workqueue: add queue_work and activate_work trace points

These two tracepoints allow tracking when and how a work is queued and
activated.  This patch is based on Frederic's patch to add queue_work
trace point.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>

include/trace/events/workqueue.h

@@ -24,6 +24,59 @@ DECLARE_EVENT_CLASS(workqueue_work,
 	TP_printk("work struct %p", __entry->work)
 );
 
+/**
+ * workqueue_queue_work - called when a work gets queued
+ * @req_cpu:	the requested cpu
+ * @cwq:	pointer to struct cpu_workqueue_struct
+ * @work:	pointer to struct work_struct
+ *
+ * This event occurs when a work is queued immediately or once a
+ * delayed work is actually queued on a workqueue (ie: once the delay
+ * has been reached).
+ */
+TRACE_EVENT(workqueue_queue_work,
+
+	TP_PROTO(unsigned int req_cpu, struct cpu_workqueue_struct *cwq,
+		 struct work_struct *work),
+
+	TP_ARGS(req_cpu, cwq, work),
+
+	TP_STRUCT__entry(
+		__field( void *,	work	)
+		__field( void *,	function)
+		__field( void *,	workqueue)
+		__field( unsigned int,	req_cpu	)
+		__field( unsigned int,	cpu	)
+	),
+
+	TP_fast_assign(
+		__entry->work		= work;
+		__entry->function	= work->func;
+		__entry->workqueue	= cwq->wq;
+		__entry->req_cpu	= req_cpu;
+		__entry->cpu		= cwq->gcwq->cpu;
+	),
+
+	TP_printk("work struct=%p function=%pf workqueue=%p req_cpu=%u cpu=%u",
+		  __entry->work, __entry->function, __entry->workqueue,
+		  __entry->req_cpu, __entry->cpu)
+);
+
+/**
+ * workqueue_activate_work - called when a work gets activated
+ * @work:	pointer to struct work_struct
+ *
+ * This event occurs when a queued work is put on the active queue,
+ * which happens immediately after queueing unless @max_active limit
+ * is reached.
+ */
+DEFINE_EVENT(workqueue_work, workqueue_activate_work,
+
+	TP_PROTO(struct work_struct *work),
+
+	TP_ARGS(work)
+);
+
 /**
  * workqueue_execute_start - called immediately before the workqueue callback
  * @work:	pointer to struct work_struct

kernel/workqueue.c

@@ -997,6 +997,7 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq,
 
 	/* gcwq determined, get cwq and queue */
 	cwq = get_cwq(gcwq->cpu, wq);
+	trace_workqueue_queue_work(cpu, cwq, work);
 
 	BUG_ON(!list_empty(&work->entry));
 
@@ -1004,6 +1005,7 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq,
 	work_flags = work_color_to_flags(cwq->work_color);
 
 	if (likely(cwq->nr_active < cwq->max_active)) {
+		trace_workqueue_activate_work(work);
 		cwq->nr_active++;
 		worklist = gcwq_determine_ins_pos(gcwq, cwq);
 	} else {
@@ -1679,6 +1681,7 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq)
 						    struct work_struct, entry);
 	struct list_head *pos = gcwq_determine_ins_pos(cwq->gcwq, cwq);
 
+	trace_workqueue_activate_work(work);
 	move_linked_works(work, pos, NULL);
 	__clear_bit(WORK_STRUCT_DELAYED_BIT, work_data_bits(work));
 	cwq->nr_active++;