tracing: Show the preempt count of when the event was called

Because tracepoint callbacks are done with preemption enabled, the trace
events are always called with preempt disable due to the
rcu_read_lock_sched_notrace() in __DO_TRACE(). This causes the preempt count
shown in the recorded trace event to be inaccurate. It is always one more
that what the preempt_count was when the tracepoint was called.

If CONFIG_PREEMPT is enabled, subtract 1 from the preempt_count before
recording it in the trace buffer.

Link: http://lkml.kernel.org/r/20160525132537.GA10808@linutronix.deReported-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

kernel/trace/trace_events.c

@@ -261,6 +261,14 @@ void *trace_event_buffer_reserve(struct trace_event_buffer *fbuffer,
 
 	local_save_flags(fbuffer->flags);
 	fbuffer->pc = preempt_count();
+	/*
+	 * If CONFIG_PREEMPT is enabled, then the tracepoint itself disables
+	 * preemption (adding one to the preempt_count). Since we are
+	 * interested in the preempt_count at the time the tracepoint was
+	 * hit, we need to subtract one to offset the increment.
+	 */
+	if (IS_ENABLED(CONFIG_PREEMPT))
+		fbuffer->pc--;
 	fbuffer->trace_file = trace_file;
 
 	fbuffer->event =