Drop WebRTC audio data if OS has skipped frames.

content/renderer/media/webrtc_audio_renderer.cc

Index: content/renderer/media/webrtc_audio_renderer.cc
diff --git a/content/renderer/media/webrtc_audio_renderer.cc b/content/renderer/media/webrtc_audio_renderer.cc
index 640d8dc4ad689ba6db96781017231662b532d0be..6689222262dac1be09a3c8330de50cddbee3c557 100644
--- a/content/renderer/media/webrtc_audio_renderer.cc
+++ b/content/renderer/media/webrtc_audio_renderer.cc
@@ -187,7 +187,6 @@ WebRtcAudioRenderer::WebRtcAudioRenderer(
       play_ref_count_(0),
       start_ref_count_(0),
       audio_delay_milliseconds_(0),
-      fifo_delay_milliseconds_(0),
       sink_params_(media::AudioParameters::AUDIO_PCM_LOW_LATENCY,
                    media::CHANNEL_LAYOUT_STEREO,
                    0,
@@ -434,6 +433,26 @@ int WebRtcAudioRenderer::Render(media::AudioBus* audio_bus,
   DCHECK_LE(audio_delay_milliseconds, static_cast<uint32_t>(INT_MAX));
   audio_delay_milliseconds_ = static_cast<int>(audio_delay_milliseconds);
 
+  // If there are skipped frames, pull and throw away the same amount.
+  if (frames_skipped > 0) {
+    const uint32_t frames_per_10ms =
+        static_cast<uint32_t>(sink_params_.sample_rate() / 100);
+    if (!audio_fifo_ && frames_skipped != frames_per_10ms) {
+      audio_fifo_.reset(new media::AudioPullFifo(
+          sink_params_.channels(), frames_per_10ms,
+          base::Bind(&WebRtcAudioRenderer::SourceCallback,
+                     base::Unretained(this))));
+    }
+
+    scoped_ptr<media::AudioBus> drop_bus =
+        media::AudioBus::Create(audio_bus->channels(), frames_skipped);
+    if (audio_fifo_)
+      audio_fifo_->Consume(drop_bus.get(), drop_bus->frames());
+    else
+      SourceCallback(0, drop_bus.get());
+  }
+
+  // Pull the data we shall deliver.

[tommi (sloooow) - chröme, 2016/01/15 18:50:03]

nit: s/shall/will

[Henrik Grunell, 2016/01/18 13:00:54]

Done.

   if (audio_fifo_)
     audio_fifo_->Consume(audio_bus, audio_bus->frames());
   else
@@ -457,7 +476,9 @@ void WebRtcAudioRenderer::SourceCallback(
            << audio_bus->frames() << ")";
 
   int output_delay_milliseconds = audio_delay_milliseconds_;
-  output_delay_milliseconds += fifo_delay_milliseconds_;
+  output_delay_milliseconds += fifo_frame_delay *
+                               base::Time::kMillisecondsPerSecond /
+                               sink_params_.sample_rate();
   DVLOG(2) << "output_delay_milliseconds: " << output_delay_milliseconds;
 
   // We need to keep render data for the |source_| regardless of |state_|,
@@ -627,34 +648,28 @@ void WebRtcAudioRenderer::PrepareSink() {
   // Create a FIFO if re-buffering is required to match the source input with
   // the sink request. The source acts as provider here and the sink as
   // consumer.
-  int new_fifo_delay_milliseconds = 0;
   scoped_ptr<media::AudioPullFifo> new_audio_fifo;
-  if (source_params.frames_per_buffer() !=
-      new_sink_params.frames_per_buffer()) {
+  DCHECK_EQ(source_params.channels(),
+            media::ChannelLayoutToChannelCount(media::CHANNEL_LAYOUT_STEREO));

[tommi (sloooow) - chröme, 2016/01/15 18:50:03]

new assumption or did the previous code assume this?

[Henrik Grunell, 2016/01/18 13:00:54]

Actually, the assumption is that source and sink operates on the same. But the
source params is always initialized with the sink params' channel layout. In
addition, the sink layout is never changed. So this assert is unnecessary.
Though the code in this function is a bit hard to read. Removed.

+  const bool different_source_sink_frames =
+      source_params.frames_per_buffer() != new_sink_params.frames_per_buffer();
+  if (different_source_sink_frames) {
     DVLOG(1) << "Rebuffering from " << source_params.frames_per_buffer()
              << " to " << new_sink_params.frames_per_buffer();
-    new_audio_fifo.reset(new media::AudioPullFifo(
-        source_params.channels(), source_params.frames_per_buffer(),
-        base::Bind(&WebRtcAudioRenderer::SourceCallback,
-                   base::Unretained(this))));
-
-    if (new_sink_params.frames_per_buffer() >
-        source_params.frames_per_buffer()) {
-      int frame_duration_milliseconds =
-          base::Time::kMillisecondsPerSecond /
-          static_cast<double>(source_params.sample_rate());
-      new_fifo_delay_milliseconds = (new_sink_params.frames_per_buffer() -
-                                     source_params.frames_per_buffer()) *
-                                    frame_duration_milliseconds;
-    }
   }
 
   {
     base::AutoLock lock(lock_);
+    const bool source_frames_changed =
+        audio_fifo_->SizeInFrames() != source_params.frames_per_buffer();
+    if ((!audio_fifo_ && different_source_sink_frames) ||
+        (audio_fifo_ && source_frames_changed)) {
+      audio_fifo_.reset(new media::AudioPullFifo(
+          source_params.channels(), source_params.frames_per_buffer(),
+          base::Bind(&WebRtcAudioRenderer::SourceCallback,
+                     base::Unretained(this))));
+    }
     sink_params_ = new_sink_params;
-    fifo_delay_milliseconds_ = new_fifo_delay_milliseconds;
-    if (new_audio_fifo.get())
-      audio_fifo_ = std::move(new_audio_fifo);
   }
 
   sink_->Initialize(new_sink_params, this);