1 files changed, 51 insertions, 2 deletions

diff --git a/kernel/kthread.c b/kernel/kthread.c
index e29773c82b70..a5eceecd4513 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -704,8 +704,15 @@ repeat:
 	raw_spin_unlock_irq(&worker->lock);
 
 	if (work) {
+		kthread_work_func_t func = work->func;
 		__set_current_state(TASK_RUNNING);
+		trace_sched_kthread_work_execute_start(work);
 		work->func(work);
+		/*
+		 * Avoid dereferencing work after this point.  The trace
+		 * event only cares about the address.
+		 */
+		trace_sched_kthread_work_execute_end(work, func);
 	} else if (!freezing(current))
 		schedule();
 
@@ -786,7 +793,25 @@ EXPORT_SYMBOL(kthread_create_worker);
  * A good practice is to add the cpu number also into the worker name.
  * For example, use kthread_create_worker_on_cpu(cpu, "helper/%d", cpu).
  *
- * Returns a pointer to the allocated worker on success, ERR_PTR(-ENOMEM)
+ * CPU hotplug:
+ * The kthread worker API is simple and generic. It just provides a way
+ * to create, use, and destroy workers.
+ *
+ * It is up to the API user how to handle CPU hotplug. They have to decide
+ * how to handle pending work items, prevent queuing new ones, and
+ * restore the functionality when the CPU goes off and on. There are a
+ * few catches:
+ *
+ *    - CPU affinity gets lost when it is scheduled on an offline CPU.
+ *
+ *    - The worker might not exist when the CPU was off when the user
+ *      created the workers.
+ *
+ * Good practice is to implement two CPU hotplug callbacks and to
+ * destroy/create the worker when the CPU goes down/up.
+ *
+ * Return:
+ * The pointer to the allocated worker on success, ERR_PTR(-ENOMEM)
  * when the needed structures could not get allocated, and ERR_PTR(-EINTR)
  * when the worker was SIGKILLed.
  */
@@ -834,6 +859,8 @@ static void kthread_insert_work(struct kthread_worker *worker,
 {
 	kthread_insert_work_sanity_check(worker, work);
 
+	trace_sched_kthread_work_queue_work(worker, work);
+
 	list_add_tail(&work->node, pos);
 	work->worker = worker;
 	if (!worker->current_work && likely(worker->task))
@@ -897,7 +924,8 @@ void kthread_delayed_work_timer_fn(struct timer_list *t)
 	/* Move the work from worker->delayed_work_list. */
 	WARN_ON_ONCE(list_empty(&work->node));
 	list_del_init(&work->node);
-	kthread_insert_work(worker, work, &worker->work_list);
+	if (!work->canceling)
+		kthread_insert_work(worker, work, &worker->work_list);
 
 	raw_spin_unlock_irqrestore(&worker->lock, flags);
 }
@@ -1248,6 +1276,7 @@ void kthread_use_mm(struct mm_struct *mm)
 		tsk->active_mm = mm;
 	}
 	tsk->mm = mm;
+	membarrier_update_current_mm(mm);
 	switch_mm_irqs_off(active_mm, mm, tsk);
 	local_irq_enable();
 	task_unlock(tsk);
@@ -1255,8 +1284,19 @@ void kthread_use_mm(struct mm_struct *mm)
 	finish_arch_post_lock_switch();
 #endif
 
+	/*
+	 * When a kthread starts operating on an address space, the loop
+	 * in membarrier_{private,global}_expedited() may not observe
+	 * that tsk->mm, and not issue an IPI. Membarrier requires a
+	 * memory barrier after storing to tsk->mm, before accessing
+	 * user-space memory. A full memory barrier for membarrier
+	 * {PRIVATE,GLOBAL}_EXPEDITED is implicitly provided by
+	 * mmdrop(), or explicitly with smp_mb().
+	 */
 	if (active_mm != mm)
 		mmdrop(active_mm);
+	else
+		smp_mb();
 
 	to_kthread(tsk)->oldfs = force_uaccess_begin();
 }
@@ -1276,9 +1316,18 @@ void kthread_unuse_mm(struct mm_struct *mm)
 	force_uaccess_end(to_kthread(tsk)->oldfs);
 
 	task_lock(tsk);
+	/*
+	 * When a kthread stops operating on an address space, the loop
+	 * in membarrier_{private,global}_expedited() may not observe
+	 * that tsk->mm, and not issue an IPI. Membarrier requires a
+	 * memory barrier after accessing user-space memory, before
+	 * clearing tsk->mm.
+	 */
+	smp_mb__after_spinlock();
 	sync_mm_rss(mm);
 	local_irq_disable();
 	tsk->mm = NULL;
+	membarrier_update_current_mm(NULL);
 	/* active_mm is still 'mm' */
 	enter_lazy_tlb(mm, tsk);
 	local_irq_enable();