[PATCH] perf/x86/intel: Fix n_metric for the canceled group

* [PATCH] perf/x86/intel: Fix n_metric for the canceled group
@ 2020-09-30 14:29 kan.liang
  2020-10-02 11:02 ` Peter Zijlstra
  0 siblings, 1 reply; 8+ messages in thread
From: kan.liang @ 2020-09-30 14:29 UTC (permalink / raw)
  To: peterz, mingo, linux-kernel; +Cc: ak, Kan Liang

From: Kan Liang <kan.liang@linux.intel.com>

When a group that has TopDown members is failed to be scheduled, any
later TopDown groups will not return valid values.

Here is an example.

A background perf that occupies all the GP counters and the fixed
counter 1.
 $perf stat -e "{cycles,cycles,cycles,cycles,cycles,cycles,cycles,
                 cycles,cycles}:D" -a

A user monitors a TopDown group. It works well, because the fixed
counter 3 and the PERF_METRICS are available.
 $perf stat -x, --topdown -- ./workload
   retiring,bad speculation,frontend bound,backend bound,
   18.0,16.1,40.4,25.5,

Then the user tries to monitor a group that has TopDown members.
Because of the cycles event, the group is failed to be scheduled.
 $perf stat -x, -e '{slots,topdown-retiring,topdown-be-bound,
                     topdown-fe-bound,topdown-bad-spec,cycles}'
                     -- ./workload
    <not counted>,,slots,0,0.00,,
    <not counted>,,topdown-retiring,0,0.00,,
    <not counted>,,topdown-be-bound,0,0.00,,
    <not counted>,,topdown-fe-bound,0,0.00,,
    <not counted>,,topdown-bad-spec,0,0.00,,
    <not counted>,,cycles,0,0.00,,

The user tries to monitor a TopDown group again. It doesn't work anymore.
 $perf stat -x, --topdown -- ./workload

    ,,,,,

In a txn, cancel_txn() is to truncate the event_list for a canceled
group and update the number of events added in this transaction.
However, the number of TopDown events added in this transaction is not
updated. The kernel will probably fail to add new Topdown events.

Check if the canceled group has Topdown events. If so, subtract the
TopDown events from n_metric accordingly.

Fixes: 7b2c05a15d29 ("perf/x86/intel: Generic support for hardware TopDown metrics")
Reported-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
---
 arch/x86/events/core.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index 0f3d01562ded..4cb3ccbe2d62 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -2017,6 +2017,7 @@ static void x86_pmu_cancel_txn(struct pmu *pmu)
 {
 	unsigned int txn_flags;
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int i;
 
 	WARN_ON_ONCE(!cpuc->txn_flags);	/* no txn in flight */
 
@@ -2031,6 +2032,15 @@ static void x86_pmu_cancel_txn(struct pmu *pmu)
 	 */
 	__this_cpu_sub(cpu_hw_events.n_added, __this_cpu_read(cpu_hw_events.n_txn));
 	__this_cpu_sub(cpu_hw_events.n_events, __this_cpu_read(cpu_hw_events.n_txn));
+
+	/* Subtract Topdown events in the canceled group from n_metric */
+	if (x86_pmu.intel_cap.perf_metrics && cpuc->n_metric) {
+		for (i = 0; i < cpuc->n_txn; i++) {
+			if (is_metric_event(cpuc->event_list[i + cpuc->n_events]))
+				__this_cpu_dec(cpu_hw_events.n_metric);
+		}
+		WARN_ON_ONCE(__this_cpu_read(cpu_hw_events.n_metric) < 0);
+	}
 	perf_pmu_enable(pmu);
 }
 
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 8+ messages in thread