Enable renderer-side mixing output of different audio sources basing on finch

content/renderer/media/audio_device_factory.cc

Index: content/renderer/media/audio_device_factory.cc
diff --git a/content/renderer/media/audio_device_factory.cc b/content/renderer/media/audio_device_factory.cc
index af6ddbe18517603da9da3e48b18ab1fac458e51b..d6fcbaf9c55e30faf02614be1ccb14e799f495c1 100644
--- a/content/renderer/media/audio_device_factory.cc
+++ b/content/renderer/media/audio_device_factory.cc
@@ -15,6 +15,7 @@
 #include "media/audio/audio_input_device.h"
 #include "media/audio/audio_output_device.h"
 #include "media/base/audio_renderer_mixer_input.h"
+#include "media/base/media_switches.h"
 #include "url/origin.h"
 
 namespace content {
@@ -48,8 +49,9 @@ bool IsMixable(AudioDeviceFactory::SourceType source_type) {
   if (source_type == AudioDeviceFactory::kSourceMediaElement)
     return true;  // Must ALWAYS go through mixer.
 
-  // TODO(olka): make a decision for the rest of the sources basing on OS
-  // type and configuration parameters.
+  if (base::FeatureList::IsEnabled(media::kNewAudioRenderingMixingStrategy))

[tommi (sloooow) - chröme, 2016/06/29 14:03:40]

return base::FeatureList::IsEnabled(media::kNewAudioRenderingMixingStrategy);

[o1ka, 2016/06/29 14:28:35]

Done.

+    return true;  // Mix everything if experiment is enabled.
+
   return false;
 }