[FFmpeg-devel] [RFC] Are those workarounds really needed?

[Måns Rullgård]

Luca Barbato <lu_zero at gentoo.org> writes:

> Today I got a strange bug report, somebody was using ffmpeg since old
> age and it was working quite fine for his purposes, today he updated and
> got a severe desync.
>
> I got a sample and it broke ffplay/ffmpeg severely while it plays
> correctly on vlc.
>
> I dug a bit and I found the culprit (patch attached).
>
> Now, is that really needed?
>
> diff --git a/libavformat/utils.c b/libavformat/utils.c
> index 870a287..25fba4c 100644
> --- a/libavformat/utils.c
> +++ b/libavformat/utils.c
> @@ -950,10 +950,10 @@ static void compute_pkt_fields(AVFormatContext *s, AVStream *st,
>      // some mpeg2 in mpeg-ps lack dts (issue171 / input_file.mpg)
>      // we take the conservative approach and discard both
>      // Note, if this is misbehaving for a H.264 file then possibly presentation_delayed is not set correctly.
> -    if(delay==1 && pkt->dts == pkt->pts && pkt->dts != AV_NOPTS_VALUE && presentation_delayed){
> +/*    if(delay==1 && pkt->dts == pkt->pts && pkt->dts != AV_NOPTS_VALUE && presentation_delayed){
>          av_log(s, AV_LOG_DEBUG, "invalid dts/pts combination\n");
>          pkt->dts= pkt->pts= AV_NOPTS_VALUE;
> -    }
> +    }*/
>
>      if (pkt->duration == 0) {
>          compute_frame_duration(&num, &den, st, pc, pkt);

Did you figure out why this causes problems with your sample?  Can you
share the sample?

-- 
M?ns Rullg?rd
mans at mansr.com