Merge branch '3.15-fixes' into mips-for-linux-next

15 files changed, 337 insertions, 231 deletions

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 11a03d67635a..3f1ca934a237 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -348,7 +348,7 @@ struct cgrp_cset_link {
  * reference-counted, to improve performance when child cgroups
  * haven't been created.
  */
-static struct css_set init_css_set = {
+struct css_set init_css_set = {
 	.refcount		= ATOMIC_INIT(1),
 	.cgrp_links		= LIST_HEAD_INIT(init_css_set.cgrp_links),
 	.tasks			= LIST_HEAD_INIT(init_css_set.tasks),
diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c
index 2bc4a2256444..345628c78b5b 100644
--- a/kernel/cgroup_freezer.c
+++ b/kernel/cgroup_freezer.c
@@ -21,6 +21,7 @@
 #include <linux/uaccess.h>
 #include <linux/freezer.h>
 #include <linux/seq_file.h>
+#include <linux/mutex.h>
 
 /*
  * A cgroup is freezing if any FREEZING flags are set.  FREEZING_SELF is
@@ -42,9 +43,10 @@ enum freezer_state_flags {
 struct freezer {
 	struct cgroup_subsys_state	css;
 	unsigned int			state;
-	spinlock_t			lock;
 };
 
+static DEFINE_MUTEX(freezer_mutex);
+
 static inline struct freezer *css_freezer(struct cgroup_subsys_state *css)
 {
 	return css ? container_of(css, struct freezer, css) : NULL;
@@ -93,7 +95,6 @@ freezer_css_alloc(struct cgroup_subsys_state *parent_css)
 	if (!freezer)
 		return ERR_PTR(-ENOMEM);
 
-	spin_lock_init(&freezer->lock);
 	return &freezer->css;
 }
 
@@ -110,14 +111,7 @@ static int freezer_css_online(struct cgroup_subsys_state *css)
 	struct freezer *freezer = css_freezer(css);
 	struct freezer *parent = parent_freezer(freezer);
 
-	/*
-	 * The following double locking and freezing state inheritance
-	 * guarantee that @cgroup can never escape ancestors' freezing
-	 * states.  See css_for_each_descendant_pre() for details.
-	 */
-	if (parent)
-		spin_lock_irq(&parent->lock);
-	spin_lock_nested(&freezer->lock, SINGLE_DEPTH_NESTING);
+	mutex_lock(&freezer_mutex);
 
 	freezer->state |= CGROUP_FREEZER_ONLINE;
 
@@ -126,10 +120,7 @@ static int freezer_css_online(struct cgroup_subsys_state *css)
 		atomic_inc(&system_freezing_cnt);
 	}
 
-	spin_unlock(&freezer->lock);
-	if (parent)
-		spin_unlock_irq(&parent->lock);
-
+	mutex_unlock(&freezer_mutex);
 	return 0;
 }
 
@@ -144,14 +135,14 @@ static void freezer_css_offline(struct cgroup_subsys_state *css)
 {
 	struct freezer *freezer = css_freezer(css);
 
-	spin_lock_irq(&freezer->lock);
+	mutex_lock(&freezer_mutex);
 
 	if (freezer->state & CGROUP_FREEZING)
 		atomic_dec(&system_freezing_cnt);
 
 	freezer->state = 0;
 
-	spin_unlock_irq(&freezer->lock);
+	mutex_unlock(&freezer_mutex);
 }
 
 static void freezer_css_free(struct cgroup_subsys_state *css)
@@ -175,7 +166,7 @@ static void freezer_attach(struct cgroup_subsys_state *new_css,
 	struct task_struct *task;
 	bool clear_frozen = false;
 
-	spin_lock_irq(&freezer->lock);
+	mutex_lock(&freezer_mutex);
 
 	/*
 	 * Make the new tasks conform to the current state of @new_css.
@@ -197,21 +188,13 @@ static void freezer_attach(struct cgroup_subsys_state *new_css,
 		}
 	}
 
-	spin_unlock_irq(&freezer->lock);
-
-	/*
-	 * Propagate FROZEN clearing upwards.  We may race with
-	 * update_if_frozen(), but as long as both work bottom-up, either
-	 * update_if_frozen() sees child's FROZEN cleared or we clear the
-	 * parent's FROZEN later.  No parent w/ !FROZEN children can be
-	 * left FROZEN.
-	 */
+	/* propagate FROZEN clearing upwards */
 	while (clear_frozen && (freezer = parent_freezer(freezer))) {
-		spin_lock_irq(&freezer->lock);
 		freezer->state &= ~CGROUP_FROZEN;
 		clear_frozen = freezer->state & CGROUP_FREEZING;
-		spin_unlock_irq(&freezer->lock);
 	}
+
+	mutex_unlock(&freezer_mutex);
 }
 
 /**
@@ -228,9 +211,6 @@ static void freezer_fork(struct task_struct *task)
 {
 	struct freezer *freezer;
 
-	rcu_read_lock();
-	freezer = task_freezer(task);
-
 	/*
 	 * The root cgroup is non-freezable, so we can skip locking the
 	 * freezer.  This is safe regardless of race with task migration.
@@ -238,24 +218,18 @@ static void freezer_fork(struct task_struct *task)
 	 * to do.  If we lost and root is the new cgroup, noop is still the
 	 * right thing to do.
 	 */
-	if (!parent_freezer(freezer))
-		goto out;
+	if (task_css_is_root(task, freezer_cgrp_id))
+		return;
 
-	/*
-	 * Grab @freezer->lock and freeze @task after verifying @task still
-	 * belongs to @freezer and it's freezing.  The former is for the
-	 * case where we have raced against task migration and lost and
-	 * @task is already in a different cgroup which may not be frozen.
-	 * This isn't strictly necessary as freeze_task() is allowed to be
-	 * called spuriously but let's do it anyway for, if nothing else,
-	 * documentation.
-	 */
-	spin_lock_irq(&freezer->lock);
-	if (freezer == task_freezer(task) && (freezer->state & CGROUP_FREEZING))
+	mutex_lock(&freezer_mutex);
+	rcu_read_lock();
+
+	freezer = task_freezer(task);
+	if (freezer->state & CGROUP_FREEZING)
 		freeze_task(task);
-	spin_unlock_irq(&freezer->lock);
-out:
+
 	rcu_read_unlock();
+	mutex_unlock(&freezer_mutex);
 }
 
 /**
@@ -281,22 +255,24 @@ static void update_if_frozen(struct cgroup_subsys_state *css)
 	struct css_task_iter it;
 	struct task_struct *task;
 
-	WARN_ON_ONCE(!rcu_read_lock_held());
-
-	spin_lock_irq(&freezer->lock);
+	lockdep_assert_held(&freezer_mutex);
 
 	if (!(freezer->state & CGROUP_FREEZING) ||
 	    (freezer->state & CGROUP_FROZEN))
-		goto out_unlock;
+		return;
 
 	/* are all (live) children frozen? */
+	rcu_read_lock();
 	css_for_each_child(pos, css) {
 		struct freezer *child = css_freezer(pos);
 
 		if ((child->state & CGROUP_FREEZER_ONLINE) &&
-		    !(child->state & CGROUP_FROZEN))
-			goto out_unlock;
+		    !(child->state & CGROUP_FROZEN)) {
+			rcu_read_unlock();
+			return;
+		}
 	}
+	rcu_read_unlock();
 
 	/* are all tasks frozen? */
 	css_task_iter_start(css, &it);
@@ -317,21 +293,29 @@ static void update_if_frozen(struct cgroup_subsys_state *css)
 	freezer->state |= CGROUP_FROZEN;
 out_iter_end:
 	css_task_iter_end(&it);
-out_unlock:
-	spin_unlock_irq(&freezer->lock);
 }
 
 static int freezer_read(struct seq_file *m, void *v)
 {
 	struct cgroup_subsys_state *css = seq_css(m), *pos;
 
+	mutex_lock(&freezer_mutex);
 	rcu_read_lock();
 
 	/* update states bottom-up */
-	css_for_each_descendant_post(pos, css)
+	css_for_each_descendant_post(pos, css) {
+		if (!css_tryget(pos))
+			continue;
+		rcu_read_unlock();
+
 		update_if_frozen(pos);
 
+		rcu_read_lock();
+		css_put(pos);
+	}
+
 	rcu_read_unlock();
+	mutex_unlock(&freezer_mutex);
 
 	seq_puts(m, freezer_state_strs(css_freezer(css)->state));
 	seq_putc(m, '\n');
@@ -373,7 +357,7 @@ static void freezer_apply_state(struct freezer *freezer, bool freeze,
 				unsigned int state)
 {
 	/* also synchronizes against task migration, see freezer_attach() */
-	lockdep_assert_held(&freezer->lock);
+	lockdep_assert_held(&freezer_mutex);
 
 	if (!(freezer->state & CGROUP_FREEZER_ONLINE))
 		return;
@@ -414,31 +398,29 @@ static void freezer_change_state(struct freezer *freezer, bool freeze)
 	 * descendant will try to inherit its parent's FREEZING state as
 	 * CGROUP_FREEZING_PARENT.
 	 */
+	mutex_lock(&freezer_mutex);
 	rcu_read_lock();
 	css_for_each_descendant_pre(pos, &freezer->css) {
 		struct freezer *pos_f = css_freezer(pos);
 		struct freezer *parent = parent_freezer(pos_f);
 
-		spin_lock_irq(&pos_f->lock);
+		if (!css_tryget(pos))
+			continue;
+		rcu_read_unlock();
 
-		if (pos_f == freezer) {
+		if (pos_f == freezer)
 			freezer_apply_state(pos_f, freeze,
 					    CGROUP_FREEZING_SELF);
-		} else {
-			/*
-			 * Our update to @parent->state is already visible
-			 * which is all we need.  No need to lock @parent.
-			 * For more info on synchronization, see
-			 * freezer_post_create().
-			 */
+		else
 			freezer_apply_state(pos_f,
 					    parent->state & CGROUP_FREEZING,
 					    CGROUP_FREEZING_PARENT);
-		}
 
-		spin_unlock_irq(&pos_f->lock);
+		rcu_read_lock();
+		css_put(pos);
 	}
 	rcu_read_unlock();
+	mutex_unlock(&freezer_mutex);
 }
 
 static int freezer_write(struct cgroup_subsys_state *css, struct cftype *cft,
diff --git a/kernel/cpu.c b/kernel/cpu.c
index a9e710eef0e2..247979a1b815 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -726,10 +726,12 @@ void set_cpu_present(unsigned int cpu, bool present)
 
 void set_cpu_online(unsigned int cpu, bool online)
 {
-	if (online)
+	if (online) {
 		cpumask_set_cpu(cpu, to_cpumask(cpu_online_bits));
-	else
+		cpumask_set_cpu(cpu, to_cpumask(cpu_active_bits));
+	} else {
 		cpumask_clear_cpu(cpu, to_cpumask(cpu_online_bits));
+	}
 }
 
 void set_cpu_active(unsigned int cpu, bool active)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index f83a71a3e46d..440eefc67397 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1443,6 +1443,11 @@ group_sched_out(struct perf_event *group_event,
 		cpuctx->exclusive = 0;
 }
 
+struct remove_event {
+	struct perf_event *event;
+	bool detach_group;
+};
+
 /*
  * Cross CPU call to remove a performance event
  *
@@ -1451,12 +1456,15 @@ group_sched_out(struct perf_event *group_event,
  */
 static int __perf_remove_from_context(void *info)
 {
-	struct perf_event *event = info;
+	struct remove_event *re = info;
+	struct perf_event *event = re->event;
 	struct perf_event_context *ctx = event->ctx;
 	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	raw_spin_lock(&ctx->lock);
 	event_sched_out(event, cpuctx, ctx);
+	if (re->detach_group)
+		perf_group_detach(event);
 	list_del_event(event, ctx);
 	if (!ctx->nr_events && cpuctx->task_ctx == ctx) {
 		ctx->is_active = 0;
@@ -1481,10 +1489,14 @@ static int __perf_remove_from_context(void *info)
  * When called from perf_event_exit_task, it's OK because the
  * context has been detached from its task.
  */
-static void perf_remove_from_context(struct perf_event *event)
+static void perf_remove_from_context(struct perf_event *event, bool detach_group)
 {
 	struct perf_event_context *ctx = event->ctx;
 	struct task_struct *task = ctx->task;
+	struct remove_event re = {
+		.event = event,
+		.detach_group = detach_group,
+	};
 
 	lockdep_assert_held(&ctx->mutex);
 
@@ -1493,12 +1505,12 @@ static void perf_remove_from_context(struct perf_event *event)
 		 * Per cpu events are removed via an smp call and
 		 * the removal is always successful.
 		 */
-		cpu_function_call(event->cpu, __perf_remove_from_context, event);
+		cpu_function_call(event->cpu, __perf_remove_from_context, &re);
 		return;
 	}
 
 retry:
-	if (!task_function_call(task, __perf_remove_from_context, event))
+	if (!task_function_call(task, __perf_remove_from_context, &re))
 		return;
 
 	raw_spin_lock_irq(&ctx->lock);
@@ -1515,6 +1527,8 @@ retry:
 	 * Since the task isn't running, its safe to remove the event, us
 	 * holding the ctx->lock ensures the task won't get scheduled in.
 	 */
+	if (detach_group)
+		perf_group_detach(event);
 	list_del_event(event, ctx);
 	raw_spin_unlock_irq(&ctx->lock);
 }
@@ -3178,7 +3192,8 @@ static void free_event_rcu(struct rcu_head *head)
 }
 
 static void ring_buffer_put(struct ring_buffer *rb);
-static void ring_buffer_detach(struct perf_event *event, struct ring_buffer *rb);
+static void ring_buffer_attach(struct perf_event *event,
+			       struct ring_buffer *rb);
 
 static void unaccount_event_cpu(struct perf_event *event, int cpu)
 {
@@ -3238,8 +3253,6 @@ static void free_event(struct perf_event *event)
 	unaccount_event(event);
 
 	if (event->rb) {
-		struct ring_buffer *rb;
-
 		/*
 		 * Can happen when we close an event with re-directed output.
 		 *
@@ -3247,12 +3260,7 @@ static void free_event(struct perf_event *event)
 		 * over us; possibly making our ring_buffer_put() the last.
 		 */
 		mutex_lock(&event->mmap_mutex);
-		rb = event->rb;
-		if (rb) {
-			rcu_assign_pointer(event->rb, NULL);
-			ring_buffer_detach(event, rb);
-			ring_buffer_put(rb); /* could be last */
-		}
+		ring_buffer_attach(event, NULL);
 		mutex_unlock(&event->mmap_mutex);
 	}
 
@@ -3281,10 +3289,7 @@ int perf_event_release_kernel(struct perf_event *event)
 	 *     to trigger the AB-BA case.
 	 */
 	mutex_lock_nested(&ctx->mutex, SINGLE_DEPTH_NESTING);
-	raw_spin_lock_irq(&ctx->lock);
-	perf_group_detach(event);
-	raw_spin_unlock_irq(&ctx->lock);
-	perf_remove_from_context(event);
+	perf_remove_from_context(event, true);
 	mutex_unlock(&ctx->mutex);
 
 	free_event(event);
@@ -3839,28 +3844,47 @@ unlock:
 static void ring_buffer_attach(struct perf_event *event,
 			       struct ring_buffer *rb)
 {
+	struct ring_buffer *old_rb = NULL;
 	unsigned long flags;
 
-	if (!list_empty(&event->rb_entry))
-		return;
+	if (event->rb) {
+		/*
+		 * Should be impossible, we set this when removing
+		 * event->rb_entry and wait/clear when adding event->rb_entry.
+		 */
+		WARN_ON_ONCE(event->rcu_pending);
 
-	spin_lock_irqsave(&rb->event_lock, flags);
-	if (list_empty(&event->rb_entry))
-		list_add(&event->rb_entry, &rb->event_list);
-	spin_unlock_irqrestore(&rb->event_lock, flags);
-}
+		old_rb = event->rb;
+		event->rcu_batches = get_state_synchronize_rcu();
+		event->rcu_pending = 1;
 
-static void ring_buffer_detach(struct perf_event *event, struct ring_buffer *rb)
-{
-	unsigned long flags;
+		spin_lock_irqsave(&old_rb->event_lock, flags);
+		list_del_rcu(&event->rb_entry);
+		spin_unlock_irqrestore(&old_rb->event_lock, flags);
+	}
 
-	if (list_empty(&event->rb_entry))
-		return;
+	if (event->rcu_pending && rb) {
+		cond_synchronize_rcu(event->rcu_batches);
+		event->rcu_pending = 0;
+	}
+
+	if (rb) {
+		spin_lock_irqsave(&rb->event_lock, flags);
+		list_add_rcu(&event->rb_entry, &rb->event_list);
+		spin_unlock_irqrestore(&rb->event_lock, flags);
+	}
+
+	rcu_assign_pointer(event->rb, rb);
 
-	spin_lock_irqsave(&rb->event_lock, flags);
-	list_del_init(&event->rb_entry);
-	wake_up_all(&event->waitq);
-	spin_unlock_irqrestore(&rb->event_lock, flags);
+	if (old_rb) {
+		ring_buffer_put(old_rb);
+		/*
+		 * Since we detached before setting the new rb, so that we
+		 * could attach the new rb, we could have missed a wakeup.
+		 * Provide it now.
+		 */
+		wake_up_all(&event->waitq);
+	}
 }
 
 static void ring_buffer_wakeup(struct perf_event *event)
@@ -3929,7 +3953,7 @@ static void perf_mmap_close(struct vm_area_struct *vma)
 {
 	struct perf_event *event = vma->vm_file->private_data;
 
-	struct ring_buffer *rb = event->rb;
+	struct ring_buffer *rb = ring_buffer_get(event);
 	struct user_struct *mmap_user = rb->mmap_user;
 	int mmap_locked = rb->mmap_locked;
 	unsigned long size = perf_data_size(rb);
@@ -3937,18 +3961,14 @@ static void perf_mmap_close(struct vm_area_struct *vma)
 	atomic_dec(&rb->mmap_count);
 
 	if (!atomic_dec_and_mutex_lock(&event->mmap_count, &event->mmap_mutex))
-		return;
+		goto out_put;
 
-	/* Detach current event from the buffer. */
-	rcu_assign_pointer(event->rb, NULL);
-	ring_buffer_detach(event, rb);
+	ring_buffer_attach(event, NULL);
 	mutex_unlock(&event->mmap_mutex);
 
 	/* If there's still other mmap()s of this buffer, we're done. */
-	if (atomic_read(&rb->mmap_count)) {
-		ring_buffer_put(rb); /* can't be last */
-		return;
-	}
+	if (atomic_read(&rb->mmap_count))
+		goto out_put;
 
 	/*
 	 * No other mmap()s, detach from all other events that might redirect
@@ -3978,11 +3998,9 @@ again:
 		 * still restart the iteration to make sure we're not now
 		 * iterating the wrong list.
 		 */
-		if (event->rb == rb) {
-			rcu_assign_pointer(event->rb, NULL);
-			ring_buffer_detach(event, rb);
-			ring_buffer_put(rb); /* can't be last, we still have one */
-		}
+		if (event->rb == rb)
+			ring_buffer_attach(event, NULL);
+
 		mutex_unlock(&event->mmap_mutex);
 		put_event(event);
 
@@ -4007,6 +4025,7 @@ again:
 	vma->vm_mm->pinned_vm -= mmap_locked;
 	free_uid(mmap_user);
 
+out_put:
 	ring_buffer_put(rb); /* could be last */
 }
 
@@ -4124,7 +4143,6 @@ again:
 	vma->vm_mm->pinned_vm += extra;
 
 	ring_buffer_attach(event, rb);
-	rcu_assign_pointer(event->rb, rb);
 
 	perf_event_init_userpage(event);
 	perf_event_update_userpage(event);
@@ -5408,6 +5426,9 @@ struct swevent_htable {
 
 	/* Recursion avoidance in each contexts */
 	int				recursion[PERF_NR_CONTEXTS];
+
+	/* Keeps track of cpu being initialized/exited */
+	bool				online;
 };
 
 static DEFINE_PER_CPU(struct swevent_htable, swevent_htable);
@@ -5654,8 +5675,14 @@ static int perf_swevent_add(struct perf_event *event, int flags)
 	hwc->state = !(flags & PERF_EF_START);
 
 	head = find_swevent_head(swhash, event);
-	if (WARN_ON_ONCE(!head))
+	if (!head) {
+		/*
+		 * We can race with cpu hotplug code. Do not
+		 * WARN if the cpu just got unplugged.
+		 */
+		WARN_ON_ONCE(swhash->online);
 		return -EINVAL;
+	}
 
 	hlist_add_head_rcu(&event->hlist_entry, head);
 
@@ -6914,7 +6941,7 @@ err_size:
 static int
 perf_event_set_output(struct perf_event *event, struct perf_event *output_event)
 {
-	struct ring_buffer *rb = NULL, *old_rb = NULL;
+	struct ring_buffer *rb = NULL;
 	int ret = -EINVAL;
 
 	if (!output_event)
@@ -6942,8 +6969,6 @@ set:
 	if (atomic_read(&event->mmap_count))
 		goto unlock;
 
-	old_rb = event->rb;
-
 	if (output_event) {
 		/* get the rb we want to redirect to */
 		rb = ring_buffer_get(output_event);
@@ -6951,23 +6976,7 @@ set:
 			goto unlock;
 	}
 
-	if (old_rb)
-		ring_buffer_detach(event, old_rb);
-
-	if (rb)
-		ring_buffer_attach(event, rb);
-
-	rcu_assign_pointer(event->rb, rb);
-
-	if (old_rb) {
-		ring_buffer_put(old_rb);
-		/*
-		 * Since we detached before setting the new rb, so that we
-		 * could attach the new rb, we could have missed a wakeup.
-		 * Provide it now.
-		 */
-		wake_up_all(&event->waitq);
-	}
+	ring_buffer_attach(event, rb);
 
 	ret = 0;
 unlock:
@@ -7018,6 +7027,9 @@ SYSCALL_DEFINE5(perf_event_open,
 	if (attr.freq) {
 		if (attr.sample_freq > sysctl_perf_event_sample_rate)
 			return -EINVAL;
+	} else {
+		if (attr.sample_period & (1ULL << 63))
+			return -EINVAL;
 	}
 
 	/*
@@ -7165,7 +7177,7 @@ SYSCALL_DEFINE5(perf_event_open,
 		struct perf_event_context *gctx = group_leader->ctx;
 
 		mutex_lock(&gctx->mutex);
-		perf_remove_from_context(group_leader);
+		perf_remove_from_context(group_leader, false);
 
 		/*
 		 * Removing from the context ends up with disabled
@@ -7175,7 +7187,7 @@ SYSCALL_DEFINE5(perf_event_open,
 		perf_event__state_init(group_leader);
 		list_for_each_entry(sibling, &group_leader->sibling_list,
 				    group_entry) {
-			perf_remove_from_context(sibling);
+			perf_remove_from_context(sibling, false);
 			perf_event__state_init(sibling);
 			put_ctx(gctx);
 		}
@@ -7305,7 +7317,7 @@ void perf_pmu_migrate_context(struct pmu *pmu, int src_cpu, int dst_cpu)
 	mutex_lock(&src_ctx->mutex);
 	list_for_each_entry_safe(event, tmp, &src_ctx->event_list,
 				 event_entry) {
-		perf_remove_from_context(event);
+		perf_remove_from_context(event, false);
 		unaccount_event_cpu(event, src_cpu);
 		put_ctx(src_ctx);
 		list_add(&event->migrate_entry, &events);
@@ -7367,13 +7379,7 @@ __perf_event_exit_task(struct perf_event *child_event,
 			 struct perf_event_context *child_ctx,
 			 struct task_struct *child)
 {
-	if (child_event->parent) {
-		raw_spin_lock_irq(&child_ctx->lock);
-		perf_group_detach(child_event);
-		raw_spin_unlock_irq(&child_ctx->lock);
-	}
-
-	perf_remove_from_context(child_event);
+	perf_remove_from_context(child_event, !!child_event->parent);
 
 	/*
 	 * It can happen that the parent exits first, and has events
@@ -7724,6 +7730,8 @@ int perf_event_init_context(struct task_struct *child, int ctxn)
 	 * swapped under us.
 	 */
 	parent_ctx = perf_pin_task_context(parent, ctxn);
+	if (!parent_ctx)
+		return 0;
 
 	/*
 	 * No need to check if parent_ctx != NULL here; since we saw
@@ -7835,6 +7843,7 @@ static void perf_event_init_cpu(int cpu)
 	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
 	mutex_lock(&swhash->hlist_mutex);
+	swhash->online = true;
 	if (swhash->hlist_refcount > 0) {
 		struct swevent_hlist *hlist;
 
@@ -7857,14 +7866,14 @@ static void perf_pmu_rotate_stop(struct pmu *pmu)
 
 static void __perf_event_exit_context(void *__info)
 {
+	struct remove_event re = { .detach_group = false };
 	struct perf_event_context *ctx = __info;
-	struct perf_event *event;
 
 	perf_pmu_rotate_stop(ctx->pmu);
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(event, &ctx->event_list, event_entry)
-		__perf_remove_from_context(event);
+	list_for_each_entry_rcu(re.event, &ctx->event_list, event_entry)
+		__perf_remove_from_context(&re);
 	rcu_read_unlock();
 }
 
@@ -7892,6 +7901,7 @@ static void perf_event_exit_cpu(int cpu)
 	perf_event_exit_cpu_context(cpu);
 
 	mutex_lock(&swhash->hlist_mutex);
+	swhash->online = false;
 	swevent_hlist_release(swhash);
 	mutex_unlock(&swhash->hlist_mutex);
 }
diff --git a/kernel/futex.c b/kernel/futex.c
index 5f589279e462..81dbe773ce4c 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -745,7 +745,8 @@ void exit_pi_state_list(struct task_struct *curr)
 
 static int
 lookup_pi_state(u32 uval, struct futex_hash_bucket *hb,
-		union futex_key *key, struct futex_pi_state **ps)
+		union futex_key *key, struct futex_pi_state **ps,
+		struct task_struct *task)
 {
 	struct futex_pi_state *pi_state = NULL;
 	struct futex_q *this, *next;
@@ -786,6 +787,16 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb,
 					return -EINVAL;
 			}
 
+			/*
+			 * Protect against a corrupted uval. If uval
+			 * is 0x80000000 then pid is 0 and the waiter
+			 * bit is set. So the deadlock check in the
+			 * calling code has failed and we did not fall
+			 * into the check above due to !pid.
+			 */
+			if (task && pi_state->owner == task)
+				return -EDEADLK;
+
 			atomic_inc(&pi_state->refcount);
 			*ps = pi_state;
 
@@ -803,6 +814,11 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb,
 	if (!p)
 		return -ESRCH;
 
+	if (!p->mm) {
+		put_task_struct(p);
+		return -EPERM;
+	}
+
 	/*
 	 * We need to look at the task state flags to figure out,
 	 * whether the task is exiting. To protect against the do_exit
@@ -935,7 +951,7 @@ retry:
 	 * We dont have the lock. Look up the PI state (or create it if
 	 * we are the first waiter):
 	 */
-	ret = lookup_pi_state(uval, hb, key, ps);
+	ret = lookup_pi_state(uval, hb, key, ps, task);
 
 	if (unlikely(ret)) {
 		switch (ret) {
@@ -1347,7 +1363,7 @@ void requeue_pi_wake_futex(struct futex_q *q, union futex_key *key,
  *
  * Return:
  *  0 - failed to acquire the lock atomically;
- *  1 - acquired the lock;
+ * >0 - acquired the lock, return value is vpid of the top_waiter
  * <0 - error
  */
 static int futex_proxy_trylock_atomic(u32 __user *pifutex,
@@ -1358,7 +1374,7 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex,
 {
 	struct futex_q *top_waiter = NULL;
 	u32 curval;
-	int ret;
+	int ret, vpid;
 
 	if (get_futex_value_locked(&curval, pifutex))
 		return -EFAULT;
@@ -1386,11 +1402,13 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex,
 	 * the contended case or if set_waiters is 1.  The pi_state is returned
 	 * in ps in contended cases.
 	 */
+	vpid = task_pid_vnr(top_waiter->task);
 	ret = futex_lock_pi_atomic(pifutex, hb2, key2, ps, top_waiter->task,
 				   set_waiters);
-	if (ret == 1)
+	if (ret == 1) {
 		requeue_pi_wake_futex(top_waiter, key2, hb2);
-
+		return vpid;
+	}
 	return ret;
 }
 
@@ -1421,7 +1439,6 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags,
 	struct futex_pi_state *pi_state = NULL;
 	struct futex_hash_bucket *hb1, *hb2;
 	struct futex_q *this, *next;
-	u32 curval2;
 
 	if (requeue_pi) {
 		/*
@@ -1509,16 +1526,25 @@ retry_private:
 		 * At this point the top_waiter has either taken uaddr2 or is
 		 * waiting on it.  If the former, then the pi_state will not
 		 * exist yet, look it up one more time to ensure we have a
-		 * reference to it.
+		 * reference to it. If the lock was taken, ret contains the
+		 * vpid of the top waiter task.
 		 */
-		if (ret == 1) {
+		if (ret > 0) {
 			WARN_ON(pi_state);
 			drop_count++;
 			task_count++;
-			ret = get_futex_value_locked(&curval2, uaddr2);
-			if (!ret)
-				ret = lookup_pi_state(curval2, hb2, &key2,
-						      &pi_state);
+			/*
+			 * If we acquired the lock, then the user
+			 * space value of uaddr2 should be vpid. It
+			 * cannot be changed by the top waiter as it
+			 * is blocked on hb2 lock if it tries to do
+			 * so. If something fiddled with it behind our
+			 * back the pi state lookup might unearth
+			 * it. So we rather use the known value than
+			 * rereading and handing potential crap to
+			 * lookup_pi_state.
+			 */
+			ret = lookup_pi_state(ret, hb2, &key2, &pi_state, NULL);
 		}
 
 		switch (ret) {
diff --git a/kernel/kexec.c b/kernel/kexec.c
index c8380ad203bc..28c57069ef68 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -1683,6 +1683,14 @@ int kernel_kexec(void)
 		kexec_in_progress = true;
 		kernel_restart_prepare(NULL);
 		migrate_to_reboot_cpu();
+
+		/*
+		 * migrate_to_reboot_cpu() disables CPU hotplug assuming that
+		 * no further code needs to use CPU hotplug (which is true in
+		 * the reboot case). However, the kexec path depends on using
+		 * CPU hotplug again; so re-enable it here.
+		 */
+		cpu_hotplug_enable();
 		printk(KERN_EMERG "Starting new kernel\n");
 		machine_shutdown();
 	}
diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c
index aa4dff04b594..a620d4d08ca6 100644
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -343,9 +343,16 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task,
 	 * top_waiter can be NULL, when we are in the deboosting
 	 * mode!
 	 */
-	if (top_waiter && (!task_has_pi_waiters(task) ||
-			   top_waiter != task_top_pi_waiter(task)))
-		goto out_unlock_pi;
+	if (top_waiter) {
+		if (!task_has_pi_waiters(task))
+			goto out_unlock_pi;
+		/*
+		 * If deadlock detection is off, we stop here if we
+		 * are not the top pi waiter of the task.
+		 */
+		if (!detect_deadlock && top_waiter != task_top_pi_waiter(task))
+			goto out_unlock_pi;
+	}
 
 	/*
 	 * When deadlock detection is off then we check, if further
@@ -361,7 +368,12 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task,
 		goto retry;
 	}
 
-	/* Deadlock detection */
+	/*
+	 * Deadlock detection. If the lock is the same as the original
+	 * lock which caused us to walk the lock chain or if the
+	 * current lock is owned by the task which initiated the chain
+	 * walk, we detected a deadlock.
+	 */
 	if (lock == orig_lock || rt_mutex_owner(lock) == top_task) {
 		debug_rt_mutex_deadlock(deadlock_detect, orig_waiter, lock);
 		raw_spin_unlock(&lock->wait_lock);
@@ -527,6 +539,18 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock,
 	unsigned long flags;
 	int chain_walk = 0, res;
 
+	/*
+	 * Early deadlock detection. We really don't want the task to
+	 * enqueue on itself just to untangle the mess later. It's not
+	 * only an optimization. We drop the locks, so another waiter
+	 * can come in before the chain walk detects the deadlock. So
+	 * the other will detect the deadlock and return -EDEADLOCK,
+	 * which is wrong, as the other waiter is not in a deadlock
+	 * situation.
+	 */
+	if (detect_deadlock && owner == task)
+		return -EDEADLK;
+
 	raw_spin_lock_irqsave(&task->pi_lock, flags);
 	__rt_mutex_adjust_prio(task);
 	waiter->task = task;
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d9d8ece46a15..0a7251678982 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2592,8 +2592,14 @@ pick_next_task(struct rq *rq, struct task_struct *prev)
 	if (likely(prev->sched_class == class &&
 		   rq->nr_running == rq->cfs.h_nr_running)) {
 		p = fair_sched_class.pick_next_task(rq, prev);
-		if (likely(p && p != RETRY_TASK))
-			return p;
+		if (unlikely(p == RETRY_TASK))
+			goto again;
+
+		/* assumes fair_sched_class->next == idle_sched_class */
+		if (unlikely(!p))
+			p = idle_sched_class.pick_next_task(rq, prev);
+
+		return p;
 	}
 
 again:
@@ -3124,6 +3130,7 @@ __setparam_dl(struct task_struct *p, const struct sched_attr *attr)
 	dl_se->dl_bw = to_ratio(dl_se->dl_period, dl_se->dl_runtime);
 	dl_se->dl_throttled = 0;
 	dl_se->dl_new = 1;
+	dl_se->dl_yielded = 0;
 }
 
 static void __setscheduler_params(struct task_struct *p,
@@ -3188,17 +3195,40 @@ __getparam_dl(struct task_struct *p, struct sched_attr *attr)
  * We ask for the deadline not being zero, and greater or equal
  * than the