H.264 packetization mode 0 (try 3)

webrtc/modules/video_coding/codecs/h264/h264_encoder_impl.cc

 /*
  *  Copyright (c) 2015 The WebRTC project authors. All Rights Reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
  *  tree. An additional intellectual property rights grant can be found
  *  in the file PATENTS.  All contributing project authors may
  *  be found in the AUTHORS file in the root of the source tree.
  *
  */
 
 #include "webrtc/modules/video_coding/codecs/h264/h264_encoder_impl.h"
 
 #include <limits>
+#include <string>
 
 #include "third_party/openh264/src/codec/api/svc/codec_api.h"
 #include "third_party/openh264/src/codec/api/svc/codec_app_def.h"
 #include "third_party/openh264/src/codec/api/svc/codec_def.h"
 #include "third_party/openh264/src/codec/api/svc/codec_ver.h"
 
 #include "webrtc/base/checks.h"
 #include "webrtc/base/logging.h"
 #include "webrtc/common_video/libyuv/include/webrtc_libyuv.h"
+#include "webrtc/media/base/mediaconstants.h"
 #include "webrtc/system_wrappers/include/metrics.h"
 
 namespace webrtc {
 
 namespace {
 
 const bool kOpenH264EncoderDetailedLogging = false;
 
 // Used by histograms. Values of entries should not be changed.
 enum H264EncoderImplEvent {
@@ skipping 109 matching lines @@
       layer_len += layerInfo.pNalLengthInByte[nal];
     }
     // Copy the entire layer's data (including start codes).
     memcpy(encoded_image->_buffer + encoded_image->_length,
            layerInfo.pBsBuf,
            layer_len);
     encoded_image->_length += layer_len;
   }
 }
 
-H264EncoderImpl::H264EncoderImpl()
+H264EncoderImpl::H264EncoderImpl(const cricket::VideoCodec& codec)
     : openh264_encoder_(nullptr),
       width_(0),
       height_(0),
       max_frame_rate_(0.0f),
       target_bps_(0),
       max_bps_(0),
       mode_(kRealtimeVideo),
       frame_dropping_on_(false),
       key_frame_interval_(0),
+      packetization_mode_(H264PacketizationMode::SingleNalUnit),
+      max_payload_size_(0),
       number_of_cores_(0),
       encoded_image_callback_(nullptr),
       has_reported_init_(false),
-      has_reported_error_(false) {}
+      has_reported_error_(false) {
+  RTC_CHECK(cricket::CodecNamesEq(codec.name, cricket::kH264CodecName));
+  std::string packetization_mode_string;

[magjed_webrtc, 2016/12/02 12:34:26]

nit: It feels unnecessary to do std::string copies all the time. Can you write
it like this instead:
auto it = codec.params.find(cricket::kH264FmtpPacketizationMode);
if (it != codec.params.end() && strcmp(it->second, "1") == 0)
  packetization_mode_ = H264PacketizationMode::NonInterleaved;

[hta-webrtc, 2016/12/02 12:48:47]

Only by bypassing codec.GetParam. I don't want this piece of code to depend on
the internal representation of parameters inside cricket::Codec.

+  if (codec.GetParam(cricket::kH264FmtpPacketizationMode,
+                     &packetization_mode_string) &&
+      packetization_mode_string == "1") {
+    packetization_mode_ = H264PacketizationMode::NonInterleaved;
+  }
+}
 
 H264EncoderImpl::~H264EncoderImpl() {
   Release();
 }
 
 int32_t H264EncoderImpl::InitEncode(const VideoCodec* codec_settings,
                                     int32_t number_of_cores,
-                                    size_t /*max_payload_size*/) {
+                                    size_t max_payload_size) {
   ReportInit();
   if (!codec_settings ||
       codec_settings->codecType != kVideoCodecH264) {
     ReportError();
     return WEBRTC_VIDEO_CODEC_ERR_PARAMETER;
   }
   if (codec_settings->maxFramerate == 0) {
     ReportError();
     return WEBRTC_VIDEO_CODEC_ERR_PARAMETER;
   }
@@ skipping 26 matching lines @@
   // else WELS_LOG_DEFAULT is used by default.
 
   number_of_cores_ = number_of_cores;
   // Set internal settings from codec_settings
   width_ = codec_settings->width;
   height_ = codec_settings->height;
   max_frame_rate_ = static_cast<float>(codec_settings->maxFramerate);
   mode_ = codec_settings->mode;
   frame_dropping_on_ = codec_settings->H264().frameDroppingOn;
   key_frame_interval_ = codec_settings->H264().keyFrameInterval;
+  max_payload_size_ = max_payload_size;
 
   // Codec_settings uses kbits/second; encoder uses bits/second.
   max_bps_ = codec_settings->maxBitrate * 1000;
   if (codec_settings->targetBitrate == 0)
     target_bps_ = codec_settings->startBitrate * 1000;
   else
     target_bps_ = codec_settings->targetBitrate * 1000;
 
   SEncParamExt encoder_params = CreateEncoderParams();
+
   // Initialize.
   if (openh264_encoder_->InitializeExt(&encoder_params) != 0) {
     LOG(LS_ERROR) << "Failed to initialize OpenH264 encoder";
     Release();
     ReportError();
     return WEBRTC_VIDEO_CODEC_ERROR;
   }
   // TODO(pbos): Base init params on these values before submitting.
   quality_scaler_.Init(codec_settings->codecType, codec_settings->startBitrate,
                        codec_settings->width, codec_settings->height,
@@ skipping 149 matching lines @@
                                           encoded_image_._length);
     int qp = -1;
     if (h264_bitstream_parser_.GetLastSliceQp(&qp)) {
       quality_scaler_.ReportQP(qp);
       encoded_image_.qp_ = qp;
     }
 
     // Deliver encoded image.
     CodecSpecificInfo codec_specific;
     codec_specific.codecType = kVideoCodecH264;
+    codec_specific.codecSpecific.H264.packetization_mode = packetization_mode_;
     encoded_image_callback_->OnEncodedImage(encoded_image_, &codec_specific,
                                             &frag_header);
   } else {
     quality_scaler_.ReportDroppedFrame();
   }
   return WEBRTC_VIDEO_CODEC_OK;
 }
 
 const char* H264EncoderImpl::ImplementationName() const {
   return "OpenH264";
@@ skipping 41 matching lines @@
   encoder_params.iMultipleThreadIdc = NumberOfThreads(
       encoder_params.iPicWidth, encoder_params.iPicHeight, number_of_cores_);
   // The base spatial layer 0 is the only one we use.
   encoder_params.sSpatialLayers[0].iVideoWidth = encoder_params.iPicWidth;
   encoder_params.sSpatialLayers[0].iVideoHeight = encoder_params.iPicHeight;
   encoder_params.sSpatialLayers[0].fFrameRate = encoder_params.fMaxFrameRate;
   encoder_params.sSpatialLayers[0].iSpatialBitrate =
       encoder_params.iTargetBitrate;
   encoder_params.sSpatialLayers[0].iMaxSpatialBitrate =
       encoder_params.iMaxBitrate;
+  LOG(INFO) << "OpenH264 version is " << OPENH264_MAJOR << "."
+            << OPENH264_MINOR;
+  switch (packetization_mode_) {
+    case H264PacketizationMode::SingleNalUnit:
+// Limit the size of the packets produced.

[hbos, 2016/12/02 11:58:42]

nit: indent this:
    case H264PacketizationMode::SingleNalUnit:
      // Limit the size of the packets produced.

[hta-webrtc, 2016/12/02 12:48:47]

git cl format did this. I haven't figured out where to file the bug yet.

 #if (OPENH264_MAJOR == 1) && (OPENH264_MINOR <= 5)
-  // Slice num according to number of threads.
-  encoder_params.sSpatialLayers[0].sSliceCfg.uiSliceMode = SM_AUTO_SLICE;
+      encoder_params.sSpatialLayers[0].sSliceCfg.uiSliceMode = SM_DYN_SLICE;
+      // The slice size is max payload size - room for a NAL header.
+      // The constant 50 is NAL_HEADER_ADD_0X30BYTES in openh264 source,
+      // but is not exported.
+      const kNalHeaderSizeAllocation = 50;
+      encoder_params.sSpatialLayers[0]
+          .sSliceCfg.sSliceArgument.uiSliceSizeConstraint =
+          static_cast<unsigned int>(max_payload_size_ -
+                                    kNalHeaderSizeAllocation);
+      encoder_params.uiMaxNalSize =
+          static_cast<unsigned int>(max_payload_size_);
 #else
-  // When uiSliceMode = SM_FIXEDSLCNUM_SLICE, uiSliceNum = 0 means auto design
-  // it with cpu core number.
-  // TODO(sprang): Set to 0 when we understand why the rate controller borks
-  //               when uiSliceNum > 1.
-  encoder_params.sSpatialLayers[0].sSliceArgument.uiSliceNum = 1;
-  encoder_params.sSpatialLayers[0].sSliceArgument.uiSliceMode =
-      SM_FIXEDSLCNUM_SLICE;
+      encoder_params.sSpatialLayers[0].sSliceArgument.uiSliceNum = 1;
+      encoder_params.sSpatialLayers[0].sSliceArgument.uiSliceMode =
+          SM_SIZELIMITED_SLICE;
+      encoder_params.sSpatialLayers[0].sSliceArgument.uiSliceSizeConstraint =
+          static_cast<unsigned int>(max_payload_size_);
 #endif
-
+      break;
+    case H264PacketizationMode::NonInterleaved:
+#if (OPENH264_MAJOR == 1) && (OPENH264_MINOR <= 5)
+      // Slice num according to number of threads.
+      encoder_params.sSpatialLayers[0].sSliceCfg.uiSliceMode = SM_AUTO_SLICE;
+#else
+      // When uiSliceMode = SM_FIXEDSLCNUM_SLICE, uiSliceNum = 0 means auto
+      // design it with cpu core number.
+      // TODO(sprang): Set to 0 when we understand why the rate controller borks
+      //               when uiSliceNum > 1.
+      encoder_params.sSpatialLayers[0].sSliceArgument.uiSliceNum = 1;
+      encoder_params.sSpatialLayers[0].sSliceArgument.uiSliceMode =
+          SM_FIXEDSLCNUM_SLICE;
+#endif
+      break;
+  }
   return encoder_params;
 }
 
 void H264EncoderImpl::ReportInit() {
   if (has_reported_init_)
     return;
   RTC_HISTOGRAM_ENUMERATION("WebRTC.Video.H264EncoderImpl.Event",
                             kH264EncoderEventInit,
                             kH264EncoderEventMax);
   has_reported_init_ = true;
@@ skipping 15 matching lines @@
 
 int32_t H264EncoderImpl::SetPeriodicKeyFrames(bool enable) {
   return WEBRTC_VIDEO_CODEC_OK;
 }
 
 void H264EncoderImpl::OnDroppedFrame() {
   quality_scaler_.ReportDroppedFrame();
 }
 
 }  // namespace webrtc