SurfaceStats / Telemetry: calculate FPS in different frame buckets.

build/android/pylib/surface_stats_collector.py

Index: build/android/pylib/surface_stats_collector.py
diff --git a/build/android/pylib/surface_stats_collector.py b/build/android/pylib/surface_stats_collector.py
index d5709fffe7d72159f45e5824cfcf075fc8f949e4..00516fedd3d13d4a405f1d45e5ec4ee479dc0564 100644
--- a/build/android/pylib/surface_stats_collector.py
+++ b/build/android/pylib/surface_stats_collector.py
@@ -85,6 +85,15 @@ class SurfaceStatsCollector(object):
     deltas = [t2 - t1 for t1, t2 in zip(data, data[1:])]
     return (deltas, [delta / refresh_period for delta in deltas])
 
+  def _CalcFpsInBuckets(self, timestamps):
+    for pct in [0.9, 0.5]:
+      sliced = timestamps[int(-pct * len(timestamps)) + 3 : ]

[Sami, 2013/07/03 15:11:50]

Probably want to bail out here if there were too few samples.

[bulach, 2013/07/03 18:09:15]

when is the "legacy" method used? the non-legacy guarantees there'll be at least
three samples.. I can check it in here too, or...remove the legacy method ? :)

[Sami, 2013/07/03 18:21:37]

Ah, good point. The legacy path is only used on pre-JB devices. I think we want
to keep that around since devices like S3 are still on ICS. I don't think we
need to implement windowing there, though.

+      seconds = sliced[-1] - sliced[0]
+      frame_count = len(sliced)
+      self._results.append(SurfaceStatsCollector.Result(
+          'avg_surface_fps_' + str(pct).replace('.', '_'),

[Sami, 2013/07/03 15:11:50]

int(pct * 100) would gives nicer names I think :)

[bulach, 2013/07/03 18:09:15]

indeed! done.

+          int(round(frame_count / seconds)), 'fps'))
+
   def _StorePerfResults(self):
     if self._use_legacy_method:
       surface_after = self._GetSurfaceStatsLegacy()
@@ -122,6 +131,7 @@ class SurfaceStatsCollector(object):
           'frame_lengths', normalized_frame_lengths, 'vsyncs'))
     self._results.append(SurfaceStatsCollector.Result(
         'avg_surface_fps', int(round(frame_count / seconds)), 'fps'))
+    self._CalcFpsInBuckets(timestamps)

[Sami, 2013/07/03 15:11:50]

If we decide to go ahead with this, let's include jank_count, max_frame_delay
and frame_lengths too.

[bulach, 2013/07/03 18:09:15]

I need to understand this better though: they all depend on "refresh_period",
which seems to be for the entire set?
that is, jank_count depends on normalized_changes which depends on this
"refresh_period"... can we derive "refresh_period" from the trimmed timestamps
instead, i.e., "seconds"?

[Sami, 2013/07/03 18:21:37]

|refresh_period| is a constant given by the display panel and is used to
normalize/threshold the data in |timestamps|. That is, it's used to derive these
metrics and not the other way around.

One way this could work is to move the loop here:

  for pct in [0.5, 0.9]:
    trimmed_timestamps = timestamps[int(-pct * len(timestamps)) + 3 : ]
    seconds = trimmed_timestamps[-1] - trimmed_timestamps[0]
    frame_count = len(trimmed_timestamps)
    self._results.append(SurfaceStatsCollector.Result(
        'avg_surface_fps' + suffix, int(round(frame_count / seconds)), 'fps'))
    frame_lengths, normalized_frame_lengths = \
        self._GetNormalizedDeltas(trimmed_timestamps, refresh_period)
    ...
    self._results.append(SurfaceStatsCollector.Result(
          'frame_lengths' + suffix, normalized_frame_lengths, 'vsyncs'))
    [etc.]

 
   def _CollectorThread(self):
     last_timestamp = 0