Express audio delay more precisely in frames rather than milliseconds.

media/audio/audio_output_stream_sink.cc

Index: media/audio/audio_output_stream_sink.cc
diff --git a/media/audio/audio_output_stream_sink.cc b/media/audio/audio_output_stream_sink.cc
index 92739b31e3a771f5978201e887fb01b9ca32eaef..c94abf6f0a7e2e099e80c696858283cb00566f61 100644
--- a/media/audio/audio_output_stream_sink.cc
+++ b/media/audio/audio_output_stream_sink.cc
@@ -83,8 +83,10 @@ int AudioOutputStreamSink::OnMoreData(AudioBus* dest,
   if (!active_render_callback_)
     return 0;
 
-  return active_render_callback_->Render(
-      dest, total_bytes_delay * 1000.0 / active_params_.GetBytesPerSecond(), 0);
+  uint32_t frames_delayed = std::round(static_cast<double>(total_bytes_delay) /
+                                       active_params_.GetBytesPerFrame());
+
+  return active_render_callback_->Render(dest, frames_delayed, 0);

[chcunningham, 2016/02/11 01:25:42]

frames_skipped is not propagated? Anyone know why?

[DaleCurtis, 2016/02/11 01:57:15]

This is only used for Mojo demos currently I think. Feel free to forward it.

[chcunningham, 2016/02/11 21:02:52]

Done.

[Henrik Grunell, 2016/02/12 17:47:01]

|frames_skipped| is only used for WebRTC. Other clients may possibly use it in
the future.

[chcunningham, 2016/02/17 01:56:52]

Acknowledged.

 }
 
 void AudioOutputStreamSink::OnError(AudioOutputStream* stream) {