[PATCH 5/9] perf/x86/intel: Set correct weight for TopDown metrics events

From: kan . liang
Date: Tue May 21 2019 - 17:44:33 EST


From: Andi Kleen <ak@xxxxxxxxxxxxxxx>

The topdown metrics and slots events are mapped to a fixed counter,
but should have the normal weight for the scheduler.
So special case this.

Signed-off-by: Andi Kleen <ak@xxxxxxxxxxxxxxx>
Signed-off-by: Kan Liang <kan.liang@xxxxxxxxxxxxxxx>
---
arch/x86/events/intel/core.c | 9 +++++++++
1 file changed, 9 insertions(+)

diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index 2eec172765f4..6de9249acb28 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -5281,6 +5281,15 @@ __init int intel_pmu_init(void)
* counter, so do not extend mask to generic counters
*/
for_each_event_constraint(c, x86_pmu.event_constraints) {
+ /*
+ * Don't limit the event mask for TopDown
+ * metrics and slots events.
+ */
+ if (x86_pmu.num_counters_fixed >= 3 &&
+ c->idxmsk64 & INTEL_PMC_MSK_ANY_SLOTS) {
+ c->weight = hweight64(c->idxmsk64);
+ continue;
+ }
if (c->cmask == FIXED_EVENT_FLAGS
&& c->idxmsk64 != INTEL_PMC_MSK_FIXED_REF_CYCLES) {
c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1;
--
2.14.5