From: Andreas Rheinhardt <andreas.rheinhardt@outlook.com> To: ffmpeg-devel@ffmpeg.org Cc: Andreas Rheinhardt <andreas.rheinhardt@outlook.com> Subject: [FFmpeg-devel] [PATCH v2 5/6] avformat/mux: Peek into the muxing queue for avoid_negative_ts Date: Wed, 19 Jan 2022 22:29:39 +0100 Message-ID: <AM7PR03MB6660641EF5BDC44969B6260C8F599@AM7PR03MB6660.eurprd03.prod.outlook.com> (raw) In-Reply-To: <20220119212940.1071477-1-andreas.rheinhardt@outlook.com> Peeking into the muxing queue can improve the estimate of the lowest timestamp needed for avoid_negative_ts in case the lowest timestamp is in a packet other than the first packet to be muxed. This fixes tickets #4536 and #5784 as well as the output from the matroska-avoid-negative-ts FATE-test. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com> --- libavformat/avformat.h | 2 +- libavformat/mux.c | 21 +++++++++++++++++++-- tests/fate/matroska.mak | 2 -- tests/ref/fate/matroska-avoid-negative-ts | 2 +- 4 files changed, 21 insertions(+), 6 deletions(-) diff --git a/libavformat/avformat.h b/libavformat/avformat.h index cd253fb28e..b4b8075ae6 100644 --- a/libavformat/avformat.h +++ b/libavformat/avformat.h @@ -1527,7 +1527,7 @@ typedef struct AVFormatContext { /** * Avoid negative timestamps during muxing. * Any value of the AVFMT_AVOID_NEG_TS_* constants. - * Note, this only works when using av_interleaved_write_frame. (interleave_packet_per_dts is in use) + * Note, this works better when using av_interleaved_write_frame(). * - muxing: Set by user * - demuxing: unused */ diff --git a/libavformat/mux.c b/libavformat/mux.c index 0810b674a7..53eb56f0af 100644 --- a/libavformat/mux.c +++ b/libavformat/mux.c @@ -655,16 +655,33 @@ static void handle_avoid_negative_ts(FFFormatContext *si, FFStream *sti, if (si->avoid_negative_ts_status == AVOID_NEGATIVE_TS_UNKNOWN) { int use_pts = si->avoid_negative_ts_use_pts; int64_t ts = use_pts ? pkt->pts : pkt->dts; + AVRational tb = sti->pub.time_base; if (ts == AV_NOPTS_VALUE) return; + + /* Peek into the muxing queue to improve our estimate + * of the lowest timestamp if av_interleaved_write_frame() is used. */ + for (const PacketListEntry *pktl = si->packet_buffer.head; + pktl; pktl = pktl->next) { + AVRational cmp_tb = s->streams[pktl->pkt.stream_index]->time_base; + int64_t cmp_ts = use_pts ? pktl->pkt.pts : pktl->pkt.dts; + if (cmp_ts == AV_NOPTS_VALUE) + continue; + if (s->output_ts_offset) + cmp_ts += av_rescale_q(s->output_ts_offset, AV_TIME_BASE_Q, cmp_tb); + if (av_compare_ts(cmp_ts, cmp_tb, ts, tb) < 0) { + ts = cmp_ts; + tb = cmp_tb; + } + } + if (ts < 0 || ts > 0 && s->avoid_negative_ts == AVFMT_AVOID_NEG_TS_MAKE_ZERO) { for (unsigned i = 0; i < s->nb_streams; i++) { AVStream *const st2 = s->streams[i]; FFStream *const sti2 = ffstream(st2); - sti2->mux_ts_offset = av_rescale_q_rnd(-ts, - sti->pub.time_base, + sti2->mux_ts_offset = av_rescale_q_rnd(-ts, tb, st2->time_base, AV_ROUND_UP); } diff --git a/tests/fate/matroska.mak b/tests/fate/matroska.mak index da1fdbd5ea..784b55f6e0 100644 --- a/tests/fate/matroska.mak +++ b/tests/fate/matroska.mak @@ -94,8 +94,6 @@ fate-matroska-dovi-write-config7: CMD = transcode mov $(TARGET_SAMPLES)/mov/dovi # the first packet (with the overall lowest dts) is a video packet, # whereas an audio packet to be muxed later has the overall lowest pts # which happens to be negative and therefore needs to be shifted. -# This is currently buggy (the timestamps are not shifted properly: -# the first audio packet has negative pts). # (-ss 1.09 ensures that a video frame has the lowest dts of all packets; # yet there is an audio packet with the overall lowest pts. output_ts_offset # makes the pts of the audio packet, but not the leading video packet negative diff --git a/tests/ref/fate/matroska-avoid-negative-ts b/tests/ref/fate/matroska-avoid-negative-ts index 1b9b2f2786..02790a3985 100644 --- a/tests/ref/fate/matroska-avoid-negative-ts +++ b/tests/ref/fate/matroska-avoid-negative-ts @@ -1,4 +1,4 @@ -90cf5a330659140d47ec11208f525908 *tests/data/fate/matroska-avoid-negative-ts.matroska +804842437b2be0a1604ce33c6b08c800 *tests/data/fate/matroska-avoid-negative-ts.matroska 973070 tests/data/fate/matroska-avoid-negative-ts.matroska #extradata 0: 22, 0x2885037c #tb 0: 1/1000 -- 2.32.0 _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
next prev parent reply other threads:[~2022-01-19 21:30 UTC|newest] Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-01-19 21:19 [FFmpeg-devel] [PATCH v2 1/6] avformat/mux: Remove assert based on faulty assumptions Andreas Rheinhardt 2022-01-19 21:29 ` [FFmpeg-devel] [PATCH v2 2/6] fate/matroska: Add test for avoiding negative timestamps Andreas Rheinhardt [not found] ` <20220119212940.1071477-1-andreas.rheinhardt@outlook.com> 2022-01-19 21:29 ` [FFmpeg-devel] [PATCH v2 3/6] avformat/avformat: Add AVFMT_AVOID_NEG_TS_DISABLED Andreas Rheinhardt 2022-01-19 21:29 ` [FFmpeg-devel] [PATCH v2 4/6] avformat/mux: Preserve sync even if later packet has negative ts Andreas Rheinhardt 2022-01-19 21:29 ` Andreas Rheinhardt [this message] 2022-01-19 21:29 ` [FFmpeg-devel] [PATCH v2 6/6] avformat/hls: Remove redundant cast Andreas Rheinhardt 2022-01-21 12:14 ` [FFmpeg-devel] [PATCH v2 1/6] avformat/mux: Remove assert based on faulty assumptions Andreas Rheinhardt
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=AM7PR03MB6660641EF5BDC44969B6260C8F599@AM7PR03MB6660.eurprd03.prod.outlook.com \ --to=andreas.rheinhardt@outlook.com \ --cc=ffmpeg-devel@ffmpeg.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: link
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel This inbox may be cloned and mirrored by anyone: git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git # If you have public-inbox 1.1+ installed, you may # initialize and index your mirror using the following commands: public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \ ffmpegdev@gitmailbox.com public-inbox-index ffmpegdev Example config snippet for mirrors. AGPL code for this site: git clone https://public-inbox.org/public-inbox.git