On Wed, 5 Nov 2014, Anton Khirnov wrote:
Quoting Martin Storsjö (2014-11-04 15:38:10)
From: Michael Niedermayer <[email protected]>
---
libavformat/mux.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/libavformat/mux.c b/libavformat/mux.c
index 94b4391..f4c6e40 100644
--- a/libavformat/mux.c
+++ b/libavformat/mux.c
@@ -342,6 +342,14 @@ static int write_packet(AVFormatContext *s, AVPacket *pkt)
pkt->dts += offset;
if (pkt->pts != AV_NOPTS_VALUE)
pkt->pts += offset;
+
+ if (pkt->dts != AV_NOPTS_VALUE && pkt->dts < 0) {
+ av_log(s, AV_LOG_WARNING,
+ "Packets poorly interleaved, failed to avoid negative "
+ "timestamp %"PRId64" in stream %d.\n"
+ "Try -max_interleave_delta 0 as a possible workaround.\n",
+ pkt->dts, pkt->stream_index);
+ }
}
ret = s->oformat->write_packet(s, pkt);
--
1.9.3 (Apple Git-50)
Won't the caller already get a warning/error about non-monotonous DTS
in this case?
Not necessarily; you only get those within one individual stream, but if
you write e.g. all packets in one stream first, it will probably trigger
the max_interleave_delta so some packets from the first stream are written
before any of the packets from the other stream even is seen.
// Martin
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
