Support Opus Ambisonics playback

media/renderers/audio_renderer_impl.cc

Index: media/renderers/audio_renderer_impl.cc
diff --git a/media/renderers/audio_renderer_impl.cc b/media/renderers/audio_renderer_impl.cc
index a351b9a84ec0a7ae9207f94af19cc412237dc0ee..7abe17163b9e01ac2aedcb3c5dae479a76b7d2e8 100644
--- a/media/renderers/audio_renderer_impl.cc
+++ b/media/renderers/audio_renderer_impl.cc
@@ -48,6 +48,7 @@ AudioRendererImpl::AudioRendererImpl(
       last_decoded_sample_rate_(0),
       last_decoded_channel_layout_(CHANNEL_LAYOUT_NONE),
       is_encrypted_(false),
+      last_decoded_channels_(0),
       playback_rate_(0.0),
       state_(kUninitialized),
       create_audio_decoders_cb_(create_audio_decoders_cb),
@@ -369,9 +370,17 @@ void AudioRendererImpl::Initialize(DemuxerStream* stream,
   auto output_device_info = sink_->GetOutputDeviceInfo();
   const AudioParameters& hw_params = output_device_info.output_params();
   expecting_config_changes_ = stream->SupportsConfigChanges();
-  if (!expecting_config_changes_ || !hw_params.IsValid() ||
-      hw_params.format() == AudioParameters::AUDIO_FAKE ||
-      !sink_->IsOptimizedForHardwareParameters()) {
+
+  bool use_stream_params = !expecting_config_changes_ || !hw_params.IsValid() ||
+                           hw_params.format() == AudioParameters::AUDIO_FAKE ||
+                           !sink_->IsOptimizedForHardwareParameters();
+
+  if (stream->audio_decoder_config().channel_layout() ==

[DaleCurtis, 2017/05/25 00:29:16]

Why do we need this? I think what you had before is correct. I.e. never do
conversion if WebAudio is attached?

[flim-chromium, 2017/05/25 00:46:06]

Yep, that's still there in line 376. But if webaudio isn't attached, then using
the stream params will lead to an error at audio_converter.cc(35) (has DISCRETE
layout but 0 channels)

[DaleCurtis, 2017/05/25 00:53:38]

Seems like you could just drop this and always call set_channels_for_discrete()
on parameters then? The channels() value should be mapped to the layout already
if not discrete?

[flim-chromium, 2017/05/25 02:16:41]

We'd have to set the channels in AudioRendererMixerManager, AudioManagerPulse
and PulseAudioOutputStream (and maybe ALSA too). The html audio control would
then play but there is no sound. Not sure if this would be preferable?
Sorry, I didn't follow - this is for the discrete layout case only?

[DaleCurtis, 2017/05/25 18:51:01]

Hmm, I was proposing a simplified logic, but after looking again I don't think a
simpler form exists. (!a || !b || c || !d) && !(e && d) is the minimal form. So
this is okay, just add a {} around the conditional here.

[flim-chromium, 2017/05/25 22:00:40]

Done.

+          CHANNEL_LAYOUT_DISCRETE &&
+      sink_->IsOptimizedForHardwareParameters())
+    use_stream_params = false;
+
+  if (use_stream_params) {
     // The actual buffer size is controlled via the size of the AudioBus
     // provided to Render(), but we should choose a value here based on hardware
     // parameters if possible since it affects the initial buffer size used by
@@ -384,6 +393,8 @@ void AudioRendererImpl::Initialize(DemuxerStream* stream,
                             stream->audio_decoder_config().samples_per_second(),
                             stream->audio_decoder_config().bits_per_channel(),
                             buffer_size);
+    audio_parameters_.set_channels_for_discrete(
+        stream->audio_decoder_config().channels());
     buffer_converter_.reset();
   } else {
     // To allow for seamless sample rate adaptations (i.e. changes from say
@@ -401,8 +412,7 @@ void AudioRendererImpl::Initialize(DemuxerStream* stream,
     }
 #endif
 
-    int stream_channel_count = ChannelLayoutToChannelCount(
-        stream->audio_decoder_config().channel_layout());
+    int stream_channel_count = stream->audio_decoder_config().channels();
 
     bool try_supported_channel_layouts = false;
 #if defined(OS_WIN)
@@ -458,6 +468,8 @@ void AudioRendererImpl::Initialize(DemuxerStream* stream,
 
   is_encrypted_ = stream->audio_decoder_config().is_encrypted();
 
+  last_decoded_channels_ = stream->audio_decoder_config().channels();
+
   audio_clock_.reset(
       new AudioClock(base::TimeDelta(), audio_parameters_.sample_rate()));
 
@@ -605,6 +617,7 @@ void AudioRendererImpl::DecodedAudioReady(
 
       if (last_decoded_channel_layout_ != buffer->channel_layout()) {
         last_decoded_channel_layout_ = buffer->channel_layout();
+        last_decoded_channels_ = buffer->channel_count();
 
         // Input layouts should never be discrete.
         DCHECK_NE(last_decoded_channel_layout_, CHANNEL_LAYOUT_DISCRETE);
@@ -990,14 +1003,10 @@ void AudioRendererImpl::ConfigureChannelMask() {
   DCHECK(audio_parameters_.IsValid());
   DCHECK_NE(last_decoded_channel_layout_, CHANNEL_LAYOUT_NONE);
   DCHECK_NE(last_decoded_channel_layout_, CHANNEL_LAYOUT_UNSUPPORTED);
-  DCHECK_NE(last_decoded_channel_layout_, CHANNEL_LAYOUT_DISCRETE);
-
-  const int input_channel_count =
-      ChannelLayoutToChannelCount(last_decoded_channel_layout_);
 
   // If we're actually downmixing the signal, no mask is necessary, but ensure
   // we clear any existing mask if present.
-  if (input_channel_count >= audio_parameters_.channels()) {
+  if (last_decoded_channels_ >= audio_parameters_.channels()) {
     algorithm_->SetChannelMask(
         std::vector<bool>(audio_parameters_.channels(), true));
     return;
@@ -1005,7 +1014,7 @@ void AudioRendererImpl::ConfigureChannelMask() {
 
   // Determine the matrix used to upmix the channels.
   std::vector<std::vector<float>> matrix;
-  ChannelMixingMatrix(last_decoded_channel_layout_, input_channel_count,
+  ChannelMixingMatrix(last_decoded_channel_layout_, last_decoded_channels_,
                       audio_parameters_.channel_layout(),
                       audio_parameters_.channels())
       .CreateTransformationMatrix(&matrix);