tracing: Restore system filter behavior

Though not all events have field 'prev_pid', it was allowed to do this:

# echo 'prev_pid == 100' > events/sched/filter

but commit 75b8e98263fdb0bfbdeba60d4db463259f1fe8a2 (tracing/filter: Swap
entire filter of events) broke it without any reason.

Link: http://lkml.kernel.org/r/[e-mail]

Signed-off-by: Li Zefan <[e-mail]>
Signed-off-by: Steven Rostedt <[e-mail]>

include/linux/ftrace_event.h

···
 	TRACE_EVENT_FL_FILTERED_BIT,
 	TRACE_EVENT_FL_RECORDED_CMD_BIT,
 	TRACE_EVENT_FL_CAP_ANY_BIT,
+	TRACE_EVENT_FL_NO_SET_FILTER_BIT,
 };
 
 enum {
···
 	TRACE_EVENT_FL_FILTERED		= (1 << TRACE_EVENT_FL_FILTERED_BIT),
 	TRACE_EVENT_FL_RECORDED_CMD	= (1 << TRACE_EVENT_FL_RECORDED_CMD_BIT),
 	TRACE_EVENT_FL_CAP_ANY		= (1 << TRACE_EVENT_FL_CAP_ANY_BIT),
+	TRACE_EVENT_FL_NO_SET_FILTER	= (1 << TRACE_EVENT_FL_NO_SET_FILTER_BIT),
 };
 
 struct ftrace_event_call {

kernel/trace/trace_events_filter.c

···
 		 */
 		err = replace_preds(call, NULL, ps, filter_string, true);
 		if (err)
-			goto fail;
+			call->flags |= TRACE_EVENT_FL_NO_SET_FILTER;
+		else
+			call->flags &= ~TRACE_EVENT_FL_NO_SET_FILTER;
 	}
 
 	list_for_each_entry(call, &ftrace_events, list) {
 		struct event_filter *filter;
 
 		if (strcmp(call->class->system, system->name) != 0)
+			continue;
+
+		if (call->flags & TRACE_EVENT_FL_NO_SET_FILTER)
 			continue;
 
 		filter_item = kzalloc(sizeof(*filter_item), GFP_KERNEL);