avcodec/shorten: Fix integer overflow with offset

Fixes: signed integer overflow: -1625810908 - 582229060 cannot be represented in type 'int'
Fixes: 10977/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_SHORTEN_fuzzer-5732602018267136
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/golomb: Speed up long ur_golomb codes

Fixes: Timeout
Fixes: 10972/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_FLAC_fuzzer-5707569640243200
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

h264_metadata: Don't use inferred value of buffering frames

Using the value of buffering frames inferred from the old level is
not a sensible approach when one wants to guess the level.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@googlemail.com>
Signed-off-by: Mark Thompson <sw@jkqxz.net>

h264_levels, h264_metadata_bsf: Fix levels typo

profile_idc for level 1b should be 11, not 10.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@googlemail.com>
Signed-off-by: Mark Thompson <sw@jkqxz.net>

avcodec/prores_metadata_bsf : fix colorspace check

Report by coverity
CID 1441165

avcodec/libopenh264enc: fix class_name

Reverts some accidental change in commit e621b1ca646a2f268797adc3194b694a852548d2.

Reviewed-by: Jan Ekström <jeebjp@gmail.com>
Signed-off-by: Tobias Rapp <t.rapp@noa-archive.com>

avfilter/vf_alphamerge: do not check for ret value if all wanted frames are present

It is not needed and may be uninitialized.

avfilter/vf_fade: fix start/duration max value

A fade out (usually at the end of a video) can easily start beyond
INT32_MAX (about 36 minutes).  Regression since d40dc64173.

avfilter/af_afftfilt: use first input frame timestamp

lavc/videotoolbox: Fix cropping with HEVC and H264 videos

videotoolbox returns an already cropped stream which led to double
cropping. this issue was introduced with the refactor of the cropping
mechanism in commit 07596e4 for h264 and 000fb61 for HEVC.

to fix this we set the cropping of the frame and the output frame to 0.

Tested-by: ponpon
Fixes ticket #7544.

fate: Add test for vc1test demuxer

Signed-off-by: Jun Zhao <mypopydev@gmail.com>

avfilter/vf_alphamerge: switch to activate

avformat/hlsenc: check dirname and fix vtt problem

fix ticket: 7527
check dirname before use it
refine webvtt code in the hls_delete_old_segments

Reported-by: caspy
Signed-off-by: Steven Liu <lq@chinaffmpeg.org>

avcodec/tiff: add initial bayer and sub image support

avfilter/af_headphone: do not leak input frames on error

avfilter/af_afftfilt: calculate all channels FFT values upfront

cbs_h2645: Improve performance of writing slices

Instead of using a combination of bitreader and -writer for copying data,
one can byte-align the (obsolete and removed) bitreader to improve performance.
With the right alignment one can even use memcpy. The right alignment
normally exists for CABAC and hence for H.265 in general.
For aligned data this reduced the time to copy the slicedata from
776520 decicycles to 33889 with 262144 runs and a 6.5mb/s H.264 video.
For unaligned data the number went down from 279196 to 97739 decicycles.

Signed-off-by: Mark Thompson <sw@jkqxz.net>

fate/prores_metadata : fix md5 value

avfilter/af_afftfilt: add more window types

avfilter/af_afftfilt: extend filter functionality

fate/prores_metadata_bsf : add test for setting color property

avcodec : add prores_metadata bsf for set the color property of each prores frame

avfilter/vf_minterpolate: use common scene sad functions

Signed-off-by: Marton Balint <cus@passwd.hu>

avfilter/vf_select: use common scene sad functions

Signed-off-by: Marton Balint <cus@passwd.hu>

avfilter/vf_framerate: factorize SAD functions which compute SAD for a whole frame

Also add SIMD which works on lines because it is faster then calculating it on
8x8 blocks using pixelutils.

Signed-off-by: Marton Balint <cus@passwd.hu>

cbs_mpeg2: Improve performance of writing slices

Instead of using a combination of bitreader and -writer for copying data,
one can byte-align the (obsolete and removed) bitreader to improve performance.
One can even use memcpy in the normal case.
This improved the time needed for writing the slicedata from 33618 to
2370 decicycles when tested on a video originating from a DVD (4194394
runs).

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@googlemail.com>
Signed-off-by: Mark Thompson <sw@jkqxz.net>

cbs_h265: Add PTL parsing for sublayers

With fate test using the SLPPLP_A_VIDYO_2 conformance file, which contains
two sublayers with full PTL information.

fate/libavcodec: Fix config dependency of h264-levels test

avfilter/vf_lut3d: ignore last whitespace when comparing LUT size string

In some situations it might be tab character and in others normal space.

fftools/ffmpeg: Indent the code

Signed-off-by: Jun Zhao <mypopydev@gmail.com>

fftools/ffmpeg: Put the variable declaration at uppper for block.

move the variable declaration at start of upper for block and
remove the redundant brace.

Signed-off-by: Jun Zhao <mypopydev@gmail.com>

fftools/ffmpeg: Remove the micor like "#if 1"

They are come from 2003 and delete them.

Signed-off-by: Jun Zhao <mypopydev@gmail.com>

avcodec/imm4: Use ff_set_dimensions()

Fixes: Out of memory
Fixes: 10970/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_IMM4_fuzzer-5698750043914240
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

h264_redundant_pps: Fix logging context

The first element of H264RedundantPPSContext is not a pointer to an
AVClass as required.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@googlemail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/vf_lut3d: add spline 1D interpolation

avfilter/vf_lut3d: add cosine 1D interpolation

avcodec/hevcdec: fix non-ref frame judgement

After inspecting the source code of x265, mpv and ffmpeg, I've found that
ffmpeg mistakenly regards EVC_NAL_BLA_N_LP and HEVC_NAL_IDR_N_LP as non-
reference frames, which are acutally reference frames according to the
specification in x265, and drops them.

This patch should address the problem. I have tested it with mpv.

Signed-off-by: Mark Wu <wfwf1997@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>

avcodec/wmaprodec: improve XMA missing samples

Writes missing (delay) samples after EOF.

Signed-off-by: bnnm <bananaman255@gmail.com>

libvpxenc: extend auto-alt-ref range

vp9 now supports [0, 6]

Reviewed-by: James Almer <jamrial@gmail.com>
Signed-off-by: James Zern <jzern@google.com>

libvpxenc,vp9: add enable-tpl option

enables temporal dependency model

Signed-off-by: James Zern <jzern@google.com>

avfilter/af_headphone : fix mem leak

report by coverity
CID 1439934
CID 1439935

avfilter/af_afir: fix hang because video is first requested

avfilter/af_afir: kill useless code which increases latency for no reason

avcodec/qpeg: Optimize full width runs in qpeg_decode_intra()

This improves the speed of decoding large patches of constant color

Fixes: Timeout
Fixes: 10967/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_QPEG_fuzzer-5630803793936384
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/prores_aw : add support for prores 444 with alpha

only 16b alpha is supported (not 8 bits)

following official encoder, alpha data doesn't impact
yuv plane quality.

So the alpha data encoding is done after the yuv part.
It's also avoid to loose quality in yuv part when
alpha is not uniform.

the alpha encoding funcs is mainly take from prores_ks
encoder, except for the alpha data reorganization

avcodec/prores_aw : cosmetic indent and add comment to describe frame header value

description are based on multimedia wiki documentation

avcodec/prores_aw : change src pix_fmt and alpha description

use b64a as src pix fmt (doesn't seems to have an impact on decoding)
but it's the value use by official encoder

cbs_h264: Fix handling of auxiliary pictures

The earlier code used the most recent non-auxiliary slice to determine
whether an auxiliary slice has the syntax of an IDR slice, even when
the most recent slice was from a slice of a redundant frame. Now only
slices of the primary coded picture are used, as the specifications
mandate.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@googlemail.com>

avfilter/af_afir: add minp/maxp options to control latency and speed

avfilter/af_afir: do not leak frame and clear pointer to freed frame

avfilter/avf_showspectrum: add terrain color map

avfilter/avf_showspectrum: add cividis colormap

lavf/apngdec: Return EOF for incomplete files.

Fixes ticket #7536.

avfilter/af_aiir: implement rate option

avfilter/af_afir: implement rate option

avfilter/af_afir: return early when giving video frame

avfilter/af_afir: remove redundant code

avfilter/avf_showspectrum: add plasma color map

avfilter/avf_showspectrum: add viridis color map

avfilter/avf_showspectrum: improve magma colors

lavc/qsv_overlay: suppress code scan complain

Suppress the complain "variables 'pix_fmt' is used but maybe
uninitialized".

Signed-off-by: Zhong Li <zhong.li@intel.com>

lavc/qsvdec: flush buffered data before reinit

Flush the buffered data in libmfx before video param reinit
in case the frames drop.

Cache the first frame causing the reinit and decode zero-size
pkt to flush the buffered pkt before reinit. After all the
buffered pkts being flushed, resume to reinit and decode.

Fix the issue in ticket #7399.

[V2]: Move the definition of zero_pkt to where it is exactly
used.

Signed-off-by: Linjie Fu <linjie.fu@intel.com>
Signed-off-by: Zhong Li <zhong.li@intel.com>

avcodec: libdav1d AV1 decoder wrapper.

Originally written by Ronald S. Bultje, with fixes, optimizations and
improvements by James Almer.

Signed-off-by: James Almer <jamrial@gmail.com>

cbs_av1: Support redundant frame headers

cbs_av1: Fix header writing when already aligned

doc/filters: add tonemap_opencl document.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Signed-off-by: Gyan Doshi <gyandoshi@gmail.com>

doc/filters: add document for opencl filters

Signed-off-by: Danil Iashchenko <danyaschenko@gmail.com>
Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Signed-off-by: Gyan Doshi <gyandoshi@gmail.com>

mjpeg: Use profile names in the encoder and decoder

proresdec2: Parse codec_tag and export profile information

prores: Use profile names in the various encoders and decoders

Export FF_PROFILE_PRORES_* symbols publicly, add a missing one.

proresenc_anatoliy: Rename a profile name with the correct one

In all Apple documentation, this profile is called Prores 4444.

libopenh264enc: Handle sample_aspect_ratio

configure: Add missing V4L2 M2M decoder BSF dependencies

configure: Add missing IVF muxer BSF dependency

flvdec: Export unknown metadata packets as opaque data

Signed-off-by: Martin Storsjö <martin@martin.st>

flvdec: Rename FLV_STREAM_TYPE_DATA into FLV_STREAM_TYPE_SUBTITLE

This is always treated as a subtitle at the moment anyway.

Signed-off-by: Martin Storsjö <martin@martin.st>

libx264: Pass the reordered_opaque field through the encoder

libx264 does have a field for opaque data to pass along with frames
through the encoder, but it is a pointer, while the libavcodec
reordered_opaque field is an int64_t. Therefore, allocate an array
within the libx264 wrapper, where reordered_opaque values in flight
are stored, and pass a pointer to this array to libx264.

Update the public libavcodec documentation for the AVCodecContext
field to explain this usage, and add a codec capability that allows
detecting whether an encoder handles this field.

Signed-off-by: Martin Storsjö <martin@martin.st>

libavutil: Undeprecate the AVFrame reordered_opaque field

This was marked as deprecated (but only in the doxygen, not with an
actual deprecation attribute) in 81c623fae05 in 2011, but was
undeprecated in ad1ee5fa7.

Signed-off-by: Martin Storsjö <martin@martin.st>

avcodec/cbs_av1: fix decoder/encoder_buffer_delay variable types

buffer_delay_length_minus_1 is five bits long, meaning decode_buffer_delay and
encoder_buffer_delay can have values up to 32 bits long.

Reviewed-by: Mark Thompson <sw@jkqxz.net>
Signed-off-by: James Almer <jamrial@gmail.com>

configure: Fix av1_metadata BSF dependency

avformat/ftp: allow nonstandard 202 reply to OPTS UTF8

Fixes ticket #7481.

Signed-off-by: Marton Balint <cus@passwd.hu>

avcodec/cavsdec: Propagate error codes inside decode_mb_i()

Fixes: Timeout
Fixes: 10702/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CAVS_fuzzer-5669940938407936
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/mpeg4videodec: Clear partitioned frame in decode_studio_vop_header()

partitioned_frame is also set/cleared in decode_vop_header()

Fixes: out of array read
Fixes: 9789/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MPEG4_fuzzer-5638681627983872
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/af_afade: remove unused code

avcodec/pictordec: Error out if more than one plane is unused

Fixes: Timeout
Fixes: 9797/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_PICTOR_fuzzer-5664441659031552
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/mpegaudio_parser: Consume more than 0 bytes in case of the unsupported mp3adu case

Fixes: Timeout
Fixes: 10966/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MP3ADU_fuzzer-5348695024336896
Fixes: 10969/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MP3ADUFLOAT_fuzzer-5691669402877952
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/split: use ff_outlink_get_status()

avfilter/af_afade: do not use framequeue directly

avfilter/af_amerge: use ff_inlink_queued_samples()

avfilter/af_amultiply: make use of ff_inlink_queued_samples()

avfilter/vf_extractplanes: use ff_outlink_get_status()

fate/api-h264-slice-test: use cleaner error handling

Signed-off-by: James Almer <jamrial@gmail.com>

fate/api-h264-slice-test: don't use ssize_t

Fixes ticket #7521

Signed-off-by: James Almer <jamrial@gmail.com>

avcodec/prosumer: Simplify bit juggling of the c variable in decompress()

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/prosumer: Remove always true check in decompress()

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/prosumer: Remove unneeded ()

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/prosumer: Check for bytestream eof in decompress()

Fixes: Infinite loop
Fixes: 10685/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_PROSUMER_fuzzer-5652236881887232
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/vf_cuda_yadif: Avoid new syntax for vector initialisation

This requires a newer version of CUDA than we want to require.

avcodec/nvdec: Increase frame pool size to help deinterlacing

With the cuda yadif filter in use, the number of mapped decoder
frames could increase by two, as the filter holds on to additional
frames.

avfilter/vf_yadif_cuda: CUDA accelerated yadif deinterlacer

This is a cuda implementation of yadif, which gives us a way to
do deinterlacing when using the nvdec hwaccel. In that scenario
we don't have access to the nvidia deinterlacer.