[PATCH 1/2] perf,core: try parent pmu first when init child event

[kan . liang]

From: Kan Liang <kan.liang@xxxxxxxxx>

perf brings additional overhead when monitoring the task which
frequently generates child task.

perf_init_event is one of the hotspot for the additional overhead.
Currently, to get the pmu, it tries to search the type in pmu_idr at
first. But it is not always successful, especially for the widely used
PERF_TYPE_HARDWARE and PERF_TYPE_HW_CACHE events. So it has to go to the
slow path which go through the whole pmus list.
It will be a big performance issue, if the pmus list is long (E.g. server
with many uncore boxes) and the task frequently generates child tasks.

The child event inherits its parent event. So the child event should
tries its parent pmu first.

Here is some data from the overhead test on Broadwell server
  perf record -e $TEST_EVENTS -- ./loop.sh 50000

loop.sh
  start=$(date +%s%N)
  i=0
  while [ "$i" -le "$1" ]
  do
          date > /dev/null
          i=`expr $i + 1`
  done
  end=$(date +%s%N)
  elapsed=`expr $end - $start`

Event#	Original elapsed time	Elapsed time with patch		delta
1	196,573,192,397		189,162,029,998			-3.77%
2	257,567,753,013		241,620,788,683			-6.19%
4	398,730,726,971		370,518,938,714			-7.08%
8	824,983,761,120		740,702,489,329			-10.22%
16	1,883,411,923,498	1,672,027,508,355		-11.22%

Signed-off-by: Kan Liang <kan.liang@xxxxxxxxx>
---
 kernel/events/core.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 110b38a..924268c 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -8977,6 +8977,14 @@ static struct pmu *perf_init_event(struct perf_event *event)
 
 	idx = srcu_read_lock(&pmus_srcu);
 
+	/* try parent's pmu first */
+	if (event->parent && event->parent->pmu) {
+		pmu = event->parent->pmu;
+		ret = perf_try_init_event(pmu, event);
+		if (!ret)
+			goto unlock;
+	}
+
 	rcu_read_lock();
 	pmu = idr_find(&pmu_idr, event->attr.type);
 	rcu_read_unlock();
-- 
2.4.3