Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Nicolas George <george@nsup.org>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH 08/24] fftools/ffmpeg_filter: remove an unnecessary sub2video_push_ref() call
Date: Sat, 4 Nov 2023 15:19:00 +0100
Message-ID: <ZUZS1Am7WRKJFTW1@phare.normalesup.org> (raw)
In-Reply-To: <20231104092125.10213-9-anton@khirnov.net>


[-- Attachment #1.1: Type: text/plain, Size: 2219 bytes --]

Anton Khirnov (12023-11-04):
> It only seems to produce duplicate frames.

Yes it does, that is the point of it. Without the duplicated frame, a
filter with synchronized inputs (like overlay) will accumulate video
frames while waiting for the next sub2video frame.

It is actually super easy to test: any file with both video and bitmap
subtitles will cause a huge memory consumption if there is a long enough
time without subtitles.

Since I am feeling super helpful, here is how I just tested:

./ffmpeg_g -lavfi testsrc2=s=720x480:d=150 -preset ultrafast -y /tmp/dummy.mkv
mkvmerge -o /tmp/dummy_with_sub.mkv /tmp/dummy.mkv $ffmpeg_fate/sub/vobsub.idx
limit addressspace 2G
./ffmpeg_g -xerror -i /tmp/dummy_with_sub.mkv -preset ultrafast -lavfi '[0:v][0:s]overlay' -y /tmp/dummy_with_hardsub.mkv
mplayer /tmp/dummy_with_hardsub.mkv -ss 2:05

The limit addressspace 2G is there so that the huge memory consumption
will hit something. Another way is to log the number of queued frames:

--- a/libavfilter/framequeue.c
+++ b/libavfilter/framequeue.c
@@ -90,2 +90,3 @@ int ff_framequeue_add(FFFrameQueue *fq, AVFrame *frame)
     fq->queued++;
+    av_log(0, 16, "queued = %zd\n", fq->queued);
     fq->total_frames_head++;

ffmpeg version N-112710-g86e0dea620 Copyright (c) 2000-2023 the FFmpeg developers
  built with gcc 13 (Debian 13.2.0-5)
  configuration: --enable-shared --disable-static --enable-gpl --enable-libx264 --enable-libopus --enable-libass --enable-libfreetype --enable-opengl --assert-level=2
  libavutil      58. 31.100 / 58. 31.100
  libavcodec     60. 32.102 / 60. 32.102
  libavformat    60. 17.100 / 60. 17.100
  libavdevice    60.  4.100 / 60.  4.100
  libavfilter     9. 13.100 /  9. 13.100
  libswscale      7.  6.100 /  7.  6.100
  libswresample   4. 13.100 /  4. 13.100
  libpostproc    57.  4.100 / 57.  4.100
Input #0, matroska,webm, from '/tmp/dummy_with_sub.mkv':
[…]
queued = 2210
queued = 2211
queued = 2212
queued = 2213
[h264 @ 0x562874ac4f40] get_buffer() failed
[h264 @ 0x562874ac4f40] thread_get_buffer() failed
[h264 @ 0x562874ac4f40] decode_slice_header error
[h264 @ 0x562874ac4f40] no frame!
[…]

-- 
  Nicolas George

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[-- Attachment #2: Type: text/plain, Size: 251 bytes --]

_______________________________________________
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".

  reply	other threads:[~2023-11-04 14:19 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-04  7:56 [FFmpeg-devel] [PATCH] ffmpeg CLI multithreading Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 01/24] lavf/mux: do not apply max_interleave_delta to subtitles Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 02/24] lavfi/af_amix: make sure the output does not depend on input ordering Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 03/24] lavc/8bps: fix exporting palette after 63767b79a570404628b2521b83104108b7b6884c Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 04/24] fftools/ffmpeg: move a few inline function into a new header Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 05/24] fftools/thread_queue: do not return elements for receive-finished streams Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 06/24] fftools/thread_queue: count receive-finished streams as finished Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 07/24] fftools/ffmpeg: rework keeping track of file duration for -stream_loop Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 08/24] fftools/ffmpeg_filter: remove an unnecessary sub2video_push_ref() call Anton Khirnov
2023-11-04 14:19   ` Nicolas George [this message]
2023-11-09 10:42     ` Anton Khirnov
2023-11-09 10:47       ` Nicolas George
2023-11-09 21:29         ` Anton Khirnov
2023-11-17  9:44           ` Nicolas George
2023-11-17 11:52             ` Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 09/24] fftools/ffmpeg_filter: track input/output index in {Input, Output}FilterPriv Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 10/24] fftools/ffmpeg_filter: move filtering to a separate thread Anton Khirnov
2023-11-04 19:54   ` Michael Niedermayer
2023-11-09 11:45     ` [FFmpeg-devel] [PATCH v2 " Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 11/24] fftools/ffmpeg_filter: buffer sub2video heartbeat frames like other frames Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 12/24] fftools/ffmpeg_filter: reindent Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 13/24] fftools/ffmpeg_mux: add muxing thread private data Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 14/24] fftools/ffmpeg_mux: move bitstream filtering to the muxer thread Anton Khirnov
2023-11-04 13:39   ` James Almer
2023-11-09 11:41     ` Anton Khirnov
2023-11-09 11:47       ` James Almer
2023-11-09 12:00         ` Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 15/24] fftools/ffmpeg_demux: switch from AVThreadMessageQueue to ThreadQueue Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 16/24] fftools/ffmpeg: disable -fix_sub_duration_heartbeat Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 17/24] fftools/ffmpeg_enc: move encoding to a separate thread Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 18/24] fftools/ffmpeg: add thread-aware transcode scheduling infrastructure Anton Khirnov
2023-11-04 13:53   ` James Almer
2023-11-09 11:06     ` Anton Khirnov
2023-11-04 18:44   ` Michael Niedermayer
2023-11-09 11:36     ` [FFmpeg-devel] [PATCH v2 " Anton Khirnov
2023-11-11 15:21       ` [FFmpeg-devel] [PATCH v3 " Anton Khirnov
2023-11-09 11:39     ` [FFmpeg-devel] [PATCH " Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 19/24] fftools/ffmpeg_demux: convert to the scheduler Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 20/24] fftools/ffmpeg_dec: " Anton Khirnov
2023-11-04 18:30   ` Michael Niedermayer
2023-11-11 15:24     ` Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 21/24] fftools/ffmpeg_filter: " Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 22/24] fftools/ffmpeg_enc: " Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 23/24] fftools/ffmpeg_mux: " Anton Khirnov
2023-11-04  7:56 ` [FFmpeg-devel] [PATCH 24/24] ffmpeg: switch to scheduler Anton Khirnov
2023-11-09 11:46 ` [FFmpeg-devel] [PATCH] ffmpeg CLI multithreading Anton Khirnov
2023-11-09 12:46   ` Paul B Mahol
2023-11-11 20:10 ` Michael Niedermayer
2023-11-17 10:46   ` Anton Khirnov
2023-11-13 12:34 ` Jan Ekström
2023-11-13 12:38   ` James Almer

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=ZUZS1Am7WRKJFTW1@phare.normalesup.org \
    --to=george@nsup.org \
    --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