perf: Use generic sample reordering in perf sched

Use the new generic sample events reordering from perf sched,
this drops the need of multiplexing the buffers on record time,
improving the scalability of perf sched.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Tom Zanussi <tzanussi@gmail.com>

diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index 09ddc8e6d8e1654d2315ef6d4c7ece66856e6022..94453f1e4e01e4eb566ba786d54d802b9f0105cd 100644 (file)
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -1651,9 +1651,10 @@ static int process_lost_event(event_t *event __used,
 }
 
 static struct perf_event_ops event_ops = {
-       .sample = process_sample_event,
-       .comm   = event__process_comm,
-       .lost   = process_lost_event,
+       .sample                 = process_sample_event,
+       .comm                   = event__process_comm,
+       .lost                   = process_lost_event,
+       .ordered_samples        = true,
 };
 
 static int read_events(void)
@@ -1850,7 +1851,6 @@ static const char *record_args[] = {
        "record",
        "-a",
        "-R",
-       "-M",
        "-f",
        "-m", "1024",
        "-c", "1",