July 2024 - ffmpeg-devel - lists.ffmpeg.org

[PATCH] doc/developer: add examples to clarify code style
by Marvin Scholz 13 Sep '24

13 Sep '24

Given the frequency that new developers, myself included, get the code style wrong, it is useful to add some examples to clarify how things should be done. --- doc/developer.texi | 57 +++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 56 insertions(+), 1 deletion(-) diff --git a/doc/developer.texi b/doc/developer.texi index 63835dfa06..d7bf3f9cb8 100644 --- a/doc/developer.texi +++ b/doc/developer.texi @@ -115,7 +115,7 @@ Objective-C where required for interacting with macOS-specific interfaces. @section Code formatting conventions -There are the following guidelines regarding the indentation in files: +There are the following guidelines regarding the code style in files: @itemize @bullet @item @@ -135,6 +135,61 @@ K&R coding style is used. @end itemize The presentation is one inspired by 'indent -i4 -kr -nut'. +@subsection Examples +Some notable examples to illustrate common code style in FFmpeg: + +@itemize @bullet + +@item +Spaces around @code{if}/@code{do}/@code{while}/@code{for} conditions and assigments: + +@example c +if (condition) + av_foo(); +@end example + +@example c +for (size_t i = 0; i < len; i++) + av_bar(i); +@end example + +@example c +size_t size = 0; +@end example + +However no spaces between the parentheses and condition, unless it helps +readability of complex conditions, so the following should not be done: + +@example c +// Wrong: +if ( cond ) + av_foo(); +@end example + +@item +No unnecessary parentheses, unless it helps readability: + +@example c +flags = s->mb_x ? RIGHT_EDGE : LEFT_EDGE | RIGHT_EDGE; +@end example + +@item +No braces around single-line blocks: + +@example c +if (bits_pixel == 24) + avctx->pix_fmt = AV_PIX_FMT_BGR24; +else if (bits_pixel == 8) + avctx->pix_fmt = AV_PIX_FMT_GRAY8; +else @{ + av_log(avctx, AV_LOG_ERROR, "Invalid pixel format.\n"); + return AVERROR_INVALIDDATA; +@} +@end example + +@end itemize + + @subsection Vim configuration In order to configure Vim to follow FFmpeg formatting conventions, paste the following snippet into your @file{.vimrc}: base-commit: 86e418ffd7bbdc0530e1e4d5bd7534b6e03b5b05 -- 2.39.3 (Apple Git-145)

3 2

[PATCH 1/5] avcodec/pngdec: Check last AVFrame before deref
by Michael Niedermayer 12 Sep '24

12 Sep '24

Fixes: NULL pointer dereference Fixes: 68184/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_APNG_fuzzer-4926478069334016 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavcodec/pngdec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavcodec/pngdec.c b/libavcodec/pngdec.c index f7751223b81..b24bfa248dc 100644 --- a/libavcodec/pngdec.c +++ b/libavcodec/pngdec.c @@ -1218,7 +1218,7 @@ static int decode_fctl_chunk(AVCodecContext *avctx, PNGDecContext *s, return AVERROR_INVALIDDATA; } - if ((sequence_number == 0 || !s->last_picture.f->data[0]) && + if ((sequence_number == 0 || !s->last_picture.f || !s->last_picture.f->data[0]) && dispose_op == APNG_DISPOSE_OP_PREVIOUS) { // No previous frame to revert to for the first frame // Spec says to just treat it as a APNG_DISPOSE_OP_BACKGROUND -- 2.43.2

3 18

[PATCH v2 1/1] libavdevice/decklink: extend available actions on signal loss
by Michael Riedl 09 Sep '24

09 Sep '24

Deprecate the option 'draw_bars' in favor of the new option 'signal_loss_action', which controls the behavior when the input signal is not available (including the behavior previously available through draw_bars). The default behavior remains unchanged to be backwards compatible. The new option is more flexible for extending now and in the future. The new value 'repeat' repeats the last video frame. This is useful for very short dropouts and was not available before. Signed-off-by: Michael Riedl <michael.riedl(a)nativewaves.com> --- doc/indevs.texi | 16 ++++++++++++++++ libavdevice/decklink_common.h | 1 + libavdevice/decklink_common_c.h | 7 +++++++ libavdevice/decklink_dec.cpp | 23 ++++++++++++++++++++++- libavdevice/decklink_dec_c.c | 6 +++++- 5 files changed, 51 insertions(+), 2 deletions(-) diff --git a/doc/indevs.texi b/doc/indevs.texi index 734fc657523..cdf44a66382 100644 --- a/doc/indevs.texi +++ b/doc/indevs.texi @@ -396,6 +396,22 @@ Defaults to @samp{audio}. @item draw_bars If set to @samp{true}, color bars are drawn in the event of a signal loss. Defaults to @samp{true}. +This option is deprecated, please use the @code{signal_loss_action} option. + +@item signal_loss_action +Sets the action to take in the event of a signal loss. Accepts one of the +following values: + +@table @option +@item 1, none +Do nothing on signal loss. This usually results in black frames. +@item 2, bars +Draw color bars on signal loss. Only supported for 8-bit input signals. +@item 3, repeat +Repeat the last video frame on signal loss. +@end table + +Defaults to @samp{bars}. @item queue_size Sets maximum input buffer size in bytes. If the buffering reaches this value, diff --git a/libavdevice/decklink_common.h b/libavdevice/decklink_common.h index c54a635876c..6b32dc2d09c 100644 --- a/libavdevice/decklink_common.h +++ b/libavdevice/decklink_common.h @@ -147,6 +147,7 @@ struct decklink_ctx { DecklinkPtsSource video_pts_source; int draw_bars; BMDPixelFormat raw_format; + DecklinkSignalLossAction signal_loss_action; int frames_preroll; int frames_buffer; diff --git a/libavdevice/decklink_common_c.h b/libavdevice/decklink_common_c.h index 9c55d891494..53d2c583e7e 100644 --- a/libavdevice/decklink_common_c.h +++ b/libavdevice/decklink_common_c.h @@ -37,6 +37,12 @@ typedef enum DecklinkPtsSource { PTS_SRC_NB } DecklinkPtsSource; +typedef enum DecklinkSignalLossAction { + SIGNAL_LOSS_NONE = 1, + SIGNAL_LOSS_REPEAT = 2, + SIGNAL_LOSS_BARS = 3 +} DecklinkSignalLossAction; + struct decklink_cctx { const AVClass *cclass; @@ -68,6 +74,7 @@ struct decklink_cctx { int64_t timestamp_align; int timing_offset; int wait_for_tc; + DecklinkSignalLossAction signal_loss_action; }; #endif /* AVDEVICE_DECKLINK_COMMON_C_H */ diff --git a/libavdevice/decklink_dec.cpp b/libavdevice/decklink_dec.cpp index 671573853ba..1c2d08ea9fc 100644 --- a/libavdevice/decklink_dec.cpp +++ b/libavdevice/decklink_dec.cpp @@ -593,6 +593,7 @@ private: int no_video; int64_t initial_video_pts; int64_t initial_audio_pts; + IDeckLinkVideoInputFrame* last_video_frame; }; decklink_input_callback::decklink_input_callback(AVFormatContext *_avctx) : _refs(1) @@ -602,10 +603,13 @@ decklink_input_callback::decklink_input_callback(AVFormatContext *_avctx) : _ref ctx = (struct decklink_ctx *)cctx->ctx; no_video = 0; initial_audio_pts = initial_video_pts = AV_NOPTS_VALUE; + last_video_frame = nullptr; } decklink_input_callback::~decklink_input_callback() { + if (last_video_frame) + last_video_frame->Release(); } ULONG decklink_input_callback::AddRef(void) @@ -773,7 +777,7 @@ HRESULT decklink_input_callback::VideoInputFrameArrived( ctx->video_st->time_base.den); if (videoFrame->GetFlags() & bmdFrameHasNoInputSource) { - if (ctx->draw_bars && videoFrame->GetPixelFormat() == bmdFormat8BitYUV) { + if (ctx->signal_loss_action == SIGNAL_LOSS_BARS && videoFrame->GetPixelFormat() == bmdFormat8BitYUV) { unsigned bars[8] = { 0xEA80EA80, 0xD292D210, 0xA910A9A5, 0x90229035, 0x6ADD6ACA, 0x51EF515A, 0x286D28EF, 0x10801080 }; @@ -785,6 +789,8 @@ HRESULT decklink_input_callback::VideoInputFrameArrived( for (int x = 0; x < width; x += 2) *p++ = bars[(x * 8) / width]; } + } else if (ctx->signal_loss_action == SIGNAL_LOSS_REPEAT) { + last_video_frame->GetBytes(&frameBytes); } if (!no_video) { @@ -793,6 +799,12 @@ HRESULT decklink_input_callback::VideoInputFrameArrived( } no_video = 1; } else { + if (ctx->signal_loss_action == SIGNAL_LOSS_REPEAT) { + if (last_video_frame) + last_video_frame->Release(); + last_video_frame = videoFrame; + last_video_frame->AddRef(); + } if (no_video) { av_log(avctx, AV_LOG_WARNING, "Frame received (#%lu) - Input returned " "- Frames dropped %u\n", ctx->frameCount, ++ctx->dropped); @@ -1074,6 +1086,15 @@ av_cold int ff_decklink_read_header(AVFormatContext *avctx) ctx->audio_pts_source = cctx->audio_pts_source; ctx->video_pts_source = cctx->video_pts_source; ctx->draw_bars = cctx->draw_bars; + ctx->signal_loss_action = cctx->signal_loss_action; + if (!ctx->draw_bars && ctx->signal_loss_action == SIGNAL_LOSS_BARS) { + ctx->signal_loss_action = SIGNAL_LOSS_NONE; + av_log(avctx, AV_LOG_WARNING, "Setting signal_loss_action to none because draw_bars is false\n"); + } + if (!ctx->draw_bars && ctx->signal_loss_action != SIGNAL_LOSS_NONE) { + av_log(avctx, AV_LOG_ERROR, "options draw_bars and signal_loss_action are mutually exclusive\n"); + return AVERROR(EINVAL); + } ctx->audio_depth = cctx->audio_depth; if (cctx->raw_format > 0 && (unsigned int)cctx->raw_format < FF_ARRAY_ELEMS(decklink_raw_format_map)) ctx->raw_format = decklink_raw_format_map[cctx->raw_format]; diff --git a/libavdevice/decklink_dec_c.c b/libavdevice/decklink_dec_c.c index e211c9d3f4e..b8cdb7bd8e1 100644 --- a/libavdevice/decklink_dec_c.c +++ b/libavdevice/decklink_dec_c.c @@ -95,12 +95,16 @@ static const AVOption options[] = { { "reference", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = PTS_SRC_REFERENCE}, 0, 0, DEC, .unit = "pts_source"}, { "wallclock", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = PTS_SRC_WALLCLOCK}, 0, 0, DEC, .unit = "pts_source"}, { "abs_wallclock", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = PTS_SRC_ABS_WALLCLOCK}, 0, 0, DEC, .unit = "pts_source"}, - { "draw_bars", "draw bars on signal loss" , OFFSET(draw_bars), AV_OPT_TYPE_BOOL, { .i64 = 1}, 0, 1, DEC }, + { "draw_bars", "use option signal_loss_action instead" , OFFSET(draw_bars), AV_OPT_TYPE_BOOL, { .i64 = 1}, 0, 1, DEC | AV_OPT_FLAG_DEPRECATED }, { "queue_size", "input queue buffer size", OFFSET(queue_size), AV_OPT_TYPE_INT64, { .i64 = (1024 * 1024 * 1024)}, 0, INT64_MAX, DEC }, { "audio_depth", "audio bitdepth (16 or 32)", OFFSET(audio_depth), AV_OPT_TYPE_INT, { .i64 = 16}, 16, 32, DEC }, { "decklink_copyts", "copy timestamps, do not remove the initial offset", OFFSET(copyts), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, DEC }, { "timestamp_align", "capture start time alignment (in seconds)", OFFSET(timestamp_align), AV_OPT_TYPE_DURATION, { .i64 = 0 }, 0, INT_MAX, DEC }, { "wait_for_tc", "drop frames till a frame with timecode is received. TC format must be set", OFFSET(wait_for_tc), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, DEC }, + { "signal_loss_action", "action on signal loss", OFFSET(signal_loss_action), AV_OPT_TYPE_INT, { .i64 = SIGNAL_LOSS_BARS }, SIGNAL_LOSS_NONE, SIGNAL_LOSS_BARS, DEC, .unit = "signal_loss_action" }, + { "none", "do not do anything (usually leads to black frames)", 0, AV_OPT_TYPE_CONST, { .i64 = SIGNAL_LOSS_NONE }, 0, 0, DEC, .unit = "signal_loss_action"}, + { "bars", "draw color bars (only supported for 8-bit signals)", 0, AV_OPT_TYPE_CONST, { .i64 = SIGNAL_LOSS_BARS }, 0, 0, DEC, .unit = "signal_loss_action"}, + { "repeat", "repeat the last video frame", 0, AV_OPT_TYPE_CONST, { .i64 = SIGNAL_LOSS_REPEAT }, 0, 0, DEC, .unit = "signal_loss_action"}, { NULL }, }; -- 2.43.0

2 7

[PATCH v3] avfilter: add XPSNR filter
by Helmrich, Christian 07 Sep '24

07 Sep '24

This is a continuation of last year's version of this filter patch, see also https://ffmpeg.org/pipermail/ffmpeg-devel/2023-January/305517.html It includes a fix in one of the stride variables and some cleanup in order to adhere even more to the FFmpeg coding guidelines. Christian Helmrich Fraunhofer HHI

2 8

Add Mediacodec audio decoders support
by Matthieu Bouron 29 Aug '24

29 Aug '24

Hello, This patchset adds Mediacodec audio decoders support. Currently, only AAC, AMR, MP3, FLAC, VORBIS and OPUS are supported. This is mainly useful to avoid shipping Android builds of FFmpeg that are subjects to licensing/patents (due to AAC and AMR). Best regards, Matthieu

8 28

[PATCH 0/7] avformat/dvdvideodec: Bug fixes, seeking support, and menu chapters
by Marth64 28 Aug '24

28 Aug '24

This patch set starts with a critical bug fix to resolve packet drops, then follows on with seeking support and menu chapter marker support. In addition, some cleanup occurs and documentation/logging is improved. Seeking can be tested with mpv, ``` /mpv "av://dvdvideo:$IN" ``` Compare/view on GitHub: https://github.com/FFmpeg/FFmpeg/compare/master...Marth64x:FFmpeg:20240728_… Net addition is 69 lines across the demuxer and docs. Signed-off-by: Marth64 <marth64(a)proxyid.net>

3 11

[PATCH v3 1/2] libavformat/vapoursynth: Update to API version 4, load library at runtime
by Stefan Oltmanns 26 Aug '24

26 Aug '24

Hello, this is revised patch, this is the first part that just updates to the API v4 of VapourSynth. Changes in API v4: -All functions previously in header are now part of the "vssapi" object -Renames of different types and functions -YCoCg is not treated as different format to YUV anymore -Some pointers to arrays are now arrays inside a struct. Best regards Stefan

2 10

[PATCH 01/11] avcodec/vvcdec: thread, ensure the parse stage gets the highest priority
by Nuo Mi 21 Aug '24

21 Aug '24

The parser stage is not parallelizable. We need to schedule it as soon as possible to create later stages, which are more parallelizable clips | before | after | delta --------------------------------------------|--------|-------|------ RitualDance_1920x1080_60_10_420_37_RA.266 | 342.7 | 365.3 | 6.59% NovosobornayaSquare_1920x1080.bin | 321.7 | 400 | 24.34% Tango2_3840x2160_60_10_420_27_LD.266 | 82.3 | 91.7 | 11.42% RitualDance_1920x1080_60_10_420_32_LD.266 | 323.7 | 319.3 | -1.36% Chimera_8bit_1080P_1000_frames.vvc | 364 | 411.3 | 12.99% BQTerrace_1920x1080_60_10_420_22_RA.vvc | 162.7 | 185.7 | 14.14% --- libavcodec/vvc/thread.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/libavcodec/vvc/thread.c b/libavcodec/vvc/thread.c index 8dd9a36fd5..28065d726f 100644 --- a/libavcodec/vvc/thread.c +++ b/libavcodec/vvc/thread.c @@ -385,14 +385,15 @@ static int task_priority_higher(const AVTask *_a, const AVTask *_b) const VVCTask *a = (const VVCTask*)_a; const VVCTask *b = (const VVCTask*)_b; - CHECK(a->fc->decode_order, b->fc->decode_order); //decode order - if (a->stage == VVC_TASK_STAGE_PARSE || b->stage == VVC_TASK_STAGE_PARSE) { + if (a->stage <= VVC_TASK_STAGE_PARSE || b->stage <= VVC_TASK_STAGE_PARSE) { CHECK(a->stage, b->stage); + CHECK(a->fc->decode_order, b->fc->decode_order); //decode order CHECK(a->ry, b->ry); return a->rx < b->rx; } + CHECK(a->fc->decode_order, b->fc->decode_order); //decode order CHECK(a->rx + a->ry + a->stage, b->rx + b->ry + b->stage); //zigzag with type CHECK(a->rx + a->ry, b->rx + b->ry); //zigzag return a->ry < b->ry; -- 2.34.1

2 1

[PATCH 1/5] avutil/timecode: Use a 64bit framenum internally
by Michael Niedermayer 21 Aug '24

21 Aug '24

Fixes: negation of -2147483648 cannot be represented in type 'int'; cast to an unsigned type to negate this value to itself Fixes: 68550/clusterfuzz-testcase-minimized-ffmpeg_dem_MXF_fuzzer-6424065930756096 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavutil/timecode.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/libavutil/timecode.c b/libavutil/timecode.c index bd879bd3cc0..f40a10eb385 100644 --- a/libavutil/timecode.c +++ b/libavutil/timecode.c @@ -100,11 +100,12 @@ uint32_t av_timecode_get_smpte(AVRational rate, int drop, int hh, int mm, int ss return tc; } -char *av_timecode_make_string(const AVTimecode *tc, char *buf, int framenum) +char *av_timecode_make_string(const AVTimecode *tc, char *buf, int framenum_arg) { int fps = tc->fps; int drop = tc->flags & AV_TIMECODE_FLAG_DROPFRAME; int hh, mm, ss, ff, ff_len, neg = 0; + int64_t framenum = framenum_arg; framenum += tc->start; if (drop) -- 2.45.2

2 6

[PATCH] fate/jpeg2000dec: add support for p0_10.j2k
by pal＠sandflow.com 21 Aug '24

21 Aug '24

From: Pierre-Anthony Lemieux <pal(a)palemieux.com> p0_10.j2k is one of the reference codestreams included in Rec. ITU-T T.803 | ISO/IEC 15444-4. --- tests/fate/jpeg2000.mak | 3 +++ tests/ref/fate/jpeg2000dec-p0_10 | 6 ++++++ 2 files changed, 9 insertions(+) create mode 100644 tests/ref/fate/jpeg2000dec-p0_10 diff --git a/tests/fate/jpeg2000.mak b/tests/fate/jpeg2000.mak index 2969d2cf0a..a99b0c4e0c 100644 --- a/tests/fate/jpeg2000.mak +++ b/tests/fate/jpeg2000.mak @@ -42,6 +42,9 @@ fate-jpeg2000dec-p0_08: CMD = framecrc -flags +bitexact -auto_conversion_filters FATE_JPEG2000DEC += fate-jpeg2000dec-p0_09 fate-jpeg2000dec-p0_09: CMD = framecrc -flags +bitexact -i $(TARGET_SAMPLES)/jpeg2000/itu-iso/codestreams_profile0/p0_09.j2k +FATE_JPEG2000DEC += fate-jpeg2000dec-p0_10 +fate-jpeg2000dec-p0_10: CMD = framecrc -flags +bitexact -i $(TARGET_SAMPLES)/jpeg2000/itu-iso/codestreams_profile0/p0_10.j2k + FATE_JPEG2000DEC += fate-jpeg2000dec-p0_11 fate-jpeg2000dec-p0_11: CMD = framecrc -flags +bitexact -i $(TARGET_SAMPLES)/jpeg2000/itu-iso/codestreams_profile0/p0_11.j2k diff --git a/tests/ref/fate/jpeg2000dec-p0_10 b/tests/ref/fate/jpeg2000dec-p0_10 new file mode 100644 index 0000000000..16c4e5e39d --- /dev/null +++ b/tests/ref/fate/jpeg2000dec-p0_10 @@ -0,0 +1,6 @@ +#tb 0: 1/25 +#media_type 0: video +#codec_id 0: rawvideo +#dimensions 0: 64x64 +#sar 0: 0/1 +0, 0, 0, 1, 12288, 0x68638483 -- 2.25.1

3 5