diff options
| author | Liqiang TAO <taolq@outlook.com> | 2022-01-25 19:14:20 +0900 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2022-01-25 10:14:20 +0000 |
| commit | d0fbf8ac238d3bc3a111334ab3553173812d78e7 (patch) | |
| tree | 589bdc6d72203e72080ad9b3e128f8d4ce9580b0 /src/benchmark_runner.cc | |
| parent | 57b2bfa33bbc4f171a7fc2807de71583b1e2fd2a (diff) | |
| download | google-benchmark-d0fbf8ac238d3bc3a111334ab3553173812d78e7.tar.gz | |
Cache PerfCounters instance in PerfCountersMeasurement (#1308)
This patch fixes #1306, by reducing the pinned instances of
PerfCounters.
The issue is caused by creating multiple pinned events in the
same thread, doing so results in the Snapshot(PerfCounterValues* values)
failing, and that's now discoverable.
Creating multile pinned events is an unsupported behavior currently.
The error would be detected at read() time, not
perf_event_open() / iotcl() time.
The unsupported benavior above is confirmed by Stephane Eranian @seranian,
and he also pointed the dectection method.
Finished this patch under the guidance of Mircea Trofin @mtrofin.
Diffstat (limited to 'src/benchmark_runner.cc')
| -rw-r--r-- | src/benchmark_runner.cc | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/src/benchmark_runner.cc b/src/benchmark_runner.cc index eac807b..ec7c127 100644 --- a/src/benchmark_runner.cc +++ b/src/benchmark_runner.cc @@ -152,8 +152,7 @@ BenchmarkRunner::BenchmarkRunner( has_explicit_iteration_count(b.iterations() != 0), pool(b.threads() - 1), iters(has_explicit_iteration_count ? b.iterations() : 1), - perf_counters_measurement( - PerfCounters::Create(StrSplit(FLAGS_benchmark_perf_counters, ','))), + perf_counters_measurement(StrSplit(FLAGS_benchmark_perf_counters, ',')), perf_counters_measurement_ptr(perf_counters_measurement.IsValid() ? &perf_counters_measurement : nullptr) { |