admin/webrtc_m130
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Remove usage of AVCodecContext::reordered_opaque

Browse Source 
FFmpeg has removed this field and usage of it in chromium must be
removed before the ffmpeg dependency is updated. The chromium media
change can be found here:
https://chromium-review.googlesource.com/c/chromium/src/+/5384308

The usage of the field in webrtc seems only to be for sanity checking,
so it should be just safe to remove entirely, since webrtc does not
expect re-ordering at all.

Bug: chromium:330573128
Change-Id: I9c5854ec82c3ad2d55374ea4eaa0c571437f8267
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/343840
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Ted (Chromium) Meyer <tmathmeyer@chromium.org>
Cr-Commit-Position: refs/heads/main@{#41935}
This commit is contained in:
Ted Meyer 2024-03-20 15:32:19 -07:00 committed by WebRTC LUCI CQ
parent 8aaf85a687
commit e7d1004709
1 changed files with 0 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
modules/video_coding/codecs/h264/h264_decoder_impl.cc
View File

@ -233,7 +233,6 @@ int H264DecoderImpl::AVGetBuffer2(AVCodecContext* context,
int total_size = y_size + 2 * uv_size;
av_frame->format = context->pix_fmt;
av_frame->reordered_opaque = context->reordered_opaque;
// Create a VideoFrame object, to keep a reference to the buffer.
// TODO(nisse): The VideoFrame's timestamp and rotation info is not used.
@ -381,8 +380,6 @@ int32_t H264DecoderImpl::Decode(const EncodedImage& input_image,
return WEBRTC_VIDEO_CODEC_ERROR;
}
packet->size = static_cast<int>(input_image.size());
int64_t frame_timestamp_us = input_image.ntp_time_ms_ * 1000; // ms -> μs
av_context_->reordered_opaque = frame_timestamp_us;
int result = avcodec_send_packet(av_context_.get(), packet.get());
@ -399,10 +396,6 @@ int32_t H264DecoderImpl::Decode(const EncodedImage& input_image,
return WEBRTC_VIDEO_CODEC_ERROR;
}
// We don't expect reordering. Decoded frame timestamp should match
// the input one.
RTC_DCHECK_EQ(av_frame_->reordered_opaque, frame_timestamp_us);
// TODO(sakal): Maybe it is possible to get QP directly from FFmpeg.
h264_bitstream_parser_.ParseBitstream(input_image);
absl::optional<int> qp = h264_bitstream_parser_.GetLastSliceQp();