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

Pull scheduler updates from Ingo Molnar:
 "The main changes in this cycle were:

   - Optimized support for Intel "Cluster-on-Die" (CoD) topologies (Dave
     Hansen)

   - Various sched/idle refinements for better idle handling (Nicolas
     Pitre, Daniel Lezcano, Chuansheng Liu, Vincent Guittot)

   - sched/numa updates and optimizations (Rik van Riel)

   - sysbench speedup (Vincent Guittot)

   - capacity calculation cleanups/refactoring (Vincent Guittot)

   - Various cleanups to thread group iteration (Oleg Nesterov)

   - Double-rq-lock removal optimization and various refactorings
     (Kirill Tkhai)

   - various sched/deadline fixes

  ... and lots of other changes"

* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (72 commits)
  sched/dl: Use dl_bw_of() under rcu_read_lock_sched()
  sched/fair: Delete resched_cpu() from idle_balance()
  sched, time: Fix build error with 64 bit cputime_t on 32 bit systems
  sched: Improve sysbench performance by fixing spurious active migration
  sched/x86: Fix up typo in topology detection
  x86, sched: Add new topology for multi-NUMA-node CPUs
  sched/rt: Use resched_curr() in task_tick_rt()
  sched: Use rq->rd in sched_setaffinity() under RCU read lock
  sched: cleanup: Rename 'out_unlock' to 'out_free_new_mask'
  sched: Use dl_bw_of() under RCU read lock
  sched/fair: Remove duplicate code from can_migrate_task()
  sched, mips, ia64: Remove __ARCH_WANT_UNLOCKED_CTXSW
  sched: print_rq(): Don't use tasklist_lock
  sched: normalize_rt_tasks(): Don't use _irqsave for tasklist_lock, use task_rq_lock()
  sched: Fix the task-group check in tg_has_rt_tasks()
  sched/fair: Leverage the idle state info when choosing the "idlest" cpu
  sched: Let the scheduler see CPU idle states
  sched/deadline: Fix inter- exclusive cpusets migrations
  sched/deadline: Clear dl_entity params when setscheduling to different class
  sched/numa: Kill the wrong/dead TASK_DEAD check in task_numa_fault()
  ...

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);

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);
@@ -1604,6 +1618,25 @@ static void ttwu_queue_remote(struct task_struct *p, int cpu)
 	}
 }
 
+void wake_up_if_idle(int cpu)
+{
+	struct rq *rq = cpu_rq(cpu);
+	unsigned long flags;
+
+	if (!is_idle_task(rq->curr))
+		return;
+
+	if (set_nr_if_polling(rq->idle)) {
+		trace_sched_wake_idle_without_ipi(cpu);
+	} else {
+		raw_spin_lock_irqsave(&rq->lock, flags);
+		if (is_idle_task(rq->curr))
+			smp_send_reschedule(cpu);
+		/* Else cpu is not in idle, do nothing here */
+		raw_spin_unlock_irqrestore(&rq->lock, flags);
+	}
+}
+
 bool cpus_share_cache(int this_cpu, int that_cpu)
 {
 	return per_cpu(sd_llc_id, this_cpu) == per_cpu(sd_llc_id, that_cpu);
@@ -1726,7 +1759,7 @@ 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);
@@ -1759,6 +1792,20 @@ int wake_up_state(struct task_struct *p, unsigned int state)
 	return try_to_wake_up(p, state, 0);
 }
 
+/*
+ * This function clears the sched_dl_entity static params.
+ */
+void __dl_clear_params(struct task_struct *p)
+{
+	struct sched_dl_entity *dl_se = &p->dl;
+
+	dl_se->dl_runtime = 0;
+	dl_se->dl_deadline = 0;
+	dl_se->dl_period = 0;
+	dl_se->flags = 0;
+	dl_se->dl_bw = 0;
+}
+
 /*
  * Perform scheduler related setup for a newly forked process p.
  * p is forked by current.
@@ -1783,10 +1830,7 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p)
 
 	RB_CLEAR_NODE(&p->dl.rb_node);
 	hrtimer_init(&p->dl.dl_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-	p->dl.dl_runtime = p->dl.runtime = 0;
-	p->dl.dl_deadline = p->dl.deadline = 0;
-	p->dl.dl_period = 0;
-	p->dl.flags = 0;
+	__dl_clear_params(p);
 
 	INIT_LIST_HEAD(&p->rt.run_list);
 
@@ -1961,6 +2005,8 @@ unsigned long to_ratio(u64 period, u64 runtime)
 #ifdef CONFIG_SMP
 inline struct dl_bw *dl_bw_of(int i)
 {
+	rcu_lockdep_assert(rcu_read_lock_sched_held(),
+			   "sched RCU must be held");
 	return &cpu_rq(i)->rd->dl_bw;
 }
 
@@ -1969,6 +2015,8 @@ static inline int dl_bw_cpus(int i)
 	struct root_domain *rd = cpu_rq(i)->rd;
 	int cpus = 0;
 
+	rcu_lockdep_assert(rcu_read_lock_sched_held(),
+			   "sched RCU must be held");
 	for_each_cpu_and(i, rd->span, cpu_active_mask)
 		cpus++;
 
@@ -2079,7 +2127,7 @@ 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
@@ -2271,10 +2319,6 @@ asmlinkage __visible void schedule_tail(struct task_struct *prev)
 	 */
 	post_schedule(rq);
 
-#ifdef __ARCH_WANT_UNLOCKED_CTXSW
-	/* In this case, finish_task_switch does not reenable preemption */
-	preempt_enable();
-#endif
 	if (current->set_child_tid)
 		put_user(task_pid_vnr(current), current->set_child_tid);
 }
@@ -2317,9 +2361,7 @@ context_switch(struct rq *rq, struct task_struct *prev,
 	 * of the scheduler it's an obvious special-case), so we
 	 * do an early lockdep release here:
 	 */
-#ifndef __ARCH_WANT_UNLOCKED_CTXSW
 	spin_release(&rq->lock.dep_map, 1, _THIS_IP_);
-#endif
 
 	context_tracking_task_switch(prev, next);
 	/* Here we just switch the register state and the stack. */
@@ -2447,7 +2489,7 @@ 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)
@@ -2493,7 +2535,7 @@ 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
 
@@ -2656,6 +2698,9 @@ static noinline void __schedule_bug(struct task_struct *prev)
  */
 static inline void schedule_debug(struct task_struct *prev)
 {
+#ifdef CONFIG_SCHED_STACK_END_CHECK
+	BUG_ON(unlikely(task_stack_end_corrupted(prev)));
+#endif
 	/*
 	 * Test if we are atomic. Since do_exit() needs to call into
 	 * schedule() atomically, we ignore that path. Otherwise whine
@@ -2797,7 +2842,7 @@ 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);
@@ -2962,7 +3007,7 @@ 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;
 
@@ -2991,12 +3036,12 @@ 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);
+		put_prev_task(rq, p);
 
 	/*
 	 * Boosting condition are:
@@ -3033,7 +3078,7 @@ 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);
@@ -3044,7 +3089,7 @@ 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;
 
@@ -3065,8 +3110,8 @@ 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);
@@ -3075,7 +3120,7 @@ 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
@@ -3347,7 +3392,7 @@ 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;
@@ -3544,19 +3589,19 @@ 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);
+		put_prev_task(rq, p);
 
 	prev_class = p->sched_class;
 	__setscheduler(rq, p, attr);
 
 	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).
@@ -3980,14 +4025,14 @@ long sched_setaffinity(pid_t pid, const struct cpumask *in_mask)
 		rcu_read_lock();
 		if (!ns_capable(__task_cred(p)->user_ns, CAP_SYS_NICE)) {
 			rcu_read_unlock();
-			goto out_unlock;
+			goto out_free_new_mask;
 		}
 		rcu_read_unlock();
 	}
 
 	retval = security_task_setscheduler(p);
 	if (retval)
-		goto out_unlock;
+		goto out_free_new_mask;
 
 
 	cpuset_cpus_allowed(p, cpus_allowed);
@@ -4000,13 +4045,14 @@ long sched_setaffinity(pid_t pid, const struct cpumask *in_mask)
 	 * root_domain.
 	 */
 #ifdef CONFIG_SMP
-	if (task_has_dl_policy(p)) {
-		const struct cpumask *span = task_rq(p)->rd->span;
-
-		if (dl_bandwidth_enabled() && !cpumask_subset(span, new_mask)) {
+	if (task_has_dl_policy(p) && dl_bandwidth_enabled()) {
+		rcu_read_lock();
+		if (!cpumask_subset(task_rq(p)->rd->span, new_mask)) {
 			retval = -EBUSY;
-			goto out_unlock;
+			rcu_read_unlock();
+			goto out_free_new_mask;
 		}
+		rcu_read_unlock();
 	}
 #endif
 again:
@@ -4024,7 +4070,7 @@ again:
 			goto again;
 		}
 	}
-out_unlock:
+out_free_new_mask:
 	free_cpumask_var(new_mask);
 out_free_cpus_allowed:
 	free_cpumask_var(cpus_allowed);
@@ -4508,7 +4554,7 @@ 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:
@@ -4516,7 +4562,7 @@ 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();
 
@@ -4571,7 +4617,7 @@ 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
@@ -4592,6 +4638,33 @@ void init_idle(struct task_struct *idle, int cpu)
 }
 
 #ifdef CONFIG_SMP
+/*
+ * move_queued_task - move a queued task to new rq.
+ *
+ * Returns (locked) new rq. Old rq's lock is released.
+ */
+static struct rq *move_queued_task(struct task_struct *p, int new_cpu)
+{
+	struct rq *rq = task_rq(p);
+
+	lockdep_assert_held(&rq->lock);
+
+	dequeue_task(rq, p, 0);
+	p->on_rq = TASK_ON_RQ_MIGRATING;
+	set_task_cpu(p, new_cpu);
+	raw_spin_unlock(&rq->lock);
+
+	rq = cpu_rq(new_cpu);
+
+	raw_spin_lock(&rq->lock);
+	BUG_ON(task_cpu(p) != new_cpu);
+	p->on_rq = TASK_ON_RQ_QUEUED;
+	enqueue_task(rq, p, 0);
+	check_preempt_curr(rq, p, 0);
+
+	return rq;
+}
+
 void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask)
 {
 	if (p->sched_class && p->sched_class->set_cpus_allowed)
@@ -4648,14 +4721,15 @@ 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_running(rq, p) || p->state == TASK_WAKING) {
 		struct migration_arg arg = { p, dest_cpu };
 		/* Need help from migration thread: drop lock and wait. */
 		task_rq_unlock(rq, p, &flags);
 		stop_one_cpu(cpu_of(rq), migration_cpu_stop, &arg);
 		tlb_migrate_finish(p->mm);
 		return 0;
-	}
+	} else if (task_on_rq_queued(p))
+		rq = move_queued_task(p, dest_cpu);
 out:
 	task_rq_unlock(rq, p, &flags);
 
@@ -4676,20 +4750,20 @@ 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;
@@ -4698,16 +4772,12 @@ 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);
-		set_task_cpu(p, dest_cpu);
-		enqueue_task(rq_dest, p, 0);
-		check_preempt_curr(rq_dest, p, 0);
-	}
+	if (task_on_rq_queued(p))
+		rq = move_queued_task(p, dest_cpu);
 done:
 	ret = 1;
 fail:
-	double_rq_unlock(rq_src, rq_dest);
+	raw_spin_unlock(&rq->lock);
 	raw_spin_unlock(&p->pi_lock);
 	return ret;
 }
@@ -4739,22 +4809,22 @@ 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);
+		put_prev_task(rq, p);
 
 	p->numa_preferred_nid = 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);
 }
@@ -4774,6 +4844,12 @@ static int migration_cpu_stop(void *data)
 	 * be on another cpu but it doesn't matter.
 	 */
 	local_irq_disable();
+	/*
+	 * We need to explicitly wake pending tasks before running
+	 * __migrate_task() such that we will not miss enforcing cpus_allowed
+	 * during wakeups, see set_cpus_allowed_ptr()'s TASK_WAKING test.
+	 */
+	sched_ttwu_pending();
 	__migrate_task(arg->task, raw_smp_processor_id(), arg->dest_cpu);
 	local_irq_enable();
 	return 0;
@@ -5184,6 +5260,7 @@ static int sched_cpu_inactive(struct notifier_block *nfb,
 {
 	unsigned long flags;
 	long cpu = (long)hcpu;
+	struct dl_bw *dl_b;
 
 	switch (action & ~CPU_TASKS_FROZEN) {
 	case CPU_DOWN_PREPARE:
@@ -5191,15 +5268,19 @@ static int sched_cpu_inactive(struct notifier_block *nfb,
 
 		/* explicitly allow suspend */
 		if (!(action & CPU_TASKS_FROZEN)) {
-			struct dl_bw *dl_b = dl_bw_of(cpu);
 			bool overflow;
 			int cpus;
 
+			rcu_read_lock_sched();
+			dl_b = dl_bw_of(cpu);
+
 			raw_spin_lock_irqsave(&dl_b->lock, flags);
 			cpus = dl_bw_cpus(cpu);
 			overflow = __dl_overflow(dl_b, cpus, 0, 0);
 			raw_spin_unlock_irqrestore(&dl_b->lock, flags);
 
+			rcu_read_unlock_sched();
+
 			if (overflow)
 				return notifier_from_errno(-EBUSY);
 		}
@@ -5742,7 +5823,7 @@ 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);
@@ -5753,10 +5834,10 @@ 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(),
@@ -5766,10 +5847,9 @@ 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);
@@ -7120,13 +7200,13 @@ 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);
 	}
@@ -7140,12 +7220,12 @@ void normalize_rt_tasks(void)
 	unsigned long flags;
 	struct rq *rq;
 
-	read_lock_irqsave(&tasklist_lock, flags);
-	do_each_thread(g, p) {
+	read_lock(&tasklist_lock);
+	for_each_process_thread(g, p) {
 		/*
 		 * Only normalize user tasks:
 		 */
-		if (!p->mm)
+		if (p->flags & PF_KTHREAD)
 			continue;
 
 		p->se.exec_start		= 0;
@@ -7160,21 +7240,16 @@ void normalize_rt_tasks(void)
 			 * Renice negative nice level userspace
 			 * tasks back to 0:
 			 */
-			if (task_nice(p) < 0 && p->mm)
+			if (task_nice(p) < 0)
 				set_user_nice(p, 0);
 			continue;
 		}
 
-		raw_spin_lock(&p->pi_lock);
-		rq = __task_rq_lock(p);
-
+		rq = task_rq_lock(p, &flags);
 		normalize_task(rq, p);
-
-		__task_rq_unlock(rq);
-		raw_spin_unlock(&p->pi_lock);
-	} while_each_thread(g, p);
-
-	read_unlock_irqrestore(&tasklist_lock, flags);
+		task_rq_unlock(rq, p, &flags);
+	}
+	read_unlock(&tasklist_lock);
 }
 
 #endif /* CONFIG_MAGIC_SYSRQ */
@@ -7314,19 +7389,19 @@ 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);
+		put_prev_task(rq, tsk);
 
 	tg = container_of(task_css_check(tsk, cpu_cgrp_id,
 				lockdep_is_held(&tsk->sighand->siglock)),
@@ -7336,14 +7411,14 @@ 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);
@@ -7361,10 +7436,10 @@ static inline int tg_has_rt_tasks(struct task_group *tg)
 {
 	struct task_struct *g, *p;
 
-	do_each_thread(g, p) {
-		if (rt_task(p) && task_rq(p)->rt.tg == tg)
+	for_each_process_thread(g, p) {
+		if (rt_task(p) && task_group(p) == tg)
 			return 1;
-	} while_each_thread(g, p);
+	}
 
 	return 0;
 }
@@ -7573,6 +7648,7 @@ static int sched_dl_global_constraints(void)
 	u64 runtime = global_rt_runtime();
 	u64 period = global_rt_period();
 	u64 new_bw = to_ratio(period, runtime);
+	struct dl_bw *dl_b;
 	int cpu, ret = 0;
 	unsigned long flags;
 
@@ -7586,13 +7662,16 @@ static int sched_dl_global_constraints(void)
 	 * solutions is welcome!
 	 */
 	for_each_possible_cpu(cpu) {
-		struct dl_bw *dl_b = dl_bw_of(cpu);
+		rcu_read_lock_sched();
+		dl_b = dl_bw_of(cpu);
 
 		raw_spin_lock_irqsave(&dl_b->lock, flags);
 		if (new_bw < dl_b->total_bw)
 			ret = -EBUSY;
 		raw_spin_unlock_irqrestore(&dl_b->lock, flags);
 
+		rcu_read_unlock_sched();
+
 		if (ret)
 			break;
 	}
@@ -7603,6 +7682,7 @@ static int sched_dl_global_constraints(void)
 static void sched_dl_do_global(void)
 {
 	u64 new_bw = -1;
+	struct dl_bw *dl_b;
 	int cpu;
 	unsigned long flags;
 
@@ -7616,11 +7696,14 @@ static void sched_dl_do_global(void)
 	 * FIXME: As above...
 	 */
 	for_each_possible_cpu(cpu) {
-		struct dl_bw *dl_b = dl_bw_of(cpu);
+		rcu_read_lock_sched();
+		dl_b = dl_bw_of(cpu);
 
 		raw_spin_lock_irqsave(&dl_b->lock, flags);
 		dl_b->bw = new_bw;
 		raw_spin_unlock_irqrestore(&dl_b->lock, flags);
+
+		rcu_read_unlock_sched();
 	}
 }
 
@@ -8001,7 +8084,7 @@ static int tg_cfs_schedulable_down(struct task_group *tg, void *data)
 		struct cfs_bandwidth *parent_b = &tg->parent->cfs_bandwidth;
 
 		quota = normalize_cfs_quota(tg, d);
-		parent_quota = parent_b->hierarchal_quota;
+		parent_quota = parent_b->hierarchical_quota;
 
 		/*
 		 * ensure max(child_quota) <= parent_quota, inherit when no
@@ -8012,7 +8095,7 @@ static int tg_cfs_schedulable_down(struct task_group *tg, void *data)
 		else if (parent_quota != RUNTIME_INF && quota > parent_quota)
 			return -EINVAL;
 	}
-	cfs_b->hierarchal_quota = quota;
+	cfs_b->hierarchical_quota = quota;
 
 	return 0;
 }

kernel/sched/cpudeadline.c

@@ -107,9 +107,7 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
 	int best_cpu = -1;
 	const struct sched_dl_entity *dl_se = &p->dl;
 
-	if (later_mask && cpumask_and(later_mask, cp->free_cpus,
-			&p->cpus_allowed) && cpumask_and(later_mask,
-			later_mask, cpu_active_mask)) {
+	if (later_mask && cpumask_and(later_mask, later_mask, cp->free_cpus)) {
 		best_cpu = cpumask_any(later_mask);
 		goto out;
 	} else if (cpumask_test_cpu(cpudl_maximum(cp), &p->cpus_allowed) &&

kernel/sched/cputime.c

@@ -288,24 +288,29 @@ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times)
 	struct signal_struct *sig = tsk->signal;
 	cputime_t utime, stime;
 	struct task_struct *t;
-
-	times->utime = sig->utime;
-	times->stime = sig->stime;
-	times->sum_exec_runtime = sig->sum_sched_runtime;
+	unsigned int seq, nextseq;
+	unsigned long flags;
 
 	rcu_read_lock();
-	/* make sure we can trust tsk->thread_group list */
-	if (!likely(pid_alive(tsk)))
-		goto out;
-
-	t = tsk;
+	/* Attempt a lockless read on the first round. */
+	nextseq = 0;
 	do {
-		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:
+		seq = nextseq;
+		flags = read_seqbegin_or_lock_irqsave(&sig->stats_lock, &seq);
+		times->utime = sig->utime;
+		times->stime = sig->stime;
+		times->sum_exec_runtime = sig->sum_sched_runtime;
+
+		for_each_thread(tsk, t) {
+			task_cputime(t, &utime, &stime);
+			times->utime += utime;
+			times->stime += stime;
+			times->sum_exec_runtime += task_sched_runtime(t);
+		}
+		/* If lockless access failed, take the lock. */
+		nextseq = 1;
+	} while (need_seqretry(&sig->stats_lock, seq));
+	done_seqretry_irqrestore(&sig->stats_lock, seq, flags);
 	rcu_read_unlock();
 }
 
@@ -549,6 +554,23 @@ drop_precision:
 	return (__force cputime_t) scaled;
 }
 
+/*
+ * Atomically advance counter to the new value. Interrupts, vcpu
+ * scheduling, and scaling inaccuracies can cause cputime_advance
+ * to be occasionally called with a new value smaller than counter.
+ * Let's enforce atomicity.
+ *
+ * Normally a caller will only go through this loop once, or not
+ * at all in case a previous caller updated counter the same jiffy.
+ */
+static void cputime_advance(cputime_t *counter, cputime_t new)
+{
+	cputime_t old;
+
+	while (new > (old = ACCESS_ONCE(*counter)))
+		cmpxchg_cputime(counter, old, new);
+}
+
 /*
  * Adjust tick based cputime random precision against scheduler
  * runtime accounting.
@@ -594,13 +616,8 @@ static void cputime_adjust(struct task_cputime *curr,
 		utime = rtime - stime;
 	}
 
-	/*
-	 * If the tick based count grows faster than the scheduler one,
-	 * the result of the scaling may go backward.
-	 * Let's enforce monotonicity.
-	 */
-	prev->stime = max(prev->stime, stime);
-	prev->utime = max(prev->utime, utime);
+	cputime_advance(&prev->stime, stime);
+	cputime_advance(&prev->utime, utime);
 
 out:
 	*ut = prev->utime;
@@ -617,9 +634,6 @@ void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
 	cputime_adjust(&cputime, &p->prev_cputime, ut, st);
 }
 
-/*
- * Must be called with siglock held.
- */
 void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
 {
 	struct task_cputime cputime;

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;
 	}
 
@@ -1124,10 +1121,8 @@ static void set_curr_task_dl(struct rq *rq)
 static int pick_dl_task(struct rq *rq, struct task_struct *p, int cpu)
 {
 	if (!task_running(rq, p) &&
-	    (cpu < 0 || cpumask_test_cpu(cpu, &p->cpus_allowed)) &&
-	    (p->nr_cpus_allowed > 1))
+	    cpumask_test_cpu(cpu, tsk_cpus_allowed(p)))
 		return 1;
-
 	return 0;
 }
 
@@ -1169,6 +1164,13 @@ static int find_later_rq(struct task_struct *task)
 	if (task->nr_cpus_allowed == 1)
 		return -1;
 
+	/*
+	 * We have to consider system topology and task affinity
+	 * first, then we can look for a suitable cpu.
+	 */
+	cpumask_copy(later_mask, task_rq(task)->rd->span);
+	cpumask_and(later_mask, later_mask, cpu_active_mask);
+	cpumask_and(later_mask, later_mask, &task->cpus_allowed);
 	best_cpu = cpudl_find(&task_rq(task)->rd->cpudl,
 			task, later_mask);
 	if (best_cpu == -1)
@@ -1257,7 +1259,8 @@ 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 +1299,7 @@ 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 +1446,7 @@ 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
@@ -1569,6 +1572,8 @@ static void switched_from_dl(struct rq *rq, struct task_struct *p)
 	if (hrtimer_active(&p->dl.dl_timer) && !dl_policy(p->policy))
 		hrtimer_try_to_cancel(&p->dl.dl_timer);
 
+	__dl_clear_params(p);
+
 #ifdef CONFIG_SMP
 	/*
 	 * Since this might be the only -deadline task on the rq,
@@ -1596,7 +1601,7 @@ 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 +1619,7 @@ 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

kernel/sched/debug.c

@@ -150,7 +150,6 @@ print_task(struct seq_file *m, struct rq *rq, struct task_struct *p)
 static void print_rq(struct seq_file *m, struct rq *rq, int rq_cpu)
 {
 	struct task_struct *g, *p;
-	unsigned long flags;
 
 	SEQ_printf(m,
 	"\nrunnable tasks:\n"
@@ -159,16 +158,14 @@ 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) {
+	rcu_read_lock();
+	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);
+	}
+	rcu_read_unlock();
 }
 
 void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
@@ -333,9 +330,7 @@ do {									\
 	print_cfs_stats(m, cpu);
 	print_rt_stats(m, cpu);
 
-	rcu_read_lock();
 	print_rq(m, rq, cpu);
-	rcu_read_unlock();
 	spin_unlock_irqrestore(&sched_debug_lock, flags);
 	SEQ_printf(m, "\n");
 }

kernel/sched/fair.c

@@ -23,6 +23,7 @@
 #include <linux/latencytop.h>
 #include <linux/sched.h>
 #include <linux/cpumask.h>
+#include <linux/cpuidle.h>
 #include <linux/slab.h>
 #include <linux/profile.h>
 #include <linux/interrupt.h>
@@ -665,6 +666,7 @@ static u64 sched_vslice(struct cfs_rq *cfs_rq, struct sched_entity *se)
 }
 
 #ifdef CONFIG_SMP
+static int select_idle_sibling(struct task_struct *p, int cpu);
 static unsigned long task_h_load(struct task_struct *p);
 
 static inline void __update_task_entity_contrib(struct sched_entity *se);
@@ -1038,7 +1040,8 @@ 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 +1065,12 @@ 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 +1213,7 @@ 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;
 
@@ -1252,6 +1259,13 @@ balance:
 	if (load_too_imbalanced(src_load, dst_load, env))
 		goto unlock;
 
+	/*
+	 * One idle CPU per node is evaluated for a task numa move.
+	 * Call select_idle_sibling to maybe find a better one.
+	 */
+	if (!cur)
+		env->dst_cpu = select_idle_sibling(env->p, env->dst_cpu);
+
 assign:
 	task_numa_assign(env, cur, imp);
 unlock:
@@ -1775,7 +1789,7 @@ 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);
 	}
 
@@ -1804,10 +1818,6 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags)
 	if (!p->mm)
 		return;
 
-	/* Do not worry about placement if exiting */
-	if (p->state == TASK_DEAD)
-		return;
-
 	/* Allocate buffer to track faults on a per-node basis */
 	if (unlikely(!p->numa_faults_memory)) {
 		int size = sizeof(*p->numa_faults_memory) *
@@ -2211,8 +2221,8 @@ static __always_inline u64 decay_load(u64 val, u64 n)
 
 	/*
 	 * As y^PERIOD = 1/2, we can combine
-	 *    y^n = 1/2^(n/PERIOD) * k^(n%PERIOD)
-	 * With a look-up table which covers k^n (n<PERIOD)
+	 *    y^n = 1/2^(n/PERIOD) * y^(n%PERIOD)
+	 * With a look-up table which covers y^n (n<PERIOD)
 	 *
 	 * To achieve constant time decay_load.
 	 */
@@ -2377,6 +2387,9 @@ static inline void __update_cfs_rq_tg_load_contrib(struct cfs_rq *cfs_rq,
 	tg_contrib = cfs_rq->runnable_load_avg + cfs_rq->blocked_load_avg;
 	tg_contrib -= cfs_rq->tg_load_contrib;
 
+	if (!tg_contrib)
+		return;
+
 	if (force_update || abs(tg_contrib) > cfs_rq->tg_load_contrib / 8) {
 		atomic_long_add(tg_contrib, &tg->load_avg);
 		cfs_rq->tg_load_contrib += tg_contrib;
@@ -3892,14 +3905,6 @@ 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);
 	}
 }
@@ -4087,7 +4092,7 @@ static unsigned long capacity_of(int cpu)
 static unsigned long cpu_avg_load_per_task(int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
-	unsigned long nr_running = ACCESS_ONCE(rq->nr_running);
+	unsigned long nr_running = ACCESS_ONCE(rq->cfs.h_nr_running);
 	unsigned long load_avg = rq->cfs.runnable_load_avg;
 
 	if (nr_running)
@@ -4276,8 +4281,8 @@ static int wake_wide(struct task_struct *p)
 static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
 {
 	s64 this_load, load;
+	s64 this_eff_load, prev_eff_load;
 	int idx, this_cpu, prev_cpu;
-	unsigned long tl_per_task;
 	struct task_group *tg;
 	unsigned long weight;
 	int balanced;
@@ -4320,47 +4325,30 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
 	 * Otherwise check if either cpus are near enough in load to allow this
 	 * task to be woken on this_cpu.
 	 */
-	if (this_load > 0) {
-		s64 this_eff_load, prev_eff_load;
+	this_eff_load = 100;
+	this_eff_load *= capacity_of(prev_cpu);
 
-		this_eff_load = 100;
-		this_eff_load *= capacity_of(prev_cpu);
+	prev_eff_load = 100 + (sd->imbalance_pct - 100) / 2;
+	prev_eff_load *= capacity_of(this_cpu);
+
+	if (this_load > 0) {
 		this_eff_load *= this_load +
 			effective_load(tg, this_cpu, weight, weight);
 
-		prev_eff_load = 100 + (sd->imbalance_pct - 100) / 2;
-		prev_eff_load *= capacity_of(this_cpu);
 		prev_eff_load *= load + effective_load(tg, prev_cpu, 0, weight);
+	}
 
-		balanced = this_eff_load <= prev_eff_load;
-	} else
-		balanced = true;
-
-	/*
-	 * If the currently running task will sleep within
-	 * a reasonable amount of time then attract this newly
-	 * woken task:
-	 */
-	if (sync && balanced)
-		return 1;
+	balanced = this_eff_load <= prev_eff_load;
 
 	schedstat_inc(p, se.statistics.nr_wakeups_affine_attempts);
-	tl_per_task = cpu_avg_load_per_task(this_cpu);
 
-	if (balanced ||
-	    (this_load <= load &&
-	     this_load + target_load(prev_cpu, idx) <= tl_per_task)) {
-		/*
-		 * This domain has SD_WAKE_AFFINE and
-		 * p is cache cold in this domain, and
-		 * there is no bad imbalance.
-		 */
-		schedstat_inc(sd, ttwu_move_affine);
-		schedstat_inc(p, se.statistics.nr_wakeups_affine);
+	if (!balanced)
+		return 0;
 
-		return 1;
-	}
-	return 0;
+	schedstat_inc(sd, ttwu_move_affine);
+	schedstat_inc(p, se.statistics.nr_wakeups_affine);
+
+	return 1;
 }
 
 /*
@@ -4428,20 +4416,46 @@ static int
 find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 {
 	unsigned long load, min_load = ULONG_MAX;
-	int idlest = -1;
+	unsigned int min_exit_latency = UINT_MAX;
+	u64 latest_idle_timestamp = 0;
+	int least_loaded_cpu = this_cpu;
+	int shallowest_idle_cpu = -1;
 	int i;
 
 	/* Traverse only the allowed CPUs */
 	for_each_cpu_and(i, sched_group_cpus(group), tsk_cpus_allowed(p)) {
-		load = weighted_cpuload(i);
-
-		if (load < min_load || (load == min_load && i == this_cpu)) {
-			min_load = load;
-			idlest = i;
+		if (idle_cpu(i)) {
+			struct rq *rq = cpu_rq(i);
+			struct cpuidle_state *idle = idle_get_state(rq);
+			if (idle && idle->exit_latency < min_exit_latency) {
+				/*
+				 * We give priority to a CPU whose idle state
+				 * has the smallest exit latency irrespective
+				 * of any idle timestamp.
+				 */
+				min_exit_latency = idle->exit_latency;
+				latest_idle_timestamp = rq->idle_stamp;
+				shallowest_idle_cpu = i;
+			} else if ((!idle || idle->exit_latency == min_exit_latency) &&
+				   rq->idle_stamp > latest_idle_timestamp) {
+				/*
+				 * If equal or no active idle state, then
+				 * the most recently idled CPU might have
+				 * a warmer cache.
+				 */
+				latest_idle_timestamp = rq->idle_stamp;
+				shallowest_idle_cpu = i;
+			}
+		} else {
+			load = weighted_cpuload(i);
+			if (load < min_load || (load == min_load && i == this_cpu)) {
+				min_load = load;
+				least_loaded_cpu = i;
+			}
 		}
 	}
 
-	return idlest;
+	return shallowest_idle_cpu != -1 ? shallowest_idle_cpu : least_loaded_cpu;
 }
 
 /*
@@ -4513,11 +4527,8 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f
 	if (p->nr_cpus_allowed == 1)
 		return prev_cpu;
 
-	if (sd_flag & SD_BALANCE_WAKE) {
-		if (cpumask_test_cpu(cpu, tsk_cpus_allowed(p)))
-			want_affine = 1;
-		new_cpu = prev_cpu;
-	}
+	if (sd_flag & SD_BALANCE_WAKE)
+		want_affine = cpumask_test_cpu(cpu, tsk_cpus_allowed(p));
 
 	rcu_read_lock();
 	for_each_domain(cpu, tmp) {
@@ -4704,7 +4715,7 @@ 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.
@@ -5112,20 +5123,9 @@ 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:
  */
@@ -5133,6 +5133,8 @@ 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 +5254,9 @@ 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
@@ -5310,24 +5315,12 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env)
 	if (!tsk_cache_hot)
 		tsk_cache_hot = migrate_degrades_locality(p, env);
 
-	if (migrate_improves_locality(p, env)) {
-#ifdef CONFIG_SCHEDSTATS
+	if (migrate_improves_locality(p, env) || !tsk_cache_hot ||
+	    env->sd->nr_balance_failed > env->sd->cache_nice_tries) {
 		if (tsk_cache_hot) {
 			schedstat_inc(env->sd, lb_hot_gained[env->idle]);
 			schedstat_inc(p, se.statistics.nr_forced_migrations);
 		}
-#endif
-		return 1;
-	}
-
-	if (!tsk_cache_hot ||
-		env->sd->nr_balance_failed > env->sd->cache_nice_tries) {
-
-		if (tsk_cache_hot) {
-			schedstat_inc(env->sd, lb_hot_gained[env->idle]);
-			schedstat_inc(p, se.statistics.nr_forced_migrations);
-		}
-
 		return 1;
 	}
 
@@ -5336,47 +5329,63 @@ 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
- * part of active balancing operations within "domain".
- * Returns 1 if successful and 0 otherwise.
- *
- * Called with both runqueues locked.
+ * detach_task() -- detach the task for the migration specified in env
  */
-static int move_one_task(struct lb_env *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 a task if successful and NULL otherwise.
+ */
+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 +5416,16 @@ 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 +5445,58 @@ 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 pulled;
+	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);
+
+	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 +5615,13 @@ 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 +5635,7 @@ 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 +5673,8 @@ 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,
 		},
 	};
 }
@@ -5652,19 +5717,17 @@ unsigned long __weak arch_scale_freq_capacity(struct sched_domain *sd, int cpu)
 	return default_scale_capacity(sd, cpu);
 }
 
-static unsigned long default_scale_smt_capacity(struct sched_domain *sd, int cpu)
+static unsigned long default_scale_cpu_capacity(struct sched_domain *sd, int cpu)
 {
-	unsigned long weight = sd->span_weight;
-	unsigned long smt_gain = sd->smt_gain;
+	if ((sd->flags & SD_SHARE_CPUCAPACITY) && (sd->span_weight > 1))
+		return sd->smt_gain / sd->span_weight;
 
-	smt_gain /= weight;
-
-	return smt_gain;
+	return SCHED_CAPACITY_SCALE;
 }
 
-unsigned long __weak arch_scale_smt_capacity(struct sched_domain *sd, int cpu)
+unsigned long __weak arch_scale_cpu_capacity(struct sched_domain *sd, int cpu)
 {
-	return default_scale_smt_capacity(sd, cpu);
+	return default_scale_cpu_capacity(sd, cpu);
 }
 
 static unsigned long scale_rt_capacity(int cpu)
@@ -5703,18 +5766,15 @@ static unsigned long scale_rt_capacity(int cpu)
 
 static void update_cpu_capacity(struct sched_domain *sd, int cpu)
 {
-	unsigned long weight = sd->span_weight;
 	unsigned long capacity = SCHED_CAPACITY_SCALE;
 	struct sched_group *sdg = sd->groups;
 
-	if ((sd->flags & SD_SHARE_CPUCAPACITY) && weight > 1) {
-		if (sched_feat(ARCH_CAPACITY))
-			capacity *= arch_scale_smt_capacity(sd, cpu);
-		else
-			capacity *= default_scale_smt_capacity(sd, cpu);
+	if (sched_feat(ARCH_CAPACITY))
+		capacity *= arch_scale_cpu_capacity(sd, cpu);
+	else
+		capacity *= default_scale_cpu_capacity(sd, cpu);
 
-		capacity >>= SCHED_CAPACITY_SHIFT;
-	}
+	capacity >>= SCHED_CAPACITY_SHIFT;
 
 	sdg->sgc->capacity_orig = capacity;
 
@@ -5891,6 +5951,18 @@ 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.
@@ -5920,7 +5992,7 @@ static inline void update_sg_lb_stats(struct lb_env *env,
 			load = source_load(i, load_idx);
 
 		sgs->group_load += load;
-		sgs->sum_nr_running += rq->nr_running;
+		sgs->sum_nr_running += rq->cfs.h_nr_running;
 
 		if (rq->nr_running > 1)
 			*overload = true;
@@ -5942,9 +6014,8 @@ 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 +6039,19 @@ 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 +6059,7 @@ 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 +6304,7 @@ 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 +6324,11 @@ 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 +6412,7 @@ 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 */
@@ -6346,7 +6421,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 		goto force_balance;
 
 	/*
-	 * If the local group is more busy than the selected busiest group
+	 * If the local group is busier than the selected busiest group
 	 * don't try and pull any tasks.
 	 */
 	if (local->avg_load >= busiest->avg_load)
@@ -6361,13 +6436,14 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 
 	if (env->idle == CPU_IDLE) {
 		/*
-		 * This cpu is idle. If the busiest group load doesn't
-		 * have more tasks than the number of available cpu's and
-		 * there is no imbalance between this and busiest group
-		 * wrt to idle cpu's, it is balanced.
+		 * This cpu is idle. If the busiest group is not overloaded
+		 * and there is no imbalance between this and busiest group
+		 * wrt idle cpus, it is balanced. The imbalance becomes
+		 * significant if the diff is greater than 1 otherwise we
+		 * might end up to just move the imbalance on another group
 		 */
-		if ((local->idle_cpus < busiest->idle_cpus) &&
-		    busiest->sum_nr_running <= busiest->group_weight)
+		if ((busiest->group_type != group_overloaded) &&
+				(local->idle_cpus <= (busiest->idle_cpus + 1)))
 			goto out_balanced;
 	} else {
 		/*
@@ -6550,6 +6626,7 @@ 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),
 	};
 
 	/*
@@ -6599,23 +6676,30 @@ 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);
-		local_irq_restore(flags);
+		cur_ld_moved = detach_tasks(&env);
 
 		/*
-		 * some other cpu did the load balance for us.
+		 * 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.
 		 */
-		if (cur_ld_moved && env.dst_cpu != smp_processor_id())
-			resched_cpu(env.dst_cpu);
+
+		raw_spin_unlock(&busiest->lock);
+
+		if (cur_ld_moved) {
+			attach_tasks(&env);
+			ld_moved += cur_ld_moved;
+		}
+
+		local_irq_restore(flags);
 
 		if (env.flags & LBF_NEED_BREAK) {
 			env.flags &= ~LBF_NEED_BREAK;
@@ -6665,10 +6749,8 @@ more_balance:
 		if (sd_parent) {
 			int *group_imbalance = &sd_parent->groups->sgc->imbalance;
 
-			if ((env.flags & LBF_SOME_PINNED) && env.imbalance > 0) {
+			if ((env.flags & LBF_SOME_PINNED) && env.imbalance > 0)
 				*group_imbalance = 1;
-			} else if (*group_imbalance)
-				*group_imbalance = 0;
 		}
 
 		/* All tasks on this runqueue were pinned by CPU affinity */
@@ -6679,7 +6761,7 @@ more_balance:
 				env.loop_break = sched_nr_migrate_break;
 				goto redo;
 			}
-			goto out_balanced;
+			goto out_all_pinned;
 		}
 	}
 
@@ -6744,7 +6826,7 @@ 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;
@@ -6753,6 +6835,23 @@ more_balance:
 	goto out;
 
 out_balanced:
+	/*
+	 * We reach balance although we may have faced some affinity
+	 * constraints. Clear the imbalance flag if it was set.
+	 */
+	if (sd_parent) {
+		int *group_imbalance = &sd_parent->groups->sgc->imbalance;
+
+		if (*group_imbalance)
+			*group_imbalance = 0;
+	}
+
+out_all_pinned:
+	/*
+	 * We reach balance because all tasks are pinned at this level so
+	 * we can't migrate them. Let the imbalance flag set so parent level
+	 * can try to migrate them.
+	 */
 	schedstat_inc(sd, lb_balanced[idle]);
 
 	sd->nr_balance_failed = 0;
@@ -6914,6 +7013,7 @@ 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 +7033,6 @@ 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 +7053,22 @@ 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 +7568,7 @@ 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 +7593,11 @@ 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 +7624,15 @@ 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 +7678,7 @@ 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 +7697,7 @@ 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 +7708,14 @@ 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

kernel/sched/idle.c

@@ -147,6 +147,9 @@ use_default:
 	    clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &dev->cpu))
 		goto use_default;
 
+	/* Take note of the planned idle state. */
+	idle_set_state(this_rq(), &drv->states[next_state]);
+
 	/*
 	 * Enter the idle state previously returned by the governor decision.
 	 * This function will block until an interrupt occurs and will take
@@ -154,6 +157,9 @@ use_default:
 	 */
 	entered_state = cpuidle_enter(drv, dev, next_state);
 
+	/* The cpu is no longer idle or about to enter idle. */
+	idle_set_state(this_rq(), NULL);
+
 	if (broadcast)
 		clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_EXIT, &dev->cpu);
 

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;
 	}
@@ -1468,8 +1468,7 @@ pick_next_task_rt(struct rq *rq, struct task_struct *prev)
 	p = _pick_next_task_rt(rq);
 
 	/* The running task is never eligible for pushing */
-	if (p)
-		dequeue_pushable_task(rq, p);
+	dequeue_pushable_task(rq, p);
 
 	set_post_schedule(rq);
 
@@ -1624,7 +1623,7 @@ 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 +1657,7 @@ 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 +1808,7 @@ 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 +1869,7 @@ 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 +1935,7 @@ 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 +1969,7 @@ 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 +1988,7 @@ 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) {
@@ -2073,7 +2072,7 @@ static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued)
 	for_each_sched_rt_entity(rt_se) {
 		if (rt_se->run_list.prev != rt_se->run_list.next) {
 			requeue_task_rt(rq, p, 0);
-			set_tsk_need_resched(p);
+			resched_curr(rq);
 			return;
 		}
 	}

kernel/sched/sched.h

@@ -14,6 +14,11 @@
 #include "cpuacct.h"
 
 struct rq;
+struct cpuidle_state;
+
+/* task_struct::on_rq states: */
+#define TASK_ON_RQ_QUEUED	1
+#define TASK_ON_RQ_MIGRATING	2
 
 extern __read_mostly int scheduler_running;
 
@@ -126,6 +131,9 @@ struct rt_bandwidth {
 	u64			rt_runtime;
 	struct hrtimer		rt_period_timer;
 };
+
+void __dl_clear_params(struct task_struct *p);
+
 /*
  * To keep the bandwidth of -deadline tasks and groups under control
  * we need some place where:
@@ -184,7 +192,7 @@ struct cfs_bandwidth {
 	raw_spinlock_t lock;
 	ktime_t period;
 	u64 quota, runtime;
-	s64 hierarchal_quota;
+	s64 hierarchical_quota;
 	u64 runtime_expires;
 
 	int idle, timer_active;
@@ -636,6 +644,11 @@ struct rq {
 #ifdef CONFIG_SMP
 	struct llist_head wake_list;
 #endif
+
+#ifdef CONFIG_CPU_IDLE
+	/* Must be inspected within a rcu lock section */
+	struct cpuidle_state *idle_state;
+#endif
 };
 
 static inline int cpu_of(struct rq *rq)
@@ -647,7 +660,7 @@ 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 +955,15 @@ 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)
@@ -953,7 +975,6 @@ static inline int task_running(struct rq *rq, struct task_struct *p)
 # define finish_arch_post_lock_switch()	do { } while (0)
 #endif
 
-#ifndef __ARCH_WANT_UNLOCKED_CTXSW
 static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next)
 {
 #ifdef CONFIG_SMP
@@ -991,35 +1012,6 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev)
 	raw_spin_unlock_irq(&rq->lock);
 }
 
-#else /* __ARCH_WANT_UNLOCKED_CTXSW */
-static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next)
-{
-#ifdef CONFIG_SMP
-	/*
-	 * We can optimise this out completely for !SMP, because the
-	 * SMP rebalancing from interrupt is the only thing that cares
-	 * here.
-	 */
-	next->on_cpu = 1;
-#endif
-	raw_spin_unlock(&rq->lock);
-}
-
-static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev)
-{
-#ifdef CONFIG_SMP
-	/*
-	 * After ->on_cpu is cleared, the task can be moved to a different CPU.
-	 * We must ensure this doesn't happen until the switch is completely
-	 * finished.
-	 */
-	smp_wmb();
-	prev->on_cpu = 0;
-#endif
-	local_irq_enable();
-}
-#endif /* __ARCH_WANT_UNLOCKED_CTXSW */
-
 /*
  * wake flags
  */
@@ -1180,6 +1172,30 @@ static inline void idle_exit_fair(struct rq *rq) { }
 
 #endif
 
+#ifdef CONFIG_CPU_IDLE
+static inline void idle_set_state(struct rq *rq,
+				  struct cpuidle_state *idle_state)
+{
+	rq->idle_state = idle_state;
+}
+
+static inline struct cpuidle_state *idle_get_state(struct rq *rq)
+{
+	WARN_ON(!rcu_read_lock_held());
+	return rq->idle_state;
+}
+#else
+static inline void idle_set_state(struct rq *rq,
+				  struct cpuidle_state *idle_state)
+{
+}
+
+static inline struct cpuidle_state *idle_get_state(struct rq *rq)
+{
+	return NULL;
+}
+#endif
+
 extern void sysrq_sched_debug_show(void);
 extern void sched_init_granularity(void);
 extern void update_max_interval(void);

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);