perf trace: Add perf_event parameter to tracepoint_handler
authorStanislav Fomichev <stfomichev@yandex-team.ru>
Thu, 26 Jun 2014 16:14:24 +0000 (20:14 +0400)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Thu, 26 Jun 2014 19:00:45 +0000 (16:00 -0300)
It will be used by next pagefault tracing patches in the series.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1403799268-1367-2-git-send-email-stfomichev@yandex-team.ru
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/builtin-trace.c

index 5549cee61680a8d795aefe237fda89923659960e..4a9e26b731febdd334e56155d75b66830ff5ae34 100644 (file)
@@ -1538,6 +1538,7 @@ static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size,
 }
 
 typedef int (*tracepoint_handler)(struct trace *trace, struct perf_evsel *evsel,
+                                 union perf_event *event,
                                  struct perf_sample *sample);
 
 static struct syscall *trace__syscall_info(struct trace *trace,
@@ -1610,6 +1611,7 @@ static void thread__update_stats(struct thread_trace *ttrace,
 }
 
 static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
+                           union perf_event *event __maybe_unused,
                            struct perf_sample *sample)
 {
        char *msg;
@@ -1658,6 +1660,7 @@ static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
 }
 
 static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel,
+                          union perf_event *event __maybe_unused,
                           struct perf_sample *sample)
 {
        int ret;
@@ -1735,6 +1738,7 @@ out:
 }
 
 static int trace__vfs_getname(struct trace *trace, struct perf_evsel *evsel,
+                             union perf_event *event __maybe_unused,
                              struct perf_sample *sample)
 {
        trace->last_vfs_getname = perf_evsel__rawptr(evsel, sample, "pathname");
@@ -1742,6 +1746,7 @@ static int trace__vfs_getname(struct trace *trace, struct perf_evsel *evsel,
 }
 
 static int trace__sched_stat_runtime(struct trace *trace, struct perf_evsel *evsel,
+                                    union perf_event *event __maybe_unused,
                                     struct perf_sample *sample)
 {
         u64 runtime = perf_evsel__intval(evsel, sample, "runtime");
@@ -1781,7 +1786,7 @@ static bool skip_sample(struct trace *trace, struct perf_sample *sample)
 }
 
 static int trace__process_sample(struct perf_tool *tool,
-                                union perf_event *event __maybe_unused,
+                                union perf_event *event,
                                 struct perf_sample *sample,
                                 struct perf_evsel *evsel,
                                 struct machine *machine __maybe_unused)
@@ -1799,7 +1804,7 @@ static int trace__process_sample(struct perf_tool *tool,
 
        if (handler) {
                ++trace->nr_events;
-               handler(trace, evsel, sample);
+               handler(trace, evsel, event, sample);
        }
 
        return err;
@@ -1990,7 +1995,7 @@ again:
                        }
 
                        handler = evsel->handler;
-                       handler(trace, evsel, &sample);
+                       handler(trace, evsel, event, &sample);
 next_event:
                        perf_evlist__mmap_consume(evlist, i);