Merge tag 'please-pull-mce-bitmap-comment' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras into x86/ras

Pull MCE updates from Tony Luck:

 "Better comments so we understand our existing machine check
  bank bitmaps - prelude to adding another bitmap soon."

Signed-off-by: Ingo Molnar <mingo@kernel.org>

10 files changed, 275 insertions, 106 deletions

diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c
index 65349f07b878..383f8231e436 100644
--- a/kernel/context_tracking.c
+++ b/kernel/context_tracking.c
@@ -15,7 +15,6 @@
  */
 
 #include <linux/context_tracking.h>
-#include <linux/kvm_host.h>
 #include <linux/rcupdate.h>
 #include <linux/sched.h>
 #include <linux/hardirq.h>
@@ -71,6 +70,46 @@ void user_enter(void)
 	local_irq_restore(flags);
 }
 
+#ifdef CONFIG_PREEMPT
+/**
+ * preempt_schedule_context - preempt_schedule called by tracing
+ *
+ * The tracing infrastructure uses preempt_enable_notrace to prevent
+ * recursion and tracing preempt enabling caused by the tracing
+ * infrastructure itself. But as tracing can happen in areas coming
+ * from userspace or just about to enter userspace, a preempt enable
+ * can occur before user_exit() is called. This will cause the scheduler
+ * to be called when the system is still in usermode.
+ *
+ * To prevent this, the preempt_enable_notrace will use this function
+ * instead of preempt_schedule() to exit user context if needed before
+ * calling the scheduler.
+ */
+void __sched notrace preempt_schedule_context(void)
+{
+	struct thread_info *ti = current_thread_info();
+	enum ctx_state prev_ctx;
+
+	if (likely(ti->preempt_count || irqs_disabled()))
+		return;
+
+	/*
+	 * Need to disable preemption in case user_exit() is traced
+	 * and the tracer calls preempt_enable_notrace() causing
+	 * an infinite recursion.
+	 */
+	preempt_disable_notrace();
+	prev_ctx = exception_enter();
+	preempt_enable_no_resched_notrace();
+
+	preempt_schedule();
+
+	preempt_disable_notrace();
+	exception_exit(prev_ctx);
+	preempt_enable_notrace();
+}
+EXPORT_SYMBOL_GPL(preempt_schedule_context);
+#endif /* CONFIG_PREEMPT */
 
 /**
  * user_exit - Inform the context tracking that the CPU is
diff --git a/kernel/cpu/idle.c b/kernel/cpu/idle.c
index d5585f5e038e..e695c0a0bcb5 100644
--- a/kernel/cpu/idle.c
+++ b/kernel/cpu/idle.c
@@ -5,6 +5,7 @@
 #include <linux/cpu.h>
 #include <linux/tick.h>
 #include <linux/mm.h>
+#include <linux/stackprotector.h>
 
 #include <asm/tlb.h>
 
@@ -58,6 +59,7 @@ void __weak arch_cpu_idle_dead(void) { }
 void __weak arch_cpu_idle(void)
 {
 	cpu_idle_force_poll = 1;
+	local_irq_enable();
 }
 
 /*
@@ -112,6 +114,21 @@ static void cpu_idle_loop(void)
 
 void cpu_startup_entry(enum cpuhp_state state)
 {
+	/*
+	 * This #ifdef needs to die, but it's too late in the cycle to
+	 * make this generic (arm and sh have never invoked the canary
+	 * init for the non boot cpus!). Will be fixed in 3.11
+	 */
+#ifdef CONFIG_X86
+	/*
+	 * If we're the non-boot CPU, nothing set the stack canary up
+	 * for us. The boot CPU already has it initialized but no harm
+	 * in doing it again. This is a good place for updating it, as
+	 * we wont ever return from this function (so the invalid
+	 * canaries already on the stack wont ever trigger).
+	 */
+	boot_init_stack_canary();
+#endif
 	current_set_polling();
 	arch_cpu_idle_prepare();
 	cpu_idle_loop();
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 9dc297faf7c0..b391907d5352 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -196,9 +196,6 @@ static void cpu_ctx_sched_in(struct perf_cpu_context *cpuctx,
 static void update_context_time(struct perf_event_context *ctx);
 static u64 perf_event_time(struct perf_event *event);
 
-static void ring_buffer_attach(struct perf_event *event,
-			       struct ring_buffer *rb);
-
 void __weak perf_event_print_debug(void)	{ }
 
 extern __weak const char *perf_pmu_name(void)
@@ -2918,6 +2915,7 @@ 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 free_event(struct perf_event *event)
 {
@@ -2942,15 +2940,30 @@ static void free_event(struct perf_event *event)
 		if (has_branch_stack(event)) {
 			static_key_slow_dec_deferred(&perf_sched_events);
 			/* is system-wide event */
-			if (!(event->attach_state & PERF_ATTACH_TASK))
+			if (!(event->attach_state & PERF_ATTACH_TASK)) {
 				atomic_dec(&per_cpu(perf_branch_stack_events,
 						    event->cpu));
+			}
 		}
 	}
 
 	if (event->rb) {
-		ring_buffer_put(event->rb);
-		event->rb = NULL;
+		struct ring_buffer *rb;
+
+		/*
+		 * Can happen when we close an event with re-directed output.
+		 *
+		 * Since we have a 0 refcount, perf_mmap_close() will skip
+		 * 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 */
+		}
+		mutex_unlock(&event->mmap_mutex);
 	}
 
 	if (is_cgroup_event(event))
@@ -3188,30 +3201,13 @@ static unsigned int perf_poll(struct file *file, poll_table *wait)
 	unsigned int events = POLL_HUP;
 
 	/*
-	 * Race between perf_event_set_output() and perf_poll(): perf_poll()
-	 * grabs the rb reference but perf_event_set_output() overrides it.
-	 * Here is the timeline for two threads T1, T2:
-	 * t0: T1, rb = rcu_dereference(event->rb)
-	 * t1: T2, old_rb = event->rb
-	 * t2: T2, event->rb = new rb
-	 * t3: T2, ring_buffer_detach(old_rb)
-	 * t4: T1, ring_buffer_attach(rb1)
-	 * t5: T1, poll_wait(event->waitq)
-	 *
-	 * To avoid this problem, we grab mmap_mutex in perf_poll()
-	 * thereby ensuring that the assignment of the new ring buffer
-	 * and the detachment of the old buffer appear atomic to perf_poll()
+	 * Pin the event->rb by taking event->mmap_mutex; otherwise
+	 * perf_event_set_output() can swizzle our rb and make us miss wakeups.
 	 */
 	mutex_lock(&event->mmap_mutex);
-
-	rcu_read_lock();
-	rb = rcu_dereference(event->rb);
-	if (rb) {
-		ring_buffer_attach(event, rb);
+	rb = event->rb;
+	if (rb)
 		events = atomic_xchg(&rb->poll, 0);
-	}
-	rcu_read_unlock();
-
 	mutex_unlock(&event->mmap_mutex);
 
 	poll_wait(file, &event->waitq, wait);
@@ -3521,16 +3517,12 @@ static void ring_buffer_attach(struct perf_event *event,
 		return;
 
 	spin_lock_irqsave(&rb->event_lock, flags);
-	if (!list_empty(&event->rb_entry))
-		goto unlock;
-
-	list_add(&event->rb_entry, &rb->event_list);
-unlock:
+	if (list_empty(&event->rb_entry))
+		list_add(&event->rb_entry, &rb->event_list);
 	spin_unlock_irqrestore(&rb->event_lock, flags);
 }
 
-static void ring_buffer_detach(struct perf_event *event,
-			       struct ring_buffer *rb)
+static void ring_buffer_detach(struct perf_event *event, struct ring_buffer *rb)
 {
 	unsigned long flags;
 
@@ -3549,13 +3541,10 @@ static void ring_buffer_wakeup(struct perf_event *event)
 
 	rcu_read_lock();
 	rb = rcu_dereference(event->rb);
-	if (!rb)
-		goto unlock;
-
-	list_for_each_entry_rcu(event, &rb->event_list, rb_entry)
-		wake_up_all(&event->waitq);
-
-unlock:
+	if (rb) {
+		list_for_each_entry_rcu(event, &rb->event_list, rb_entry)
+			wake_up_all(&event->waitq);
+	}
 	rcu_read_unlock();
 }
 
@@ -3584,18 +3573,10 @@ static struct ring_buffer *ring_buffer_get(struct perf_event *event)
 
 static void ring_buffer_put(struct ring_buffer *rb)
 {
-	struct perf_event *event, *n;
-	unsigned long flags;
-
 	if (!atomic_dec_and_test(&rb->refcount))
 		return;
 
-	spin_lock_irqsave(&rb->event_lock, flags);
-	list_for_each_entry_safe(event, n, &rb->event_list, rb_entry) {
-		list_del_init(&event->rb_entry);
-		wake_up_all(&event->waitq);
-	}
-	spin_unlock_irqrestore(&rb->event_lock, flags);
+	WARN_ON_ONCE(!list_empty(&rb->event_list));
 
 	call_rcu(&rb->rcu_head, rb_free_rcu);
 }
@@ -3605,26 +3586,100 @@ static void perf_mmap_open(struct vm_area_struct *vma)
 	struct perf_event *event = vma->vm_file->private_data;
 
 	atomic_inc(&event->mmap_count);
+	atomic_inc(&event->rb->mmap_count);
 }
 
+/*
+ * A buffer can be mmap()ed multiple times; either directly through the same
+ * event, or through other events by use of perf_event_set_output().
+ *
+ * In order to undo the VM accounting done by perf_mmap() we need to destroy
+ * the buffer here, where we still have a VM context. This means we need
+ * to detach all events redirecting to us.
+ */
 static void perf_mmap_close(struct vm_area_struct *vma)
 {
 	struct perf_event *event = vma->vm_file->private_data;
 
-	if (atomic_dec_and_mutex_lock(&event->mmap_count, &event->mmap_mutex)) {
-		unsigned long size = perf_data_size(event->rb);
-		struct user_struct *user = event->mmap_user;
-		struct ring_buffer *rb = event->rb;
+	struct ring_buffer *rb = event->rb;
+	struct user_struct *mmap_user = rb->mmap_user;
+	int mmap_locked = rb->mmap_locked;
+	unsigned long size = perf_data_size(rb);
+
+	atomic_dec(&rb->mmap_count);
+
+	if (!atomic_dec_and_mutex_lock(&event->mmap_count, &event->mmap_mutex))
+		return;
 
-		atomic_long_sub((size >> PAGE_SHIFT) + 1, &user->locked_vm);
-		vma->vm_mm->pinned_vm -= event->mmap_locked;
-		rcu_assign_pointer(event->rb, NULL);
-		ring_buffer_detach(event, rb);
+	/* Detach current event from the buffer. */
+	rcu_assign_pointer(event->rb, NULL);
+	ring_buffer_detach(event, rb);
+	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;
+	}
+
+	/*
+	 * No other mmap()s, detach from all other events that might redirect
+	 * into the now unreachable buffer. Somewhat complicated by the
+	 * fact that rb::event_lock otherwise nests inside mmap_mutex.
+	 */
+again:
+	rcu_read_lock();
+	list_for_each_entry_rcu(event, &rb->event_list, rb_entry) {
+		if (!atomic_long_inc_not_zero(&event->refcount)) {
+			/*
+			 * This event is en-route to free_event() which will
+			 * detach it and remove it from the list.
+			 */
+			continue;
+		}
+		rcu_read_unlock();
+
+		mutex_lock(&event->mmap_mutex);
+		/*
+		 * Check we didn't race with perf_event_set_output() which can
+		 * swizzle the rb from under us while we were waiting to
+		 * acquire mmap_mutex.
+		 *
+		 * If we find a different rb; ignore this event, a next
+		 * iteration will no longer find it on the list. We have to
+		 * 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 */
+		}
 		mutex_unlock(&event->mmap_mutex);
+		put_event(event);
 
-		ring_buffer_put(rb);
-		free_uid(user);
+		/*
+		 * Restart the iteration; either we're on the wrong list or
+		 * destroyed its integrity by doing a deletion.
+		 */
+		goto again;
 	}
+	rcu_read_unlock();
+
+	/*
+	 * It could be there's still a few 0-ref events on the list; they'll
+	 * get cleaned up by free_event() -- they'll also still have their
+	 * ref on the rb and will free it whenever they are done with it.
+	 *
+	 * Aside from that, this buffer is 'fully' detached and unmapped,
+	 * undo the VM accounting.
+	 */
+
+	atomic_long_sub((size >> PAGE_SHIFT) + 1, &mmap_user->locked_vm);
+	vma->vm_mm->pinned_vm -= mmap_locked;
+	free_uid(mmap_user);
+
+	ring_buffer_put(rb); /* could be last */
 }
 
 static const struct vm_operations_struct perf_mmap_vmops = {
@@ -3674,12 +3729,24 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma)
 		return -EINVAL;
 
 	WARN_ON_ONCE(event->ctx->parent_ctx);
+again:
 	mutex_lock(&event->mmap_mutex);
 	if (event->rb) {
-		if (event->rb->nr_pages == nr_pages)
-			atomic_inc(&event->rb->refcount);
-		else
+		if (event->rb->nr_pages != nr_pages) {
 			ret = -EINVAL;
+			goto unlock;
+		}
+
+		if (!atomic_inc_not_zero(&event->rb->mmap_count)) {
+			/*
+			 * Raced against perf_mmap_close() through
+			 * perf_event_set_output(). Try again, hope for better
+			 * luck.
+			 */
+			mutex_unlock(&event->mmap_mutex);
+			goto again;
+		}
+
 		goto unlock;
 	}
 
@@ -3720,12 +3787,16 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma)
 		ret = -ENOMEM;
 		goto unlock;
 	}
-	rcu_assign_pointer(event->rb, rb);
+
+	atomic_set(&rb->mmap_count, 1);
+	rb->mmap_locked = extra;
+	rb->mmap_user = get_current_user();
 
 	atomic_long_add(user_extra, &user->locked_vm);
-	event->mmap_locked = extra;
-	event->mmap_user = get_current_user();
-	vma->vm_mm->pinned_vm += event->mmap_locked;
+	vma->vm_mm->pinned_vm += extra;
+
+	ring_buffer_attach(event, rb);
+	rcu_assign_pointer(event->rb, rb);
 
 	perf_event_update_userpage(event);
 
@@ -3734,7 +3805,11 @@ unlock:
 		atomic_inc(&event->mmap_count);
 	mutex_unlock(&event->mmap_mutex);
 
-	vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP;
+	/*
+	 * Since pinned accounting is per vm we cannot allow fork() to copy our
+	 * vma.
+	 */
+	vma->vm_flags |= VM_DONTCOPY | VM_DONTEXPAND | VM_DONTDUMP;
 	vma->vm_ops = &perf_mmap_vmops;
 
 	return ret;
@@ -6412,6 +6487,8 @@ 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);
@@ -6419,16 +6496,28 @@ set:
 			goto unlock;
 	}
 
-	old_rb = event->rb;
-	rcu_assign_pointer(event->rb, rb);
 	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);
+	}
+
 	ret = 0;
 unlock:
 	mutex_unlock(&event->mmap_mutex);
 
-	if (old_rb)
-		ring_buffer_put(old_rb);
 out:
 	return ret;
 }
diff --git a/kernel/events/internal.h b/kernel/events/internal.h
index eb675c4d59df..ca6599723be5 100644
--- a/kernel/events/internal.h
+++ b/kernel/events/internal.h
@@ -31,6 +31,10 @@ struct ring_buffer {
 	spinlock_t			event_lock;
 	struct list_head		event_list;
 
+	atomic_t			mmap_count;
+	unsigned long			mmap_locked;
+	struct user_struct		*mmap_user;
+
 	struct perf_event_mmap_page	*user_page;
 	void				*data_pages[0];
 };
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 3fed7f0cbcdf..bddf3b201a48 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -467,6 +467,7 @@ static struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr)
 /* Optimization staging list, protected by kprobe_mutex */
 static LIST_HEAD(optimizing_list);
 static LIST_HEAD(unoptimizing_list);
+static LIST_HEAD(freeing_list);
 
 static void kprobe_optimizer(struct work_struct *work);
 static DECLARE_DELAYED_WORK(optimizing_work, kprobe_optimizer);
@@ -504,7 +505,7 @@ static __kprobes void do_optimize_kprobes(void)
  * Unoptimize (replace a jump with a breakpoint and remove the breakpoint
  * if need) kprobes listed on unoptimizing_list.
  */
-static __kprobes void do_unoptimize_kprobes(struct list_head *free_list)
+static __kprobes void do_unoptimize_kprobes(void)
 {
 	struct optimized_kprobe *op, *tmp;
 
@@ -515,9 +516,9 @@ static __kprobes void do_unoptimize_kprobes(struct list_head *free_list)
 	/* Ditto to do_optimize_kprobes */
 	get_online_cpus();
 	mutex_lock(&text_mutex);
-	arch_unoptimize_kprobes(&unoptimizing_list, free_list);
+	arch_unoptimize_kprobes(&unoptimizing_list, &freeing_list);
 	/* Loop free_list for disarming */
-	list_for_each_entry_safe(op, tmp, free_list, list) {
+	list_for_each_entry_safe(op, tmp, &freeing_list, list) {
 		/* Disarm probes if marked disabled */
 		if (kprobe_disabled(&op->kp))
 			arch_disarm_kprobe(&op->kp);
@@ -536,11 +537,11 @@ static __kprobes void do_unoptimize_kprobes(struct list_head *free_list)
 }
 
 /* Reclaim all kprobes on the free_list */
-static __kprobes void do_free_cleaned_kprobes(struct list_head *free_list)
+static __kprobes void do_free_cleaned_kprobes(void)
 {
 	struct optimized_kprobe *op, *tmp;
 
-	list_for_each_entry_safe(op, tmp, free_list, list) {
+	list_for_each_entry_safe(op, tmp, &freeing_list, list) {
 		BUG_ON(!kprobe_unused(&op->kp));
 		list_del_init(&op->list);
 		free_aggr_kprobe(&op->kp);
@@ -556,8 +557,6 @@ static __kprobes void kick_kprobe_optimizer(void)
 /* Kprobe jump optimizer */
 static __kprobes void kprobe_optimizer(struct work_struct *work)
 {
-	LIST_HEAD(free_list);
-
 	mutex_lock(&kprobe_mutex);
 	/* Lock modules while optimizing kprobes */
 	mutex_lock(&module_mutex);
@@ -566,7 +565,7 @@ static __kprobes void kprobe_optimizer(struct work_struct *work)
 	 * Step 1: Unoptimize kprobes and collect cleaned (unused and disarmed)
 	 * kprobes before waiting for quiesence period.
 	 */
-	do_unoptimize_kprobes(&free_list);
+	do_unoptimize_kprobes();
 
 	/*
 	 * Step 2: Wait for quiesence period to ensure all running interrupts
@@ -581,7 +580,7 @@ static __kprobes void kprobe_optimizer(struct work_struct *work)
 	do_optimize_kprobes();
 
 	/* Step 4: Free cleaned kprobes after quiesence period */
-	do_free_cleaned_kprobes(&free_list);
+	do_free_cleaned_kprobes();
 
 	mutex_unlock(&module_mutex);
 	mutex_unlock(&kprobe_mutex);
@@ -723,8 +722,19 @@ static void __kprobes kill_optimized_kprobe(struct kprobe *p)
 	if (!list_empty(&op->list))
 		/* Dequeue from the (un)optimization queue */
 		list_del_init(&op->list);
-
 	op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED;
+
+	if (kprobe_unused(p)) {
+		/* Enqueue if it is unused */
+		list_add(&op->list, &freeing_list);
+		/*
+		 * Remove unused probes from the hash list. After waiting
+		 * for synchronization, this probe is reclaimed.
+		 * (reclaiming is done by do_free_cleaned_kprobes().)
+		 */
+		hlist_del_rcu(&op->kp.hlist);
+	}
+
 	/* Don't touch the code, because it is already freed. */
 	arch_remove_optimized_kprobe(op);
 }
diff --git a/kernel/range.c b/kernel/range.c
index eb911dbce267..322ea8e93e4b 100644
--- a/kernel/range.c
+++ b/kernel/range.c
@@ -4,7 +4,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/sort.h>
-
+#include <linux/string.h>
 #include <linux/range.h>
 
 int add_range(struct range *range, int az, int nr_range, u64 start, u64 end)
@@ -32,9 +32,8 @@ int add_range_with_merge(struct range *range, int az, int nr_range,
 	if (start >= end)
 		return nr_range;
 
-	/* Try to merge it with old one: */
+	/* get new start/end: */
 	for (i = 0; i < nr_range; i++) {
-		u64 final_start, final_end;
 		u64 common_start, common_end;
 
 		if (!range[i].end)
@@ -45,14 +44,16 @@ int add_range_with_merge(struct range *range, int az, int nr_range,
 		if (common_start > common_end)
 			continue;
 
-		final_start = min(range[i].start, start);
-		final_end = max(range[i].end, end);
+		/* new start/end, will add it back at last */
+		start = min(range[i].start, start);
+		end = max(range[i].end, end);
 
-		/* clear it and add it back for further merge */
-		range[i].start = 0;
-		range[i].end =  0;
-		return add_range_with_merge(range, az, nr_range,
-			final_start, final_end);
+		memmove(&range[i], &range[i + 1],
+			(nr_range - (i + 1)) * sizeof(range[i]));
+		range[nr_range - 1].start = 0;
+		range[nr_range - 1].end   = 0;
+		nr_range--;
+		i--;
 	}
 
 	/* Need to add it: */
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 58453b8272fd..e8b335016c52 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -633,7 +633,19 @@ void wake_up_nohz_cpu(int cpu)
 static inline bool got_nohz_idle_kick(void)
 {
 	int cpu = smp_processor_id();
-	return idle_cpu(cpu) && test_bit(NOHZ_BALANCE_KICK, nohz_flags(cpu));
+
+	if (!test_bit(NOHZ_BALANCE_KICK, nohz_flags(cpu)))
+		return false;
+
+	if (idle_cpu(cpu) && !need_resched())
+		return true;
+
+	/*
+	 * We can't run Idle Load Balance on this CPU for this time so we
+	 * cancel it and clear NOHZ_BALANCE_KICK
+	 */
+	clear_bit(NOHZ_BALANCE_KICK, nohz_flags(cpu));
+	return false;
 }
 
 #else /* CONFIG_NO_HZ_COMMON */
@@ -1393,8 +1405,9 @@ static void sched_ttwu_pending(void)
 
 void scheduler_ipi(void)
 {
-	if (llist_empty(&this_rq()->wake_list) && !got_nohz_idle_kick()
-	    && !tick_nohz_full_cpu(smp_processor_id()))
+	if (llist_empty(&this_rq()->wake_list)
+			&& !tick_nohz_full_cpu(smp_processor_id())
+			&& !got_nohz_idle_kick())
 		return;
 
 	/*
@@ -1417,7 +1430,7 @@ void scheduler_ipi(void)
 	/*
 	 * Check if someone kicked us for doing the nohz idle load balance.
 	 */
-	if (unlikely(got_nohz_idle_kick() && !need_resched())) {
+	if (unlikely(got_nohz_idle_kick())) {
 		this_rq()->idle_balance = 1;
 		raise_softirq_irqoff(SCHED_SOFTIRQ);
 	}
@@ -4745,7 +4758,7 @@ 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);
+	vtime_init_idle(idle, cpu);
 #if defined(CONFIG_SMP)
 	sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu);
 #endif
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index cc2dc3eea8a3..b5ccba22603b 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -747,17 +747,17 @@ void arch_vtime_task_switch(struct task_struct *prev)
 
 	write_seqlock(&current->vtime_seqlock);
 	current->vtime_snap_whence = VTIME_SYS;
-	current->vtime_snap = sched_clock();
+	current->vtime_snap = sched_clock_cpu(smp_processor_id());
 	write_sequnlock(&current->vtime_seqlock);
 }
 
-void vtime_init_idle(struct task_struct *t)
+void vtime_init_idle(struct task_struct *t, int cpu)
 {
 	unsigned long flags;
 
 	write_seqlock_irqsave(&t->vtime_seqlock, flags);
 	t->vtime_snap_whence = VTIME_SYS;
-	t->vtime_snap = sched_clock();
+	t->vtime_snap = sched_clock_cpu(cpu);
 	write_sequnlock_irqrestore(&t->vtime_seqlock, flags);
 }
 
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 0c739423b0f9..b4c245580b79 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -698,10 +698,6 @@ void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
 
 		bc->event_handler = tick_handle_oneshot_broadcast;
 
-		/* Take the do_timer update */
-		if (!tick_nohz_full_cpu(cpu))
-			tick_do_timer_cpu = cpu;
-
 		/*
 		 * We must be careful here. There might be other CPUs
 		 * waiting for periodic broadcast. We need to set the
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index f4208138fbf4..0cf1c1453181 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -306,7 +306,7 @@ static int __cpuinit tick_nohz_cpu_down_callback(struct notifier_block *nfb,
 		 * we can't safely shutdown that CPU.
 		 */
 		if (have_nohz_full_mask && tick_do_timer_cpu == cpu)
-			return -EINVAL;
+			return NOTIFY_BAD;
 		break;
 	}
 	return NOTIFY_OK;