Now run EndToEndTest with the WebRTC-NewVideoJitterBuffer experiment.

webrtc/modules/video_coding/video_receiver.cc

Index: webrtc/modules/video_coding/video_receiver.cc
diff --git a/webrtc/modules/video_coding/video_receiver.cc b/webrtc/modules/video_coding/video_receiver.cc
index 46373ef6cbc376260944ca459627564cc994877b..7395840feb17e99bdb23bccbe1f44981daec79cd 100644
--- a/webrtc/modules/video_coding/video_receiver.cc
+++ b/webrtc/modules/video_coding/video_receiver.cc
@@ -295,6 +295,15 @@ int32_t VideoReceiver::Decode(uint16_t maxWaitTimeMs) {
 //                 VCMEncodedFrame with FrameObject.
 int32_t VideoReceiver::Decode(const webrtc::VCMEncodedFrame* frame) {
   rtc::CritScope lock(&receive_crit_);
+  if (pre_decode_image_callback_) {

[brandtr, 2016/11/21 16:46:45]

Was this a problem found by running the E2E tests?

[philipel, 2016/11/22 14:56:23]

Yes

+    EncodedImage encoded_image(frame->EncodedImage());
+    int qp = -1;
+    if (qp_parser_.GetQp(*frame, &qp)) {
+      encoded_image.qp_ = qp;
+    }
+    pre_decode_image_callback_->OnEncodedImage(encoded_image,
+                                               frame->CodecSpecific(), nullptr);
+  }
   return Decode(*frame);
 }