Merge branch 'sched/core' into locking/core

Required for the rtmutex/sched_deadline patches which depend on both
branches

12 files changed, 648 insertions, 304 deletions

diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 3ea87fb19a94..afe5bab376c9 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -13,11 +13,12 @@
 #include <linux/bpf.h>
 #include <linux/jhash.h>
 #include <linux/filter.h>
+#include <linux/rculist_nulls.h>
 #include "percpu_freelist.h"
 #include "bpf_lru_list.h"
 
 struct bucket {
-	struct hlist_head head;
+	struct hlist_nulls_head head;
 	raw_spinlock_t lock;
 };
 
@@ -44,9 +45,14 @@ enum extra_elem_state {
 /* each htab element is struct htab_elem + key + value */
 struct htab_elem {
 	union {
-		struct hlist_node hash_node;
-		struct bpf_htab *htab;
-		struct pcpu_freelist_node fnode;
+		struct hlist_nulls_node hash_node;
+		struct {
+			void *padding;
+			union {
+				struct bpf_htab *htab;
+				struct pcpu_freelist_node fnode;
+			};
+		};
 	};
 	union {
 		struct rcu_head rcu;
@@ -162,7 +168,8 @@ skip_percpu_elems:
 				 offsetof(struct htab_elem, lru_node),
 				 htab->elem_size, htab->map.max_entries);
 	else
-		pcpu_freelist_populate(&htab->freelist, htab->elems,
+		pcpu_freelist_populate(&htab->freelist,
+				       htab->elems + offsetof(struct htab_elem, fnode),
 				       htab->elem_size, htab->map.max_entries);
 
 	return 0;
@@ -217,6 +224,11 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 	int err, i;
 	u64 cost;
 
+	BUILD_BUG_ON(offsetof(struct htab_elem, htab) !=
+		     offsetof(struct htab_elem, hash_node.pprev));
+	BUILD_BUG_ON(offsetof(struct htab_elem, fnode.next) !=
+		     offsetof(struct htab_elem, hash_node.pprev));
+
 	if (lru && !capable(CAP_SYS_ADMIN))
 		/* LRU implementation is much complicated than other
 		 * maps.  Hence, limit to CAP_SYS_ADMIN for now.
@@ -326,7 +338,7 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 		goto free_htab;
 
 	for (i = 0; i < htab->n_buckets; i++) {
-		INIT_HLIST_HEAD(&htab->buckets[i].head);
+		INIT_HLIST_NULLS_HEAD(&htab->buckets[i].head, i);
 		raw_spin_lock_init(&htab->buckets[i].lock);
 	}
 
@@ -366,20 +378,44 @@ static inline struct bucket *__select_bucket(struct bpf_htab *htab, u32 hash)
 	return &htab->buckets[hash & (htab->n_buckets - 1)];
 }
 
-static inline struct hlist_head *select_bucket(struct bpf_htab *htab, u32 hash)
+static inline struct hlist_nulls_head *select_bucket(struct bpf_htab *htab, u32 hash)
 {
 	return &__select_bucket(htab, hash)->head;
 }
 
-static struct htab_elem *lookup_elem_raw(struct hlist_head *head, u32 hash,
+/* this lookup function can only be called with bucket lock taken */
+static struct htab_elem *lookup_elem_raw(struct hlist_nulls_head *head, u32 hash,
 					 void *key, u32 key_size)
 {
+	struct hlist_nulls_node *n;
+	struct htab_elem *l;
+
+	hlist_nulls_for_each_entry_rcu(l, n, head, hash_node)
+		if (l->hash == hash && !memcmp(&l->key, key, key_size))
+			return l;
+
+	return NULL;
+}
+
+/* can be called without bucket lock. it will repeat the loop in
+ * the unlikely event when elements moved from one bucket into another
+ * while link list is being walked
+ */
+static struct htab_elem *lookup_nulls_elem_raw(struct hlist_nulls_head *head,
+					       u32 hash, void *key,
+					       u32 key_size, u32 n_buckets)
+{
+	struct hlist_nulls_node *n;
 	struct htab_elem *l;
 
-	hlist_for_each_entry_rcu(l, head, hash_node)
+again:
+	hlist_nulls_for_each_entry_rcu(l, n, head, hash_node)
 		if (l->hash == hash && !memcmp(&l->key, key, key_size))
 			return l;
 
+	if (unlikely(get_nulls_value(n) != (hash & (n_buckets - 1))))
+		goto again;
+
 	return NULL;
 }
 
@@ -387,7 +423,7 @@ static struct htab_elem *lookup_elem_raw(struct hlist_head *head, u32 hash,
 static void *__htab_map_lookup_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
-	struct hlist_head *head;
+	struct hlist_nulls_head *head;
 	struct htab_elem *l;
 	u32 hash, key_size;
 
@@ -400,7 +436,7 @@ static void *__htab_map_lookup_elem(struct bpf_map *map, void *key)
 
 	head = select_bucket(htab, hash);
 
-	l = lookup_elem_raw(head, hash, key, key_size);
+	l = lookup_nulls_elem_raw(head, hash, key, key_size, htab->n_buckets);
 
 	return l;
 }
@@ -433,8 +469,9 @@ static void *htab_lru_map_lookup_elem(struct bpf_map *map, void *key)
 static bool htab_lru_map_delete_node(void *arg, struct bpf_lru_node *node)
 {
 	struct bpf_htab *htab = (struct bpf_htab *)arg;
-	struct htab_elem *l, *tgt_l;
-	struct hlist_head *head;
+	struct htab_elem *l = NULL, *tgt_l;
+	struct hlist_nulls_head *head;
+	struct hlist_nulls_node *n;
 	unsigned long flags;
 	struct bucket *b;
 
@@ -444,9 +481,9 @@ static bool htab_lru_map_delete_node(void *arg, struct bpf_lru_node *node)
 
 	raw_spin_lock_irqsave(&b->lock, flags);
 
-	hlist_for_each_entry_rcu(l, head, hash_node)
+	hlist_nulls_for_each_entry_rcu(l, n, head, hash_node)
 		if (l == tgt_l) {
-			hlist_del_rcu(&l->hash_node);
+			hlist_nulls_del_rcu(&l->hash_node);
 			break;
 		}
 
@@ -459,7 +496,7 @@ static bool htab_lru_map_delete_node(void *arg, struct bpf_lru_node *node)
 static int htab_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
-	struct hlist_head *head;
+	struct hlist_nulls_head *head;
 	struct htab_elem *l, *next_l;
 	u32 hash, key_size;
 	int i;
@@ -473,7 +510,7 @@ static int htab_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
 	head = select_bucket(htab, hash);
 
 	/* lookup the key */
-	l = lookup_elem_raw(head, hash, key, key_size);
+	l = lookup_nulls_elem_raw(head, hash, key, key_size, htab->n_buckets);
 
 	if (!l) {
 		i = 0;
@@ -481,7 +518,7 @@ static int htab_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
 	}
 
 	/* key was found, get next key in the same bucket */
-	next_l = hlist_entry_safe(rcu_dereference_raw(hlist_next_rcu(&l->hash_node)),
+	next_l = hlist_nulls_entry_safe(rcu_dereference_raw(hlist_nulls_next_rcu(&l->hash_node)),
 				  struct htab_elem, hash_node);
 
 	if (next_l) {
@@ -500,7 +537,7 @@ find_first_elem:
 		head = select_bucket(htab, i);
 
 		/* pick first element in the bucket */
-		next_l = hlist_entry_safe(rcu_dereference_raw(hlist_first_rcu(head)),
+		next_l = hlist_nulls_entry_safe(rcu_dereference_raw(hlist_nulls_first_rcu(head)),
 					  struct htab_elem, hash_node);
 		if (next_l) {
 			/* if it's not empty, just return it */
@@ -582,9 +619,13 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key,
 	int err = 0;
 
 	if (prealloc) {
-		l_new = (struct htab_elem *)pcpu_freelist_pop(&htab->freelist);
-		if (!l_new)
+		struct pcpu_freelist_node *l;
+
+		l = pcpu_freelist_pop(&htab->freelist);
+		if (!l)
 			err = -E2BIG;
+		else
+			l_new = container_of(l, struct htab_elem, fnode);
 	} else {
 		if (atomic_inc_return(&htab->count) > htab->map.max_entries) {
 			atomic_dec(&htab->count);
@@ -661,7 +702,7 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value,
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new = NULL, *l_old;
-	struct hlist_head *head;
+	struct hlist_nulls_head *head;
 	unsigned long flags;
 	struct bucket *b;
 	u32 key_size, hash;
@@ -700,9 +741,9 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value,
 	/* add new element to the head of the list, so that
 	 * concurrent search will find it before old elem
 	 */
-	hlist_add_head_rcu(&l_new->hash_node, head);
+	hlist_nulls_add_head_rcu(&l_new->hash_node, head);
 	if (l_old) {
-		hlist_del_rcu(&l_old->hash_node);
+		hlist_nulls_del_rcu(&l_old->hash_node);
 		free_htab_elem(htab, l_old);
 	}
 	ret = 0;
@@ -716,7 +757,7 @@ static int htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value,
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new, *l_old = NULL;
-	struct hlist_head *head;
+	struct hlist_nulls_head *head;
 	unsigned long flags;
 	struct bucket *b;
 	u32 key_size, hash;
@@ -757,10 +798,10 @@ static int htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value,
 	/* add new element to the head of the list, so that
 	 * concurrent search will find it before old elem
 	 */
-	hlist_add_head_rcu(&l_new->hash_node, head);
+	hlist_nulls_add_head_rcu(&l_new->hash_node, head);
 	if (l_old) {
 		bpf_lru_node_set_ref(&l_new->lru_node);
-		hlist_del_rcu(&l_old->hash_node);
+		hlist_nulls_del_rcu(&l_old->hash_node);
 	}
 	ret = 0;
 
@@ -781,7 +822,7 @@ static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key,
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new = NULL, *l_old;
-	struct hlist_head *head;
+	struct hlist_nulls_head *head;
 	unsigned long flags;
 	struct bucket *b;
 	u32 key_size, hash;
@@ -820,7 +861,7 @@ static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key,
 			ret = PTR_ERR(l_new);
 			goto err;
 		}
-		hlist_add_head_rcu(&l_new->hash_node, head);
+		hlist_nulls_add_head_rcu(&l_new->hash_node, head);
 	}
 	ret = 0;
 err:
@@ -834,7 +875,7 @@ static int __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key,
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new = NULL, *l_old;
-	struct hlist_head *head;
+	struct hlist_nulls_head *head;
 	unsigned long flags;
 	struct bucket *b;
 	u32 key_size, hash;
@@ -882,7 +923,7 @@ static int __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key,
 	} else {
 		pcpu_copy_value(htab, htab_elem_get_ptr(l_new, key_size),
 				value, onallcpus);
-		hlist_add_head_rcu(&l_new->hash_node, head);
+		hlist_nulls_add_head_rcu(&l_new->hash_node, head);
 		l_new = NULL;
 	}
 	ret = 0;
@@ -910,7 +951,7 @@ static int htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key,
 static int htab_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
-	struct hlist_head *head;
+	struct hlist_nulls_head *head;
 	struct bucket *b;
 	struct htab_elem *l;
 	unsigned long flags;
@@ -930,7 +971,7 @@ static int htab_map_delete_elem(struct bpf_map *map, void *key)
 	l = lookup_elem_raw(head, hash, key, key_size);
 
 	if (l) {
-		hlist_del_rcu(&l->hash_node);
+		hlist_nulls_del_rcu(&l->hash_node);
 		free_htab_elem(htab, l);
 		ret = 0;
 	}
@@ -942,7 +983,7 @@ static int htab_map_delete_elem(struct bpf_map *map, void *key)
 static int htab_lru_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
-	struct hlist_head *head;
+	struct hlist_nulls_head *head;
 	struct bucket *b;
 	struct htab_elem *l;
 	unsigned long flags;
@@ -962,7 +1003,7 @@ static int htab_lru_map_delete_elem(struct bpf_map *map, void *key)
 	l = lookup_elem_raw(head, hash, key, key_size);
 
 	if (l) {
-		hlist_del_rcu(&l->hash_node);
+		hlist_nulls_del_rcu(&l->hash_node);
 		ret = 0;
 	}
 
@@ -977,12 +1018,12 @@ static void delete_all_elements(struct bpf_htab *htab)
 	int i;
 
 	for (i = 0; i < htab->n_buckets; i++) {
-		struct hlist_head *head = select_bucket(htab, i);
-		struct hlist_node *n;
+		struct hlist_nulls_head *head = select_bucket(htab, i);
+		struct hlist_nulls_node *n;
 		struct htab_elem *l;
 
-		hlist_for_each_entry_safe(l, n, head, hash_node) {
-			hlist_del_rcu(&l->hash_node);
+		hlist_nulls_for_each_entry_safe(l, n, head, hash_node) {
+			hlist_nulls_del_rcu(&l->hash_node);
 			if (l->state != HTAB_EXTRA_ELEM_USED)
 				htab_elem_free(htab, l);
 		}
diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c
index 8bfe0afaee10..b37bd9ab7f57 100644
--- a/kernel/bpf/lpm_trie.c
+++ b/kernel/bpf/lpm_trie.c
@@ -500,9 +500,15 @@ unlock:
 	raw_spin_unlock(&trie->lock);
 }
 
+static int trie_get_next_key(struct bpf_map *map, void *key, void *next_key)
+{
+	return -ENOTSUPP;
+}
+
 static const struct bpf_map_ops trie_ops = {
 	.map_alloc = trie_alloc,
 	.map_free = trie_free,
+	.map_get_next_key = trie_get_next_key,
 	.map_lookup_elem = trie_lookup_elem,
 	.map_update_elem = trie_update_elem,
 	.map_delete_elem = trie_delete_elem,
diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c
index 56eba9caa632..1dc22f6b49f5 100644
--- a/kernel/cgroup/cgroup-v1.c
+++ b/kernel/cgroup/cgroup-v1.c
@@ -1329,7 +1329,7 @@ static int cgroup_css_links_read(struct seq_file *seq, void *v)
 		struct task_struct *task;
 		int count = 0;
 
-		seq_printf(seq, "css_set %p\n", cset);
+		seq_printf(seq, "css_set %pK\n", cset);
 
 		list_for_each_entry(task, &cset->tasks, cg_list) {
 			if (count++ > MAX_TASKS_SHOWN_PER_CSS)
diff --git a/kernel/cgroup/pids.c b/kernel/cgroup/pids.c
index e756dae49300..2237201d66d5 100644
--- a/kernel/cgroup/pids.c
+++ b/kernel/cgroup/pids.c
@@ -229,7 +229,7 @@ static int pids_can_fork(struct task_struct *task)
 		/* Only log the first time events_limit is incremented. */
 		if (atomic64_inc_return(&pids->events_limit) == 1) {
 			pr_info("cgroup: fork rejected by pids controller in ");
-			pr_cont_cgroup_path(task_cgroup(current, pids_cgrp_id));
+			pr_cont_cgroup_path(css->cgroup);
 			pr_cont("\n");
 		}
 		cgroup_file_notify(&pids->events_file);
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 3b31fc05a0f1..ab9f6ac099a7 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -86,21 +86,6 @@ int sysctl_sched_rt_runtime = 950000;
 cpumask_var_t cpu_isolated_map;
 
 /*
- * this_rq_lock - lock this runqueue and disable interrupts.
- */
-static struct rq *this_rq_lock(void)
-	__acquires(rq->lock)
-{
-	struct rq *rq;
-
-	local_irq_disable();
-	rq = this_rq();
-	raw_spin_lock(&rq->lock);
-
-	return rq;
-}
-
-/*
  * __task_rq_lock - lock the rq @p resides on.
  */
 struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf)
@@ -233,8 +218,11 @@ void update_rq_clock(struct rq *rq)
 		return;
 
 #ifdef CONFIG_SCHED_DEBUG
+	if (sched_feat(WARN_DOUBLE_CLOCK))
+		SCHED_WARN_ON(rq->clock_update_flags & RQCF_UPDATED);
 	rq->clock_update_flags |= RQCF_UPDATED;
 #endif
+
 	delta = sched_clock_cpu(cpu_of(rq)) - rq->clock;
 	if (delta < 0)
 		return;
@@ -261,13 +249,14 @@ static void hrtick_clear(struct rq *rq)
 static enum hrtimer_restart hrtick(struct hrtimer *timer)
 {
 	struct rq *rq = container_of(timer, struct rq, hrtick_timer);
+	struct rq_flags rf;
 
 	WARN_ON_ONCE(cpu_of(rq) != smp_processor_id());
 
-	raw_spin_lock(&rq->lock);
+	rq_lock(rq, &rf);
 	update_rq_clock(rq);
 	rq->curr->sched_class->task_tick(rq, rq->curr, 1);
-	raw_spin_unlock(&rq->lock);
+	rq_unlock(rq, &rf);
 
 	return HRTIMER_NORESTART;
 }
@@ -287,11 +276,12 @@ static void __hrtick_restart(struct rq *rq)
 static void __hrtick_start(void *arg)
 {
 	struct rq *rq = arg;
+	struct rq_flags rf;
 
-	raw_spin_lock(&rq->lock);
+	rq_lock(rq, &rf);
 	__hrtick_restart(rq);
 	rq->hrtick_csd_pending = 0;
-	raw_spin_unlock(&rq->lock);
+	rq_unlock(rq, &rf);
 }
 
 /*
@@ -762,17 +752,23 @@ static void set_load_weight(struct task_struct *p)
 
 static inline void enqueue_task(struct rq *rq, struct task_struct *p, int flags)
 {
-	update_rq_clock(rq);
+	if (!(flags & ENQUEUE_NOCLOCK))
+		update_rq_clock(rq);
+
 	if (!(flags & ENQUEUE_RESTORE))
 		sched_info_queued(rq, p);
+
 	p->sched_class->enqueue_task(rq, p, flags);
 }
 
 static inline void dequeue_task(struct rq *rq, struct task_struct *p, int flags)
 {
-	update_rq_clock(rq);
+	if (!(flags & DEQUEUE_NOCLOCK))
+		update_rq_clock(rq);
+
 	if (!(flags & DEQUEUE_SAVE))
 		sched_info_dequeued(rq, p);
+
 	p->sched_class->dequeue_task(rq, p, flags);
 }
 
@@ -946,18 +942,19 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags)
  *
  * Returns (locked) new rq. Old rq's lock is released.
  */
-static struct rq *move_queued_task(struct rq *rq, struct task_struct *p, int new_cpu)
+static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf,
+				   struct task_struct *p, int new_cpu)
 {
 	lockdep_assert_held(&rq->lock);
 
 	p->on_rq = TASK_ON_RQ_MIGRATING;
-	dequeue_task(rq, p, 0);
+	dequeue_task(rq, p, DEQUEUE_NOCLOCK);
 	set_task_cpu(p, new_cpu);
-	raw_spin_unlock(&rq->lock);
+	rq_unlock(rq, rf);
 
 	rq = cpu_rq(new_cpu);
 
-	raw_spin_lock(&rq->lock);
+	rq_lock(rq, rf);
 	BUG_ON(task_cpu(p) != new_cpu);
 	enqueue_task(rq, p, 0);
 	p->on_rq = TASK_ON_RQ_QUEUED;
@@ -980,7 +977,8 @@ struct migration_arg {
  * So we race with normal scheduler movements, but that's OK, as long
  * as the task is no longer on this CPU.
  */
-static struct rq *__migrate_task(struct rq *rq, struct task_struct *p, int dest_cpu)
+static struct rq *__migrate_task(struct rq *rq, struct rq_flags *rf,
+				 struct task_struct *p, int dest_cpu)
 {
 	if (unlikely(!cpu_active(dest_cpu)))
 		return rq;
@@ -989,7 +987,8 @@ static struct rq *__migrate_task(struct rq *rq, struct task_struct *p, int dest_
 	if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed))
 		return rq;
 
-	rq = move_queued_task(rq, p, dest_cpu);
+	update_rq_clock(rq);
+	rq = move_queued_task(rq, rf, p, dest_cpu);
 
 	return rq;
 }
@@ -1004,6 +1003,7 @@ static int migration_cpu_stop(void *data)
 	struct migration_arg *arg = data;
 	struct task_struct *p = arg->task;
 	struct rq *rq = this_rq();
+	struct rq_flags rf;
 
 	/*
 	 * The original target CPU might have gone down and we might
@@ -1018,7 +1018,7 @@ static int migration_cpu_stop(void *data)
 	sched_ttwu_pending();
 
 	raw_spin_lock(&p->pi_lock);
-	raw_spin_lock(&rq->lock);
+	rq_lock(rq, &rf);
 	/*
 	 * If task_rq(p) != rq, it cannot be migrated here, because we're
 	 * holding rq->lock, if p->on_rq == 0 it cannot get enqueued because
@@ -1026,11 +1026,11 @@ static int migration_cpu_stop(void *data)
 	 */
 	if (task_rq(p) == rq) {
 		if (task_on_rq_queued(p))
-			rq = __migrate_task(rq, p, arg->dest_cpu);
+			rq = __migrate_task(rq, &rf, p, arg->dest_cpu);
 		else
 			p->wake_cpu = arg->dest_cpu;
 	}
-	raw_spin_unlock(&rq->lock);
+	rq_unlock(rq, &rf);
 	raw_spin_unlock(&p->pi_lock);
 
 	local_irq_enable();
@@ -1063,7 +1063,7 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask)
 		 * holding rq->lock.
 		 */
 		lockdep_assert_held(&rq->lock);
-		dequeue_task(rq, p, DEQUEUE_SAVE);
+		dequeue_task(rq, p, DEQUEUE_SAVE | DEQUEUE_NOCLOCK);
 	}
 	if (running)
 		put_prev_task(rq, p);
@@ -1071,7 +1071,7 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask)
 	p->sched_class->set_cpus_allowed(p, new_mask);
 
 	if (queued)
-		enqueue_task(rq, p, ENQUEUE_RESTORE);
+		enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK);
 	if (running)
 		set_curr_task(rq, p);
 }
@@ -1150,9 +1150,7 @@ static int __set_cpus_allowed_ptr(struct task_struct *p,
 		 * OK, since we're going to drop the lock immediately
 		 * afterwards anyway.
 		 */
-		rq_unpin_lock(rq, &rf);
-		rq = move_queued_task(rq, p, dest_cpu);
-		rq_repin_lock(rq, &rf);
+		rq = move_queued_task(rq, &rf, p, dest_cpu);
 	}
 out:
 	task_rq_unlock(rq, p, &rf);
@@ -1217,16 +1215,24 @@ static void __migrate_swap_task(struct task_struct *p, int cpu)
 {
 	if (task_on_rq_queued(p)) {
 		struct rq *src_rq, *dst_rq;
+		struct rq_flags srf, drf;
 
 		src_rq = task_rq(p);
 		dst_rq = cpu_rq(cpu);
 
+		rq_pin_lock(src_rq, &srf);
+		rq_pin_lock(dst_rq, &drf);
+
 		p->on_rq = TASK_ON_RQ_MIGRATING;
 		deactivate_task(src_rq, p, 0);
 		set_task_cpu(p, cpu);
 		activate_task(dst_rq, p, 0);
 		p->on_rq = TASK_ON_RQ_QUEUED;
 		check_preempt_curr(dst_rq, p, 0);
+
+		rq_unpin_lock(dst_rq, &drf);
+		rq_unpin_lock(src_rq, &srf);
+
 	} else {
 		/*
 		 * Task isn't running anymore; make it appear like we migrated
@@ -1680,7 +1686,7 @@ static void
 ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags,
 		 struct rq_flags *rf)
 {
-	int en_flags = ENQUEUE_WAKEUP;
+	int en_flags = ENQUEUE_WAKEUP | ENQUEUE_NOCLOCK;
 
 	lockdep_assert_held(&rq->lock);
 
@@ -1726,14 +1732,13 @@ void sched_ttwu_pending(void)
 	struct rq *rq = this_rq();
 	struct llist_node *llist = llist_del_all(&rq->wake_list);
 	struct task_struct *p;
-	unsigned long flags;
 	struct rq_flags rf;
 
 	if (!llist)
 		return;
 
-	raw_spin_lock_irqsave(&rq->lock, flags);
-	rq_pin_lock(rq, &rf);
+	rq_lock_irqsave(rq, &rf);
+	update_rq_clock(rq);
 
 	while (llist) {
 		int wake_flags = 0;
@@ -1747,8 +1752,7 @@ void sched_ttwu_pending(void)
 		ttwu_do_activate(rq, p, wake_flags, &rf);
 	}
 
-	rq_unpin_lock(rq, &rf);
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	rq_unlock_irqrestore(rq, &rf);
 }
 
 void scheduler_ipi(void)
@@ -1806,7 +1810,7 @@ static void ttwu_queue_remote(struct task_struct *p, int cpu, int wake_flags)
 void wake_up_if_idle(int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
-	unsigned long flags;
+	struct rq_flags rf;
 
 	rcu_read_lock();
 
@@ -1816,11 +1820,11 @@ void wake_up_if_idle(int cpu)
 	if (set_nr_if_polling(rq->idle)) {
 		trace_sched_wake_idle_without_ipi(cpu);
 	} else {
-		raw_spin_lock_irqsave(&rq->lock, flags);
+		rq_lock_irqsave(rq, &rf);
 		if (is_idle_task(rq->curr))
 			smp_send_reschedule(cpu);
 		/* Else CPU is not idle, do nothing here: */
-		raw_spin_unlock_irqrestore(&rq->lock, flags);
+		rq_unlock_irqrestore(rq, &rf);
 	}
 
 out:
@@ -1846,11 +1850,10 @@ static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags)
 	}
 #endif
 
-	raw_spin_lock(&rq->lock);
-	rq_pin_lock(rq, &rf);
+	rq_lock(rq, &rf);
+	update_rq_clock(rq);
 	ttwu_do_activate(rq, p, wake_flags, &rf);
-	rq_unpin_lock(rq, &rf);
-	raw_spin_unlock(&rq->lock);
+	rq_unlock(rq, &rf);
 }
 
 /*
@@ -2097,11 +2100,9 @@ static void try_to_wake_up_local(struct task_struct *p, struct rq_flags *rf)
 		 * disabled avoiding further scheduler activity on it and we've
 		 * not yet picked a replacement task.
 		 */
-		rq_unpin_lock(rq, rf);
-		raw_spin_unlock(&rq->lock);
+		rq_unlock(rq, rf);
 		raw_spin_lock(&p->pi_lock);
-		raw_spin_lock(&rq->lock);
-		rq_repin_lock(rq, rf);
+		rq_relock(rq, rf);
 	}
 
 	if (!(p->state & TASK_NORMAL))
@@ -2114,7 +2115,7 @@ static void try_to_wake_up_local(struct task_struct *p, struct rq_flags *rf)
 			delayacct_blkio_end();
 			atomic_dec(&rq->nr_iowait);
 		}
-		ttwu_activate(rq, p, ENQUEUE_WAKEUP);
+		ttwu_activate(rq, p, ENQUEUE_WAKEUP | ENQUEUE_NOCLOCK);
 	}
 
 	ttwu_do_wakeup(rq, p, 0, rf);
@@ -2555,7 +2556,7 @@ void wake_up_new_task(struct task_struct *p)
 	update_rq_clock(rq);
 	post_init_entity_util_avg(&p->se);
 
-	activate_task(rq, p, 0);
+	activate_task(rq, p, ENQUEUE_NOCLOCK);
 	p->on_rq = TASK_ON_RQ_QUEUED;
 	trace_sched_wakeup_new(p);
 	check_preempt_curr(rq, p, WF_FORK);
@@ -3093,15 +3094,18 @@ void scheduler_tick(void)
 	int cpu = smp_processor_id();
 	struct rq *rq = cpu_rq(cpu);
 	struct task_struct *curr = rq->curr;
+	struct rq_flags rf;
 
 	sched_clock_tick();
 
-	raw_spin_lock(&rq->lock);
+	rq_lock(rq, &rf);
+
 	update_rq_clock(rq);
 	curr->sched_class->task_tick(rq, curr, 0);
 	cpu_load_update_active(rq);
 	calc_global_load_tick(rq);
-	raw_spin_unlock(&rq->lock);
+
+	rq_unlock(rq, &rf);
 
 	perf_event_task_tick();
 
@@ -3386,18 +3390,18 @@ static void __sched notrace __schedule(bool preempt)
 	 * done by the caller to avoid the race with signal_wake_up().
 	 */
 	smp_mb__before_spinlock();
-	raw_spin_lock(&rq->lock);
-	rq_pin_lock(rq, &rf);
+	rq_lock(rq, &rf);
 
 	/* Promote REQ to ACT */
 	rq->clock_update_flags <<= 1;
+	update_rq_clock(rq);
 
 	switch_count = &prev->nivcsw;
 	if (!preempt && prev->state) {
 		if (unlikely(signal_pending_state(prev->state, prev))) {
 			prev->state = TASK_RUNNING;
 		} else {
-			deactivate_task(rq, prev, DEQUEUE_SLEEP);
+			deactivate_task(rq, prev, DEQUEUE_SLEEP | DEQUEUE_NOCLOCK);
 			prev->on_rq = 0;
 
 			if (prev->in_iowait) {
@@ -3421,9 +3425,6 @@ static void __sched notrace __schedule(bool preempt)
 		switch_count = &prev->nvcsw;
 	}
 
-	if (task_on_rq_queued(prev))
-		update_rq_clock(rq);
-
 	next = pick_next_task(rq, prev, &rf);
 	clear_tsk_need_resched(prev);
 	clear_preempt_need_resched();
@@ -3439,8 +3440,7 @@ static void __sched notrace __schedule(bool preempt)
 		rq = context_switch(rq, prev, next, &rf);
 	} else {
 		rq->clock_update_flags &= ~(RQCF_ACT_SKIP|RQCF_REQ_SKIP);
-		rq_unpin_lock(rq, &rf);
-		raw_spin_unlock_irq(&rq->lock);
+		rq_unlock_irq(rq, &rf);
 	}
 
 	balance_callback(rq);
@@ -3684,7 +3684,8 @@ EXPORT_SYMBOL(default_wake_function);
  */
 void rt_mutex_setprio(struct task_struct *p, int prio)
 {
-	int oldprio, queued, running, queue_flag = DEQUEUE_SAVE | DEQUEUE_MOVE;
+	int oldprio, queued, running, queue_flag =
+		DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK;
 	const struct sched_class *prev_class;
 	struct rq_flags rf;
 	struct rq *rq;
@@ -3805,7 +3806,7 @@ void set_user_nice(struct task_struct *p, long nice)
 	queued = task_on_rq_queued(p);
 	running = task_current(rq, p);
 	if (queued)
-		dequeue_task(rq, p, DEQUEUE_SAVE);
+		dequeue_task(rq, p, DEQUEUE_SAVE | DEQUEUE_NOCLOCK);
 	if (running)
 		put_prev_task(rq, p);
 
@@ -3816,7 +3817,7 @@ void set_user_nice(struct task_struct *p, long nice)
 	delta = p->prio - old_prio;
 
 	if (queued) {
-		enqueue_task(rq, p, ENQUEUE_RESTORE);
+		enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK);
 		/*
 		 * If the task increased its priority or is running and
 		 * lowered its priority, then reschedule its CPU:
@@ -4126,7 +4127,7 @@ static int __sched_setscheduler(struct task_struct *p,
 	const struct sched_class *prev_class;
 	struct rq_flags rf;
 	int reset_on_fork;
-	int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE;
+	int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK;
 	struct rq *rq;
 
 	/* May grab non-irq protected spin_locks: */
@@ -4923,7 +4924,12 @@ SYSCALL_DEFINE3(sched_getaffinity, pid_t, pid, unsigned int, len,
  */
 SYSCALL_DEFINE0(sched_yield)
 {
-	struct rq *rq = this_rq_lock();
+	struct rq_flags rf;
+	struct rq *rq;
+
+	local_irq_disable();
+	rq = this_rq();
+	rq_lock(rq, &rf);
 
 	schedstat_inc(rq->yld_count);
 	current->sched_class->yield_task(rq);
@@ -4932,9 +4938,8 @@ SYSCALL_DEFINE0(sched_yield)
 	 * Since we are going to call schedule() anyway, there's
 	 * no need to preempt or enable interrupts:
 	 */
-	__release(rq->lock);
-	spin_release(&rq->lock.dep_map, 1, _THIS_IP_);
-	do_raw_spin_unlock(&rq->lock);
+	preempt_disable();
+	rq_unlock(rq, &rf);
 	sched_preempt_enable_no_resched();
 
 	schedule();
@@ -5514,7 +5519,7 @@ void sched_setnuma(struct task_struct *p, int nid)
 	p->numa_preferred_nid = nid;
 
 	if (queued)
-		enqueue_task(rq, p, ENQUEUE_RESTORE);
+		enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK);
 	if (running)
 		set_curr_task(rq, p);
 	task_rq_unlock(rq, p, &rf);
@@ -5579,11 +5584,11 @@ static struct task_struct fake_task = {
  * there's no concurrency possible, we hold the required locks anyway
  * because of lock validation efforts.
  */
-static void migrate_tasks(struct rq *dead_rq)
+static void migrate_tasks(struct rq *dead_rq, struct rq_flags *rf)
 {
 	struct rq *rq = dead_rq;
 	struct task_struct *next, *stop = rq->stop;
-	struct rq_flags rf;
+	struct rq_flags orf = *rf;
 	int dest_cpu;
 
 	/*
@@ -5602,9 +5607,7 @@ static void migrate_tasks(struct rq *dead_rq)
 	 * class method both need to have an up-to-date
 	 * value of rq->clock[_task]
 	 */
-	rq_pin_lock(rq, &rf);
 	update_rq_clock(rq);
-	rq_unpin_lock(rq, &rf);
 
 	for (;;) {
 		/*
@@ -5617,8 +5620,7 @@ static void migrate_tasks(struct rq *dead_rq)
 		/*
 		 * pick_next_task() assumes pinned rq->lock:
 		 */
-		rq_repin_lock(rq, &rf);
-		next = pick_next_task(rq, &fake_task, &rf);
+		next = pick_next_task(rq, &fake_task, rf);
 		BUG_ON(!next);
 		next->sched_class->put_prev_task(rq, next);
 
@@ -5631,10 +5633,9 @@ static void migrate_tasks(struct rq *dead_rq)
 		 * because !cpu_active at this point, which means load-balance
 		 * will not interfere. Also, stop-machine.
 		 */
-		rq_unpin_lock(rq, &rf);
-		raw_spin_unlock(&rq->lock);
+		rq_unlock(rq, rf);
 		raw_spin_lock(&next->pi_lock);
-		raw_spin_lock(&rq->lock);
+		rq_relock(rq, rf);
 
 		/*
 		 * Since we're inside stop-machine, _nothing_ should have
@@ -5648,12 +5649,12 @@ static void migrate_tasks(struct rq *dead_rq)
 
 		/* Find suitable destination for @next, with force if needed. */
 		dest_cpu = select_fallback_rq(dead_rq->cpu, next);
-
-		rq = __migrate_task(rq, next, dest_cpu);
+		rq = __migrate_task(rq, rf, next, dest_cpu);
 		if (rq != dead_rq) {
-			raw_spin_unlock(&rq->lock);
+			rq_unlock(rq, rf);
 			rq = dead_rq;
-			raw_spin_lock(&rq->lock);
+			*rf = orf;
+			rq_relock(rq, rf);
 		}
 		raw_spin_unlock(&next->pi_lock);
 	}
@@ -5766,7 +5767,7 @@ static int cpuset_cpu_inactive(unsigned int cpu)
 int sched_cpu_activate(unsigned int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
-	unsigned long flags;
+	struct rq_flags rf;
 
 	set_cpu_active(cpu, true);
 
@@ -5784,12 +5785,12 @@ int sched_cpu_activate(unsigned int cpu)
 	 * 2) At runtime, if cpuset_cpu_active() fails to rebuild the
 	 *    domains.
 	 */
-	raw_spin_lock_irqsave(&rq->lock, flags);
+	rq_lock_irqsave(rq, &rf);
 	if (rq->rd) {
 		BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
 		set_rq_online(rq);
 	}
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	rq_unlock_irqrestore(rq, &rf);
 
 	update_max_interval();
 
@@ -5847,18 +5848,20 @@ int sched_cpu_starting(unsigned int cpu)
 int sched_cpu_dying(unsigned int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
-	unsigned long flags;
+	struct rq_flags rf;
 
 	/* Handle pending wakeups and then migrate everything off */
 	sched_ttwu_pending();
-	raw_spin_lock_irqsave(&rq->lock, flags);
+
+	rq_lock_irqsave(rq, &rf);
 	if (rq->rd) {
 		BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
 		set_rq_offline(rq);
 	}
-	migrate_tasks(rq);
+	migrate_tasks(rq, &rf);
 	BUG_ON(rq->nr_running != 1);
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	rq_unlock_irqrestore(rq, &rf);
+
 	calc_load_migrate(rq);
 	update_max_interval();
 	nohz_balance_exit_idle(cpu);
@@ -6412,7 +6415,8 @@ static void sched_change_group(struct task_struct *tsk, int type)
  */
 void sched_move_task(struct task_struct *tsk)
 {
-	int queued, running;
+	int queued, running, queue_flags =
+		DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK;
 	struct rq_flags rf;
 	struct rq *rq;
 
@@ -6423,14 +6427,14 @@ void sched_move_task(struct task_struct *tsk)
 	queued = task_on_rq_queued(tsk);
 
 	if (queued)
-		dequeue_task(rq, tsk, DEQUEUE_SAVE | DEQUEUE_MOVE);