_prio(rq);
 	} else
 		rt_rq->highest_prio.curr = MAX_RT_PRIO;
-#endif
-#ifdef CONFIG_SMP
-	if (rt_se->nr_cpus_allowed > 1)
-		rq->rt.rt_nr_migratory--;
 
-	if (rq->online && rt_rq->highest_prio.curr != highest_prio)
-		cpupri_set(&rq->rd->cpupri, rq->cpu, rt_rq->highest_prio.curr);
+	dec_rt_prio_smp(rt_rq, prio, prev_prio);
+}
+
+#else
+
+static inline void inc_rt_prio(struct rt_rq *rt_rq, int prio) {}
+static inline void dec_rt_prio(struct rt_rq *rt_rq, int prio) {}
+
+#endif /* CONFIG_SMP || CONFIG_RT_GROUP_SCHED */
 
-	update_rt_migration(rq);
-#endif /* CONFIG_SMP */
 #ifdef CONFIG_RT_GROUP_SCHED
+
+static void
+inc_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+	if (rt_se_boosted(rt_se))
+		rt_rq->rt_nr_boosted++;
+
+	if (rt_rq->tg)
+		start_rt_bandwidth(&rt_rq->tg->rt_bandwidth);
+}
+
+static void
+dec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
 	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)
+#else /* CONFIG_RT_GROUP_SCHED */
+
+static void
+inc_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+	start_rt_bandwidth(&def_rt_bandwidth);
+}
+
+static inline
+void dec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) {}
+
+#endif /* CONFIG_RT_GROUP_SCHED */
+
+static inline
+void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+	int prio = rt_se_prio(rt_se);
+
+	WARN_ON(!rt_prio(prio));
+	rt_rq->rt_nr_running++;
+
+	inc_rt_prio(rt_rq, prio);
+	inc_rt_migration(rt_se, rt_rq);
+	inc_rt_group(rt_se, rt_rq);
+}
+
+static inline
+void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+	WARN_ON(!rt_prio(rt_se_prio(rt_se)));
+	WARN_ON(!rt_rq->rt_nr_running);
+	rt_rq->rt_nr_running--;
+
+	dec_rt_prio(rt_rq, rt_se_prio(rt_se));
+	dec_rt_migration(rt_se, rt_rq);
+	dec_rt_group(rt_se, rt_rq);
+}
+
+static void __enqueue_rt_entity(struct sched_rt_entity *rt_se, bool head)
 {
 	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
 	struct rt_prio_array *array = &rt_rq->active;
@@ -703,7 +825,10 @@ static void __enqueue_rt_entity(struct sched_rt_entity *rt_se)
 	if (group_rq && (rt_rq_throttled(group_rq) || !group_rq->rt_nr_running))
 		return;
 
-	list_add_tail(&rt_se->run_list, queue);
+	if (head)
+		list_add(&rt_se->run_list, queue);
+	else
+		list_add_tail(&rt_se->run_list, queue);
 	__set_bit(rt_se_prio(rt_se), array->bitmap);
 
 	inc_rt_tasks(rt_se, rt_rq);
@@ -740,11 +865,11 @@ static void dequeue_rt_stack(struct sched_rt_entity *rt_se)
 	}
 }
 
-static void enqueue_rt_entity(struct sched_rt_entity *rt_se)
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se, bool head)
 {
 	dequeue_rt_stack(rt_se);
 	for_each_sched_rt_entity(rt_se)
-		__enqueue_rt_entity(rt_se);
+		__enqueue_rt_entity(rt_se, head);
 }
 
 static void dequeue_rt_entity(struct sched_rt_entity *rt_se)
@@ -755,29 +880,28 @@ static void dequeue_rt_entity(struct 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);
+			__enqueue_rt_entity(rt_se, false);
 	}
 }
 
 /*
  * Adding/removing a task to/from a priority array:
  */
-static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
+static void
+enqueue_task_rt(struct rq *rq, struct task_struct *p, int flags)
 {
 	struct sched_rt_entity *rt_se = &p->rt;
 
-	if (wakeup)
+	if (flags & ENQUEUE_WAKEUP)
 		rt_se->timeout = 0;
 
-	enqueue_rt_entity(rt_se);
+	enqueue_rt_entity(rt_se, flags & ENQUEUE_HEAD);
 
 	if (!task_current(rq, p) && p->rt.nr_cpus_allowed > 1)
 		enqueue_pushable_task(rq, p);
-
-	inc_cpu_load(rq, p->se.load.weight);
 }
 
-static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
+static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int flags)
 {
 	struct sched_rt_entity *rt_se = &p->rt;
 
@@ -785,8 +909,6 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
 	dequeue_rt_entity(rt_se);
 
 	dequeue_pushable_task(rq, p);
-
-	dec_cpu_load(rq, p->se.load.weight);
 }
 
 /*
@@ -826,9 +948,11 @@ static void yield_task_rt(struct rq *rq)
 #ifdef CONFIG_SMP
 static int find_lowest_rq(struct task_struct *task);
 
-static int select_task_rq_rt(struct task_struct *p, int sync)
+static int
+select_task_rq_rt(struct rq *rq, struct task_struct *p, int sd_flag, int flags)
 {
-	struct rq *rq = task_rq(p);
+	if (sd_flag != SD_BALANCE_WAKE)
+		return smp_processor_id();
 
 	/*
 	 * If the current task is an RT task, then
@@ -863,20 +987,15 @@ static int select_task_rq_rt(struct task_struct *p, int sync)
 
 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;
+	    && cpupri_find(&rq->rd->cpupri, p, NULL))
+		return;
 
-	if (!cpupri_find(&rq->rd->cpupri, rq->curr, mask))
-		goto free;
+	if (!cpupri_find(&rq->rd->cpupri, rq->curr, NULL))
+		return;
 
 	/*
 	 * There appears to be other cpus that can accept
@@ -885,8 +1004,6 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p)
 	 */
 	requeue_task_rt(rq, p, 1);
 	resched_task(rq->curr);
-free:
-	free_cpumask_var(mask);
 }
 
 #endif /* CONFIG_SMP */
@@ -894,7 +1011,7 @@ free:
 /*
  * Preempt the current task with a newly woken task if needed:
  */
-static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int sync)
+static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int flags)
 {
 	if (p->prio < rq->curr->prio) {
 		resched_task(rq->curr);
@@ -970,6 +1087,14 @@ static struct task_struct *pick_next_task_rt(struct rq *rq)
 	if (p)
 		dequeue_pushable_task(rq, p);
 
+#ifdef CONFIG_SMP
+	/*
+	 * We detect this state here so that we can avoid taking the RQ
+	 * lock again later if there is no need to push
+	 */
+	rq->post_schedule = has_pushable_tasks(rq);
+#endif
+
 	return p;
 }
 
@@ -1020,7 +1145,12 @@ static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu)
 		if (next && next->prio < idx)
 			continue;
 		list_for_each_entry(rt_se, array->queue + idx, run_list) {
-			struct task_struct *p = rt_task_of(rt_se);
+			struct task_struct *p;
+
+			if (!rt_entity_is_task(rt_se))
+				continue;
+
+			p = rt_task_of(rt_se);
 			if (pick_rt_task(rq, p, cpu)) {
 				next = p;
 				break;
@@ -1037,21 +1167,6 @@ static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu)
 
 static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask);
 
-static inline int pick_optimal_cpu(int this_cpu, cpumask_t *mask)
-{
-	int first;
-
-	/* "this_cpu" is cheaper to preempt than a remote processor */
-	if ((this_cpu != -1) && cpu_isset(this_cpu, *mask))
-		return this_cpu;
-
-	first = first_cpu(*mask);
-	if (first != NR_CPUS)
-		return first;
-
-	return -1;
-}
-
 static int find_lowest_rq(struct task_struct *task)
 {
 	struct sched_domain *sd;
@@ -1066,13 +1181,6 @@ static int find_lowest_rq(struct task_struct *task)
 		return -1; /* No targets 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.
-	 */
-	cpumask_and(lowest_mask, lowest_mask, cpu_active_mask);
-
-	/*
 	 * At this point we have built a mask of cpus representing the
 	 * lowest priority tasks in the system.  Now we want to elect
 	 * the best one based on our affinity and topology.
@@ -1087,20 +1195,24 @@ static int find_lowest_rq(struct task_struct *task)
 	 * Otherwise, we consult the sched_domains span maps to figure
 	 * out which cpu is logically closest to our hot cache data.
 	 */
-	if (this_cpu == cpu)
-		this_cpu = -1; /* Skip this_cpu opt if the same */
+	if (!cpumask_test_cpu(this_cpu, lowest_mask))
+		this_cpu = -1; /* Skip this_cpu opt if not among lowest */
 
 	for_each_domain(cpu, sd) {
 		if (sd->flags & SD_WAKE_AFFINE) {
-			cpumask_t domain_mask;
-			int       best_cpu;
+			int best_cpu;
 
-			cpumask_and(&domain_mask, sched_domain_span(sd),
-				    lowest_mask);
+			/*
+			 * "this_cpu" is cheaper to preempt than a
+			 * remote processor.
+			 */
+			if (this_cpu != -1 &&
+			    cpumask_test_cpu(this_cpu, sched_domain_span(sd)))
+				return this_cpu;
 
-			best_cpu = pick_optimal_cpu(this_cpu,
-						    &domain_mask);
-			if (best_cpu != -1)
+			best_cpu = cpumask_first_and(lowest_mask,
+						     sched_domain_span(sd));
+			if (best_cpu < nr_cpu_ids)
 				return best_cpu;
 		}
 	}
@@ -1110,7 +1222,13 @@ static int find_lowest_rq(struct task_struct *task)
 	 * just give the caller *something* to work with from the compatible
 	 * locations.
 	 */
-	return pick_optimal_cpu(this_cpu, lowest_mask);
+	if (this_cpu != -1)
+		return this_cpu;
+
+	cpu = cpumask_any(lowest_mask);
+	if (cpu < nr_cpu_ids)
+		return cpu;
+	return -1;
 }
 
 /* Will lock the rq it finds */
@@ -1142,7 +1260,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 				     task_running(rq, task) ||
 				     !task->se.on_rq)) {
 
-				spin_unlock(&lowest_rq->lock);
+				raw_spin_unlock(&lowest_rq->lock);
 				lowest_rq = NULL;
 				break;
 			}
@@ -1160,11 +1278,6 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 	return lowest_rq;
 }
 
-static inline int has_pushable_tasks(struct rq *rq)
-{
-	return !plist_head_empty(&rq->rt.pushable_tasks);
-}
-
 static struct task_struct *pick_next_pushable_task(struct rq *rq)
 {
 	struct task_struct *p;
@@ -1364,30 +1477,16 @@ static void pre_schedule_rt(struct rq *rq, struct task_struct *prev)
 		pull_rt_task(rq);
 }
 
-/*
- * assumes rq->lock is held
- */
-static int needs_post_schedule_rt(struct rq *rq)
-{
-	return has_pushable_tasks(rq);
-}
-
 static void post_schedule_rt(struct rq *rq)
 {
-	/*
-	 * This is only called if needs_post_schedule_rt() indicates that
-	 * we need to push tasks away
-	 */
-	spin_lock_irq(&rq->lock);
 	push_rt_tasks(rq);
-	spin_unlock_irq(&rq->lock);
 }
 
 /*
  * 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)
+static void task_woken_rt(struct rq *rq, struct task_struct *p)
 {
 	if (!task_running(rq, p) &&
 	    !test_tsk_need_resched(rq->curr) &&
@@ -1396,24 +1495,6 @@ static void task_wake_up_rt(struct rq *rq, struct task_struct *p)
 		push_rt_tasks(rq);
 }
 
-static unsigned long
-load_balance_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		unsigned long max_load_move,
-		struct sched_domain *sd, enum cpu_idle_type idle,
-		int *all_pinned, int *this_best_prio)
-{
-	/* don't touch RT tasks */
-	return 0;
-}
-
-static int
-move_one_task_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
-		 struct sched_domain *sd, enum cpu_idle_type idle)
-{
-	/* don't touch RT tasks */
-	return 0;
-}
-
 static void set_cpus_allowed_rt(struct task_struct *p,
 				const struct cpumask *new_mask)
 {
@@ -1453,7 +1534,7 @@ static void set_cpus_allowed_rt(struct task_struct *p,
 			rq->rt.rt_nr_migratory--;
 		}
 
-		update_rt_migration(rq);
+		update_rt_migration(&rq->rt);
 	}
 
 	cpumask_copy(&p->cpus_allowed, new_mask);
@@ -1505,7 +1586,7 @@ static inline void init_sched_rt_class(void)
 	unsigned int i;
 
 	for_each_possible_cpu(i)
-		alloc_cpumask_var_node(&per_cpu(local_cpu_mask, i),
+		zalloc_cpumask_var_node(&per_cpu(local_cpu_mask, i),
 					GFP_KERNEL, cpu_to_node(i));
 }
 #endif /* CONFIG_SMP */
@@ -1585,8 +1666,9 @@ static void watchdog(struct rq *rq, struct task_struct *p)
 	if (!p->signal)
 		return;
 
-	soft = p->signal->rlim[RLIMIT_RTTIME].rlim_cur;
-	hard = p->signal->rlim[RLIMIT_RTTIME].rlim_max;
+	/* max may change after cur was read, this will be fixed next tick */
+	soft = task_rlimit(p, RLIMIT_RTTIME);
+	hard = task_rlimit_max(p, RLIMIT_RTTIME);
 
 	if (soft != RLIM_INFINITY) {
 		unsigned long next;
@@ -1636,6 +1718,17 @@ static void set_curr_task_rt(struct rq *rq)
 	dequeue_pushable_task(rq, p);
 }
 
+static unsigned int get_rr_interval_rt(struct rq *rq, struct task_struct *task)
+{
+	/*
+	 * Time slice is 0 for SCHED_FIFO tasks
+	 */
+	if (task->policy == SCHED_RR)
+		return DEF_TIMESLICE;
+	else
+		return 0;
+}
+
 static const struct sched_class rt_sched_class = {
 	.next			= &fair_sched_class,
 	.enqueue_task		= enqueue_task_rt,
@@ -1650,21 +1743,20 @@ static const struct sched_class rt_sched_class = {
 #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,
 	.rq_online              = rq_online_rt,
 	.rq_offline             = rq_offline_rt,
 	.pre_schedule		= pre_schedule_rt,
-	.needs_post_schedule	= needs_post_schedule_rt,
 	.post_schedule		= post_schedule_rt,
-	.task_wake_up		= task_wake_up_rt,
+	.task_woken		= task_woken_rt,
 	.switched_from		= switched_from_rt,
 #endif
 
 	.set_curr_task          = set_curr_task_rt,
 	.task_tick		= task_tick_rt,
 
+	.get_rr_interval	= get_rr_interval_rt,
+
 	.prio_changed		= prio_changed_rt,
 	.switched_to		= switched_to_rt,
 };