Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Anton Khirnov <anton@khirnov.net>
To: ffmpeg-devel@ffmpeg.org
Subject: [FFmpeg-devel] [PATCH 03/49] fftools/ffmpeg: stop using OutputStream.frame_number for streamcopy
Date: Mon,  4 Apr 2022 13:29:51 +0200
Message-ID: <20220404113037.13070-4-anton@khirnov.net> (raw)
In-Reply-To: <20220404113037.13070-1-anton@khirnov.net>

This field is currently used by checks
- skipping packets before the first keyframe
- skipping packets before start time
to test whether any packets have been output already. But since
frame_number is incremented after the bitstream filters are applied
(which may involve delay), this use is incorrect. The keyframe check
works around this by adding an extra flag, the start-time check does
not.

Simplify both checks by replacing the seen_kf flag with a flag tracking
whether any packets have been output by do_streamcopy().
---
 fftools/ffmpeg.c | 10 +++++-----
 fftools/ffmpeg.h |  2 +-
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/fftools/ffmpeg.c b/fftools/ffmpeg.c
index 13be32f0cf..29b01f9d93 100644
--- a/fftools/ffmpeg.c
+++ b/fftools/ffmpeg.c
@@ -894,8 +894,6 @@ static void output_packet(OutputFile *of, AVPacket *pkt,
 
     /* apply the output bitstream filters */
     if (ost->bsf_ctx) {
-        if (pkt->flags & AV_PKT_FLAG_KEY)
-            ost->seen_kf = 1;
         ret = av_bsf_send_packet(ost->bsf_ctx, eof ? NULL : pkt);
         if (ret < 0)
             goto finish;
@@ -2043,11 +2041,11 @@ static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *p
         return;
     }
 
-    if ((!ost->frame_number && !(pkt->flags & AV_PKT_FLAG_KEY)) &&
-        !ost->copy_initial_nonkeyframes && !ost->seen_kf)
+    if (!ost->streamcopy_started && !(pkt->flags & AV_PKT_FLAG_KEY) &&
+        !ost->copy_initial_nonkeyframes)
         return;
 
-    if (!ost->frame_number && !ost->copy_prior_start) {
+    if (!ost->streamcopy_started && !ost->copy_prior_start) {
         int64_t comp_start = start_time;
         if (copy_ts && f->start_time != AV_NOPTS_VALUE)
             comp_start = FFMAX(start_time, f->start_time + f->ts_offset);
@@ -2101,6 +2099,8 @@ static void do_streamcopy(InputStream *ist, OutputStream *ost, const AVPacket *p
     ost->sync_opts += opkt->duration;
 
     output_packet(of, opkt, ost, 0);
+
+    ost->streamcopy_started = 1;
 }
 
 int guess_input_channel_layout(InputStream *ist)
diff --git a/fftools/ffmpeg.h b/fftools/ffmpeg.h
index 1e14bf9fa9..04369df139 100644
--- a/fftools/ffmpeg.h
+++ b/fftools/ffmpeg.h
@@ -536,7 +536,7 @@ typedef struct OutputStream {
     int inputs_done;
 
     const char *attachment_filename;
-    int seen_kf;
+    int streamcopy_started;
     int copy_initial_nonkeyframes;
     int copy_prior_start;
     char *disposition;
-- 
2.34.1

_______________________________________________
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".

  parent reply	other threads:[~2022-04-04 11:33 UTC|newest]

Thread overview: 81+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-04 11:29 [FFmpeg-devel] [RFC] Switching ffmpeg.c to a threaded architecture Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 01/49] fftools/ffmpeg: drop an obsolete hack Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 02/49] fftools/ffmpeg: move a comment to a more appropriate place Anton Khirnov
2022-04-06 11:20   ` James Almer
2022-04-04 11:29 ` Anton Khirnov [this message]
2022-04-06 11:20   ` [FFmpeg-devel] [PATCH 03/49] fftools/ffmpeg: stop using OutputStream.frame_number for streamcopy James Almer
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 04/49] fftools/ffmpeg: pass the muxer context explicitly to some functions Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 05/49] fftools/ffmpeg: store the output file index in OutputFile Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 06/49] fftools/ffmpeg: move some muxing-related code into a separate file Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 07/49] fftools/ffmpeg: move writing the trailer to ffmpeg_mux.c Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 08/49] fftools/ffmpeg: move freeing the output file " Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 09/49] fftools/ffmpeg: store output format separately from the muxer context Anton Khirnov
2022-04-06 12:00   ` James Almer
2022-04-13 10:18     ` Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 10/49] fftools/ffmpeg_mux: add private " Anton Khirnov
2022-04-04 11:29 ` [FFmpeg-devel] [PATCH 11/49] fftools/ffmpeg: add a helper function to access output file size Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 12/49] fftools/ffmpeg: fix the type of limit_filesize Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 13/49] fftools/ffmpeg: refactor limiting output file size with -fs Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 14/49] fftools/ffmpeg: set want_sdp when initializing the muxer Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 15/49] fftools/ffmpeg: write the header for stream-less outputs " Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 16/49] fftools/ffmpeg: move closing the file into of_write_trailer() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 17/49] fftools/ffmpeg: refactor the code checking for bitexact output Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 18/49] fftools/ffmpeg: access output file chapters through a wrapper Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 19/49] fftools/ffmpeg: do not log to the muxer context Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 20/49] fftools/ffmpeg: move the mux queue into muxer private data Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 21/49] fftools/ffmpeg_mux: split queuing packets into a separate function Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 22/49] fftools/ffmpeg_mux: split of_write_packet() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 23/49] fftools/ffmpeg: move output file opts into private context Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 24/49] fftools/ffmpeg: move processing video stats to ffmpeg_mux Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 25/49] fftools/ffmpeg_mux: drop a useless check and reduce indentation Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 26/49] fftools/ffmpeg_mux: stop using AVStream.nb_frames in do_video_stats() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 27/49] fftools/ffmpeg_mux: stop using av_stream_get_end_pts() " Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 28/49] fftools/ffmpeg_mux: merge variable declaration and initialization Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 29/49] fftools/ffmpeg_mux: move processing AV_PKT_DATA_QUALITY_STATS to do_video_stats() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 30/49] fftools/ffmpeg: share the code encoding a single frame between video and audio Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 31/49] fftools/ffmpeg: reuse the encoding code for flushing encoders Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 32/49] fftools/ffmpeg: reindent after previous commit Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 33/49] fftools/ffmpeg: use refcounted packets for encoded subtitles Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 34/49] fftools/ffmpeg: rework -shortest implementation Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 35/49] fftools/ffmpeg: use the sync queues to handle -frames Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 36/49] fftools/ffmpeg: stop using OutputStream.frame_number in print_report() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 37/49] fftools/ffmpeg: only set OutputStream.frame_number for video encoding Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 38/49] fftools/ffmpeg: make the muxer AVFormatContext private to ffmpeg_mux.c Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 39/49] fftools/ffmpeg_mux: return errors from of_submit_packet() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 40/49] fftools/ffmpeg_mux: return errors from submit_packet() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 41/49] fftools/ffmpeg_mux: return errors from write_packet() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 42/49] fftools/ffmpeg_mux: simplify submit_packet() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 43/49] fftools/ffmpeg_mux: return errors from update_video_stats() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 44/49] fftools/ffmpeg_mux: do not call exit_program() in print_sdp() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 45/49] fftools/ffmpeg: stop using av_stream_get_end_pts() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 46/49] fftools/ffmpeg: do not write the output file header from init_output_stream() Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 47/49] fftools/ffmpeg: depend on threads Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 48/49] fftools: add a multistream thread-safe queue Anton Khirnov
2022-04-04 11:30 ` [FFmpeg-devel] [PATCH 49/49] fftools/ffmpeg: move each muxer to a separate thread Anton Khirnov
2022-04-05  9:00 ` [FFmpeg-devel] [RFC] Switching ffmpeg.c to a threaded architecture Anton Khirnov
2022-04-05 19:15 ` Michael Niedermayer
2022-04-05 19:46   ` Anton Khirnov
2022-04-05 21:05     ` Soft Works
2022-04-05 21:18       ` Paul B Mahol
2022-04-05 21:19         ` Soft Works
2022-04-06 11:17           ` Paul B Mahol
2022-04-06 15:46             ` Soft Works
2022-04-06 15:54               ` Matt Zagrabelny
2022-04-06  8:41     ` Anton Khirnov
2022-04-06 16:29       ` Soft Works
2022-04-06 17:38         ` Paul B Mahol
2022-04-06 17:56           ` Kieran Kunhya
2022-04-06 18:53             ` Soft Works
2022-04-07  8:32       ` Anton Khirnov
2022-04-08 15:27         ` Soft Works
2022-04-11  8:28         ` Anton Khirnov
2022-04-11 20:09           ` Soft Works
2022-04-11 20:52             ` Paul B Mahol
2022-04-11 20:58               ` Soft Works
2022-04-12  9:29                 ` Paul B Mahol
2022-04-12 22:43                   ` Soft Works
2022-04-13  9:42                     ` Nicolas George
2022-04-13 22:14                       ` Soft Works
2022-04-14 10:02                     ` Paul B Mahol
2022-04-14 21:37                       ` Soft Works
2022-04-06 10:02 ` Anton Khirnov

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=20220404113037.13070-4-anton@khirnov.net \
    --to=anton@khirnov.net \
    --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