[telemetry] Add TodoMVC examples based test for V8 to telemetry

tools/perf/benchmarks/v8.py

 # Copyright 2014 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 import shlex
 
 from core import perf_benchmark
 
 from measurements import v8_detached_context_age_in_gc
 from measurements import v8_gc_times
 import page_sets
@@ skipping 100 matching lines @@
       return (value.name.startswith('memory_') and
               'v8_renderer' in value.name) or \
              (value.name.startswith('v8_') and not
               value.name.startswith('v8_gc'))
     else:
       return value.tir_label == 'Scrolling'
 
   @classmethod
   def ShouldTearDownStateAfterEachStoryRun(cls):
     return True
+
+
+class V8TodoMVC(perf_benchmark.PerfBenchmark):
+  """Measures V8 Execution metrics on the TodoMVC examples."""
+  page_set = page_sets.TodoMVCPageSet
+
+  def CreateTimelineBasedMeasurementOptions(self):
+    category_filter = tracing_category_filter.TracingCategoryFilter('v8')
+    options = timeline_based_measurement.Options(category_filter)
+    options.SetLegacyTimelineBasedMetrics([v8_execution.V8ExecutionMetric()])
+    return options
+
+  @classmethod
+  def Name(cls):
+    return 'v8.todomvc'
+
+  @classmethod
+  def ShouldTearDownStateAfterEachStoryRun(cls):
+    return True

[Sami, 2016/03/09 18:39:02]

Any particular reason for restarting? It probably makes sense for the memory
cases above but not sure about the numbers you're collecting here.

[jochen (gone - plz use gerrit), 2016/03/09 18:44:42]

using the same renderer will result in some of the stuff coming from the memory
cache, eg. I'd expect that the analytics scripts on all those pages eventually
will come out of the compilation cache.