[tools/perf] Mark stories as failed in the case of metrics failure

Telemetry marks stories as failed when there is a problem with metrics
computation. This CL reproduces the same behavior in Results Processor.

Bug: 1010041
Change-Id: Idb0b86fc30c2cc4a9e8fd18feb56737501f34487
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1840654
Reviewed-by: Juan Antonio Navarro Pérez <perezju@chromium.org>
Commit-Queue: Mikhail Khokhlov <khokhlov@google.com>
Cr-Commit-Position: refs/heads/master@{#702867}

tools/perf/core/results_processor/compute_metrics.py

@@ -48,6 +48,7 @@ def _PoolWorker(test_result):
         test_result['testPath'], time.time() - start))
 
     if mre_result.failures:
+      test_result['status'] = 'FAIL'
       for f in mre_result.failures:
         logging.error('Failure recorded for test %s: %s',
                       test_result['testPath'], f)
@@ -92,6 +93,7 @@ def ComputeTBMv2Metrics(intermediate_results):
     # details.
     # TODO(crbug.com/1010041): Return a non-zero exit code in this case.
     if trace_size_in_mib > 400:
+      test_result['status'] = 'FAIL'
       logging.error('%s: Trace size is too big: %s MiB',
                     test_result['testPath'], trace_size_in_mib)
       continue

tools/perf/core/results_processor/compute_metrics_unittest.py

@@ -7,6 +7,8 @@
 from core.results_processor import compute_metrics
 from core.results_processor import testing
 
+from tracing.mre import failure
+from tracing.mre import job
 from tracing.mre import mre_result
 from tracing.value import histogram
 
@@ -47,6 +49,8 @@ def testComputeTBMv2Metrics(self):
         histogram_dicts = compute_metrics.ComputeTBMv2Metrics(in_results)
 
     self.assertEqual(histogram_dicts, [test_dict, test_dict])
+    self.assertEqual(in_results['testResults'][0]['status'], 'PASS')
+    self.assertEqual(in_results['testResults'][1]['status'], 'PASS')
 
   def testComputeTBMv2MetricsTraceTooBig(self):
     in_results = testing.IntermediateResults([
@@ -66,3 +70,27 @@ def testComputeTBMv2MetricsTraceTooBig(self):
         run_metrics_mock.assert_not_called()
 
     self.assertEqual(histogram_dicts, [])
+    self.assertEqual(in_results['testResults'][0]['status'], 'FAIL')
+
+  def testComputeTBMv2MetricsFailure(self):
+    in_results = testing.IntermediateResults([
+        testing.TestResult(
+            'benchmark/story1',
+            artifacts={
+                compute_metrics.HTML_TRACE_NAME:
+                    testing.Artifact('/trace1.html', 'gs://trace1.html')},
+            tags=['tbmv2:metric1'],
+        ),
+    ])
+
+    metrics_result = mre_result.MreResult()
+    metrics_result.AddFailure(failure.Failure(job.Job(0), 0, 0, 0, 0, 0))
+
+    with mock.patch(GETSIZE_METHOD) as getsize_mock:
+      with mock.patch(RUN_METRICS_METHOD) as run_metrics_mock:
+        getsize_mock.return_value = 100
+        run_metrics_mock.return_value = metrics_result
+        histogram_dicts = compute_metrics.ComputeTBMv2Metrics(in_results)
+
+    self.assertEqual(histogram_dicts, [])
+    self.assertEqual(in_results['testResults'][0]['status'], 'FAIL')