perf: Optimize perf_tp_event_match()
authorPeter Zijlstra <a.p.zijlstra@chello.nl>
Thu, 20 May 2010 18:54:31 +0000 (20:54 +0200)
committerIngo Molnar <mingo@elte.hu>
Fri, 21 May 2010 09:38:00 +0000 (11:38 +0200)
Since we know tracepoints come from kernel context,
avoid conditionals that try and establish that very
fact.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <20100521090710.904944001@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/perf_event.c

index d25c864cadbf9f3522211630a92428cea71f5d98..e099650cd249a7fb728e00f409b9557d598ceb8f 100644 (file)
@@ -4496,7 +4496,10 @@ static int perf_tp_event_match(struct perf_event *event,
                                struct perf_sample_data *data,
                                struct pt_regs *regs)
 {
-       if (perf_exclude_event(event, regs))
+       /*
+        * All tracepoints are from kernel-space.
+        */
+       if (event->attr.exclude_kernel)
                return 0;
 
        if (!perf_tp_filter_match(event, data))