On Wed, Feb 02, 2022 at 03:45:10PM -0500, Patrick Keroulas wrote: > In previous state, a new frame was allocated on each timestamp step, > i.e. each frame/field transition. However, for interlace, a new frame > should be allocated on 1st field, completed with the 2nd and finally > freed. > > This commit fixes the frame allocation and the detection of missing RTP > markers. > > Signed-off-by: Patrick Keroulas <patrick.kerou...@radio-canada.ca> > --- > libavformat/rtpdec_rfc4175.c | 15 ++++++++------- > 1 file changed, 8 insertions(+), 7 deletions(-) > > diff --git a/libavformat/rtpdec_rfc4175.c b/libavformat/rtpdec_rfc4175.c > index 8e73c07838..83abe499f8 100644 > --- a/libavformat/rtpdec_rfc4175.c > +++ b/libavformat/rtpdec_rfc4175.c > @@ -234,20 +234,21 @@ static int rfc4175_handle_packet(AVFormatContext *ctx, > PayloadContext *data, > uint8_t *dest; > > if (*timestamp != data->timestamp) { > - if (data->frame) { > + if (data->frame && (!data->interlaced || data->field)) { > /* > - * if we're here, it means that two RTP packets didn't have the > - * same timestamp, which is a sign that they were packets from > two > - * different frames, but we didn't get the flag RTP_FLAG_MARKER > on > - * the first one of these frames (last packet of a frame). > - * Finalize the previous frame anyway by filling the AVPacket. > + * if we're here, it means that we missed the cue to return > + * the previous AVPacket, that cue being the RTP_FLAG_MARKER > + * in the last packet of either the previous frame (progressive) > + * or the previous second field (interlace). Let's finalize the > + * previous frame (or pair of fields) anyway by filling the > AVPacket. > */ > av_log(ctx, AV_LOG_ERROR, "Missed previous RTP Marker\n"); > missed_last_packet = 1; > rfc4175_finalize_packet(data, pkt, st->index); > } > > - data->frame = av_malloc(data->frame_size); > + if (!data->frame) > + data->frame = av_malloc(data->frame_size); > > data->timestamp = *timestamp;
LGTM > > -- > 2.25.1 > -- Thanks, Limin Wang _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".