[FFmpeg-devel] [PATCH 2/3] lavf/utils: add flag to discard timestamps on corrupted frames

[Michael Niedermayer]

On Fri, Dec 08, 2017 at 10:34:39PM -0600, Rodger Combs wrote:
> 
> 
> > On Dec 8, 2017, at 11:06, Michael Niedermayer <michael at niedermayer.cc> wrote:
> > 
> > On Thu, Dec 07, 2017 at 10:23:15PM -0600, Rodger Combs wrote:
> >> ---
> >> libavformat/avformat.h      | 1 +
> >> libavformat/options_table.h | 1 +
> >> libavformat/utils.c         | 8 ++++++++
> >> 3 files changed, 10 insertions(+)
> >> 
> >> diff --git a/libavformat/avformat.h b/libavformat/avformat.h
> >> index 4f2798a871..d10d583dff 100644
> >> --- a/libavformat/avformat.h
> >> +++ b/libavformat/avformat.h
> >> @@ -1450,6 +1450,7 @@ typedef struct AVFormatContext {
> >> #define AVFMT_FLAG_FAST_SEEK   0x80000 ///< Enable fast, but inaccurate seeks for some formats
> >> #define AVFMT_FLAG_SHORTEST   0x100000 ///< Stop muxing when the shortest stream stops.
> >> #define AVFMT_FLAG_AUTO_BSF   0x200000 ///< Add bitstream filters as requested by the muxer
> >> +#define AVFMT_FLAG_DISCARD_CORRUPT_TS 0x400000 ///< Discard timestamps of frames marked corrupt
> >> 
> >>     /**
> >>      * Maximum size of the data read from input for determining
> >> diff --git a/libavformat/options_table.h b/libavformat/options_table.h
> >> index b8fa47c6fd..515574d3e0 100644
> >> --- a/libavformat/options_table.h
> >> +++ b/libavformat/options_table.h
> >> @@ -58,6 +58,7 @@ static const AVOption avformat_options[] = {
> >> {"bitexact", "do not write random/volatile data", 0, AV_OPT_TYPE_CONST, { .i64 = AVFMT_FLAG_BITEXACT }, 0, 0, E, "fflags" },
> >> {"shortest", "stop muxing with the shortest stream", 0, AV_OPT_TYPE_CONST, { .i64 = AVFMT_FLAG_SHORTEST }, 0, 0, E, "fflags" },
> >> {"autobsf", "add needed bsfs automatically", 0, AV_OPT_TYPE_CONST, { .i64 = AVFMT_FLAG_AUTO_BSF }, 0, 0, E, "fflags" },
> >> +{"discardcorruptts", "discard timestamps on corrupted frames", 0, AV_OPT_TYPE_CONST, { .i64 = AVFMT_FLAG_DISCARD_CORRUPT_TS }, 0, 0, E, "fflags" },
> >> {"analyzeduration", "specify how many microseconds are analyzed to probe the input", OFFSET(max_analyze_duration), AV_OPT_TYPE_INT64, {.i64 = 0 }, 0, INT64_MAX, D},
> >> {"cryptokey", "decryption key", OFFSET(key), AV_OPT_TYPE_BINARY, {.dbl = 0}, 0, 0, D},
> >> {"indexmem", "max memory used for timestamp index (per stream)", OFFSET(max_index_size), AV_OPT_TYPE_INT, {.i64 = 1<<20 }, 0, INT_MAX, D},
> >> diff --git a/libavformat/utils.c b/libavformat/utils.c
> >> index 84e49208b8..98af941e9f 100644
> >> --- a/libavformat/utils.c
> >> +++ b/libavformat/utils.c
> > 
> >> @@ -873,6 +873,14 @@ int ff_read_packet(AVFormatContext *s, AVPacket *pkt)
> >>                 st->cur_dts = wrap_timestamp(st, st->cur_dts);
> >>         }
> >> 
> >> +        if ((s->flags & AVFMT_FLAG_DISCARD_CORRUPT_TS) &&
> >> +            (pkt->flags & AV_PKT_FLAG_CORRUPT)) {
> >> +            pkt->pts = pkt->dts = AV_NOPTS_VALUE;
> >> +            av_log(s, AV_LOG_WARNING,
> >> +                   "Discarded timestamp on corrupted packet (stream = %d)\n",
> >> +                   pkt->stream_index);
> >> +        }
> > 
> > how many of the cases that set AV_PKT_FLAG_CORRUPT can even be due to
> > the timestamp ?
> 
> Pretty much just the new TEI check, or potentially other CRC checks. I don't _think_ the continuity-counter check failing could indicate an incorrect timestamp. I suppose we could add a second flag just for corruption that could affect timestamps?

I think thats a great idea, maybe if we already change it we should
also split the truncated case out as it is very common.

AV_PKT_FLAG_DATA_CORRUPT                // the packet data may be corrupted
AV_PKT_FLAG_DATA_TRUNCATED              // the packet size may be corrupted, data available is undamaged unless another flag indicates otherwise. Any headers or sub-packets which are complete are non corrupted-
AV_PKT_FLAG_TS_CORRUPT                  // the packets timestamps and or duration may be corrupted

#define AV_PKT_FLAG_CORRUPT (AV_PKT_FLAG_DATA_CORRUPT | AV_PKT_FLAG_DATA_TRUNCATED | AV_PKT_FLAG_TS_CORRUPT)

or some variation of this

Thanks

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

There will always be a question for which you do not know the correct answer.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20171209/0e06adcc/attachment.sig>