Merge branch 'sched/core'

10 files changed, 318 insertions, 199 deletions

diff --git a/include/linux/wait.h b/include/linux/wait.h
index 6fb1ba5f9b2f4..034f6fc7c65f8 100644
--- a/include/linux/wait.h
+++ b/include/linux/wait.h
@@ -280,9 +280,11 @@ do {									\
  * wake_up() has to be called after changing any variable that could
  * change the result of the wait condition.
  *
- * The function returns 0 if the @timeout elapsed, or the remaining
- * jiffies (at least 1) if the @condition evaluated to %true before
- * the @timeout elapsed.
+ * Returns:
+ * 0 if the @condition evaluated to %false after the @timeout elapsed,
+ * 1 if the @condition evaluated to %true after the @timeout elapsed,
+ * or the remaining jiffies (at least 1) if the @condition evaluated
+ * to %true before the @timeout elapsed.
  */
 #define wait_event_timeout(wq, condition, timeout)			\
 ({									\
@@ -363,9 +365,11 @@ do {									\
  * change the result of the wait condition.
  *
  * Returns:
- * 0 if the @timeout elapsed, -%ERESTARTSYS if it was interrupted by
- * a signal, or the remaining jiffies (at least 1) if the @condition
- * evaluated to %true before the @timeout elapsed.
+ * 0 if the @condition evaluated to %false after the @timeout elapsed,
+ * 1 if the @condition evaluated to %true after the @timeout elapsed,
+ * the remaining jiffies (at least 1) if the @condition evaluated
+ * to %true before the @timeout elapsed, or -%ERESTARTSYS if it was
+ * interrupted by a signal.
  */
 #define wait_event_interruptible_timeout(wq, condition, timeout)	\
 ({									\
diff --git a/kernel/sched/auto_group.c b/kernel/sched/auto_group.c
index e73efba98301f..8a2e230fb86ad 100644
--- a/kernel/sched/auto_group.c
+++ b/kernel/sched/auto_group.c
@@ -148,11 +148,8 @@ autogroup_move_group(struct task_struct *p, struct autogroup *ag)
 	if (!ACCESS_ONCE(sysctl_sched_autogroup_enabled))
 		goto out;
 
-	t = p;
-	do {
+	for_each_thread(p, t)
 		sched_move_task(t);
-	} while_each_thread(p, t);
-
 out:
 	unlock_task_sighand(p, &flags);
 	autogroup_kref_put(prev);
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index c5a0c93854aa7..53985a609963a 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -317,9 +317,12 @@ static inline struct rq *__task_rq_lock(struct task_struct *p)
 	for (;;) {
 		rq = task_rq(p);
 		raw_spin_lock(&rq->lock);
-		if (likely(rq == task_rq(p)))
+		if (likely(rq == task_rq(p) && !task_on_rq_migrating(p)))
 			return rq;
 		raw_spin_unlock(&rq->lock);
+
+		while (unlikely(task_on_rq_migrating(p)))
+			cpu_relax();
 	}
 }
 
@@ -336,10 +339,13 @@ static struct rq *task_rq_lock(struct task_struct *p, unsigned long *flags)
 		raw_spin_lock_irqsave(&p->pi_lock, *flags);
 		rq = task_rq(p);
 		raw_spin_lock(&rq->lock);
-		if (likely(rq == task_rq(p)))
+		if (likely(rq == task_rq(p) && !task_on_rq_migrating(p)))
 			return rq;
 		raw_spin_unlock(&rq->lock);
 		raw_spin_unlock_irqrestore(&p->pi_lock, *flags);
+
+		while (unlikely(task_on_rq_migrating(p)))
+			cpu_relax();
 	}
 }
 
@@ -433,7 +439,15 @@ static void __hrtick_start(void *arg)
 void hrtick_start(struct rq *rq, u64 delay)
 {
 	struct hrtimer *timer = &rq->hrtick_timer;
-	ktime_t time = ktime_add_ns(timer->base->get_time(), delay);
+	ktime_t time;
+	s64 delta;
+
+	/*
+	 * Don't schedule slices shorter than 10000ns, that just
+	 * doesn't make sense and can cause timer DoS.
+	 */
+	delta = max_t(s64, delay, 10000LL);
+	time = ktime_add_ns(timer->base->get_time(), delta);
 
 	hrtimer_set_expires(timer, time);
 
@@ -1027,7 +1041,7 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags)
 	 * A queue event has occurred, and we're going to schedule.  In
 	 * this case, we can save a useless back to back clock update.
 	 */
-	if (rq->curr->on_rq && test_tsk_need_resched(rq->curr))
+	if (task_on_rq_queued(rq->curr) && test_tsk_need_resched(rq->curr))
 		rq->skip_clock_update = 1;
 }
 
@@ -1072,7 +1086,7 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
 
 static void __migrate_swap_task(struct task_struct *p, int cpu)
 {
-	if (p->on_rq) {
+	if (task_on_rq_queued(p)) {
 		struct rq *src_rq, *dst_rq;
 
 		src_rq = task_rq(p);
@@ -1198,7 +1212,7 @@ static int migration_cpu_stop(void *data);
 unsigned long wait_task_inactive(struct task_struct *p, long match_state)
 {
 	unsigned long flags;
-	int running, on_rq;
+	int running, queued;
 	unsigned long ncsw;
 	struct rq *rq;
 
@@ -1236,7 +1250,7 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state)
 		rq = task_rq_lock(p, &flags);
 		trace_sched_wait_task(p);
 		running = task_running(rq, p);
-		on_rq = p->on_rq;
+		queued = task_on_rq_queued(p);
 		ncsw = 0;
 		if (!match_state || p->state == match_state)
 			ncsw = p->nvcsw | LONG_MIN; /* sets MSB */
@@ -1268,7 +1282,7 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state)
 		 * running right now), it's preempted, and we should
 		 * yield - it could be a while.
 		 */
-		if (unlikely(on_rq)) {
+		if (unlikely(queued)) {
 			ktime_t to = ktime_set(0, NSEC_PER_SEC/HZ);
 
 			set_current_state(TASK_UNINTERRUPTIBLE);
@@ -1462,7 +1476,7 @@ ttwu_stat(struct task_struct *p, int cpu, int wake_flags)
 static void ttwu_activate(struct rq *rq, struct task_struct *p, int en_flags)
 {
 	activate_task(rq, p, en_flags);
-	p->on_rq = 1;
+	p->on_rq = TASK_ON_RQ_QUEUED;
 
 	/* if a worker is waking up, notify workqueue */
 	if (p->flags & PF_WQ_WORKER)
@@ -1521,7 +1535,7 @@ static int ttwu_remote(struct task_struct *p, int wake_flags)
 	int ret = 0;
 
 	rq = __task_rq_lock(p);
-	if (p->on_rq) {
+	if (task_on_rq_queued(p)) {
 		/* check_preempt_curr() may use rq clock */
 		update_rq_clock(rq);
 		ttwu_do_wakeup(rq, p, wake_flags);
@@ -1726,7 +1740,7 @@ static void try_to_wake_up_local(struct task_struct *p)
 	if (!(p->state & TASK_NORMAL))
 		goto out;
 
-	if (!p->on_rq)
+	if (!task_on_rq_queued(p))
 		ttwu_activate(rq, p, ENQUEUE_WAKEUP);
 
 	ttwu_do_wakeup(rq, p, 0);
@@ -2079,7 +2093,7 @@ void wake_up_new_task(struct task_struct *p)
 	init_task_runnable_average(p);
 	rq = __task_rq_lock(p);
 	activate_task(rq, p, 0);
-	p->on_rq = 1;
+	p->on_rq = TASK_ON_RQ_QUEUED;
 	trace_sched_wakeup_new(p, true);
 	check_preempt_curr(rq, p, WF_FORK);
 #ifdef CONFIG_SMP
@@ -2435,7 +2449,7 @@ static u64 do_task_delta_exec(struct task_struct *p, struct rq *rq)
 	 * project cycles that may never be accounted to this
 	 * thread, breaking clock_gettime().
 	 */
-	if (task_current(rq, p) && p->on_rq) {
+	if (task_current(rq, p) && task_on_rq_queued(p)) {
 		update_rq_clock(rq);
 		ns = rq_clock_task(rq) - p->se.exec_start;
 		if ((s64)ns < 0)
@@ -2481,7 +2495,7 @@ unsigned long long task_sched_runtime(struct task_struct *p)
 	 * If we see ->on_cpu without ->on_rq, the task is leaving, and has
 	 * been accounted, so we're correct here as well.
 	 */
-	if (!p->on_cpu || !p->on_rq)
+	if (!p->on_cpu || !task_on_rq_queued(p))
 		return p->se.sum_exec_runtime;
 #endif
 
@@ -2785,7 +2799,7 @@ need_resched:
 		switch_count = &prev->nvcsw;
 	}
 
-	if (prev->on_rq || rq->skip_clock_update < 0)
+	if (task_on_rq_queued(prev) || rq->skip_clock_update < 0)
 		update_rq_clock(rq);
 
 	next = pick_next_task(rq, prev);
@@ -2950,7 +2964,7 @@ EXPORT_SYMBOL(default_wake_function);
  */
 void rt_mutex_setprio(struct task_struct *p, int prio)
 {
-	int oldprio, on_rq, running, enqueue_flag = 0;
+	int oldprio, queued, running, enqueue_flag = 0;
 	struct rq *rq;
 	const struct sched_class *prev_class;
 
@@ -2979,9 +2993,9 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 	trace_sched_pi_setprio(p, prio);
 	oldprio = p->prio;
 	prev_class = p->sched_class;
-	on_rq = p->on_rq;
+	queued = task_on_rq_queued(p);
 	running = task_current(rq, p);
-	if (on_rq)
+	if (queued)
 		dequeue_task(rq, p, 0);
 	if (running)
 		p->sched_class->put_prev_task(rq, p);
@@ -3021,7 +3035,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 
 	if (running)
 		p->sched_class->set_curr_task(rq);
-	if (on_rq)
+	if (queued)
 		enqueue_task(rq, p, enqueue_flag);
 
 	check_class_changed(rq, p, prev_class, oldprio);
@@ -3032,7 +3046,7 @@ out_unlock:
 
 void set_user_nice(struct task_struct *p, long nice)
 {
-	int old_prio, delta, on_rq;
+	int old_prio, delta, queued;
 	unsigned long flags;
 	struct rq *rq;
 
@@ -3053,8 +3067,8 @@ void set_user_nice(struct task_struct *p, long nice)
 		p->static_prio = NICE_TO_PRIO(nice);
 		goto out_unlock;
 	}
-	on_rq = p->on_rq;
-	if (on_rq)
+	queued = task_on_rq_queued(p);
+	if (queued)
 		dequeue_task(rq, p, 0);
 
 	p->static_prio = NICE_TO_PRIO(nice);
@@ -3063,7 +3077,7 @@ void set_user_nice(struct task_struct *p, long nice)
 	p->prio = effective_prio(p);
 	delta = p->prio - old_prio;
 
-	if (on_rq) {
+	if (queued) {
 		enqueue_task(rq, p, 0);
 		/*
 		 * If the task increased its priority or is running and
@@ -3335,7 +3349,7 @@ static int __sched_setscheduler(struct task_struct *p,
 {
 	int newprio = dl_policy(attr->sched_policy) ? MAX_DL_PRIO - 1 :
 		      MAX_RT_PRIO - 1 - attr->sched_priority;
-	int retval, oldprio, oldpolicy = -1, on_rq, running;
+	int retval, oldprio, oldpolicy = -1, queued, running;
 	int policy = attr->sched_policy;
 	unsigned long flags;
 	const struct sched_class *prev_class;
@@ -3532,9 +3546,9 @@ change:
 		return 0;
 	}
 
-	on_rq = p->on_rq;
+	queued = task_on_rq_queued(p);
 	running = task_current(rq, p);
-	if (on_rq)
+	if (queued)
 		dequeue_task(rq, p, 0);
 	if (running)
 		p->sched_class->put_prev_task(rq, p);
@@ -3544,7 +3558,7 @@ change:
 
 	if (running)
 		p->sched_class->set_curr_task(rq);
-	if (on_rq) {
+	if (queued) {
 		/*
 		 * We enqueue to tail when the priority of a task is
 		 * increased (user space view).
@@ -4496,7 +4510,7 @@ void show_state_filter(unsigned long state_filter)
 		"  task                        PC stack   pid father\n");
 #endif
 	rcu_read_lock();
-	do_each_thread(g, p) {
+	for_each_process_thread(g, p) {
 		/*
 		 * reset the NMI-timeout, listing all files on a slow
 		 * console might take a lot of time:
@@ -4504,7 +4518,7 @@ void show_state_filter(unsigned long state_filter)
 		touch_nmi_watchdog();
 		if (!state_filter || (p->state & state_filter))
 			sched_show_task(p);
-	} while_each_thread(g, p);
+	}
 
 	touch_all_softlockup_watchdogs();
 
@@ -4559,7 +4573,7 @@ void init_idle(struct task_struct *idle, int cpu)
 	rcu_read_unlock();
 
 	rq->curr = rq->idle = idle;
-	idle->on_rq = 1;
+	idle->on_rq = TASK_ON_RQ_QUEUED;
 #if defined(CONFIG_SMP)
 	idle->on_cpu = 1;
 #endif
@@ -4636,7 +4650,7 @@ int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask)
 		goto out;
 
 	dest_cpu = cpumask_any_and(cpu_active_mask, new_mask);
-	if (p->on_rq) {
+	if (task_on_rq_queued(p)) {
 		struct migration_arg arg = { p, dest_cpu };
 		/* Need help from migration thread: drop lock and wait. */
 		task_rq_unlock(rq, p, &flags);
@@ -4664,20 +4678,20 @@ EXPORT_SYMBOL_GPL(set_cpus_allowed_ptr);
  */
 static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu)
 {
-	struct rq *rq_dest, *rq_src;
+	struct rq *rq;
 	int ret = 0;
 
 	if (unlikely(!cpu_active(dest_cpu)))
 		return ret;
 
-	rq_src = cpu_rq(src_cpu);
-	rq_dest = cpu_rq(dest_cpu);
+	rq = cpu_rq(src_cpu);
 
 	raw_spin_lock(&p->pi_lock);
-	double_rq_lock(rq_src, rq_dest);
+	raw_spin_lock(&rq->lock);
 	/* Already moved. */
 	if (task_cpu(p) != src_cpu)
 		goto done;
+
 	/* Affinity changed (again). */
 	if (!cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p)))
 		goto fail;
@@ -4686,16 +4700,23 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu)
 	 * If we're not on a rq, the next wake-up will ensure we're
 	 * placed properly.
 	 */
-	if (p->on_rq) {
-		dequeue_task(rq_src, p, 0);
+	if (task_on_rq_queued(p)) {
+		dequeue_task(rq, p, 0);
+		p->on_rq = TASK_ON_RQ_MIGRATING;
 		set_task_cpu(p, dest_cpu);
-		enqueue_task(rq_dest, p, 0);
-		check_preempt_curr(rq_dest, p, 0);
+		raw_spin_unlock(&rq->lock);
+
+		rq = cpu_rq(dest_cpu);
+		raw_spin_lock(&rq->lock);
+		BUG_ON(task_rq(p) != rq);
+		p->on_rq = TASK_ON_RQ_QUEUED;
+		enqueue_task(rq, p, 0);
+		check_preempt_curr(rq, p, 0);
 	}
 done:
 	ret = 1;
 fail:
-	double_rq_unlock(rq_src, rq_dest);
+	raw_spin_unlock(&rq->lock);
 	raw_spin_unlock(&p->pi_lock);
 	return ret;
 }
@@ -4727,13 +4748,13 @@ void sched_setnuma(struct task_struct *p, int nid)
 {
 	struct rq *rq;
 	unsigned long flags;
-	bool on_rq, running;
+	bool queued, running;
 
 	rq = task_rq_lock(p, &flags);
-	on_rq = p->on_rq;
+	queued = task_on_rq_queued(p);
 	running = task_current(rq, p);
 
-	if (on_rq)
+	if (queued)
 		dequeue_task(rq, p, 0);
 	if (running)
 		p->sched_class->put_prev_task(rq, p);
@@ -4742,7 +4763,7 @@ void sched_setnuma(struct task_struct *p, int nid)
 
 	if (running)
 		p->sched_class->set_curr_task(rq);
-	if (on_rq)
+	if (queued)
 		enqueue_task(rq, p, 0);
 	task_rq_unlock(rq, p, &flags);
 }
@@ -5730,7 +5751,7 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu)
 	const struct cpumask *span = sched_domain_span(sd);
 	struct cpumask *covered = sched_domains_tmpmask;
 	struct sd_data *sdd = sd->private;
-	struct sched_domain *child;
+	struct sched_domain *sibling;
 	int i;
 
 	cpumask_clear(covered);
@@ -5741,10 +5762,10 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu)
 		if (cpumask_test_cpu(i, covered))
 			continue;
 
-		child = *per_cpu_ptr(sdd->sd, i);
+		sibling = *per_cpu_ptr(sdd->sd, i);
 
 		/* See the comment near build_group_mask(). */
-		if (!cpumask_test_cpu(i, sched_domain_span(child)))
+		if (!cpumask_test_cpu(i, sched_domain_span(sibling)))
 			continue;
 
 		sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(),
@@ -5754,10 +5775,9 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu)
 			goto fail;
 
 		sg_span = sched_group_cpus(sg);
-		if (child->child) {
-			child = child->child;
-			cpumask_copy(sg_span, sched_domain_span(child));
-		} else
+		if (sibling->child)
+			cpumask_copy(sg_span, sched_domain_span(sibling->child));
+		else
 			cpumask_set_cpu(i, sg_span);
 
 		cpumask_or(covered, covered, sg_span);
@@ -7108,13 +7128,13 @@ static void normalize_task(struct rq *rq, struct task_struct *p)
 		.sched_policy = SCHED_NORMAL,
 	};
 	int old_prio = p->prio;
-	int on_rq;
+	int queued;
 
-	on_rq = p->on_rq;
-	if (on_rq)
+	queued = task_on_rq_queued(p);
+	if (queued)
 		dequeue_task(rq, p, 0);
 	__setscheduler(rq, p, &attr);
-	if (on_rq) {
+	if (queued) {
 		enqueue_task(rq, p, 0);
 		resched_curr(rq);
 	}
@@ -7129,7 +7149,7 @@ void normalize_rt_tasks(void)
 	struct rq *rq;
 
 	read_lock_irqsave(&tasklist_lock, flags);
-	do_each_thread(g, p) {
+	for_each_process_thread(g, p) {
 		/*
 		 * Only normalize user tasks:
 		 */
@@ -7160,8 +7180,7 @@ void normalize_rt_tasks(void)
 
 		__task_rq_unlock(rq);
 		raw_spin_unlock(&p->pi_lock);
-	} while_each_thread(g, p);
-
+	}
 	read_unlock_irqrestore(&tasklist_lock, flags);
 }
 
@@ -7302,16 +7321,16 @@ void sched_offline_group(struct task_group *tg)
 void sched_move_task(struct task_struct *tsk)
 {
 	struct task_group *tg;
-	int on_rq, running;
+	int queued, running;
 	unsigned long flags;
 	struct rq *rq;
 
 	rq = task_rq_lock(tsk, &flags);
 
 	running = task_current(rq, tsk);
-	on_rq = tsk->on_rq;
+	queued = task_on_rq_queued(tsk);
 
-	if (on_rq)
+	if (queued)
 		dequeue_task(rq, tsk, 0);
 	if (unlikely(running))
 		tsk->sched_class->put_prev_task(rq, tsk);
@@ -7324,14 +7343,14 @@ void sched_move_task(struct task_struct *tsk)
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	if (tsk->sched_class->task_move_group)
-		tsk->sched_class->task_move_group(tsk, on_rq);
+		tsk->sched_class->task_move_group(tsk, queued);
 	else
 #endif
 		set_task_rq(tsk, task_cpu(tsk));
 
 	if (unlikely(running))
 		tsk->sched_class->set_curr_task(rq);
-	if (on_rq)
+	if (queued)
 		enqueue_task(rq, tsk, 0);
 
 	task_rq_unlock(rq, tsk, &flags);
@@ -7349,10 +7368,10 @@ static inline int tg_has_rt_tasks(struct task_group *tg)
 {
 	struct task_struct *g, *p;
 
-	do_each_thread(g, p) {
+	for_each_process_thread(g, p) {
 		if (rt_task(p) && task_rq(p)->rt.tg == tg)
 			return 1;
-	} while_each_thread(g, p);
+	}
 
 	return 0;
 }
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index 72fdf06ef8652..3e52836359baf 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -294,18 +294,12 @@ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times)
 	times->sum_exec_runtime = sig->sum_sched_runtime;
 
 	rcu_read_lock();
-	/* make sure we can trust tsk->thread_group list */
-	if (!likely(pid_alive(tsk)))
-		goto out;
-
-	t = tsk;
-	do {
+	for_each_thread(tsk, t) {
 		task_cputime(t, &utime, &stime);
 		times->utime += utime;
 		times->stime += stime;
 		times->sum_exec_runtime += task_sched_runtime(t);
-	} while_each_thread(tsk, t);
-out:
+	}
 	rcu_read_unlock();
 }
 
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 255ce138b6523..cc4eb89019c18 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -530,7 +530,7 @@ again:
 	update_rq_clock(rq);
 	dl_se->dl_throttled = 0;
 	dl_se->dl_yielded = 0;
-	if (p->on_rq) {
+	if (task_on_rq_queued(p)) {
 		enqueue_task_dl(rq, p, ENQUEUE_REPLENISH);
 		if (task_has_dl_policy(rq->curr))
 			check_preempt_curr_dl(rq, p, 0);
@@ -997,10 +997,7 @@ static void check_preempt_curr_dl(struct rq *rq, struct task_struct *p,
 #ifdef CONFIG_SCHED_HRTICK
 static void start_hrtick_dl(struct rq *rq, struct task_struct *p)
 {
-	s64 delta = p->dl.dl_runtime - p->dl.runtime;
-
-	if (delta > 10000)
-		hrtick_start(rq, p->dl.runtime);
+	hrtick_start(rq, p->dl.runtime);
 }
 #endif
 
@@ -1030,7 +1027,7 @@ struct task_struct *pick_next_task_dl(struct rq *rq, struct task_struct *prev)
 		 * means a stop task can slip in, in which case we need to
 		 * re-start task selection.
 		 */
-		if (rq->stop && rq->stop->on_rq)
+		if (rq->stop && task_on_rq_queued(rq->stop))
 			return RETRY_TASK;
 	}
 
@@ -1257,7 +1254,8 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq)
 			if (unlikely(task_rq(task) != rq ||
 				     !cpumask_test_cpu(later_rq->cpu,
 				                       &task->cpus_allowed) ||
-				     task_running(rq, task) || !task->on_rq)) {
+				     task_running(rq, task) ||
+				     !task_on_rq_queued(task))) {
 				double_unlock_balance(rq, later_rq);
 				later_rq = NULL;
 				break;
@@ -1296,7 +1294,7 @@ static struct task_struct *pick_next_pushable_dl_task(struct rq *rq)
 	BUG_ON(task_current(rq, p));
 	BUG_ON(p->nr_cpus_allowed <= 1);
 
-	BUG_ON(!p->on_rq);
+	BUG_ON(!task_on_rq_queued(p));
 	BUG_ON(!dl_task(p));
 
 	return p;
@@ -1443,7 +1441,7 @@ static int pull_dl_task(struct rq *this_rq)
 		     dl_time_before(p->dl.deadline,
 				    this_rq->dl.earliest_dl.curr))) {
 			WARN_ON(p == src_rq->curr);
-			WARN_ON(!p->on_rq);
+			WARN_ON(!task_on_rq_queued(p));
 
 			/*
 			 * Then we pull iff p has actually an earlier
@@ -1596,7 +1594,7 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p)
 	if (unlikely(p->dl.dl_throttled))
 		return;
 
-	if (p->on_rq && rq->curr != p) {
+	if (task_on_rq_queued(p) && rq->curr != p) {
 #ifdef CONFIG_SMP
 		if (rq->dl.overloaded && push_dl_task(rq) && rq != task_rq(p))
 			/* Only reschedule if pushing failed */
@@ -1614,7 +1612,7 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p)
 static void prio_changed_dl(struct rq *rq, struct task_struct *p,
 			    int oldprio)
 {
-	if (p->on_rq || rq->curr == p) {
+	if (task_on_rq_queued(p) || rq->curr == p) {
 #ifdef CONFIG_SMP
 		/*
 		 * This might be too much, but unfortunately
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 627b3c34b821d..c7fe1ea0e8ab0 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -160,14 +160,12 @@ static void print_rq(struct seq_file *m, struct rq *rq, int rq_cpu)
 	"----------------------------------------------------\n");
 
 	read_lock_irqsave(&tasklist_lock, flags);
-
-	do_each_thread(g, p) {
+	for_each_process_thread(g, p) {
 		if (task_cpu(p) != rq_cpu)
 			continue;
 
 		print_task(m, rq, p);
-	} while_each_thread(g, p);
-
+	}
 	read_unlock_irqrestore(&tasklist_lock, flags);
 }
 
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index bfa3c86d0d686..50d2025c1777b 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -1038,7 +1038,8 @@ struct numa_stats {
  */
 static void update_numa_stats(struct numa_stats *ns, int nid)
 {
-	int cpu, cpus = 0;
+	int smt, cpu, cpus = 0;
+	unsigned long capacity;
 
 	memset(ns, 0, sizeof(*ns));
 	for_each_cpu(cpu, cpumask_of_node(nid)) {
@@ -1062,8 +1063,12 @@ static void update_numa_stats(struct numa_stats *ns, int nid)
 	if (!cpus)
 		return;
 
-	ns->task_capacity =
-		DIV_ROUND_CLOSEST(ns->compute_capacity, SCHED_CAPACITY_SCALE);
+	/* smt := ceil(cpus / capacity), assumes: 1 < smt_power < 2 */
+	smt = DIV_ROUND_UP(SCHED_CAPACITY_SCALE * cpus, ns->compute_capacity);
+	capacity = cpus / smt; /* cores */
+
+	ns->task_capacity = min_t(unsigned, capacity,
+		DIV_ROUND_CLOSEST(ns->compute_capacity, SCHED_CAPACITY_SCALE));
 	ns->has_free_capacity = (ns->nr_running < ns->task_capacity);
 }
 
@@ -1206,7 +1211,7 @@ static void task_numa_compare(struct task_numa_env *env,
 
 	if (!cur) {
 		/* Is there capacity at our destination? */
-		if (env->src_stats.has_free_capacity &&
+		if (env->src_stats.nr_running <= env->src_stats.task_capacity &&
 		    !env->dst_stats.has_free_capacity)
 			goto unlock;
 
@@ -1775,7 +1780,7 @@ void task_numa_free(struct task_struct *p)
 		list_del(&p->numa_entry);
 		grp->nr_tasks--;
 		spin_unlock_irqrestore(&grp->lock, flags);
-		rcu_assign_pointer(p->numa_group, NULL);
+		RCU_INIT_POINTER(p->numa_group, NULL);
 		put_numa_group(grp);
 	}
 
@@ -3892,14 +3897,6 @@ static void hrtick_start_fair(struct rq *rq, struct task_struct *p)
 				resched_curr(rq);
 			return;
 		}
-
-		/*
-		 * Don't schedule slices shorter than 10000ns, that just
-		 * doesn't make sense. Rely on vruntime for fairness.
-		 */
-		if (rq->curr != p)
-			delta = max_t(s64, 10000LL, delta);
-
 		hrtick_start(rq, delta);
 	}
 }
@@ -4704,7 +4701,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_
 		return;
 
 	/*
-	 * This is possible from callers such as move_task(), in which we
+	 * This is possible from callers such as attach_tasks(), in which we
 	 * unconditionally check_prempt_curr() after an enqueue (which may have
 	 * lead to a throttle).  This both saves work and prevents false
 	 * next-buddy nomination below.
@@ -5112,27 +5109,18 @@ struct lb_env {
 	unsigned int		loop_max;
 
 	enum fbq_type		fbq_type;
+	struct list_head	tasks;
 };
 
 /*
- * move_task - move a task from one runqueue to another runqueue.
- * Both runqueues must be locked.
- */
-static void move_task(struct task_struct *p, struct lb_env *env)
-{
-	deactivate_task(env->src_rq, p, 0);
-	set_task_cpu(p, env->dst_cpu);
-	activate_task(env->dst_rq, p, 0);
-	check_preempt_curr(env->dst_rq, p, 0);
-}
-
-/*
  * Is this task likely cache-hot:
  */
 static int task_hot(struct task_struct *p, struct lb_env *env)
 {
 	s64 delta;
 
+	lockdep_assert_held(&env->src_rq->lock);
+
 	if (p->sched_class != &fair_sched_class)
 		return 0;
 
@@ -5252,6 +5240,9 @@ static
 int can_migrate_task(struct task_struct *p, struct lb_env *env)
 {
 	int tsk_cache_hot = 0;
+
+	lockdep_assert_held(&env->src_rq->lock);
+
 	/*
 	 * We do not migrate tasks that are:
 	 * 1) throttled_lb_pair, or
@@ -5336,47 +5327,63 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env)
 }
 
 /*
- * move_one_task tries to move exactly one task from busiest to this_rq, as
+ * detach_task() -- detach the task for the migration specified in env
+ */
+static void detach_task(struct task_struct *p, struct lb_env *env)
+{
+	lockdep_assert_held(&env->src_rq->lock);
+
+	deactivate_task(env->src_rq, p, 0);
+	p->on_rq = TASK_ON_RQ_MIGRATING;
+	set_task_cpu(p, env->dst_cpu);
+}
+
+/*
+ * detach_one_task() -- tries to dequeue exactly one task from env->src_rq, as
  * part of active balancing operations within "domain".
- * Returns 1 if successful and 0 otherwise.
  *
- * Called with both runqueues locked.
+ * Returns a task if successful and NULL otherwise.
  */
-static int move_one_task(struct lb_env *env)
+static struct task_struct *detach_one_task(struct lb_env *env)
 {
 	struct task_struct *p, *n;
 
+	lockdep_assert_held(&env->src_rq->lock);
+
 	list_for_each_entry_safe(p, n, &env->src_rq->cfs_tasks, se.group_node) {
 		if (!can_migrate_task(p, env))
 			continue;
 
-		move_task(p, env);
+		detach_task(p, env);
+
 		/*
-		 * Right now, this is only the second place move_task()
-		 * is called, so we can safely collect move_task()
-		 * stats here rather than inside move_task().
+		 * Right now, this is only the second place where
+		 * lb_gained[env->idle] is updated (other is detach_tasks)
+		 * so we can safely collect stats here rather than
+		 * inside detach_tasks().
 		 */
 		schedstat_inc(env->sd, lb_gained[env->idle]);
-		return 1;
+		return p;
 	}
-	return 0;
+	return NULL;
 }
 
 static const unsigned int sched_nr_migrate_break = 32;
 
 /*
- * move_tasks tries to move up to imbalance weighted load from busiest to
- * this_rq, as part of a balancing operation within domain "sd".
- * Returns 1 if successful and 0 otherwise.
+ * detach_tasks() -- tries to detach up to imbalance weighted load from
+ * busiest_rq, as part of a balancing operation within domain "sd".
  *
- * Called with both runqueues locked.
+ * Returns number of detached tasks if successful and 0 otherwise.
  */
-static int move_tasks(struct lb_env *env)
+static int detach_tasks(struct lb_env *env)
 {
 	struct list_head *tasks = &env->src_rq->cfs_tasks;
 	struct task_struct *p;
 	unsigned long load;
-	int pulled = 0;
+	int detached = 0;
+
+	lockdep_assert_held(&env->src_rq->lock);
 
 	if (env->imbalance <= 0)
 		return 0;
@@ -5407,14 +5414,16 @@ static int move_tasks(struct lb_env *env)
 		if ((load / 2) > env->imbalance)
 			goto next;
 
-		move_task(p, env);
-		pulled++;
+		detach_task(p, env);
+		list_add(&p->se.group_node, &env->tasks);
+
+		detached++;
 		env->imbalance -= load;
 
 #ifdef CONFIG_PREEMPT
 		/*
 		 * NEWIDLE balancing is a source of latency, so preemptible
-		 * kernels will stop after the first task is pulled to minimize
+		 * kernels will stop after the first task is detached to minimize
 		 * the critical section.
 		 */
 		if (env->idle == CPU_NEWLY_IDLE)
@@ -5434,13 +5443,58 @@ next:
 	}
 
 	/*
-	 * Right now, this is one of only two places move_task() is called,
-	 * so we can safely collect move_task() stats here rather than
-	 * inside move_task().
+	 * Right now, this is one of only two places we collect this stat
+	 * so we can safely collect detach_one_task() stats here rather
+	 * than inside detach_one_task().
 	 */
-	schedstat_add(env->sd, lb_gained[env->idle], pulled);
+	schedstat_add(env->sd, lb_gained[env->idle], detached);
+
+	return detached;
+}
+
+/*
+ * attach_task() -- attach the task detached by detach_task() to its new rq.
+ */
+static void attach_task(struct rq *rq, struct task_struct *p)
+{
+	lockdep_assert_held(&rq->lock);
 
-	return pulled;
+	BUG_ON(task_rq(p) != rq);
+	p->on_rq = TASK_ON_RQ_QUEUED;
+	activate_task(rq, p, 0);
+	check_preempt_curr(rq, p, 0);
+}
+
+/*
+ * attach_one_task() -- attaches the task returned from detach_one_task() to
+ * its new rq.
+ */
+static void attach_one_task(struct rq *rq, struct task_struct *p)
+{
+	raw_spin_lock(&rq->lock);
+	attach_task(rq, p);
+	raw_spin_unlock(&rq->lock);
+}
+
+/*
+ * attach_tasks() -- attaches all tasks detached by detach_tasks() to their
+ * new rq.
+ */
+static void attach_tasks(struct lb_env *env)
+{
+	struct list_head *tasks = &env->tasks;
+	struct task_struct *p;
+
+	raw_spin_lock(&env->dst_rq->lock);
+
+	while (!list_empty(tasks)) {
+		p = list_first_entry(tasks, struct task_struct, se.group_node);
+		list_del_init(&p->se.group_node);
+
+		attach_task(env->dst_rq, p);
+	}
+
+	raw_spin_unlock(&env->dst_rq->lock);
 }
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
@@ -5559,6 +5613,13 @@ static unsigned long task_h_load(struct task_struct *p)
 #endif
 
 /********** Helpers for find_busiest_group ************************/
+
+enum group_type {
+	group_other = 0,
+	group_imbalanced,
+	group_overloaded,
+};
+
 /*
  * sg_lb_stats - stats of a sched_group required for load_balancing
  */
@@ -5572,7 +5633,7 @@ struct sg_lb_stats {
 	unsigned int group_capacity_factor;
 	unsigned int idle_cpus;
 	unsigned int group_weight;
-	int group_imb; /* Is there an imbalance in the group ? */
+	enum group_type group_type;
 	int group_has_free_capacity;
 #ifdef CONFIG_NUMA_BALANCING
 	unsigned int nr_numa_running;
@@ -5610,6 +5671,8 @@ static inline void init_sd_lb_stats(struct sd_lb_stats *sds)
 		.total_capacity = 0UL,
 		.busiest_stat = {
 			.avg_load = 0UL,
+			.sum_nr_running = 0,
+			.group_type = group_other,
 		},
 	};
 }
@@ -5891,6 +5954,18 @@ static inline int sg_capacity_factor(struct lb_env *env, struct sched_group *gro
 	return capacity_factor;
 }
 
+static enum group_type
+group_classify(struct sched_group *group, struct sg_lb_stats *sgs)
+{
+	if (sgs->sum_nr_running > sgs->group_capacity_factor)
+		return group_overloaded;
+
+	if (sg_imbalanced(group))
+		return group_imbalanced;
+
+	return group_other;
+}
+
 /**
  * update_sg_lb_stats - Update sched_group's statistics for load balancing.
  * @env: The load balancing environment.
@@ -5942,9 +6017,8 @@ static inline void update_sg_lb_stats(struct lb_env *env,
 		sgs->load_per_task = sgs->sum_weighted_load / sgs->sum_nr_running;
 
 	sgs->group_weight = group->group_weight;
-
-	sgs->group_imb = sg_imbalanced(group);
 	sgs->group_capacity_factor = sg_capacity_factor(env, group);
+	sgs->group_type = group_classify(group, sgs);
 
 	if (sgs->group_capacity_factor > sgs->sum_nr_running)
 		sgs->group_has_free_capacity = 1;
@@ -5968,13 +6042,19 @@ static bool update_sd_pick_busiest(struct lb_env *env,
 				   struct sched_group *sg,
 				   struct sg_lb_stats *sgs)
 {
-	if (sgs->avg_load <= sds->busiest_stat.avg_load)
-		return false;
+	struct sg_lb_stats *busiest = &sds->busiest_stat;
 
-	if (sgs->sum_nr_running > sgs->group_capacity_factor)
+	if (sgs->group_type > busiest->group_type)
 		return true;
 
-	if (sgs->group_imb)
+	if (sgs->group_type < busiest->group_type)
+		return false;
+
+	if (sgs->avg_load <= busiest->avg_load)
+		return false;
+
+	/* This is the busiest node in its class. */
+	if (!(env->sd->flags & SD_ASYM_PACKING))
 		return true;
 
 	/*
@@ -5982,8 +6062,7 @@ static bool update_sd_pick_busiest(struct lb_env *env,
 	 * numbered CPUs in the group, therefore mark all groups
 	 * higher than ourself as busy.
 	 */
-	if ((env->sd->flags & SD_ASYM_PACKING) && sgs->sum_nr_running &&
-	    env->dst_cpu < group_first_cpu(sg)) {
+	if (sgs->sum_nr_running && env->dst_cpu < group_first_cpu(sg)) {
 		if (!sds->busiest)
 			return true;
 
@@ -6228,7 +6307,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
 	local = &sds->local_stat;
 	busiest = &sds->busiest_stat;
 
-	if (busiest->group_imb) {
+	if (busiest->group_type == group_imbalanced) {
 		/*
 		 * In the group_imb case we cannot rely on group-wide averages
 		 * to ensure cpu-load equilibrium, look at wider averages. XXX
@@ -6248,12 +6327,11 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
 		return fix_small_imbalance(env, sds);
 	}
 
-	if (!busiest->group_imb) {
-		/*
-		 * Don't want to pull so many tasks that a group would go idle.
-		 * Except of course for the group_imb case, since then we might
-		 * have to drop below capacity to reach cpu-load equilibrium.
-		 */
+	/*
+	 * If there aren't any idle cpus, avoid creating some.
+	 */
+	if (busiest->group_type == group_overloaded &&
+	    local->group_type   == group_overloaded) {
 		load_above_capacity =
 			(busiest->sum_nr_running - busiest->group_capacity_factor);
 
@@ -6337,7 +6415,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 	 * work because they assume all things are equal, which typically
 	 * isn't true due to cpus_allowed constraints and the like.
 	 */
-	if (busiest->group_imb)
+	if (busiest->group_type == group_imbalanced)
 		goto force_balance;
 
 	/* SD_BALANCE_NEWIDLE trumps SMP nice when underutilized */
@@ -6550,6 +6628,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 		.loop_break	= sched_nr_migrate_break,
 		.cpus		= cpus,
 		.fbq_type	= all,
+		.tasks		= LIST_HEAD_INIT(env.tasks),
 	};
 
 	/*
@@ -6599,16 +6678,29 @@ redo:
 		env.loop_max  = min(sysctl_sched_nr_migrate, busiest->nr_running);
 
 more_balance:
-		local_irq_save(flags);
-		double_rq_lock(env.dst_rq, busiest);
+		raw_spin_lock_irqsave(&busiest->lock, flags);
 
 		/*
 		 * cur_ld_moved - load moved in current iteration
 		 * ld_moved     - cumulative load moved across iterations
 		 */
-		cur_ld_moved = move_tasks(&env);
-		ld_moved += cur_ld_moved;
-		double_rq_unlock(env.dst_rq, busiest);
+		cur_ld_moved = detach_tasks(&env);
+
+		/*
+		 * We've detached some tasks from busiest_rq. Every
+		 * task is masked "TASK_ON_RQ_MIGRATING", so we can safely
+		 * unlock busiest->lock, and we are able to be sure
+		 * that nobody can manipulate the tasks in parallel.
+		 * See task_rq_lock() family for the details.
+		 */
+
+		raw_spin_unlock(&busiest->lock);
+
+		if (cur_ld_moved) {
+			attach_tasks(&env);
+			ld_moved += cur_ld_moved;
+		}
+
 		local_irq_restore(flags);
 
 		/*
@@ -6744,7 +6836,7 @@ more_balance:
 		 * If we've begun active balancing, start to back off. This
 		 * case may not be covered by the all_pinned logic if there
 		 * is only 1 task on the busy runqueue (because we don't call
-		 * move_tasks).
+		 * detach_tasks).
 		 */
 		if (sd->balance_interval < sd->max_interval)
 			sd->balance_interval *= 2;
@@ -6914,6 +7006,7 @@ static int active_load_balance_cpu_stop(void *data)
 	int target_cpu = busiest_rq->push_cpu;
 	struct rq *target_rq = cpu_rq(target_cpu);
 	struct sched_domain *sd;
+	struct task_struct *p = NULL;
 
 	raw_spin_lock_irq(&busiest_rq->lock);
 
@@ -6933,9 +7026,6 @@ static int active_load_balance_cpu_stop(void *data)
 	 */
 	BUG_ON(busiest_rq == target_rq);
 
-	/* move a task from busiest_rq to target_rq */
-	double_lock_balance(busiest_rq, target_rq);
-
 	/* Search for an sd spanning us and the target CPU. */
 	rcu_read_lock();
 	for_each_domain(target_cpu, sd) {
@@ -6956,16 +7046,22 @@ static int active_load_balance_cpu_stop(void *data)
 
 		schedstat_inc(sd, alb_count);
 
-		if (move_one_task(&env))
+		p = detach_one_task(&env);
+		if (p)
 			schedstat_inc(sd, alb_pushed);
 		else
 			schedstat_inc(sd, alb_failed);
 	}
 	rcu_read_unlock();
-	double_unlock_balance(busiest_rq, target_rq);
 out_unlock:
 	busiest_rq->active_balance = 0;
-	raw_spin_unlock_irq(&busiest_rq->lock);
+	raw_spin_unlock(&busiest_rq->lock);
+
+	if (p)
+		attach_one_task(target_rq, p);
+
+	local_irq_enable();
+
 	return 0;
 }
 
@@ -7465,7 +7561,7 @@ static void task_fork_fair(struct task_struct *p)
 static void
 prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio)
 {
-	if (!p->se.on_rq)
+	if (!task_on_rq_queued(p))
 		return;
 
 	/*
@@ -7490,11 +7586,11 @@ static void switched_from_fair(struct rq *rq, struct task_struct *p)
 	 * switched back to the fair class the enqueue_entity(.flags=0) will
 	 * do the right thing.
 	 *
-	 * If it's on_rq, then the dequeue_entity(.flags=0) will already
-	 * have normalized the vruntime, if it's !on_rq, then only when
+	 * If it's queued, then the dequeue_entity(.flags=0) will already
+	 * have normalized the vruntime, if it's !queued, then only when
 	 * the task is sleeping will it still have non-normalized vruntime.
 	 */
-	if (!p->on_rq && p->state != TASK_RUNNING) {
+	if (!task_on_rq_queued(p) && p->state != TASK_RUNNING) {
 		/*
 		 * Fix up our vruntime so that the current sleep doesn't
 		 * cause 'unlimited' sleep bonus.
@@ -7521,15 +7617,15 @@ static void switched_from_fair(struct rq *rq, struct task_struct *p)
  */
 static void switched_to_fair(struct rq *rq, struct task_struct *p)
 {
-	struct sched_entity *se = &p->se;
 #ifdef CONFIG_FAIR_GROUP_SCHED
+	struct sched_entity *se = &p->se;
 	/*
 	 * Since the real-depth could have been changed (only FAIR
 	 * class maintain depth value), reset depth properly.
 	 */
 	se->depth = se->parent ? se->parent->depth + 1 : 0;
 #endif
-	if (!se->on_rq)
+	if (!task_on_rq_queued(p))
 		return;
 
 	/*
@@ -7575,7 +7671,7 @@ void init_cfs_rq(struct cfs_rq *cfs_rq)
 }
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
-static void task_move_group_fair(struct task_struct *p, int on_rq)
+static void task_move_group_fair(struct task_struct *p, int queued)
 {
 	struct sched_entity *se = &p->se;
 	struct cfs_rq *cfs_rq;
@@ -7594,7 +7690,7 @@ static void task_move_group_fair(struct task_struct *p, int on_rq)
 	 * fair sleeper stuff for the first placement, but who cares.
 	 */
 	/*
-	 * When !on_rq, vruntime of the task has usually NOT been normalized.
+	 * When !queued, vruntime of the task has usually NOT been normalized.
 	 * But there are some cases where it has already been normalized:
 	 *
 	 * - Moving a forked child which is waiting for being woken up by
@@ -7605,14 +7701,14 @@ static void task_move_group_fair(struct task_struct *p, int on_rq)
 	 * To prevent boost or penalty in the new cfs_rq caused by delta
 	 * min_vruntime between the two cfs_rqs, we skip vruntime adjustment.
 	 */
-	if (!on_rq && (!se->sum_exec_runtime || p->state == TASK_WAKING))
-		on_rq = 1;
+	if (!queued && (!se->sum_exec_runtime || p->state == TASK_WAKING))
+		queued = 1;
 
-	if (!on_rq)
+	if (!queued)
 		se->vruntime -= cfs_rq_of(se)->min_vruntime;
 	set_task_rq(p, task_cpu(p));
 	se->depth = se->parent ? se->parent->depth + 1 : 0;
-	if (!on_rq) {
+	if (!queued) {
 		cfs_rq = cfs_rq_of(se);
 		se->vruntime += cfs_rq->min_vruntime;
 #ifdef CONFIG_SMP
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index 5f6edca4fafd8..4feac8fcb47f6 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1448,7 +1448,7 @@ pick_next_task_rt(struct rq *rq, struct task_struct *prev)
 		 * means a dl or stop task can slip in, in which case we need
 		 * to re-start task selection.
 		 */
-		if (unlikely((rq->stop && rq->stop->on_rq) ||
+		if (unlikely((rq->stop && task_on_rq_queued(rq->stop)) ||
 			     rq->dl.dl_nr_running))
 			return RETRY_TASK;
 	}
@@ -1624,7 +1624,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 				     !cpumask_test_cpu(lowest_rq->cpu,
 						       tsk_cpus_allowed(task)) ||
 				     task_running(rq, task) ||
-				     !task->on_rq)) {
+				     !task_on_rq_queued(task))) {
 
 				double_unlock_balance(rq, lowest_rq);
 				lowest_rq = NULL;
@@ -1658,7 +1658,7 @@ static struct task_struct *pick_next_pushable_task(struct rq *rq)
 	BUG_ON(task_current(rq, p));
 	BUG_ON(p->nr_cpus_allowed <= 1);
 
-	BUG_ON(!p->on_rq);
+	BUG_ON(!task_on_rq_queued(p));
 	BUG_ON(!rt_task(p));
 
 	return p;
@@ -1809,7 +1809,7 @@ static int pull_rt_task(struct rq *this_rq)
 		 */
 		if (p && (p->prio < this_rq->rt.highest_prio.curr)) {
 			WARN_ON(p == src_rq->curr);
-			WARN_ON(!p->on_rq);
+			WARN_ON(!task_on_rq_queued(p));
 
 			/*
 			 * There's a chance that p is higher in priority
@@ -1870,7 +1870,7 @@ static void set_cpus_allowed_rt(struct task_struct *p,
 
 	BUG_ON(!rt_task(p));
 
-	if (!p->on_rq)
+	if (!task_on_rq_queued(p))
 		return;
 
 	weight = cpumask_weight(new_mask);
@@ -1936,7 +1936,7 @@ static void switched_from_rt(struct rq *rq, struct task_struct *p)
 	 * we may need to handle the pulling of RT tasks
 	 * now.
 	 */
-	if (!p->on_rq || rq->rt.rt_nr_running)
+	if (!task_on_rq_queued(p) || rq->rt.rt_nr_running)
 		return;
 
 	if (pull_rt_task(rq))
@@ -1970,7 +1970,7 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
 	 * If that current running task is also an RT task
 	 * then see if we can move to another run queue.
 	 */
-	if (p->on_rq && rq->curr != p) {
+	if (task_on_rq_queued(p) && rq->curr != p) {
 #ifdef CONFIG_SMP
 		if (p->nr_cpus_allowed > 1 && rq->rt.overloaded &&
 		    /* Don't resched if we changed runqueues */
@@ -1989,7 +1989,7 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
 static void
 prio_changed_rt(struct rq *rq, struct task_struct *p, int oldprio)
 {
-	if (!p->on_rq)
+	if (!task_on_rq_queued(p))
 		return;
 
 	if (rq->curr == p) {
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 579712f4e9d56..aa0f73ba37771 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -15,6 +15,10 @@
 
 struct rq;
 
+/* task_struct::on_rq states: */
+#define TASK_ON_RQ_QUEUED	1
+#define TASK_ON_RQ_MIGRATING	2
+
 extern __read_mostly int scheduler_running;
 
 extern unsigned long calc_load_update;
@@ -647,7 +651,7 @@ static inline int cpu_of(struct rq *rq)
 #endif
 }
 
-DECLARE_PER_CPU(struct rq, runqueues);
+DECLARE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues);
 
 #define cpu_rq(cpu)		(&per_cpu(runqueues, (cpu)))
 #define this_rq()		(&__get_cpu_var(runqueues))
@@ -942,6 +946,15 @@ static inline int task_running(struct rq *rq, struct task_struct *p)
 #endif
 }
 
+static inline int task_on_rq_queued(struct task_struct *p)
+{
+	return p->on_rq == TASK_ON_RQ_QUEUED;
+}
+
+static inline int task_on_rq_migrating(struct task_struct *p)
+{
+	return p->on_rq == TASK_ON_RQ_MIGRATING;
+}
 
 #ifndef prepare_arch_switch
 # define prepare_arch_switch(next)	do { } while (0)
diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
index bfe0edadbfbbe..67426e529f59c 100644
--- a/kernel/sched/stop_task.c
+++ b/kernel/sched/stop_task.c
@@ -28,7 +28,7 @@ pick_next_task_stop(struct rq *rq, struct task_struct *prev)
 {
 	struct task_struct *stop = rq->stop;
 
-	if (!stop || !stop->on_rq)
+	if (!stop || !task_on_rq_queued(stop))
 		return NULL;
 
 	put_prev_task(rq, prev);