Merge branch 'next' into for-linus

Prepare first set of updates for 3.10 merge window.

10 files changed, 1846 insertions, 405 deletions

diff --git a/kernel/sched/auto_group.c b/kernel/sched/auto_group.c
index 0984a21076a..64de5f8b0c9 100644
--- a/kernel/sched/auto_group.c
+++ b/kernel/sched/auto_group.c
@@ -35,6 +35,7 @@ static inline void autogroup_destroy(struct kref *kref)
 	ag->tg->rt_se = NULL;
 	ag->tg->rt_rq = NULL;
 #endif
+	sched_offline_group(ag->tg);
 	sched_destroy_group(ag->tg);
 }
 
@@ -76,6 +77,8 @@ static inline struct autogroup *autogroup_create(void)
 	if (IS_ERR(tg))
 		goto out_free;
 
+	sched_online_group(tg, &root_task_group);
+
 	kref_init(&ag->kref);
 	init_rwsem(&ag->lock);
 	ag->id = atomic_inc_return(&autogroup_seq_nr);
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 2d8927fda71..7f12624a393 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -72,6 +72,7 @@
 #include <linux/slab.h>
 #include <linux/init_task.h>
 #include <linux/binfmts.h>
+#include <linux/context_tracking.h>
 
 #include <asm/switch_to.h>
 #include <asm/tlb.h>
@@ -82,7 +83,7 @@
 #endif
 
 #include "sched.h"
-#include "../workqueue_sched.h"
+#include "../workqueue_internal.h"
 #include "../smpboot.h"
 
 #define CREATE_TRACE_POINTS
@@ -192,23 +193,10 @@ static void sched_feat_disable(int i) { };
 static void sched_feat_enable(int i) { };
 #endif /* HAVE_JUMP_LABEL */
 
-static ssize_t
-sched_feat_write(struct file *filp, const char __user *ubuf,
-		size_t cnt, loff_t *ppos)
+static int sched_feat_set(char *cmp)
 {
-	char buf[64];
-	char *cmp;
-	int neg = 0;
 	int i;
-
-	if (cnt > 63)
-		cnt = 63;
-
-	if (copy_from_user(&buf, ubuf, cnt))
-		return -EFAULT;
-
-	buf[cnt] = 0;
-	cmp = strstrip(buf);
+	int neg = 0;
 
 	if (strncmp(cmp, "NO_", 3) == 0) {
 		neg = 1;
@@ -228,6 +216,27 @@ sched_feat_write(struct file *filp, const char __user *ubuf,
 		}
 	}
 
+	return i;
+}
+
+static ssize_t
+sched_feat_write(struct file *filp, const char __user *ubuf,
+		size_t cnt, loff_t *ppos)
+{
+	char buf[64];
+	char *cmp;
+	int i;
+
+	if (cnt > 63)
+		cnt = 63;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	buf[cnt] = 0;
+	cmp = strstrip(buf);
+
+	i = sched_feat_set(cmp);
 	if (i == __SCHED_FEAT_NR)
 		return -EINVAL;
 
@@ -922,6 +931,13 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags)
 		rq->skip_clock_update = 1;
 }
 
+static ATOMIC_NOTIFIER_HEAD(task_migration_notifier);
+
+void register_task_migration_notifier(struct notifier_block *n)
+{
+	atomic_notifier_chain_register(&task_migration_notifier, n);
+}
+
 #ifdef CONFIG_SMP
 void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
 {
@@ -952,8 +968,18 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
 	trace_sched_migrate_task(p, new_cpu);
 
 	if (task_cpu(p) != new_cpu) {
+		struct task_migration_notifier tmn;
+
+		if (p->sched_class->migrate_task_rq)
+			p->sched_class->migrate_task_rq(p, new_cpu);
 		p->se.nr_migrations++;
 		perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, 1, NULL, 0);
+
+		tmn.task = p;
+		tmn.from_cpu = task_cpu(p);
+		tmn.to_cpu = new_cpu;
+
+		atomic_notifier_call_chain(&task_migration_notifier, 0, &tmn);
 	}
 
 	__set_task_cpu(p, new_cpu);
@@ -1106,18 +1132,28 @@ EXPORT_SYMBOL_GPL(kick_process);
  */
 static int select_fallback_rq(int cpu, struct task_struct *p)
 {
-	const struct cpumask *nodemask = cpumask_of_node(cpu_to_node(cpu));
+	int nid = cpu_to_node(cpu);
+	const struct cpumask *nodemask = NULL;
 	enum { cpuset, possible, fail } state = cpuset;
 	int dest_cpu;
 
-	/* Look for allowed, online CPU in same node. */
-	for_each_cpu(dest_cpu, nodemask) {
-		if (!cpu_online(dest_cpu))
-			continue;
-		if (!cpu_active(dest_cpu))
-			continue;
-		if (cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p)))
-			return dest_cpu;
+	/*
+	 * If the node that the cpu is on has been offlined, cpu_to_node()
+	 * will return -1. There is no cpu on the node, and we should
+	 * select the cpu on the other node.
+	 */
+	if (nid != -1) {
+		nodemask = cpumask_of_node(nid);
+
+		/* Look for allowed, online CPU in same node. */
+		for_each_cpu(dest_cpu, nodemask) {
+			if (!cpu_online(dest_cpu))
+				continue;
+			if (!cpu_active(dest_cpu))
+				continue;
+			if (cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p)))
+				return dest_cpu;
+		}
 	}
 
 	for (;;) {
@@ -1497,7 +1533,8 @@ out:
  */
 int wake_up_process(struct task_struct *p)
 {
-	return try_to_wake_up(p, TASK_ALL, 0);
+	WARN_ON(task_is_stopped_or_traced(p));
+	return try_to_wake_up(p, TASK_NORMAL, 0);
 }
 EXPORT_SYMBOL(wake_up_process);
 
@@ -1524,6 +1561,15 @@ static void __sched_fork(struct task_struct *p)
 	p->se.vruntime			= 0;
 	INIT_LIST_HEAD(&p->se.group_node);
 
+/*
+ * Load-tracking only depends on SMP, FAIR_GROUP_SCHED dependency below may be
+ * removed when useful for applications beyond shares distribution (e.g.
+ * load-balance).
+ */
+#if defined(CONFIG_SMP) && defined(CONFIG_FAIR_GROUP_SCHED)
+	p->se.avg.runnable_avg_period = 0;
+	p->se.avg.runnable_avg_sum = 0;
+#endif
 #ifdef CONFIG_SCHEDSTATS
 	memset(&p->se.statistics, 0, sizeof(p->se.statistics));
 #endif
@@ -1533,7 +1579,40 @@ static void __sched_fork(struct task_struct *p)
 #ifdef CONFIG_PREEMPT_NOTIFIERS
 	INIT_HLIST_HEAD(&p->preempt_notifiers);
 #endif
+
+#ifdef CONFIG_NUMA_BALANCING
+	if (p->mm && atomic_read(&p->mm->mm_users) == 1) {
+		p->mm->numa_next_scan = jiffies;
+		p->mm->numa_next_reset = jiffies;
+		p->mm->numa_scan_seq = 0;
+	}
+
+	p->node_stamp = 0ULL;
+	p->numa_scan_seq = p->mm ? p->mm->numa_scan_seq : 0;
+	p->numa_migrate_seq = p->mm ? p->mm->numa_scan_seq - 1 : 0;
+	p->numa_scan_period = sysctl_numa_balancing_scan_delay;
+	p->numa_work.next = &p->numa_work;
+#endif /* CONFIG_NUMA_BALANCING */
+}
+
+#ifdef CONFIG_NUMA_BALANCING
+#ifdef CONFIG_SCHED_DEBUG
+void set_numabalancing_state(bool enabled)
+{
+	if (enabled)
+		sched_feat_set("NUMA");
+	else
+		sched_feat_set("NO_NUMA");
+}
+#else
+__read_mostly bool numabalancing_enabled;
+
+void set_numabalancing_state(bool enabled)
+{
+	numabalancing_enabled = enabled;
 }
+#endif /* CONFIG_SCHED_DEBUG */
+#endif /* CONFIG_NUMA_BALANCING */
 
 /*
  * fork()/clone()-time setup:
@@ -1673,9 +1752,8 @@ EXPORT_SYMBOL_GPL(preempt_notifier_unregister);
 static void fire_sched_in_preempt_notifiers(struct task_struct *curr)
 {
 	struct preempt_notifier *notifier;
-	struct hlist_node *node;
 
-	hlist_for_each_entry(notifier, node, &curr->preempt_notifiers, link)
+	hlist_for_each_entry(notifier, &curr->preempt_notifiers, link)
 		notifier->ops->sched_in(notifier, raw_smp_processor_id());
 }
 
@@ -1684,9 +1762,8 @@ fire_sched_out_preempt_notifiers(struct task_struct *curr,
 				 struct task_struct *next)
 {
 	struct preempt_notifier *notifier;
-	struct hlist_node *node;
 
-	hlist_for_each_entry(notifier, node, &curr->preempt_notifiers, link)
+	hlist_for_each_entry(notifier, &curr->preempt_notifiers, link)
 		notifier->ops->sched_out(notifier, next);
 }
 
@@ -1886,8 +1963,8 @@ context_switch(struct rq *rq, struct task_struct *prev,
 	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. */
-	rcu_switch(prev, next);
 	switch_to(prev, next, prev);
 
 	barrier();
@@ -1900,11 +1977,10 @@ context_switch(struct rq *rq, struct task_struct *prev,
 }
 
 /*
- * nr_running, nr_uninterruptible and nr_context_switches:
+ * nr_running and nr_context_switches:
  *
  * externally visible scheduler statistics: current number of runnable
- * threads, current number of uninterruptible-sleeping threads, total
- * number of context switches performed since bootup.
+ * threads, total number of context switches performed since bootup.
  */
 unsigned long nr_running(void)
 {
@@ -1916,23 +1992,6 @@ unsigned long nr_running(void)
 	return sum;
 }
 
-unsigned long nr_uninterruptible(void)
-{
-	unsigned long i, sum = 0;
-
-	for_each_possible_cpu(i)
-		sum += cpu_rq(i)->nr_uninterruptible;
-
-	/*
-	 * Since we read the counters lockless, it might be slightly
-	 * inaccurate. Do not allow it to go below zero though:
-	 */
-	if (unlikely((long)sum < 0))
-		sum = 0;
-
-	return sum;
-}
-
 unsigned long long nr_context_switches(void)
 {
 	int i;
@@ -2717,7 +2776,7 @@ static noinline void __schedule_bug(struct task_struct *prev)
 	if (irqs_disabled())
 		print_irqtrace_events(prev);
 	dump_stack();
-	add_taint(TAINT_WARN);
+	add_taint(TAINT_WARN, LOCKDEP_STILL_OK);
 }
 
 /*
@@ -2911,7 +2970,7 @@ asmlinkage void __sched schedule(void)
 }
 EXPORT_SYMBOL(schedule);
 
-#ifdef CONFIG_RCU_USER_QS
+#ifdef CONFIG_CONTEXT_TRACKING
 asmlinkage void __sched schedule_user(void)
 {
 	/*
@@ -2920,9 +2979,9 @@ asmlinkage void __sched schedule_user(void)
 	 * we haven't yet exited the RCU idle mode. Do it here manually until
 	 * we find a better solution.
 	 */
-	rcu_user_exit();
+	user_exit();
 	schedule();
-	rcu_user_enter();
+	user_enter();
 }
 #endif
 
@@ -3027,7 +3086,7 @@ asmlinkage void __sched preempt_schedule_irq(void)
 	/* Catch callers which need to be fixed */
 	BUG_ON(ti->preempt_count || !irqs_disabled());
 
-	rcu_user_exit();
+	user_exit();
 	do {
 		add_preempt_count(PREEMPT_ACTIVE);
 		local_irq_enable();
@@ -3199,7 +3258,8 @@ void complete_all(struct completion *x)
 EXPORT_SYMBOL(complete_all);
 
 static inline long __sched
-do_wait_for_common(struct completion *x, long timeout, int state)
+do_wait_for_common(struct completion *x,
+		   long (*action)(long), long timeout, int state)
 {
 	if (!x->done) {
 		DECLARE_WAITQUEUE(wait, current);
@@ -3212,7 +3272,7 @@ do_wait_for_common(struct completion *x, long timeout, int state)
 			}
 			__set_current_state(state);
 			spin_unlock_irq(&x->wait.lock);
-			timeout = schedule_timeout(timeout);
+			timeout = action(timeout);
 			spin_lock_irq(&x->wait.lock);
 		} while (!x->done && timeout);
 		__remove_wait_queue(&x->wait, &wait);
@@ -3223,17 +3283,30 @@ do_wait_for_common(struct completion *x, long timeout, int state)
 	return timeout ?: 1;
 }
 
-static long __sched
-wait_for_common(struct completion *x, long timeout, int state)
+static inline long __sched
+__wait_for_common(struct completion *x,
+		  long (*action)(long), long timeout, int state)
 {
 	might_sleep();
 
 	spin_lock_irq(&x->wait.lock);
-	timeout = do_wait_for_common(x, timeout, state);
+	timeout = do_wait_for_common(x, action, timeout, state);
 	spin_unlock_irq(&x->wait.lock);
 	return timeout;
 }
 
+static long __sched
+wait_for_common(struct completion *x, long timeout, int state)
+{
+	return __wait_for_common(x, schedule_timeout, timeout, state);
+}
+
+static long __sched
+wait_for_common_io(struct completion *x, long timeout, int state)
+{
+	return __wait_for_common(x, io_schedule_timeout, timeout, state);
+}
+
 /**
  * wait_for_completion: - waits for completion of a task
  * @x:  holds the state of this particular completion
@@ -3270,6 +3343,39 @@ wait_for_completion_timeout(struct completion *x, unsigned long timeout)
 EXPORT_SYMBOL(wait_for_completion_timeout);
 
 /**
+ * wait_for_completion_io: - waits for completion of a task
+ * @x:  holds the state of this particular completion
+ *
+ * This waits to be signaled for completion of a specific task. It is NOT
+ * interruptible and there is no timeout. The caller is accounted as waiting
+ * for IO.
+ */
+void __sched wait_for_completion_io(struct completion *x)
+{
+	wait_for_common_io(x, MAX_SCHEDULE_TIMEOUT, TASK_UNINTERRUPTIBLE);
+}
+EXPORT_SYMBOL(wait_for_completion_io);
+
+/**
+ * wait_for_completion_io_timeout: - waits for completion of a task (w/timeout)
+ * @x:  holds the state of this particular completion
+ * @timeout:  timeout value in jiffies
+ *
+ * This waits for either a completion of a specific task to be signaled or for a
+ * specified timeout to expire. The timeout is in jiffies. It is not
+ * interruptible. The caller is accounted as waiting for IO.
+ *
+ * The return value is 0 if timed out, and positive (at least 1, or number of
+ * jiffies left till timeout) if completed.
+ */
+unsigned long __sched
+wait_for_completion_io_timeout(struct completion *x, unsigned long timeout)
+{
+	return wait_for_common_io(x, timeout, TASK_UNINTERRUPTIBLE);
+}
+EXPORT_SYMBOL(wait_for_completion_io_timeout);
+
+/**
  * wait_for_completion_interruptible: - waits for completion of a task (w/intr)
  * @x:  holds the state of this particular completion
  *
@@ -4029,8 +4135,14 @@ long sched_setaffinity(pid_t pid, const struct cpumask *in_mask)
 		goto out_free_cpus_allowed;
 	}
 	retval = -EPERM;
-	if (!check_same_owner(p) && !ns_capable(task_user_ns(p), CAP_SYS_NICE))
-		goto out_unlock;
+	if (!check_same_owner(p)) {
+		rcu_read_lock();
+		if (!ns_capable(__task_cred(p)->user_ns, CAP_SYS_NICE)) {
+			rcu_read_unlock();
+			goto out_unlock;
+		}
+		rcu_read_unlock();
+	}
 
 	retval = security_task_setscheduler(p);
 	if (retval)
@@ -4289,20 +4401,32 @@ EXPORT_SYMBOL(yield);
  * It's the caller's job to ensure that the target task struct
  * can't go away on us before we can do any checks.
  *
- * Returns true if we indeed boosted the target task.
+ * Returns:
+ *	true (>0) if we indeed boosted the target task.
+ *	false (0) if we failed to boost the target.
+ *	-ESRCH if there's no task to yield to.
  */
 bool __sched yield_to(struct task_struct *p, bool preempt)
 {
 	struct task_struct *curr = current;
 	struct rq *rq, *p_rq;
 	unsigned long flags;
-	bool yielded = 0;
+	int yielded = 0;
 
 	local_irq_save(flags);
 	rq = this_rq();
 
 again:
 	p_rq = task_rq(p);
+	/*
+	 * If we're the only runnable task on the rq and target rq also
+	 * has only one task, there's absolutely no point in yielding.
+	 */
+	if (rq->nr_running == 1 && p_rq->nr_running == 1) {
+		yielded = -ESRCH;
+		goto out_irq;
+	}
+
 	double_rq_lock(rq, p_rq);
 	while (task_rq(p) != p_rq) {
 		double_rq_unlock(rq, p_rq);
@@ -4310,13 +4434,13 @@ again:
 	}
 
 	if (!curr->sched_class->yield_to_task)
-		goto out;
+		goto out_unlock;
 
 	if (curr->sched_class != p->sched_class)
-		goto out;
+		goto out_unlock;
 
 	if (task_running(p_rq, p) || p->state)
-		goto out;
+		goto out_unlock;
 
 	yielded = curr->sched_class->yield_to_task(rq, p, preempt);
 	if (yielded) {
@@ -4329,11 +4453,12 @@ again:
 			resched_task(p_rq->curr);
 	}
 
-out:
+out_unlock:
 	double_rq_unlock(rq, p_rq);
+out_irq:
 	local_irq_restore(flags);
 
-	if (yielded)
+	if (yielded > 0)
 		schedule();
 
 	return yielded;
@@ -4474,6 +4599,7 @@ static const char stat_nam[] = TASK_STATE_TO_CHAR_STR;
 void sched_show_task(struct task_struct *p)
 {
 	unsigned long free = 0;
+	int ppid;
 	unsigned state;
 
 	state = p->state ? __ffs(p->state) + 1 : 0;
@@ -4493,8 +4619,11 @@ void sched_show_task(struct task_struct *p)
 #ifdef CONFIG_DEBUG_STACK_USAGE
 	free = stack_not_used(p);
 #endif
+	rcu_read_lock();
+	ppid = task_pid_nr(rcu_dereference(p->real_parent));
+	rcu_read_unlock();
 	printk(KERN_CONT "%5lu %5d %6d 0x%08lx\n", free,
-		task_pid_nr(p), task_pid_nr(rcu_dereference(p->real_parent)),
+		task_pid_nr(p), ppid,
 		(unsigned long)task_thread_info(p)->flags);
 
 	show_stack(p, NULL);
@@ -4588,6 +4717,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
 	 */
 	idle->sched_class = &idle_sched_class;
 	ftrace_graph_init_idle_task(idle, cpu);
+	vtime_init_idle(idle);
 #if defined(CONFIG_SMP)
 	sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu);
 #endif
@@ -7081,7 +7211,6 @@ static void free_sched_group(struct task_group *tg)
 struct task_group *sched_create_group(struct task_group *parent)
 {
 	struct task_group *tg;
-	unsigned long flags;
 
 	tg = kzalloc(sizeof(*tg), GFP_KERNEL);
 	if (!tg)
@@ -7093,6 +7222,17 @@ struct task_group *sched_create_group(struct task_group *parent)
 	if (!alloc_rt_sched_group(tg, parent))
 		goto err;
 
+	return tg;
+
+err:
+	free_sched_group(tg);
+	return ERR_PTR(-ENOMEM);
+}
+
+void sched_online_group(struct task_group *tg, struct task_group *parent)
+{
+	unsigned long flags;
+
 	spin_lock_irqsave(&task_group_lock, flags);
 	list_add_rcu(&tg->list, &task_groups);
 
@@ -7102,12 +7242,6 @@ struct task_group *sched_create_group(struct task_group *parent)
 	INIT_LIST_HEAD(&tg->children);
 	list_add_rcu(&tg->siblings, &parent->children);
 	spin_unlock_irqrestore(&task_group_lock, flags);
-
-	return tg;
-
-err:
-	free_sched_group(tg);
-	return ERR_PTR(-ENOMEM);
 }
 
 /* rcu callback to free various structures associated with a task group */
@@ -7120,6 +7254,12 @@ static void free_sched_group_rcu(struct rcu_head *rhp)
 /* Destroy runqueue etc associated with a task group */
 void sched_destroy_group(struct task_group *tg)
 {
+	/* wait for possible concurrent references to cfs_rqs complete */
+	call_rcu(&tg->rcu, free_sched_group_rcu);
+}
+
+void sched_offline_group(struct task_group *tg)
+{
 	unsigned long flags;
 	int i;
 
@@ -7131,9 +7271,6 @@ void sched_destroy_group(struct task_group *tg)
 	list_del_rcu(&tg->list);
 	list_del_rcu(&tg->siblings);
 	spin_unlock_irqrestore(&task_group_lock, flags);
-
-	/* wait for possible concurrent references to cfs_rqs complete */
-	call_rcu(&tg->rcu, free_sched_group_rcu);
 }
 
 /* change task's runqueue when it moves between groups.
@@ -7429,6 +7566,25 @@ static int sched_rt_global_constraints(void)
 }
 #endif /* CONFIG_RT_GROUP_SCHED */
 
+int sched_rr_handler(struct ctl_table *table, int write,
+		void __user *buffer, size_t *lenp,
+		loff_t *ppos)
+{
+	int ret;
+	static DEFINE_MUTEX(mutex);
+
+	mutex_lock(&mutex);
+	ret = proc_dointvec(table, write, buffer, lenp, ppos);
+	/* make sure that internally we keep jiffies */
+	/* also, writing zero resets timeslice to default */
+	if (!ret && write) {
+		sched_rr_timeslice = sched_rr_timeslice <= 0 ?
+			RR_TIMESLICE : msecs_to_jiffies(sched_rr_timeslice);
+	}
+	mutex_unlock(&mutex);
+	return ret;
+}
+
 int sched_rt_handler(struct ctl_table *table, int write,
 		void __user *buffer, size_t *lenp,
 		loff_t *ppos)
@@ -7468,7 +7624,7 @@ static inline struct task_group *cgroup_tg(struct cgroup *cgrp)
 			    struct task_group, css);
 }
 
-static struct cgroup_subsys_state *cpu_cgroup_create(struct cgroup *cgrp)
+static struct cgroup_subsys_state *cpu_cgroup_css_alloc(struct cgroup *cgrp)
 {
 	struct task_group *tg, *parent;
 
@@ -7485,13 +7641,33 @@ static struct cgroup_subsys_state *cpu_cgroup_create(struct cgroup *cgrp)
 	return &tg->css;
 }
 
-static void cpu_cgroup_destroy(struct cgroup *cgrp)
+static int cpu_cgroup_css_online(struct cgroup *cgrp)
+{
+	struct task_group *tg = cgroup_tg(cgrp);
+	struct task_group *parent;
+
+	if (!cgrp->parent)
+		return 0;
+
+	parent = cgroup_tg(cgrp->parent);
+	sched_online_group(tg, parent);
+	return 0;
+}
+
+static void cpu_cgroup_css_free(struct cgroup *cgrp)
 {
 	struct task_group *tg = cgroup_tg(cgrp);
 
 	sched_destroy_group(tg);
 }
 
+static void cpu_cgroup_css_offline(struct cgroup *cgrp)
+{
+	struct task_group *tg = cgroup_tg(cgrp);
+
+	sched_offline_group(tg);
+}
+
 static int cpu_cgroup_can_attach(struct cgroup *cgrp,
 				 struct cgroup_taskset *tset)
 {
@@ -7845,8 +8021,10 @@ static struct cftype cpu_files[] = {
 
 struct cgroup_subsys cpu_cgroup_subsys = {
 	.name		= "cpu",
-	.create		= cpu_cgroup_create,
-	.destroy	= cpu_cgroup_destroy,
+	.css_alloc	= cpu_cgroup_css_alloc,
+	.css_free	= cpu_cgroup_css_free,
+	.css_online	= cpu_cgroup_css_online,
+	.css_offline	= cpu_cgroup_css_offline,
 	.can_attach	= cpu_cgroup_can_attach,
 	.attach		= cpu_cgroup_attach,
 	.exit		= cpu_cgroup_exit,
@@ -7869,7 +8047,7 @@ struct cgroup_subsys cpu_cgroup_subsys = {
 struct cpuacct root_cpuacct;
 
 /* create a new cpu accounting group */
-static struct cgroup_subsys_state *cpuacct_create(struct cgroup *cgrp)
+static struct cgroup_subsys_state *cpuacct_css_alloc(struct cgroup *cgrp)
 {
 	struct cpuacct *ca;
 
@@ -7899,7 +8077,7 @@ out:
 }
 
 /* destroy an existing cpu accounting group */
-static void cpuacct_destroy(struct cgroup *cgrp)
+static void cpuacct_css_free(struct cgroup *cgrp)
 {
 	struct cpuacct *ca = cgroup_ca(cgrp);
 
@@ -8070,9 +8248,15 @@ void cpuacct_charge(struct task_struct *tsk, u64 cputime)
 
 struct cgroup_subsys cpuacct_subsys = {
 	.name = "cpuacct",
-	.create = cpuacct_create,
-	.destroy = cpuacct_destroy,
+	.css_alloc = cpuacct_css_alloc,
+	.css_free = cpuacct_css_free,
 	.subsys_id = cpuacct_subsys_id,
 	.base_cftypes = files,
 };
 #endif	/* CONFIG_CGROUP_CPUACCT */
+
+void dump_cpu_task(int cpu)
+{
+	pr_info("Task dump for CPU %d:\n", cpu);
+	sched_show_task(cpu_curr(cpu));
+}
diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c
index 23aa789c53e..1095e878a46 100644
--- a/kernel/sched/cpupri.c
+++ b/kernel/sched/cpupri.c
@@ -28,6 +28,8 @@
  */
 
 #include <linux/gfp.h>
+#include <linux/sched.h>
+#include <linux/sched/rt.h>
 #include "cpupri.h"
 
 /* Convert between a 140 based task->prio, and our 102 based cpupri */
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index 81b763ba58a..ed12cbb135f 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -3,6 +3,7 @@
 #include <linux/tsacct_kern.h>
 #include <linux/kernel_stat.h>
 #include <linux/static_key.h>
+#include <linux/context_tracking.h>
 #include "sched.h"
 
 
@@ -43,7 +44,7 @@ DEFINE_PER_CPU(seqcount_t, irq_time_seq);
  * Called before incrementing preempt_count on {soft,}irq_enter
  * and before decrementing preempt_count on {soft,}irq_exit.
  */
-void vtime_account(struct task_struct *curr)
+void irqtime_account_irq(struct task_struct *curr)
 {
 	unsigned long flags;
 	s64 delta;
@@ -73,7 +74,7 @@ void vtime_account(struct task_struct *curr)
 	irq_time_write_end();
 	local_irq_restore(flags);
 }
-EXPORT_SYMBOL_GPL(vtime_account);
+EXPORT_SYMBOL_GPL(irqtime_account_irq);
 
 static int irqtime_account_hi_update(void)
 {
@@ -163,7 +164,7 @@ void account_user_time(struct task_struct *p, cputime_t cputime,
 	task_group_account_field(p, index, (__force u64) cputime);
 
 	/* Account for user time used */
-	acct_update_integrals(p);
+	acct_account_cputime(p);
 }
 
 /*
@@ -213,7 +214,7 @@ void __account_system_time(struct task_struct *p, cputime_t cputime,
 	task_group_account_field(p, index, (__force u64) cputime);
 
 	/* Account for system time used */
-	acct_update_integrals(p);
+	acct_account_cputime(p);
 }
 
 /*
@@ -288,7 +289,35 @@ static __always_inline bool steal_account_process_tick(void)
 	return false;
 }
 
-#ifndef CONFIG_VIRT_CPU_ACCOUNTING
+/*
+ * Accumulate raw cputime values of dead tasks (sig->[us]time) and live
+ * tasks (sum on group iteration) belonging to @tsk's group.
+ */
+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;
+
+	rcu_read_lock();
+	/* make sure we can trust tsk->thread_group list */
+	if (!likely(pid_alive(tsk)))
+		goto out;
+
+	t = tsk;
+	do {
+		task_cputime(tsk, &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();
+}
 
 #ifdef CONFIG_IRQ_TIME_ACCOUNTING
 /*
@@ -354,11 +383,12 @@ static void irqtime_account_idle_ticks(int ticks)
 		irqtime_account_process_tick(current, 0, rq);
 }
 #else /* CONFIG_IRQ_TIME_ACCOUNTING */
-static void irqtime_account_idle_ticks(int ticks) {}
-static void irqtime_account_process_tick(struct task_struct *p, int user_tick,
+static inline void irqtime_account_idle_ticks(int ticks) {}
+static inline void irqtime_account_process_tick(struct task_struct *p, int user_tick,
 						struct rq *rq) {}
 #endif /* CONFIG_IRQ_TIME_ACCOUNTING */
 
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
 /*
  * Account a single tick of cpu time.
  * @p: the process that the cpu time gets accounted to
@@ -369,6 +399,9 @@ void account_process_tick(struct task_struct *p, int user_tick)
 	cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy);
 	struct rq *rq = this_rq();
 
+	if (vtime_accounting_enabled())
+		return;
+
 	if (sched_clock_irqtime) {
 		irqtime_account_process_tick(p, user_tick, rq);
 		return;
@@ -410,20 +443,19 @@ void account_idle_ticks(unsigned long ticks)
 
 	account_idle_time(jiffies_to_cputime(ticks));
 }
-
-#endif
+#endif /* !CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */
 
 /*
  * Use precise platform statistics if available:
  */
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
-void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
+void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
 {
 	*ut = p->utime;
 	*st = p->stime;
 }
 
-void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
+void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
 {
 	struct task_cputime cputime;
 
@@ -433,6 +465,24 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
 	*st = cputime.stime;
 }
 
+#ifndef __ARCH_HAS_VTIME_TASK_SWITCH
+void vtime_task_switch(struct task_struct *prev)
+{
+	if (!vtime_accounting_enabled())
+		return;
+
+	if (is_idle_task(prev))
+		vtime_account_idle(prev);
+	else
+		vtime_account_system(prev);
+
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
+	vtime_account_user(prev);
+#endif
+	arch_vtime_task_switch(prev);
+}
+#endif
+
 /*
  * Archs that account the whole time spent in the idle task
  * (outside irq) as idle time can rely on this and just implement
@@ -442,33 +492,40 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
  * vtime_account().
  */
 #ifndef __ARCH_HAS_VTIME_ACCOUNT
-void vtime_account(struct task_struct *tsk)
+void vtime_account_irq_enter(struct task_struct *tsk)
 {
-	unsigned long flags;
-
-	local_irq_save(flags);
-
-	if (in_interrupt() || !is_idle_task(tsk))
-		vtime_account_system(tsk);
-	else
-		vtime_account_idle(tsk);
+	if (!vtime_accounting_enabled())
+		return;
 
-	local_irq_restore(flags);
+	if (!in_interrupt()) {
+		/*
+		 * If we interrupted user, context_tracking_in_user()
+		 * is 1 because the context tracking don't hook
+		 * on irq entry/exit. This way we know if
+		 * we need to flush user time on kernel entry.
+		 */
+		if (context_tracking_in_user()) {
+			vtime_account_user(tsk);
+			return;
+		}
+
+		if (is_idle_task(tsk)) {
+			vtime_account_idle(tsk);
+			return;
+		}
+	}
+	vtime_account_system(tsk);
 }
-EXPORT_SYMBOL_GPL(vtime_account);
+EXPORT_SYMBOL_GPL(vtime_account_irq_enter);
 #endif /* __ARCH_HAS_VTIME_ACCOUNT */
 
-#else
+#else /* !CONFIG_VIRT_CPU_ACCOUNTING */
 
-#ifndef nsecs_to_cputime
-# define nsecs_to_cputime(__nsecs)	nsecs_to_jiffies(__nsecs)
-#endif
-
-static cputime_t scale_utime(cputime_t utime, cputime_t rtime, cputime_t total)
+static cputime_t scale_stime(cputime_t stime, cputime_t rtime, cputime_t total)
 {
 	u64 temp = (__force u64) rtime;
 
-	temp *= (__force u64) utime;
+	temp *= (__force u64) stime;
 
 	if (sizeof(cputime_t) == 4)
 		temp = div_u64(temp, (__force u32) total);
@@ -478,53 +535,283 @@ static cputime_t scale_utime(cputime_t utime, cputime_t rtime, cputime_t total)
 	return (__force cputime_t) temp;
 }
 
-void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
+/*
+ * Adjust tick based cputime random precision against scheduler
+ * runtime accounting.
+ */
+static void cputime_adjust(struct task_cputime *curr,
+			   struct cputime *prev,
+			   cputime_t *ut, cputime_t *st)
 {
-	cputime_t rtime, utime = p->utime, total = utime + p->stime;
+	cputime_t rtime, stime, total;
+
+	stime = curr->stime;
+	total = stime + curr->utime;
 
 	/*
-	 * Use CFS's precise accounting:
+	 * Tick based cputime accounting depend on random scheduling
+	 * timeslices of a task to be interrupted or not by the timer.
+	 * Depending on these circumstances, the number of these interrupts
+	 * may be over or under-optimistic, matching the real user and system
+	 * cputime with a variable precision.
+	 *
+	 * Fix this by scaling these tick based values against the total
+	 * runtime accounted by the CFS scheduler.
 	 */
-	rtime = nsecs_to_cputime(p->se.sum_exec_runtime);
+	rtime = nsecs_to_cputime(curr->sum_exec_runtime);
 
 	if (total)
-		utime = scale_utime(utime, rtime, total);
+		stime = scale_stime(stime, rtime, total);
 	else
-		utime = rtime;
+		stime = rtime;
 
 	/*
-	 * Compare with previous values, to keep monotonicity:
+	 * If the tick based count grows faster than the scheduler one,
+	 * the result of the scaling may go backward.
+	 * Let's enforce monotonicity.
 	 */
-	p->prev_utime = max(p->prev_utime, utime);
-	p->prev_stime = max(p->prev_stime, rtime - p->prev_utime);
+	prev->stime = max(prev->stime, stime);
+	prev->utime = max(prev->utime, rtime - prev->stime);
+
+	*ut = prev->utime;
+	*st = prev->stime;
+}
 
-	*ut = p->prev_utime;
-	*st = p->prev_stime;
+void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
+{
+	struct task_cputime cputime = {
+		.sum_exec_runtime = p->se.sum_exec_runtime,
+	};
+
+	task_cputime(p, &cputime.utime, &cputime.stime);
+	cputime_adjust(&cputime, &p->prev_cputime, ut, st);
 }
 
 /*
  * Must be called with siglock held.
  */
-void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
+void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
 {
-	struct signal_struct *sig = p->signal;
 	struct task_cputime cputime;
-	cputime_t rtime, utime, total;
 
 	thread_group_cputime(p, &cputime);
+	cputime_adjust(&cputime, &p->signal->prev_cputime, ut, st);
+}
+#endif /* !CONFIG_VIRT_CPU_ACCOUNTING */
 
-	total = cputime.utime + cputime.stime;
-	rtime = nsecs_to_cputime(cputime.sum_exec_runtime);
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
+static unsigned long long vtime_delta(struct task_struct *tsk)
+{
+	unsigned long long clock;
 
-	if (total)
-		utime = scale_utime(cputime.utime, rtime, total);
-	else
-		utime = rtime;
+	clock = local_clock();
+	if (clock < tsk->vtime_snap)
+		return 0;
 
-	sig->prev_utime = max(sig->prev_utime, utime);
-	sig->prev_stime = max(sig->prev_stime, rtime - sig->prev_utime);
+	return clock - tsk->vtime_snap;
+}
 
-	*ut = sig->prev_utime;
-	*st = sig->prev_stime;
+static cputime_t get_vtime_delta(struct task_struct *tsk)
+{
+	unsigned long long delta = vtime_delta(tsk);
+
+	WARN_ON_ONCE(tsk->vtime_snap_whence == VTIME_SLEEPING);
+	tsk->vtime_snap += delta;
+
+	/* CHECKME: always safe to convert nsecs to cputime? */
+	return nsecs_to_cputime(delta);
 }
-#endif
+
+static void __vtime_account_system(struct task_struct *tsk)
+{
+	cputime_t delta_cpu = get_vtime_delta(tsk);
+
+	account_system_time(tsk, irq_count(), delta_cpu, cputime_to_scaled(delta_cpu));
+}
+
+void vtime_account_system(struct task_struct *tsk)
+{
+	if (!vtime_accounting_enabled())
+		return;
+
+	write_seqlock(&tsk->vtime_seqlock);
+	__vtime_account_system(tsk);
+	write_sequnlock(&tsk->vtime_seqlock);
+}
+
+void vtime_account_irq_exit(struct task_struct *tsk)
+{
+	if (!vtime_accounting_enabled())
+		return;
+
+	write_seqlock(&tsk->vtime_seqlock);
+	if (context_tracking_in_user())
+		tsk->vtime_snap_whence = VTIME_USER;
+	__vtime_account_system(tsk);
+	write_sequnlock(&tsk->vtime_seqlock);
+}
+
+void vtime_account_user(struct task_struct *tsk)
+{
+	cputime_t delta_cpu;
+
+	if (!vtime_accounting_enabled())
+		return;
+
+	delta_cpu = get_vtime_delta(tsk);
+
+	write_seqlock(&tsk->vtime_seqlock);
+	tsk->vtime_snap_whence = VTIME_SYS;
+	account_user_time(tsk, delta_cpu, cputime_to_scaled(delta_cpu));
+	write_sequnlock(&tsk->vtime_seqlock);
+}
+
+void vtime_user_enter(struct task_struct *tsk)
+{
+	if (!vtime_accounting_enabled())
+		return;
+
+	write_seqlock(&tsk->vtime_seqlock);
+	tsk->vtime_snap_whence = VTIME_USER;
+	__vtime_account_system(tsk);
+	write_sequnlock(&tsk->vtime_seqlock);
+}
+
+void vtime_guest_enter(struct task_struct *tsk)
+{
+	write_seqlock(&tsk->vtime_seqlock);
+	__vtime_account_system(tsk);
+	current->flags |= PF_VCPU;
+	write_sequnlock(&tsk->vtime_seqlock);
+}
+
+void vtime_guest_exit(struct task_struct *tsk)
+{
+	write_seqlock(&tsk->vtime_seqlock);
+	__vtime_account_system(tsk);
+	current->flags &= ~PF_VCPU;
+	write_sequnlock(&tsk->vtime_seqlock);
+}
+
+void vtime_account_idle(struct task_struct *tsk)
+{
+	cputime_t delta_cpu = get_vtime_delta(tsk);
+
+	account_idle_time(delta_cpu);
+}
+
+bool vtime_accounting_enabled(void)
+{
+	return context_tracking_active();
+}
+
+void arch_vtime_task_switch(struct task_struct *prev)
+{
+	write_seqlock(&prev->vtime_seqlock);
+	prev->vtime_snap_whence = VTIME_SLEEPING;
+	write_sequnlock(&prev->vtime_seqlock);
+
+	write_seqlock(&current->vtime_seqlock);
+	current->vtime_snap_whence = VTIME_SYS;
+	current->vtime_snap = sched_clock();
+	write_sequnlock(&current->vtime_seqlock);
+}
+
+void vtime_init_idle(struct task_struct *t)
+{
+	unsigned long flags;
+
+	write_seqlock_irqsave(&t->vtime_seqlock, flags);
+	t->vtime_snap_whence = VTIME_SYS;
+	t->vtime_snap = sched_clock();
+	write_sequnlock_irqrestore(&t->vtime_seqlock, flags);
+}
+
+cputime_t task_gtime(struct task_struct *t)
+{
+	unsigned int seq;
+	cputime_t gtime;
+
+	do {
+		seq = read_seqbegin(&t->vtime_seqlock);
+
+		gtime = t->gtime;
+		if (t->flags & PF_VCPU)
+			gtime += vtime_delta(t);
+
+	} while (read_seqretry(&t->vtime_seqlock, seq));
+
+	return gtime;
+}
+
+/*
+ * Fetch cputime raw values from fields of task_struct and
+ * add up the pending nohz execution time since the last
+ * cputime snapshot.
+ */
+static void
+fetch_task_cputime(struct task_struct *t,
+		   cputime_t *u_dst, cputime_t *s_dst,
+		   cputime_t *u_src, cputime_t *s_src,
+		   cputime_t *udelta, cputime_t *sdelta)
+{
+	unsigned int seq;
+	unsigned long long delta;
+
+	do {
+		*udelta = 0;
+		*sdelta = 0;
+
+		seq = read_seqbegin(&t->vtime_seqlock);
+
+		if (u_dst)
+			*u_dst = *u_src;
+		if (s_dst)
+			*s_dst = *s_src;
+
+		/* Task is sleeping, nothing to add */
+		if (t->vtime_snap_whence == VTIME_SLEEPING ||
+		    is_idle_task(t))
+			continue;
+
+		delta = vtime_delta(t);
+
+		/*
+		 * Task runs either in user or kernel space, add pending nohz time to
+		 * the right place.
+		 */
+		if (t->vtime_snap_whence == VTIME_USER || t->flags & PF_VCPU) {
+			*udelta = delta;
+		} else {
+			if (t->vtime_snap_whence == VTIME_SYS)
+				*sdelta = delta;
+		}
+	} while (read_seqretry(&t->vtime_seqlock, seq));
+}
+
+
+void task_cputime(struct task_struct *t, cputime_t *utime, cputime_t *stime)
+{
+	cputime_t udelta, sdelta;
+
+	fetch_task_cputime(t, utime, stime, &t->utime,
+			   &t->stime, &udelta, &sdelta);
+	if (utime)
+		*utime += udelta;
+	if (stime)
+		*stime += sdelta;
+}
+
+void task_cputime_scaled(struct task_struct *t,
+			 cputime_t *utimescaled, cputime_t *stimescaled)
+{
+	cputime_t udelta, sdelta;
+
+	fetch_task_cputime(t, utimescaled, stimescaled,
+			   &t->utimescaled, &t->stimescaled, &udelta, &sdelta);
+	if (utimescaled)
+		*utimescaled += cputime_to_scaled(udelta);
+	if (stimescaled)
+		*stimescaled += cputime_to_scaled(sdelta);
+}
+#endif /* CONFIG_VIRT_CPU_ACCOUNTING_GEN */
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 6f79596e0ea..75024a67352 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -61,14 +61,20 @@ static unsigned long nsec_low(unsigned long long nsec)
 static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group *tg)
 {
 	struct sched_entity *se = tg->se[cpu];
-	if (!se)
-		return;
 
 #define P(F) \
 	SEQ_printf(m, "  .%-30s: %lld\n", #F, (long long)F)
 #define PN(F) \
 	SEQ_printf(m, "  .%-30s: %lld.%06ld\n", #F, SPLIT_NS((long long)F))
 
+	if (!se) {
+		struct sched_avg *avg = &cpu_rq(cpu)->avg;
+		P(avg->runnable_avg_sum);
+		P(avg->runnable_avg_period);
+		return;
+	}
+
+
 	PN(se->exec_start);
 	PN(se->vruntime);
 	PN(se->sum_exec_runtime);
@@ -85,6 +91,12 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group
 	P(se->statistics.wait_count);
 #endif
 	P(se->load.weight);
+#ifdef CONFIG_SMP
+	P(se->avg.runnable_avg_sum);
+	P(se->avg.runnable_avg_period);
+	P(se->avg.load_avg_contrib);
+	P(se->avg.decay_count);
+#endif
 #undef PN
 #undef P
 }
@@ -98,13 +110,6 @@ static char *task_group_path(struct task_group *tg)
 	if (autogroup_path(tg, group_path, PATH_MAX))
 		return group_path;
 
-	/*
-	 * May be NULL if the underlying cgroup isn't fully-created yet
-	 */
-	if (!tg->css.cgroup) {
-		group_path[0] = '\0';
-		return group_path;
-	}
 	cgroup_path(tg->css.cgroup, group_path, PATH_MAX);
 	return group_path;
 }
@@ -206,14 +211,18 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
 	SEQ_printf(m, "  .%-30s: %ld\n", "load", cfs_rq->load.weight);
 #ifdef CONFIG_FAIR_GROUP_SCHED
 #ifdef CONFIG_SMP
-	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "load_avg",
-			SPLIT_NS(cfs_rq->load_avg));
-	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "load_period",
-			SPLIT_NS(cfs_rq->load_period));
-	SEQ_printf(m, "  .%-30s: %ld\n", "load_contrib",
-			cfs_rq->load_contribution);
-	SEQ_printf(m, "  .%-30s: %d\n", "load_tg",
-			atomic_read(&cfs_rq->tg->load_weight));
+	SEQ_printf(m, "  .%-30s: %lld\n", "runnable_load_avg",
+			cfs_rq->runnable_load_avg);
+	SEQ_printf(m, "  .%-30s: %lld\n", "blocked_load_avg",
+			cfs_rq->blocked_load_avg);
+	SEQ_printf(m, "  .%-30s: %lld\n", "tg_load_avg",
+			(unsigned long long)atomic64_read(&cfs_rq->tg->load_avg));
+	SEQ_printf(m, "  .%-30s: %lld\n", "tg_load_contrib",
+			cfs_rq->tg_load_contrib);
+	SEQ_printf(m, "  .%-30s: %d\n", "tg_runnable_contrib",
+			cfs_rq->tg_runnable_contrib);
+	SEQ_printf(m, "  .%-30s: %d\n", "tg->runnable_avg",
+			atomic_read(&cfs_rq->tg->runnable_avg));
 #endif
 
 	print_cfs_group_stats(m, cpu, cfs_rq->tg);
@@ -253,11 +262,11 @@ static void print_cpu(struct seq_file *m, int cpu)
 	{
 		unsigned int freq = cpu_khz ? : 1;
 
-		SEQ_printf(m, "\ncpu#%d, %u.%03u MHz\n",
+		SEQ_printf(m, "cpu#%d, %u.%03u MHz\n",
 			   cpu, freq / 1000, (freq % 1000));
 	}
 #else
-	SEQ_printf(m, "\ncpu#%d\n", cpu);
+	SEQ_printf(m, "cpu#%d\n", cpu);
 #endif
 
 #define P(x)								\
@@ -314,6 +323,7 @@ do {									\
 	print_rq(m, rq, cpu);
 	rcu_read_unlock();
 	spin_unlock_irqrestore(&sched_debug_lock, flags);
+	SEQ_printf(m, "\n");
 }
 
 static const char *sched_tunable_scaling_names[] = {
@@ -322,11 +332,10 @@ static const char *sched_tunable_scaling_names[] = {
 	"linear"
 };
 
-static int sched_debug_show(struct seq_file *m, void *v)
+static void sched_debug_header(struct seq_file *m)
 {
 	u64 ktime, sched_clk, cpu_clk;
 	unsigned long flags;
-	int cpu;
 
 	local_irq_save(flags);
 	ktime = ktime_to_ns(ktime_get());
@@ -368,33 +377,101 @@ static int sched_debug_show(struct seq_file *m, void *v)
 #undef PN
 #undef P
 
-	SEQ_printf(m, "  .%-40s: %d (%s)\n", "sysctl_sched_tunable_scaling",
+	SEQ_printf(m, "  .%-40s: %d (%s)\n",
+		"sysctl_sched_tunable_scaling",
 		sysctl_sched_tunable_scaling,
 		sched_tunable_scaling_names[sysctl_sched_tunable_scaling]);
+	SEQ_printf(m, "\n");
+}
 
-	for_each_online_cpu(cpu)
-		print_cpu(m, cpu);
+static int sched_debug_show(struct seq_file *m, void *v)
+{
+	int cpu = (unsigned long)(v - 2);
 
-	SEQ_printf(m, "\n");
+	if (cpu != -1)
+		print_cpu(m, cpu);
+	else
+		sched_debug_header(m);
 
 	return 0;
 }
 
 void sysrq_sched_debug_show(void)
 {
-	sched_debug_show(NULL, NULL);
+	int cpu;
+
+	sched_debug_header(NULL);
+	for_each_online_cpu(cpu)
+		print_cpu(NULL, cpu);
+
+}
+
+/*
+ * This itererator needs some explanation.
+ * It returns 1 for the header position.
+ * This means 2 is cpu 0.
+ * In a hotplugged system some cpus, including cpu 0, may be missing so we have
+ * to use cpumask_* to iterate over the cpus.
+ */
+static void *sched_debug_start(struct seq_file *file, loff_t *offset)
+{
+	unsigned long n = *offset;
+
+	if (n == 0)
+		return (void *) 1;
+
+	n--;
+
+	if (n > 0)
+		n = cpumask_next(n - 1, cpu_online_mask);
+	else
+		n = cpumask_first(cpu_online_mask);
+
+	*offset = n + 1;
+
+	if (n < nr_cpu_ids)
+		return (void *)(unsigned long)(n + 2);
+	return NULL;
+}
+
+static void *sched_debug_next(struct seq_file *file, void *data, loff_t *offset)
+{
+	(*offset)++;
+	return sched_debug_start(file, offset);
+}
+
+static void sched_debug_stop(struct seq_file *file, void *data)
+{
+}
+
+static const struct seq_operations sched_debug_sops = {
+	.start = sched_debug_start,
+	.next = sched_debug_next,
+	.stop = sched_debug_stop,
+	.show = sched_debug_show,
+};
+
+static int sched_debug_release(struct inode *inode, struct file *file)
+{
+	seq_release(inode, file);
+
+	return 0;
 }
 
 static int sched_debug_open(struct inode *inode, struct file *filp)
 {
-	return single_open(filp, sched_debug_show, NULL);
+	int ret = 0;
+
+	ret = seq_open(filp, &sched_debug_sops);
+
+	return ret;
 }
 
 static const struct file_operations sched_debug_fops = {
 	.open		= sched_debug_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= single_release,
+	.release	= sched_debug_release,
 };
 
 static int __init init_sched_debug_procfs(void)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 6b800a14b99..7a33e5986fc 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -26,6 +26,9 @@
 #include <linux/slab.h>
 #include <linux/profile.h>
 #include <linux/interrupt.h>
+#include <linux/mempolicy.h>
+#include <linux/migrate.h>
+#include <linux/task_work.h>
 
 #include <trace/events/sched.h>
 
@@ -259,6 +262,9 @@ static inline struct cfs_rq *group_cfs_rq(struct sched_entity *grp)
 	return grp->my_q;
 }
 
+static void update_cfs_rq_blocked_load(struct cfs_rq *cfs_rq,
+				       int force_update);
+
 static inline void list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq)
 {
 	if (!cfs_rq->on_list) {
@@ -278,6 +284,8 @@ static inline void list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq)
 		}
 
 		cfs_rq->on_list = 1;
+		/* We should have no load, but we need to update last_decay. */
+		update_cfs_rq_blocked_load(cfs_rq, 0);
 	}
 }
 
@@ -653,9 +661,6 @@ static u64 sched_vslice(struct cfs_rq *cfs_rq, struct sched_entity *se)
 	return calc_delta_fair(sched_slice(cfs_rq, se), se);
 }
 
-static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update);
-static void update_cfs_shares(struct cfs_rq *cfs_rq);
-
 /*
  * Update the current task's runtime statistics. Skip current tasks that
  * are not in our scheduling class.
@@ -675,10 +680,6 @@ __update_curr(struct cfs_rq *cfs_rq, struct sched_entity *curr,
 
 	curr->vruntime += delta_exec_weighted;
 	update_min_vruntime(cfs_rq);
-
-#if defined CONFIG_SMP && defined CONFIG_FAIR_GROUP_SCHED
-	cfs_rq->load_unacc_exec_time += delta_exec;
-#endif
 }
 
 static void update_curr(struct cfs_rq *cfs_rq)
@@ -776,6 +777,230 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se)
  * Scheduling class queueing methods:
  */
 
+#ifdef CONFIG_NUMA_BALANCING
+/*
+ * numa task sample period in ms
+ */
+unsigned int sysctl_numa_balancing_scan_period_min = 100;
+unsigned int sysctl_numa_balancing_scan_period_max = 100*50;
+unsigned int sysctl_numa_balancing_scan_period_reset = 100*600;
+
+/* Portion of address space to scan in MB */
+unsigned int sysctl_numa_balancing_scan_size = 256;
+
+/* Scan @scan_size MB every @scan_period after an initial @scan_delay in ms */
+unsigned int sysctl_numa_balancing_scan_delay = 1000;
+
+static void task_numa_placement(struct task_struct *p)
+{
+	int seq;
+
+	if (!p->mm)	/* for example, ksmd faulting in a user's mm */
+		return;
+	seq = ACCESS_ONCE(p->mm->numa_scan_seq);
+	if (p->numa_scan_seq == seq)
+		return;
+	p->numa_scan_seq = seq;
+
+	/* FIXME: Scheduling placement policy hints go here */
+}
+
+/*
+ * Got a PROT_NONE fault for a page on @node.
+ */
+void task_numa_fault(int node, int pages, bool migrated)
+{
+	struct task_struct *p = current;
+
+	if (!sched_feat_numa(NUMA))
+		return;
+
+	/* FIXME: Allocate task-specific structure for placement policy here */
+
+	/*
+	 * If pages are properly placed (did not migrate) then scan slower.
+	 * This is reset periodically in case of phase changes
+	 */
+        if (!migrated)
+		p->numa_scan_period = min(sysctl_numa_balancing_scan_period_max,
+			p->numa_scan_period + jiffies_to_msecs(10));
+
+	task_numa_placement(p);
+}
+
+static void reset_ptenuma_scan(struct task_struct *p)
+{
+	ACCESS_ONCE(p->mm->numa_scan_seq)++;
+	p->mm->numa_scan_offset = 0;
+}
+
+/*
+ * The expensive part of numa migration is done from task_work context.
+ * Triggered from task_tick_numa().
+ */
+void task_numa_work(struct callback_head *work)
+{
+	unsigned long migrate, next_scan, now = jiffies;
+	struct task_struct *p = current;
+	struct mm_struct *mm = p->mm;
+	struct vm_area_struct *vma;
+	unsigned long start, end;
+	long pages;
+
+	WARN_ON_ONCE(p != container_of(work, struct task_struct, numa_work));
+
+	work->next = work; /* protect against double add */
+	/*
+	 * Who cares about NUMA placement when they're dying.
+	 *
+	 * NOTE: make sure not to dereference p->mm before this check,
+	 * exit_task_work() happens _after_ exit_mm() so we could be called
+	 * without p->mm even though we still had it when we enqueued this
+	 * work.
+	 */
+	if (p->flags & PF_EXITING)
+		return;
+
+	/*
+	 * We do not care about task placement until a task runs on a node
+	 * other than the first one used by the address space. This is
+	 * largely because migrations are driven by what CPU the task
+	 * is running on. If it's never scheduled on another node, it'll
+	 * not migrate so why bother trapping the fault.
+	 */
+	if (mm->first_nid == NUMA_PTE_SCAN_INIT)
+		mm->first_nid = numa_node_id();
+	if (mm->first_nid != NUMA_PTE_SCAN_ACTIVE) {
+		/* Are we running on a new node yet? */
+		if (numa_node_id() == mm->first_nid &&
+		    !sched_feat_numa(NUMA_FORCE))
+			return;
+
+		mm->first_nid = NUMA_PTE_SCAN_ACTIVE;
+	}
+
+	/*
+	 * Reset the scan period if enough time has gone by. Objective is that
+	 * scanning will be reduced if pages are properly placed. As tasks
+	 * can enter different phases this needs to be re-examined. Lacking
+	 * proper tracking of reference behaviour, this blunt hammer is used.
+	 */
+	migrate = mm->numa_next_reset;
+	if (time_after(now, migrate)) {
+		p->numa_scan_period = sysctl_numa_balancing_scan_period_min;
+		next_scan = now + msecs_to_jiffies(sysctl_numa_balancing_scan_period_reset);
+		xchg(&mm->numa_next_reset, next_scan);
+	}
+
+	/*
+	 * Enforce maximal scan/migration frequency..
+	 */
+	migrate = mm->numa_next_scan;
+	if (time_before(now, migrate))
+		return;
+
+	if (p->numa_scan_period == 0)
+		p->numa_scan_period = sysctl_numa_balancing_scan_period_min;
+
+	next_scan = now + msecs_to_jiffies(p->numa_scan_period);
+	if (cmpxchg(&mm->numa_next_scan, migrate, next_scan) != migrate)
+		return;
+
+	/*
+	 * Do not set pte_numa if the current running node is rate-limited.
+	 * This loses statistics on the fault but if we are unwilling to
+	 * migrate to this node, it is less likely we can do useful work
+	 */
+	if (migrate_ratelimited(numa_node_id()))
+		return;
+
+	start = mm->numa_scan_offset;
+	pages = sysctl_numa_balancing_scan_size;
+	pages <<= 20 - PAGE_SHIFT; /* MB in pages */
+	if (!pages)
+		return;
+
+	down_read(&mm->mmap_sem);
+	vma = find_vma(mm, start);
+	if (!vma) {
+		reset_ptenuma_scan(p);
+		start = 0;
+		vma = mm->mmap;
+	}
+	for (; vma; vma = vma->vm_next) {
+		if (!vma_migratable(vma))
+			continue;
+
+		/* Skip small VMAs. They are not likely to be of relevance */
+		if (vma->vm_end - vma->vm_start < HPAGE_SIZE)
+			continue;
+
+		do {
+			start = max(start, vma->vm_start);
+			end = ALIGN(start + (pages << PAGE_SHIFT), HPAGE_SIZE);
+			end = min(end, vma->vm_end);
+			pages -= change_prot_numa(vma, start, end);
+
+			start = end;
+			if (pages <= 0)
+				goto out;
+		} while (end != vma->vm_end);
+	}
+
+out:
+	/*
+	 * It is possible to reach the end of the VMA list but the last few VMAs are
+	 * not guaranteed to the vma_migratable. If they are not, we would find the
+	 * !migratable VMA on the next scan but not reset the scanner to the start
+	 * so check it now.
+	 */
+	if (vma)
+		mm->numa_scan_offset = start;
+	else
+		reset_ptenuma_scan(p);
+	up_read(&mm->mmap_sem);
+}
+
+/*
+ * Drive the periodic memory faults..
+ */
+void task_tick_numa(struct rq *rq, struct task_struct *curr)
+{
+	struct callback_head *work = &curr->numa_work;
+	u64 period, now;
+
+	/*
+	 * We don't care about NUMA placement if we don't have memory.
+	 */
+	if (!curr->mm || (curr->flags & PF_EXITING) || work->next != work)
+		return;
+
+	/*
+	 * Using runtime rather than walltime has the dual advantage that
+	 * we (mostly) drive the selection from busy threads and that the
+	 * task needs to have done some actual work before we bother with
+	 * NUMA placement.
+	 */
+	now = curr->se.sum_exec_runtime;
+	period = (u64)curr->numa_scan_period * NSEC_PER_MSEC;
+
+	if (now - curr->node_stamp > period) {
+		if (!curr->node_stamp)
+			curr->numa_scan_period = sysctl_numa_balancing_scan_period_min;
+		curr->node_stamp = now;
+
+		if (!time_before(jiffies, curr->mm->numa_next_scan)) {
+			init_task_work(work, task_numa_work); /* TODO: move this into sched_fork() */
+			task_work_add(curr, work, true);
+		}
+	}
+}
+#else
+static void task_tick_numa(struct rq *rq, struct task_struct *curr)
+{
+}
+#endif /* CONFIG_NUMA_BALANCING */
+
 static void
 account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
@@ -801,72 +1026,7 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se)
 }
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
-/* we need this in update_cfs_load and load-balance functions below */
-static inline int throttled_hierarchy(struct cfs_rq *cfs_rq);
 # ifdef CONFIG_SMP
-static void update_cfs_rq_load_contribution(struct cfs_rq *cfs_rq,
-					    int global_update)
-{
-	struct task_group *tg = cfs_rq->tg;
-	long load_avg;
-
-	load_avg = div64_u64(cfs_rq->load_avg, cfs_rq->load_period+1);
-	load_avg -= cfs_rq->load_contribution;
-
-	if (global_update || abs(load_avg) > cfs_rq->load_contribution / 8) {
-		atomic_add(load_avg, &tg->load_weight);
-		cfs_rq->load_contribution += load_avg;
-	}
-}
-
-static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update)
-{
-	u64 period = sysctl_sched_shares_window;
-	u64 now, delta;
-	unsigned long load = cfs_rq->load.weight;
-
-	if (cfs_rq->tg == &root_task_group || throttled_hierarchy(cfs_rq))
-		return;
-
-	now = rq_of(cfs_rq)->clock_task;
-	delta = now - cfs_rq->load_stamp;
-
-	/* truncate load history at 4 idle periods */
-	if (cfs_rq->load_stamp > cfs_rq->load_last &&
-	    now - cfs_rq->load_last > 4 * period) {
-		cfs_rq->load_period = 0;
-		cfs_rq->load_avg = 0;
-		delta = period - 1;
-	}
-
-	cfs_rq->load_stamp = now;
-	cfs_rq->load_unacc_exec_time = 0;
-	cfs_rq->load_period += delta;
-	if (load) {
-		cfs_rq->load_last = now;
-		cfs_rq->load_avg += delta * load;
-	}
-
-	/* consider updating load contribution on each fold or truncate */
-	if (global_update || cfs_rq->load_period > period
-	    || !cfs_rq->load_period)
-		update_cfs_rq_load_contribution(cfs_rq, global_update);
-
-	while (cfs_rq->load_period > period) {
-		/*
-		 * Inline assembly required to prevent the compiler
-		 * optimising this loop into a divmod call.
-		 * See __iter_div_u64_rem() for another example of this.
-		 */
-		asm("" : "+rm" (cfs_rq->load_period));
-		cfs_rq->load_period /= 2;
-		cfs_rq->load_avg /= 2;
-	}
-
-	if (!cfs_rq->curr && !cfs_rq->nr_running && !cfs_rq->load_avg)
-		list_del_leaf_cfs_rq(cfs_rq);
-}
-
 static inline long calc_tg_weight(struct task_group *tg, struct cfs_rq *cfs_rq)
 {
 	long tg_weight;
@@ -876,8 +1036,8 @@ static inline long calc_tg_weight(struct task_group *tg, struct cfs_rq *cfs_rq)
 	 * to gain a more accurate current total weight. See
 	 * update_cfs_rq_load_contribution().
 	 */
-	tg_weight = atomic_read(&tg->load_weight);
-	tg_weight -= cfs_rq->load_contribution;
+	tg_weight = atomic64_read(&tg->load_avg);
+	tg_weight -= cfs_rq->tg_load_contrib;
 	tg_weight += cfs_rq->load.weight;
 
 	return tg_weight;
@@ -901,27 +1061,11 @@ static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
 
 	return shares;
 }
-
-static void update_entity_shares_tick(struct cfs_rq *cfs_rq)
-{
-	if (cfs_rq->load_unacc_exec_time > sysctl_sched_shares_window) {
-		update_cfs_load(cfs_rq, 0);
-		update_cfs_shares(cfs_rq);
-	}
-}
 # else /* CONFIG_SMP */
-static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update)
-{
-}
-
 static inline long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
 {
 	return tg->shares;
 }
-
-static inline void update_entity_shares_tick(struct cfs_rq *cfs_rq)
-{
-}
 # endif /* CONFIG_SMP */
 static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se,
 			    unsigned long weight)
@@ -939,6 +1083,8 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se,
 		account_entity_enqueue(cfs_rq, se);
 }
 
+static inline int throttled_hierarchy(struct cfs_rq *cfs_rq);
+
 static void update_cfs_shares(struct cfs_rq *cfs_rq)
 {
 	struct task_group *tg;
@@ -958,18 +1104,477 @@ static void update_cfs_shares(struct cfs_rq *cfs_rq)
 	reweight_entity(cfs_rq_of(se), se, shares);
 }
 #else /* CONFIG_FAIR_GROUP_SCHED */
-static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update)
+static inline void update_cfs_shares(struct cfs_rq *cfs_rq)
 {
 }
+#endif /* CONFIG_FAIR_GROUP_SCHED */
 
-static inline void update_cfs_shares(struct cfs_rq *cfs_rq)
+/* Only depends on SMP, FAIR_GROUP_SCHED may be removed when useful in lb */
+#if defined(CONFIG_SMP) && defined(CONFIG_FAIR_GROUP_SCHED)
+/*
+ * We choose a half-life close to 1 scheduling period.
+ * Note: The tables below are dependent on this value.
+ */
+#define LOAD_AVG_PERIOD 32
+#define LOAD_AVG_MAX 47742 /* maximum possible load avg */
+#define LOAD_AVG_MAX_N 345 /* number of full periods to produce LOAD_MAX_AVG */
+
+/* Precomputed fixed inverse multiplies for multiplication by y^n */
+static const u32 runnable_avg_yN_inv[] = {
+	0xffffffff, 0xfa83b2da, 0xf5257d14, 0xefe4b99a, 0xeac0c6e6, 0xe5b906e6,
+	0xe0ccdeeb, 0xdbfbb796, 0xd744fcc9, 0xd2a81d91, 0xce248c14, 0xc9b9bd85,
+	0xc5672a10, 0xc12c4cc9, 0xbd08a39e, 0xb8fbaf46, 0xb504f333, 0xb123f581,
+	0xad583ee9, 0xa9a15ab4, 0xa5fed6a9, 0xa2704302, 0x9ef5325f, 0x9b8d39b9,
+	0x9837f050, 0x94f4efa8, 0x91c3d373, 0x8ea4398a, 0x8b95c1e3, 0x88980e80,
+	0x85aac367, 0x82cd8698,
+};
+
+/*
+ * Precomputed \Sum y^k { 1<=k<=n }.  These are floor(true_value) to prevent
+ * over-estimates when re-combining.
+ */
+static const u32 runnable_avg_yN_sum[] = {
+	    0, 1002, 1982, 2941, 3880, 4798, 5697, 6576, 7437, 8279, 9103,
+	 9909,10698,11470,12226,12966,13690,14398,15091,15769,16433,17082,
+	17718,18340,18949,19545,20128,20698,21256,21802,22336,22859,23371,
+};
+
+/*
+ * Approximate:
+ *   val * y^n,    where y^32 ~= 0.5 (~1 scheduling period)
+ */
+static __always_inline u64 decay_load(u64 val, u64 n)
+{
+	unsigned int local_n;
+
+	if (!n)
+		return val;
+	else if (unlikely(n > LOAD_AVG_PERIOD * 63))
+		return 0;
+
+	/* after bounds checking we can collapse to 32-bit */
+	local_n = 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)
+	 *
+	 * To achieve constant time decay_load.
+	 */
+	if (unlikely(local_n >= LOAD_AVG_PERIOD)) {
+		val >>= local_n / LOAD_AVG_PERIOD;
+		local_n %= LOAD_AVG_PERIOD;
+	}
+
+	val *= runnable_avg_yN_inv[local_n];
+	/* We don't use SRR here since we always want to round down. */
+	return val >> 32;
+}
+
+/*
+ * For updates fully spanning n periods, the contribution to runnable
+ * average will be: \Sum 1024*y^n
+ *
+ * We can compute this reasonably efficiently by combining:
+ *   y^PERIOD = 1/2 with precomputed \Sum 1024*y^n {for  n <PERIOD}
+ */
+static u32 __compute_runnable_contrib(u64 n)
+{
+	u32 contrib = 0;
+
+	if (likely(n <= LOAD_AVG_PERIOD))
+		return runnable_avg_yN_sum[n];
+	else if (unlikely(n >= LOAD_AVG_MAX_N))
+		return LOAD_AVG_MAX;
+
+	/* Compute \Sum k^n combining precomputed values for k^i, \Sum k^j */
+	do {
+		contrib /= 2; /* y^LOAD_AVG_PERIOD = 1/2 */
+		contrib += runnable_avg_yN_sum[LOAD_AVG_PERIOD];
+
+		n -= LOAD_AVG_PERIOD;
+	} while (n > LOAD_AVG_PERIOD);
+
+	contrib = decay_load(contrib, n);
+	return contrib + runnable_avg_yN_sum[n];
+}
+
+/*
+ * We can represent the historical contribution to runnable average as the
+ * coefficients of a geometric series.  To do this we sub-divide our runnable
+ * history into segments of approximately 1ms (1024us); label the segment that
+ * occurred N-ms ago p_N, with p_0 corresponding to the current period, e.g.
+ *
+ * [<- 1024us ->|<- 1024us ->|<- 1024us ->| ...
+ *      p0            p1           p2
+ *     (now)       (~1ms ago)  (~2ms ago)
+ *
+ * Let u_i denote the fraction of p_i that the entity was runnable.
+ *
+ * We then designate the fractions u_i as our co-efficients, yielding the
+ * following representation of historical load:
+ *   u_0 + u_1*y + u_2*y^2 + u_3*y^3 + ...
+ *
+ * We choose y based on the with of a reasonably scheduling period, fixing:
+ *   y^32 = 0.5
+ *
+ * This means that the contribution to load ~32ms ago (u_32) will be weighted
+ * approximately half as much as the contribution to load within the last ms
+ * (u_0).
+ *
+ * When a period "rolls over" and we have new u_0`, multiplying the previous
+ * sum again by y is sufficient to update:
+ *   load_avg = u_0` + y*(u_0 + u_1*y + u_2*y^2 + ... )
+ *            = u_0 + u_1*y + u_2*y^2 + ... [re-labeling u_i --> u_{i+1}]
+ */
+static __always_inline int __update_entity_runnable_avg(u64 now,
+							struct sched_avg *sa,
+							int runnable)
 {
+	u64 delta, periods;
+	u32 runnable_contrib;
+	int delta_w, decayed = 0;
+
+	delta = now - sa->last_runnable_update;
+	/*
+	 * This should only happen when time goes backwards, which it
+	 * unfortunately does during sched clock init when we swap over to TSC.
+	 */
+	if ((s64)delta < 0) {
+		sa->last_runnable_update = now;
+		return 0;
+	}
+
+	/*
+	 * Use 1024ns as the unit of measurement since it's a reasonable
+	 * approximation of 1us and fast to compute.
+	 */
+	delta >>= 10;
+	if (!delta)
+		return 0;
+	sa->last_runnable_update = now;
+
+	/* delta_w is the amount already accumulated against our next period */
+	delta_w = sa->runnable_avg_period % 1024;
+	if (delta + delta_w >= 1024) {
+		/* period roll-over */
+		decayed = 1;
+
+		/*
+		 * Now that we know we're crossing a period boundary, figure
+		 * out how much from delta we need to complete the current
+		 * period and accrue it.
+		 */
+		delta_w = 1024 - delta_w;
+		if (runnable)
+			sa->runnable_avg_sum += delta_w;
+		sa->runnable_avg_period += delta_w;
+
+		delta -= delta_w;
+
+		/* Figure out how many additional periods this update spans */
+		periods = delta / 1024;
+		delta %= 1024;
+
+		sa->runnable_avg_sum = decay_load(sa->runnable_avg_sum,
+						  periods + 1);
+		sa->runnable_avg_period = decay_load(sa->runnable_avg_period,
+						     periods + 1);
+
+		/* Efficiently calculate \sum (1..n_period) 1024*y^i */
+		runnable_contrib = __compute_runnable_contrib(periods);
+		if (runnable)
+			sa->runnable_avg_sum += runnable_contrib;
+		sa->runnable_avg_period += runnable_contrib;
+	}
+
+	/* Remainder of delta accrued against u_0` */
+	if (runnable)
+		sa->runnable_avg_sum += delta;
+	sa->runnable_avg_period += delta;
+
+	return decayed;
 }
 
-static inline void update_entity_shares_tick(struct cfs_rq *cfs_rq)
+/* Synchronize an entity's decay with its parenting cfs_rq.*/
+static inline u64 __synchronize_entity_decay(struct sched_entity *se)
 {
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+	u64 decays = atomic64_read(&cfs_rq->decay_counter);
+
+	decays -= se->avg.decay_count;
+	if (!decays)
+		return 0;
+
+	se->avg.load_avg_contrib = decay_load(se->avg.load_avg_contrib, decays);
+	se->avg.decay_count = 0;
+
+	return decays;
 }
-#endif /* CONFIG_FAIR_GROUP_SCHED */
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+static inline void __update_cfs_rq_tg_load_contrib(struct cfs_rq *cfs_rq,
+						 int force_update)
+{
+	struct task_group *tg = cfs_rq->tg;
+	s64 tg_contrib;
+
+	tg_contrib = cfs_rq->runnable_load_avg + cfs_rq->blocked_load_avg;
+	tg_contrib -= cfs_rq->tg_load_contrib;
+
+	if (force_update || abs64(tg_contrib) > cfs_rq->tg_load_contrib / 8) {
+		atomic64_add(tg_contrib, &tg->load_avg);
+		cfs_rq->tg_load_contrib += tg_contrib;
+	}
+}
+
+/*
+ * Aggregate cfs_rq runnable averages into an equivalent task_group
+ * representation for computing load contributions.
+ */
+static inline void __update_tg_runnable_avg(struct sched_avg *sa,
+						  struct cfs_rq *cfs_rq)
+{
+	struct task_group *tg = cfs_rq->tg;
+	long contrib;
+
+	/* The fraction of a cpu used by this cfs_rq */
+	contrib = div_u64(sa->runnable_avg_sum << NICE_0_SHIFT,
+			  sa->runnable_avg_period + 1);
+	contrib -= cfs_rq->tg_runnable_contrib;
+
+	if (abs(contrib) > cfs_rq->tg_runnable_contrib / 64) {
+		atomic_add(contrib, &tg->runnable_avg);
+		cfs_rq->tg_runnable_contrib += contrib;
+	}
+}
+
+static inline void __update_group_entity_contrib(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq = group_cfs_rq(se);
+	struct task_group *tg = cfs_rq->tg;
+	int runnable_avg;
+
+	u64 contrib;
+
+	contrib = cfs_rq->tg_load_contrib * tg->shares;
+	se->avg.load_avg_contrib = div64_u64(contrib,
+					     atomic64_read(&tg->load_avg) + 1);
+
+	/*
+	 * For group entities we need to compute a correction term in the case
+	 * that they are consuming <1 cpu so that we would contribute the same
+	 * load as a task of equal weight.
+	 *
+	 * Explicitly co-ordinating this measurement would be expensive, but
+	 * fortunately the sum of each cpus contribution forms a usable
+	 * lower-bound on the true value.
+	 *
+	 * Consider the aggregate of 2 contributions.  Either they are disjoint
+	 * (and the sum represents true value) or they are disjoint and we are
+	 * understating by the aggregate of their overlap.
+	 *
+	 * Extending this to N cpus, for a given overlap, the maximum amount we
+	 * understand is then n_i(n_i+1)/2 * w_i where n_i is the number of
+	 * cpus that overlap for this interval and w_i is the interval width.
+	 *
+	 * On a small machine; the first term is well-bounded which bounds the
+	 * total error since w_i is a subset of the period.  Whereas on a
+	 * larger machine, while this first term can be larger, if w_i is the
+	 * of consequential size guaranteed to see n_i*w_i quickly converge to
+	 * our upper bound of 1-cpu.
+	 */
+	runnable_avg = atomic_read(&tg->runnable_avg);
+	if (runnable_avg < NICE_0_LOAD) {
+		se->avg.load_avg_contrib *= runnable_avg;
+		se->avg.load_avg_contrib >>= NICE_0_SHIFT;
+	}
+}
+#else
+static inline void __update_cfs_rq_tg_load_contrib(struct cfs_rq *cfs_rq,
+						 int force_update) {}
+static inline void __update_tg_runnable_avg(struct sched_avg *sa,
+						  struct cfs_rq *cfs_rq) {}
+static inline void __update_group_entity_contrib(struct sched_entity *se) {}
+#endif
+
+static inline void __update_task_entity_contrib(struct sched_entity *se)
+{
+	u32 contrib;
+
+	/* avoid overflowing a 32-bit type w/ SCHED_LOAD_SCALE */
+	contrib = se->avg.runnable_avg_sum * scale_load_down(se->load.weight);
+	contrib /= (se->avg.runnable_avg_period + 1);
+	se->avg.load_avg_contrib = scale_load(contrib);
+}
+
+/* Compute the current contribution to load_avg by se, return any delta */
+static long __update_entity_load_avg_contrib(struct sched_entity *se)
+{
+	long old_contrib = se->avg.load_avg_contrib;
+
+	if (entity_is_task(se)) {
+		__update_task_entity_contrib(se);
+	} else {
+		__update_tg_runnable_avg(&se->avg, group_cfs_rq(se));
+		__update_group_entity_contrib(se);
+	}
+
+	return se->avg.load_avg_contrib - old_contrib;
+}
+
+static inline void subtract_blocked_load_contrib(struct cfs_rq *cfs_rq,
+						 long load_contrib)
+{
+	if (likely(load_contrib < cfs_rq->blocked_load_avg))
+		cfs_rq->blocked_load_avg -= load_contrib;
+	else
+		cfs_rq->blocked_load_avg = 0;
+}
+
+static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq);
+
+/* Update a sched_entity's runnable average */
+static inline void update_entity_load_avg(struct sched_entity *se,
+					  int update_cfs_rq)
+{
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+	long contrib_delta;
+	u64 now;
+
+	/*
+	 * For a group entity we need to use their owned cfs_rq_clock_task() in
+	 * case they are the parent of a throttled hierarchy.
+	 */
+	if (entity_is_task(se))
+		now = cfs_rq_clock_task(cfs_rq);
+	else
+		now = cfs_rq_clock_task(group_cfs_rq(se));
+
+	if (!__update_entity_runnable_avg(now, &se->avg, se->on_rq))
+		return;
+
+	contrib_delta = __update_entity_load_avg_contrib(se);
+
+	if (!update_cfs_rq)
+		return;
+
+	if (se->on_rq)
+		cfs_rq->runnable_load_avg += contrib_delta;
+	else
+		subtract_blocked_load_contrib(cfs_rq, -contrib_delta);
+}
+
+/*
+ * Decay the load contributed by all blocked children and account this so that
+ * their contribution may appropriately discounted when they wake up.
+ */
+static void update_cfs_rq_blocked_load(struct cfs_rq *cfs_rq, int force_update)
+{
+	u64 now = cfs_rq_clock_task(cfs_rq) >> 20;
+	u64 decays;
+
+	decays = now - cfs_rq->last_decay;
+	if (!decays && !force_update)
+		return;
+
+	if (atomic64_read(&cfs_rq->removed_load)) {
+		u64 removed_load = atomic64_xchg(&cfs_rq->removed_load, 0);
+		subtract_blocked_load_contrib(cfs_rq, removed_load);
+	}
+
+	if (decays) {
+		cfs_rq->blocked_load_avg = decay_load(cfs_rq->blocked_load_avg,
+						      decays);
+		atomic64_add(decays, &cfs_rq->decay_counter);
+		cfs_rq->last_decay = now;
+	}
+
+	__update_cfs_rq_tg_load_contrib(cfs_rq, force_update);
+}
+
+static inline void update_rq_runnable_avg(struct rq *rq, int runnable)
+{
+	__update_entity_runnable_avg(rq->clock_task, &rq->avg, runnable);
+	__update_tg_runnable_avg(&rq->avg, &rq->cfs);
+}
+
+/* Add the load generated by se into cfs_rq's child load-average */
+static inline void enqueue_entity_load_avg(struct cfs_rq *cfs_rq,
+						  struct sched_entity *se,
+						  int wakeup)
+{
+	/*
+	 * We track migrations using entity decay_count <= 0, on a wake-up
+	 * migration we use a negative decay count to track the remote decays
+	 * accumulated while sleeping.
+	 */
+	if (unlikely(se->avg.decay_count <= 0)) {
+		se->avg.last_runnable_update = rq_of(cfs_rq)->clock_task;
+		if (se->avg.decay_count) {
+			/*
+			 * In a wake-up migration we have to approximate the
+			 * time sleeping.  This is because we can't synchronize
+			 * clock_task between the two cpus, and it is not
+			 * guaranteed to be read-safe.  Instead, we can
+			 * approximate this using our carried decays, which are
+			 * explicitly atomically readable.
+			 */
+			se->avg.last_runnable_update -= (-se->avg.decay_count)
+							<< 20;
+			update_entity_load_avg(se, 0);
+			/* Indicate that we're now synchronized and on-rq */
+			se->avg.decay_count = 0;
+		}
+		wakeup = 0;
+	} else {
+		__synchronize_entity_decay(se);
+	}
+
+	/* migrated tasks did not contribute to our blocked load */
+	if (wakeup) {
+		subtract_blocked_load_contrib(cfs_rq, se->avg.load_avg_contrib);
+		update_entity_load_avg(se, 0);
+	}
+
+	cfs_rq->runnable_load_avg += se->avg.load_avg_contrib;
+	/* we force update consideration on load-balancer moves */
+	update_cfs_rq_blocked_load(cfs_rq, !wakeup);
+}
+
+/*
+ * Remove se's load from this cfs_rq child load-average, if the entity is
+ * transitioning to a blocked state we track its projected decay using
+ * blocked_load_avg.
+ */
+static inline void dequeue_entity_load_avg(struct cfs_rq *cfs_rq,
+						  struct sched_entity *se,
+						  int sleep)
+{
+	update_entity_load_avg(se, 1);
+	/* we force update consideration on load-balancer moves */
+	update_cfs_rq_blocked_load(cfs_rq, !sleep);
+
+	cfs_rq->runnable_load_avg -= se->avg.load_avg_contrib;
+	if (sleep) {
+		cfs_rq->blocked_load_avg += se->avg.load_avg_contrib;
+		se->avg.decay_count = atomic64_read(&cfs_rq->decay_counter);
+	} /* migrations, e.g. sleep=0 leave decay_count == 0 */
+}
+#else
+static inline void update_entity_load_avg(struct sched_entity *se,
+					  int update_cfs_rq) {}
+static inline void update_rq_runnable_avg(struct rq *rq, int runnable) {}
+static inline void enqueue_entity_load_avg(struct cfs_rq *cfs_rq,
+					   struct sched_entity *se,
+					   int wakeup) {}
+static inline void dequeue_entity_load_avg(struct cfs_rq *cfs_rq,
+					   struct sched_entity *se,
+					   int sleep) {}
+static inline void update_cfs_rq_blocked_load(struct cfs_rq *cfs_rq,
+					      int force_update) {}
+#endif
 
 static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
@@ -1075,9 +1680,7 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial)
 	}
 
 	/* ensure we never gain time by being placed backwards. */
-	vruntime = max_vruntime(se->vruntime, vruntime);
-
-	se->vruntime = vruntime;
+	se->vruntime = max_vruntime(se->vruntime, vruntime);
 }
 
 static void check_enqueue_throttle(struct cfs_rq *cfs_rq);
@@ -1096,7 +1699,7 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
 	 * Update run-time statistics of the 'current'.
 	 */
 	update_curr(cfs_rq);
-	update_cfs_load(cfs_rq, 0);
+	enqueue_entity_load_avg(cfs_rq, se, flags & ENQUEUE_WAKEUP);
 	account_entity_enqueue(cfs_rq, se);
 	update_cfs_shares(cfs_rq);
 
@@ -1171,6 +1774,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
 	 * Update run-time statistics of the 'current'.
 	 */
 	update_curr(cfs_rq);
+	dequeue_entity_load_avg(cfs_rq, se, flags & DEQUEUE_SLEEP);
 
 	update_stats_dequeue(cfs_rq, se);
 	if (flags & DEQUEUE_SLEEP) {
@@ -1191,7 +1795,6 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
 	if (se != cfs_rq->curr)
 		__dequeue_entity(cfs_rq, se);
 	se->on_rq = 0;
-	update_cfs_load(cfs_rq, 0);
 	account_entity_dequeue(cfs_rq, se);
 
 	/*
@@ -1340,6 +1943,8 @@ static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *prev)
 		update_stats_wait_start(cfs_rq, prev);
 		/* Put 'current' back into the tree. */
 		__enqueue_entity(cfs_rq, prev);
+		/* in !on_rq case, update occurred at dequeue */
+		update_entity_load_avg(prev, 1);
 	}
 	cfs_rq->curr = NULL;
 }
@@ -1353,9 +1958,10 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued)
 	update_curr(cfs_rq);
 
 	/*
-	 * Update share accounting for long-running entities.
+	 * Ensure that runnable average is periodically updated.
 	 */
-	update_entity_shares_tick(cfs_rq);
+	update_entity_load_avg(curr, 1);
+	update_cfs_rq_blocked_load(cfs_rq, 1);
 
 #ifdef CONFIG_SCHED_HRTICK
 	/*
@@ -1448,6 +2054,15 @@ static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg)
 	return &tg->cfs_bandwidth;
 }
 
+/* rq->task_clock normalized against any time this cfs_rq has spent throttled */
+static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq)
+{
+	if (unlikely(cfs_rq->throttle_count))
+		return cfs_rq->throttled_clock_task;
+
+	return rq_of(cfs_rq)->clock_task - cfs_rq->throttled_clock_task_time;
+}
+
 /* returns 0 on failure to allocate runtime */
 static int assign_cfs_rq_runtime(struct cfs_rq *cfs_rq)
 {
@@ -1592,14 +2207,9 @@ static int tg_unthrottle_up(struct task_group *tg, void *data)
 	cfs_rq->throttle_count--;
 #ifdef CONFIG_SMP
 	if (!cfs_rq->throttle_count) {
-		u64 delta = rq->clock_task - cfs_rq->load_stamp;
-
-		/* leaving throttled state, advance shares averaging windows */
-		cfs_rq->load_stamp += delta;
-		cfs_rq->load_last += delta;
-
-		/* update entity weight now that we are on_rq again */
-		update_cfs_shares(cfs_rq);
+		/* adjust cfs_rq_clock_task() */
+		cfs_rq->throttled_clock_task_time += rq->clock_task -
+					     cfs_rq->throttled_clock_task;
 	}
 #endif
 
@@ -1611,9 +2221,9 @@ static int tg_throttle_down(struct task_group *tg, void *data)
 	struct rq *rq = data;
 	struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)];
 
-	/* group is entering throttled state, record last load */
+	/* group is entering throttled state, stop time */
 	if (!cfs_rq->throttle_count)
-		update_cfs_load(cfs_rq, 0);
+		cfs_rq->throttled_clock_task = rq->clock_task;
 	cfs_rq->throttle_count++;
 
 	return 0;
@@ -1628,7 +2238,7 @@ static void throttle_cfs_rq(struct cfs_rq *cfs_rq)
 
 	se = cfs_rq->tg->se[cpu_of(rq_of(cfs_rq))];
 
-	/* account load preceding throttle */
+	/* freeze hierarchy runnable averages while throttled */
 	rcu_read_lock();
 	walk_tg_tree_from(cfs_rq->tg, tg_throttle_down, tg_nop, (void *)rq);
 	rcu_read_unlock();
@@ -1652,7 +2262,7 @@ static void throttle_cfs_rq(struct cfs_rq *cfs_rq)
 		rq->nr_running -= task_delta;
 
 	cfs_rq->throttled = 1;
-	cfs_rq->throttled_timestamp = rq->clock;
+	cfs_rq->throttled_clock = rq->clock;
 	raw_spin_lock(&cfs_b->lock);
 	list_add_tail_rcu(&cfs_rq->throttled_list, &cfs_b->throttled_cfs_rq);
 	raw_spin_unlock(&cfs_b->lock);
@@ -1670,10 +2280,9 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq)
 
 	cfs_rq->throttled = 0;
 	raw_spin_lock(&cfs_b->lock);
-	cfs_b->throttled_time += rq->clock - cfs_rq->throttled_timestamp;
+	cfs_b->throttled_time += rq->clock - cfs_rq->throttled_clock;
 	list_del_rcu(&cfs_rq->throttled_list);
 	raw_spin_unlock(&cfs_b->lock);
-	cfs_rq->throttled_timestamp = 0;
 
 	update_rq_clock(rq);
 	/* update hierarchical throttle state */
@@ -2052,7 +2661,7 @@ static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b)
 	hrtimer_cancel(&cfs_b->slack_timer);
 }
 
-static void unthrottle_offline_cfs_rqs(struct rq *rq)
+static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq)
 {
 	struct cfs_rq *cfs_rq;
 
@@ -2073,8 +2682,13 @@ static void unthrottle_offline_cfs_rqs(struct rq *rq)
 }
 
 #else /* CONFIG_CFS_BANDWIDTH */
-static __always_inline
-void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, unsigned long delta_exec) {}
+static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq)
+{
+	return rq_of(cfs_rq)->clock_task;
+}
+
+static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq,
+				     unsigned long delta_exec) {}
 static void check_cfs_rq_runtime(struct cfs_rq *cfs_rq) {}
 static void check_enqueue_throttle(struct cfs_rq *cfs_rq) {}
 static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq) {}
@@ -2207,12 +2821,14 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 		if (cfs_rq_throttled(cfs_rq))
 			break;
 
-		update_cfs_load(cfs_rq, 0);
 		update_cfs_shares(cfs_rq);
+		update_entity_load_avg(se, 1);
 	}
 
-	if (!se)
+	if (!se) {
+		update_rq_runnable_avg(rq, rq->nr_running);
 		inc_nr_running(rq);
+	}
 	hrtick_update(rq);
 }
 
@@ -2266,12 +2882,14 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 		if (cfs_rq_throttled(cfs_rq))
 			break;
 
-		update_cfs_load(cfs_rq, 0);
 		update_cfs_shares(cfs_rq);
+		update_entity_load_avg(se, 1);
 	}
 
-	if (!se)
+	if (!se) {
 		dec_nr_running(rq);
+		update_rq_runnable_avg(rq, 1);
+	}
 	hrtick_update(rq);
 }
 
@@ -2634,25 +3252,18 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
  */
 static int select_idle_sibling(struct task_struct *p, int target)
 {
-	int cpu = smp_processor_id();
-	int prev_cpu = task_cpu(p);
 	struct sched_domain *sd;
 	struct sched_group *sg;
-	int i;
+	int i = task_cpu(p);
 
-	/*
-	 * If the task is going to be woken-up on this cpu and if it is
-	 * already idle, then it is the right target.
-	 */
-	if (target == cpu && idle_cpu(cpu))
-		return cpu;
+	if (idle_cpu(target))
+		return target;
 
 	/*
-	 * If the task is going to be woken-up on the cpu where it previously
-	 * ran and if it is currently idle, then it the right target.
+	 * If the prevous cpu is cache affine and idle, don't be stupid.
 	 */
-	if (target == prev_cpu && idle_cpu(prev_cpu))
-		return prev_cpu;
+	if (i != target && cpus_share_cache(i, target) && idle_cpu(i))
+		return i;
 
 	/*
 	 * Otherwise, iterate the domains and find an elegible idle cpu.
@@ -2666,7 +3277,7 @@ static int select_idle_sibling(struct task_struct *p, int target)
 				goto next;
 
 			for_each_cpu(i, sched_group_cpus(sg)) {
-				if (!idle_cpu(i))
+				if (i == target || !idle_cpu(i))
 					goto next;
 			}
 
@@ -2781,6 +3392,37 @@ unlock:
 
 	return new_cpu;
 }
+
+/*
+ * Load-tracking only depends on SMP, FAIR_GROUP_SCHED dependency below may be
+ * removed when useful for applications beyond shares distribution (e.g.
+ * load-balance).
+ */
+#ifdef CONFIG_FAIR_GROUP_SCHED
+/*
+ * Called immediately before a task is migrated to a new cpu; task_cpu(p) and
+ * cfs_rq_of(p) references at time of call are still valid and identify the
+ * previous cpu.  However, the caller only guarantees p->pi_lock is held; no
+ * other assumptions, including the state of rq->lock, should be made.
+ */
+static void
+migrate_task_rq_fair(struct task_struct *p, int next_cpu)
+{
+	struct sched_entity *se = &p->se;
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+	/*
+	 * Load tracking: accumulate removed load so that it can be processed
+	 * when we next update owning cfs_rq under rq->lock.  Tasks contribute
+	 * to blocked load iff they have a positive decay-count.  It can never
+	 * be negative here since on-rq tasks have decay-count == 0.
+	 */
+	if (se->avg.decay_count) {
+		se->avg.decay_count = -__synchronize_entity_decay(se);
+		atomic64_add(se->avg.load_avg_contrib, &cfs_rq->removed_load);
+	}
+}
+#endif
 #endif /* CONFIG_SMP */
 
 static unsigned long
@@ -2907,7 +3549,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_
 	 * Batch and idle tasks do not preempt non-idle tasks (their preemption
 	 * is driven by the tick):
 	 */
-	if (unlikely(p->policy != SCHED_NORMAL))
+	if (unlikely(p->policy != SCHED_NORMAL) || !sched_feat(WAKEUP_PREEMPTION))
 		return;
 
 	find_matching_se(&se, &pse);
@@ -3033,8 +3675,122 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p, bool preemp
 
 #ifdef CONFIG_SMP
 /**************************************************
- * Fair scheduling class load-balancing methods:
- */
+ * Fair scheduling class load-balancing methods.
+ *
+ * BASICS
+ *
+ * The purpose of load-balancing is to achieve the same basic fairness the
+ * per-cpu scheduler provides, namely provide a proportional amount of compute
+ * time to each task. This is expressed in the following equation:
+ *
+ *   W_i,n/P_i == W_j,n/P_j for all i,j                               (1)
+ *
+ * Where W_i,n is the n-th weight average for cpu i. The instantaneous weight
+ * W_i,0 is defined as:
+ *
+ *   W_i,0 = \Sum_j w_i,j                                             (2)
+ *
+ * Where w_i,j is the weight of the j-th runnable task on cpu i. This weight
+ * is derived from the nice value as per prio_to_weight[].
+ *
+ * The weight average is an exponential decay average of the instantaneous
+ * weight:
+ *
+ *   W'_i,n = (2^n - 1) / 2^n * W_i,n + 1 / 2^n * W_i,0               (3)
+ *
+ * P_i is the cpu power (or compute capacity) of cpu i, typically it is the
+ * fraction of 'recent' time available for SCHED_OTHER task execution. But it
+ * can also include other factors [XXX].
+ *
+ * To achieve this balance we define a measure of imbalance which follows
+ * directly from (1):
+ *
+ *   imb_i,j = max{ avg(W/P), W_i/P_i } - min{ avg(W/P), W_j/P_j }    (4)
+ *
+ * We them move tasks around to minimize the imbalance. In the continuous
+ * function space it is obvious this converges, in the discrete case we get
+ * a few fun cases generally called infeasible weight scenarios.
+ *
+ * [XXX expand on:
+ *     - infeasible weights;
+ *     - local vs global optima in the discrete case. ]
+ *
+ *
+ * SCHED DOMAINS
+ *
+ * In order to solve the imbalance equation (4), and avoid the obvious O(n^2)
+ * for all i,j solution, we create a tree of cpus that follows the hardware
+ * topology where each level pairs two lower groups (or better). This results
+ * in O(log n) layers. Furthermore we reduce the number of cpus going up the
+ * tree to only the first of the previous level and we decrease the frequency
+ * of load-balance at each level inv. proportional to the number of cpus in
+ * the groups.
+ *
+ * This yields:
+ *
+ *     log_2 n     1     n
+ *   \Sum       { --- * --- * 2^i } = O(n)                            (5)
+ *     i = 0      2^i   2^i
+ *                               `- size of each group
+ *         |         |     `- number of cpus doing load-balance
+ *         |         `- freq
+ *         `- sum over all levels
+ *
+ * Coupled with a limit on how many tasks we can migrate every balance pass,
+ * this makes (5) the runtime complexity of the balancer.
+ *
+ * An important property here is that each CPU is still (indirectly) connected
+ * to every other cpu in at most O(log n) steps:
+ *
+ * The adjacency matrix of the resulting graph is given by:
+ *
+ *             log_2 n     
+ *   A_i,j = \Union     (i % 2^k == 0) && i / 2^(k+1) == j / 2^(k+1)  (6)
+ *             k = 0
+ *
+ * And you'll find that:
+ *
+ *   A^(log_2 n)_i,j != 0  for all i,j                                (7)
+ *
+ * Showing there's indeed a path between every cpu in at most O(log n) steps.
+ * The task movement gives a factor of O(m), giving a convergence complexity
+ * of:
+ *
+ *   O(nm log n),  n := nr_cpus, m := nr_tasks                        (8)
+ *
+ *
+ * WORK CONSERVING
+ *
+ * In order to avoid CPUs going idle while there's still work to do, new idle
+ * balancing is more aggressive and has the newly idle cpu iterate up the domain
+ * tree itself instead of relying on other CPUs to bring it work.
+ *
+ * This adds some complexity to both (5) and (8) but it reduces the total idle
+ * time.
+ *
+ * [XXX more?]
+ *
+ *
+ * CGROUPS
+ *
+ * Cgroups make a horror show out of (2), instead of a simple sum we get:
+ *
+ *                                s_k,i
+ *   W_i,0 = \Sum_j \Prod_k w_k * -----                               (9)
+ *                                 S_k
+ *
+ * Where
+ *
+ *   s_k,i = \Sum_j w_i,j,k  and  S_k = \Sum_i s_k,i                 (10)
+ *
+ * w_i,j,k is the weight of the j-th runnable task in the k-th cgroup on cpu i.
+ *
+ * The big problem is S_k, its a global sum needed to compute a local (W_i)
+ * property.
+ *
+ * [XXX write more on how we solve this.. _after_ merging pjt's patches that
+ *      rewrite all of this once again.]
+ */ 
 
 static unsigned long __read_mostly max_load_balance_interval = HZ/10;
 
@@ -3300,52 +4056,58 @@ next:
 /*
  * update tg->load_weight by folding this cpu's load_avg
  */
-static int update_shares_cpu(struct task_group *tg, int cpu)
+static void __update_blocked_averages_cpu(struct task_group *tg, int cpu)
 {
-	struct cfs_rq *cfs_rq;
-	unsigned long flags;
-	struct rq *rq;
-
-	if (!tg->se[cpu])
-		return 0;
-
-	rq = cpu_rq(cpu);
-	cfs_rq = tg->cfs_rq[cpu];
-
-	raw_spin_lock_irqsave(&rq->lock, flags);
-
-	update_rq_clock(rq);
-	update_cfs_load(cfs_rq, 1);
+	struct sched_entity *se = tg->se[cpu];
+	struct cfs_rq *cfs_rq = tg->cfs_rq[cpu];
 
-	/*
-	 * We need to update shares after updating tg->load_weight in
-	 * order to adjust the weight of groups with long running tasks.
-	 */
-	update_cfs_shares(cfs_rq);
+	/* throttled entities do not contribute to load */
+	if (throttled_hierarchy(cfs_rq))
+		return;
 
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	update_cfs_rq_blocked_load(cfs_rq, 1);
 
-	return 0;
+	if (se) {
+		update_entity_load_avg(se, 1);
+		/*
+		 * We pivot on our runnable average having decayed to zero for
+		 * list removal.  This generally implies that all our children
+		 * have also been removed (modulo rounding error or bandwidth
+		 * control); however, such cases are rare and we can fix these
+		 * at enqueue.
+		 *
+		 * TODO: fix up out-of-order children on enqueue.
+		 */
+		if (!se->avg.runnable_avg_sum && !cfs_rq->nr_running)
+			list_del_leaf_cfs_rq(cfs_rq);
+	} else {
+		struct rq *rq = rq_of(cfs_rq);
+		update_rq_runnable_avg(rq, rq->nr_running);
+	}
 }
 
-static void update_shares(int cpu)
+static void update_blocked_averages(int cpu)
 {
-	struct cfs_rq *cfs_rq;
 	struct rq *rq = cpu_rq(cpu);
+	struct cfs_rq *cfs_rq;
+	unsigned long flags;
 
-	rcu_read_lock();
+	raw_spin_lock_irqsave(&rq->lock, flags);
+	update_rq_clock(rq);
 	/*
 	 * Iterates the task_group tree in a bottom up fashion, see
 	 * list_add_leaf_cfs_rq() for details.
 	 */
 	for_each_leaf_cfs_rq(rq, cfs_rq) {
-		/* throttled entities do not contribute to load */
-		if (throttled_hierarchy(cfs_rq))
-			continue;
-
-		update_shares_cpu(cfs_rq->tg, cpu);
+		/*
+		 * Note: We may want to consider periodically releasing
+		 * rq->lock about these updates so that creating many task
+		 * groups does not result in continually extending hold time.
+		 */
+		__update_blocked_averages_cpu(cfs_rq->tg, rq->cpu);
 	}
-	rcu_read_unlock();
+
+	raw_spin_unlock_irqrestore(&rq->lock, flags);
 }
 
 /*
@@ -3397,7 +4159,7 @@ static unsigned long task_h_load(struct task_struct *p)
 	return load;
 }
 #else
-static inline void update_shares(int cpu)
+static inline void update_blocked_averages(int cpu)
 {
 }
 
@@ -4457,12 +5219,14 @@ void idle_balance(int this_cpu, struct rq *this_rq)
 	if (this_rq->avg_idle < sysctl_sched_migration_cost)
 		return;
 
+	update_rq_runnable_avg(this_rq, 1);
+
 	/*
 	 * Drop the rq->lock, but keep IRQ/preempt disabled.
 	 */
 	raw_spin_unlock(&this_rq->lock);
 
-	update_shares(this_cpu);
+	update_blocked_averages(this_cpu);
 	rcu_read_lock();
 	for_each_domain(this_cpu, sd) {
 		unsigned long interval;
@@ -4717,7 +5481,7 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle)
 	int update_next_balance = 0;
 	int need_serialize;
 
-	update_shares(cpu);
+	update_blocked_averages(cpu);
 
 	rcu_read_lock();
 	for_each_domain(cpu, sd) {
@@ -4954,6 +5718,11 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 		cfs_rq = cfs_rq_of(se);
 		entity_tick(cfs_rq, se, queued);
 	}
+
+	if (sched_feat_numa(NUMA))
+		task_tick_numa(rq, curr);
+
+	update_rq_runnable_avg(rq, 1);
 }
 
 /*
@@ -5046,6 +5815,20 @@ static void switched_from_fair(struct rq *rq, struct task_struct *p)
 		place_entity(cfs_rq, se, 0);
 		se->vruntime -= cfs_rq->min_vruntime;
 	}
+
+#if defined(CONFIG_FAIR_GROUP_SCHED) && defined(CONFIG_SMP)
+	/*
+	* Remove our load from contribution when we leave sched_fair
+	* and ensure we don't carry in an old decay_count if we
+	* switch back.
+	*/
+	if (p->se.avg.decay_count) {
+		struct cfs_rq *cfs_rq = cfs_rq_of(&p->se);
+		__synchronize_entity_decay(&p->se);
+		subtract_blocked_load_contrib(cfs_rq,
+				p->se.avg.load_avg_contrib);
+	}
+#endif
 }
 
 /*
@@ -5092,11 +5875,16 @@ void init_cfs_rq(struct cfs_rq *cfs_rq)
 #ifndef CONFIG_64BIT
 	cfs_rq->min_vruntime_copy = cfs_rq->min_vruntime;
 #endif
+#if defined(CONFIG_FAIR_GROUP_SCHED) && defined(CONFIG_SMP)
+	atomic64_set(&cfs_rq->decay_counter, 1);
+	atomic64_set(&cfs_rq->removed_load, 0);
+#endif
 }
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 static void task_move_group_fair(struct task_struct *p, int on_rq)
 {
+	struct cfs_rq *cfs_rq;
 	/*
 	 * If the task was not on the rq at the time of this cgroup movement
 	 * it must have been asleep, sleeping tasks keep their ->vruntime
@@ -5128,8 +5916,19 @@ static void task_move_group_fair(struct task_struct *p, int on_rq)
 	if (!on_rq)
 		p->se.vruntime -= cfs_rq_of(&p->se)->min_vruntime;
 	set_task_rq(p, task_cpu(p));
-	if (!on_rq)
-		p->se.vruntime += cfs_rq_of(&p->se)->min_vruntime;
+	if (!on_rq) {
+		cfs_rq = cfs_rq_of(&p->se);
+		p->se.vruntime += cfs_rq->min_vruntime;
+#ifdef CONFIG_SMP
+		/*
+		 * migrate_task_rq_fair() will have removed our previous
+		 * contribution, but we must synchronize for ongoing future
+		 * decay.
+		 */
+		p->se.avg.decay_count = atomic64_read(&cfs_rq->decay_counter);
+		cfs_rq->blocked_load_avg += p->se.avg.load_avg_contrib;
+#endif
+	}
 }
 
 void free_fair_sched_group(struct task_group *tg)
@@ -5214,10 +6013,6 @@ void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq,
 
 	cfs_rq->tg = tg;
 	cfs_rq->rq = rq;
-#ifdef CONFIG_SMP
-	/* allow initial update_cfs_load() to truncate */
-	cfs_rq->load_stamp = 1;
-#endif
 	init_cfs_rq_runtime(cfs_rq);
 
 	tg->cfs_rq[cpu] = cfs_rq;
@@ -5297,7 +6092,7 @@ static unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task
 	 * idle runqueue:
 	 */
 	if (rq->cfs.load.weight)
-		rr_interval = NS_TO_JIFFIES(sched_slice(&rq->cfs, se));
+		rr_interval = NS_TO_JIFFIES(sched_slice(cfs_rq_of(se), se));
 
 	return rr_interval;
 }
@@ -5319,7 +6114,9 @@ const struct sched_class fair_sched_class = {
 
 #ifdef CONFIG_SMP
 	.select_task_rq		= select_task_rq_fair,
-
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	.migrate_task_rq	= migrate_task_rq_fair,
+#endif
 	.rq_online		= rq_online_fair,
 	.rq_offline		= rq_offline_fair,
 
diff --git a/kernel/sched/features.h b/kernel/sched/features.h
index eebefcad702..1ad1d2b5395 100644
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -32,6 +32,11 @@ SCHED_FEAT(LAST_BUDDY, true)
 SCHED_FEAT(CACHE_HOT_BUDDY, true)
 
 /*
+ * Allow wakeup-time preemption of the current task:
+ */
+SCHED_FEAT(WAKEUP_PREEMPTION, true)
+
+/*
  * Use arch dependent cpu power functions
  */
 SCHED_FEAT(ARCH_POWER, true)
@@ -61,3 +66,14 @@ SCHED_FEAT(TTWU_QUEUE, true)
 SCHED_FEAT(FORCE_SD_OVERLAP, false)
 SCHED_FEAT(RT_RUNTIME_SHARE, true)
 SCHED_FEAT(LB_MIN, false)
+
+/*
+ * Apply the automatic NUMA scheduling policy. Enabled automatically
+ * at runtime if running on a NUMA machine. Can be controlled via
+ * numa_balancing=. Allow PTE scanning to be forced on UMA machines
+ * for debugging the core machinery.
+ */
+#ifdef CONFIG_NUMA_BALANCING
+SCHED_FEAT(NUMA,	false)
+SCHED_FEAT(NUMA_FORCE,	false)
+#endif
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index 418feb01344..127a2c4cf4a 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -7,6 +7,8 @@
 
 #include <linux/slab.h>
 
+int sched_rr_timeslice = RR_TIMESLICE;
+
 static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun);
 
 struct rt_bandwidth def_rt_bandwidth;
@@ -566,7 +568,7 @@ static inline struct rt_bandwidth *sched_rt_bandwidth(struct rt_rq *rt_rq)
 static int do_balance_runtime(struct rt_rq *rt_rq)
 {
 	struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq);
-	struct root_domain *rd = cpu_rq(smp_processor_id())->rd;
+	struct root_domain *rd = rq_of_rt_rq(rt_rq)->rd;
 	int i, weight, more = 0;
 	u64 rt_period;
 
@@ -925,8 +927,8 @@ static void update_curr_rt(struct rq *rq)
 		return;
 
 	delta_exec = rq->clock_task - curr->se.exec_start;
-	if (unlikely((s64)delta_exec < 0))
-		delta_exec = 0;
+	if (unlikely((s64)delta_exec <= 0))
+		return;
 
 	schedstat_set(curr->se.statistics.exec_max,
 		      max(curr->se.statistics.exec_max, delta_exec));
@@ -1427,8 +1429,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
 static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 {
 	if (!task_running(rq, p) &&
-	    (cpu < 0 || cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) &&
-	    (p->nr_cpus_allowed > 1))
+	    cpumask_test_cpu(cpu, tsk_cpus_allowed(p)))
 		return 1;
 	return 0;
 }
@@ -1889,8 +1890,11 @@ 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)
-		pull_rt_task(rq);
+	if (!p->on_rq || rq->rt.rt_nr_running)
+		return;
+
+	if (pull_rt_task(rq))
+		resched_task(rq->curr);
 }
 
 void init_sched_rt_class(void)
@@ -1985,7 +1989,11 @@ static void watchdog(struct rq *rq, struct task_struct *p)
 	if (soft != RLIM_INFINITY) {
 		unsigned long next;
 
-		p->rt.timeout++;
+		if (p->rt.watchdog_stamp != jiffies) {
+			p->rt.timeout++;
+			p->rt.watchdog_stamp = jiffies;
+		}
+
 		next = DIV_ROUND_UP(min(soft, hard), USEC_PER_SEC/HZ);
 		if (p->rt.timeout > next)
 			p->cputime_expires.sched_exp = p->se.sum_exec_runtime;
@@ -2010,7 +2018,7 @@ static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued)
 	if (--p->rt.time_slice)
 		return;
 
-	p->rt.time_slice = RR_TIMESLICE;
+	p->rt.time_slice = sched_rr_timeslice;
 
 	/*
 	 * Requeue to the end of queue if we (and all of our ancestors) are the
@@ -2041,7 +2049,7 @@ static unsigned int get_rr_interval_rt(struct rq *rq, struct task_struct *task)
 	 * Time slice is 0 for SCHED_FIFO tasks
 	 */
 	if (task->policy == SCHED_RR)
-		return RR_TIMESLICE;
+		return sched_rr_timeslice;
 	else
 		return 0;
 }
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 7a7db09cfab..cc03cfdf469 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1,5 +1,7 @@
 
 #include <linux/sched.h>
+#include <linux/sched/sysctl.h>
+#include <linux/sched/rt.h>
 #include <linux/mutex.h>
 #include <linux/spinlock.h>
 #include <linux/stop_machine.h>
@@ -112,6 +114,8 @@ struct task_group {
 	unsigned long shares;
 
 	atomic_t load_weight;
+	atomic64_t load_avg;
+	atomic_t runnable_avg;
 #endif
 
 #ifdef CONFIG_RT_GROUP_SCHED
@@ -222,22 +226,29 @@ struct cfs_rq {
 	unsigned int nr_spread_over;
 #endif
 
+#ifdef CONFIG_SMP
+/*
+ * Load-tracking only depends on SMP, FAIR_GROUP_SCHED dependency below may be
+ * removed when useful for applications beyond shares distribution (e.g.
+ * load-balance).
+ */
 #ifdef CONFIG_FAIR_GROUP_SCHED
-	struct rq *rq;	/* cpu runqueue to which this cfs_rq is attached */
-
 	/*
-	 * leaf cfs_rqs are those that hold tasks (lowest schedulable entity in
-	 * a hierarchy). Non-leaf lrqs hold other higher schedulable entities
-	 * (like users, containers etc.)
-	 *
-	 * leaf_cfs_rq_list ties together list of leaf cfs_rq's in a cpu. This
-	 * list is used during load balance.
+	 * CFS Load tracking
+	 * Under CFS, load is tracked on a per-entity basis and aggregated up.
+	 * This allows for the description of both thread and group usage (in
+	 * the FAIR_GROUP_SCHED case).
 	 */
-	int on_list;
-	struct list_head leaf_cfs_rq_list;
-	struct task_group *tg;	/* group that "owns" this runqueue */
+	u64 runnable_load_avg, blocked_load_avg;
+	atomic64_t decay_counter, removed_load;
+	u64 last_decay;
+#endif /* CONFIG_FAIR_GROUP_SCHED */
+/* These always depend on CONFIG_FAIR_GROUP_SCHED */
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	u32 tg_runnable_contrib;
+	u64 tg_load_contrib;
+#endif /* CONFIG_FAIR_GROUP_SCHED */
 
-#ifdef CONFIG_SMP
 	/*
 	 *   h_load = weight * f(tg)
 	 *
@@ -245,26 +256,30 @@ struct cfs_rq {
 	 * this group.
 	 */
 	unsigned long h_load;
+#endif /* CONFIG_SMP */
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	struct rq *rq;	/* cpu runqueue to which this cfs_rq is attached */
 
 	/*
-	 * Maintaining per-cpu shares distribution for group scheduling
+	 * leaf cfs_rqs are those that hold tasks (lowest schedulable entity in
+	 * a hierarchy). Non-leaf lrqs hold other higher schedulable entities
+	 * (like users, containers etc.)
 	 *
-	 * load_stamp is the last time we updated the load average
-	 * load_last is the last time we updated the load average and saw load
-	 * load_unacc_exec_time is currently unaccounted execution time
+	 * leaf_cfs_rq_list ties together list of leaf cfs_rq's in a cpu. This
+	 * list is used during load balance.
 	 */
-	u64 load_avg;
-	u64 load_period;
-	u64 load_stamp, load_last, load_unacc_exec_time;
+	int on_list;
+	struct list_head leaf_cfs_rq_list;
+	struct task_group *tg;	/* group that "owns" this runqueue */
 
-	unsigned long load_contribution;
-#endif /* CONFIG_SMP */
 #ifdef CONFIG_CFS_BANDWIDTH
 	int runtime_enabled;
 	u64 runtime_expires;
 	s64 runtime_remaining;
 
-	u64 throttled_timestamp;
+	u64 throttled_clock, throttled_clock_task;
+	u64 throttled_clock_task_time;
 	int throttled, throttle_count;
 	struct list_head throttled_list;
 #endif /* CONFIG_CFS_BANDWIDTH */
@@ -467,6 +482,8 @@ struct rq {
 #ifdef CONFIG_SMP
 	struct llist_head wake_list;
 #endif
+
+	struct sched_avg avg;
 };
 
 static inline int cpu_of(struct rq *rq)
@@ -648,6 +665,18 @@ extern struct static_key sched_feat_keys[__SCHED_FEAT_NR];
 #define sched_feat(x) (sysctl_sched_features & (1UL << __SCHED_FEAT_##x))
 #endif /* SCHED_DEBUG && HAVE_JUMP_LABEL */
 
+#ifdef CONFIG_NUMA_BALANCING
+#define sched_feat_numa(x) sched_feat(x)
+#ifdef CONFIG_SCHED_DEBUG
+#define numabalancing_enabled sched_feat_numa(NUMA)
+#else
+extern bool numabalancing_enabled;
+#endif /* CONFIG_SCHED_DEBUG */
+#else
+#define sched_feat_numa(x) (0)
+#define numabalancing_enabled (0)
+#endif /* CONFIG_NUMA_BALANCING */
+
 static inline u64 global_rt_period(void)
 {
 	return (u64)sysctl_sched_rt_period * NSEC_PER_USEC;
@@ -1212,4 +1241,3 @@ static inline u64 irq_time_read(int cpu)
 }
 #endif /* CONFIG_64BIT */
 #endif /* CONFIG_IRQ_TIME_ACCOUNTING */
-
diff --git a/kernel/sched/stats.c b/kernel/sched/stats.c
index 903ffa9e887..e036eda1a9c 100644
--- a/kernel/sched/stats.c
+++ b/kernel/sched/stats.c
@@ -21,14 +21,17 @@ static int show_schedstat(struct seq_file *seq, void *v)
 	if (mask_str == NULL)
 		return -ENOMEM;
 
-	seq_printf(seq, "version %d\n", SCHEDSTAT_VERSION);
-	seq_printf(seq, "timestamp %lu\n", jiffies);
-	for_each_online_cpu(cpu) {
-		struct rq *rq = cpu_rq(cpu);
+	if (v == (void *)1) {
+		seq_printf(seq, "version %d\n", SCHEDSTAT_VERSION);
+		seq_printf(seq, "timestamp %lu\n", jiffies);
+	} else {
+		struct rq *rq;
 #ifdef CONFIG_SMP
 		struct sched_domain *sd;
 		int dcount = 0;
 #endif
+		cpu = (unsigned long)(v - 2);
+		rq = cpu_rq(cpu);
 
 		/* runqueue-specific stats */
 		seq_printf(seq,
@@ -77,30 +80,66 @@ static int show_schedstat(struct seq_file *seq, void *v)
 	return 0;
 }
 
-static int schedstat_open(struct inode *inode, struct file *file)
+/*
+ * This itererator needs some explanation.
+ * It returns 1 for the header position.
+ * This means 2 is cpu 0.
+ * In a hotplugged system some cpus, including cpu 0, may be missing so we have
+ * to use cpumask_* to iterate over the cpus.
+ */
+static void *schedstat_start(struct seq_file *file, loff_t *offset)
 {
-	unsigned int size = PAGE_SIZE * (1 + num_online_cpus() / 32);
-	char *buf = kmalloc(size, GFP_KERNEL);
-	struct seq_file *m;
-	int res;
+	unsigned long n = *offset;
 
-	if (!buf)
-		return -ENOMEM;
-	res = single_open(file, show_schedstat, NULL);
-	if (!res) {
-		m = file->private_data;
-		m->buf = buf;
-		m->size = size;
-	} else
-		kfree(buf);
-	return res;
+	if (n == 0)
+		return (void *) 1;
+
+	n--;
+
+	if (n > 0)
+		n = cpumask_next(n - 1, cpu_online_mask);
+	else
+		n = cpumask_first(cpu_online_mask);
+
+	*offset = n + 1;
+
+	if (n < nr_cpu_ids)
+		return (void *)(unsigned long)(n + 2);
+	return NULL;
+}
+
+static void *schedstat_next(struct seq_file *file, void *data, loff_t *offset)
+{
+	(*offset)++;
+	return schedstat_start(file, offset);
+}
+
+static void schedstat_stop(struct seq_file *file, void *data)
+{
+}
+
+static const struct seq_operations schedstat_sops = {
+	.start = schedstat_start,
+	.next  = schedstat_next,
+	.stop  = schedstat_stop,
+	.show  = show_schedstat,
+};
+
+static int schedstat_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &schedstat_sops);
 }
 
+static int schedstat_release(struct inode *inode, struct file *file)
+{
+	return 0;
+};
+
 static const struct file_operations proc_schedstat_operations = {
 	.open    = schedstat_open,
 	.read    = seq_read,
 	.llseek  = seq_lseek,
-	.release = single_release,
+	.release = schedstat_release,
 };
 
 static int __init proc_schedstat_init(void)