vp9_superframe: Avoid allocations and copies of packet structures

Up until now, the vp9_superframe bsf used distinct packets for input and
output. But at no point in the bsf are the input and output packets used
at the same time (except during a call to av_packet_move_ref()), so that
one can avoid using two packets if one switches to ff_bsf_get_packet_ref
at the same time. This also saves one malloc+free of an AVPacket
structure per filtered packet.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>

avcodec/cavsdec: Set ret on errors in decode_pic()

Fixes regression and fate failure on some machienes

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/ac3enc: Fix memleak

Fixes ticket #8294.

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

avfilter/af_arnndn: Remove superfluous qualifier

remove_doubling() returned a const float, but returning qualified types
is pointless anyway. This is all the more important given that  GCC 4.2.1
(and maybe other compilers, too) seems to treat this as an error as several
FATE units that don't compile any more show.

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

doc/filters: fixes for scale filter

Adjusted formatting and prose for option force_divisible_by in scale filter

avfilter/vf_dedot: fix crash

Fixes #8299

avfilter/vf_amplify: cosmetics

avfilter/vf_random: fix memory leaks

Fixes #8296

avfilter/af_arnndn: make sure that vad_output nb_neurons is always 1

avfilter/vf_gblur: fix undefined behaviour

Fixes #8292

avcodec/sbcdec: Fix integer overflows in sbc_synthesize_eight()

Fixes: signed integer overflow: 518484152 + 1868182638 cannot be represented in type 'int'
Fixes: 17732/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_SBC_fuzzer-5663738132168704
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/adpcm: Check initial predictor for ADPCM_IMA_EA_EACS

Fixes: signed integer overflow: -2147483360 - 631 cannot be represented in type 'int'
Fixes: 17701/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ADPCM_IMA_EA_EACS_fuzzer-5711517319692288
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/g723_1dec: Fix overflow in shift

Fixes: shift exponent 1008 is too large for 32-bit type 'int'
Fixes: 17700/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_G723_1_fuzzer-5707633436131328
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/apedec: Fix integer overflow in predictor_update_3930()

Fixes: signed integer overflow: -69555262 * 31 cannot be represented in type 'int'
Fixes: 17698/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_APE_fuzzer-5728970447781888
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/g729postfilter: Fix undefined intermediate pointers

Fixes: index -49 out of bounds for type 'int16_t [192]'
Fixes: 17689/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ACELP_KELVIN_fuzzer-5756275014500352
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/g729postfilter: Fix undefined shifts

Fixes: left shift of negative value -12
Fixes: 17689/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ACELP_KELVIN_fuzzer-5756275014500352
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/lsp: Fix undefined shifts in lsp2poly()

Fixes: left shift of negative value -30635
Fixes: 17689/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ACELP_KELVIN_fuzzer-5756275014500352
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/aacdec_template: Check decode_extension_payload() for failure

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/adpcm: Fix left shifts in AV_CODEC_ID_ADPCM_EA

Fixes: left shift of negative value -1
Fixes: 17683/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ADPCM_EA_R2_fuzzer-5111690013704192
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/smacker: cleanup on errors in smka_decode_frame()

Fixes: multiple memleaks
Fixes: 17660/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_SMACKAUD_fuzzer-5689769928949760
Fixes: 18064/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_SMACKAUD_fuzzer-5631086809317376
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

tools/target_dec_fuzzer: Adjust threshold for EATGV

Fixes: Timeout (26sec -> 9sec)
Fixes: 17645/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_EATGV_fuzzer-5717065922510848
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

tools/target_dec_fuzzer: Adjust threshold for SCPR

Fixes: Timeout (46sec -> 7sec)
Fixes: 17644/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_SCPR_fuzzer-5715704283660288
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/cavsdec: Check remaining bitstream in the main loop in decode_pic()

Fixes: Timeout (149sec ->1sec)
Fixes: 17311/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CAVS_fuzzer-5679368642232320
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/shortendec: Check k in probe

Fixes: Assertion failure
Fixes: 17640/clusterfuzz-testcase-minimized-ffmpeg_DEMUXER_fuzzer-5708767475269632
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

tools/target_dec_fuzzer: Adjust threshold for MSZH

Fixes: Timeout (250sec -> 6sec)
Fixes: 17627/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MSZH_fuzzer-5643017129558016
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

swscale/output: Avoid 64bit in Alpha in yuv2ya16_X_c_template()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

swscale/output: Correct Alpha in yuv2ya16_X_c_template()

Untested, no testcase

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

swscale/output: Implement Luma computation from yuv2ya16_X_c_template() without 64bit

This also reverts 21838cad2fc44023ad85e35d5c677e2f8d29a0ef
The revert is in this commit to avoid 2 fate updates

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

fate/source: add libavfilter/af_arnndn.c

vafilter/af_replaygain: fix undefined behaviour

Fixes #8291

avfilter/avf_abitscope: fix undefined behaviour

Fixes #8289

avfilter/vf_colorchannelmixer: do not use pointer if there is no alpha

Fixes #8288

avfilter/vf_edgedetect: fix undefined behaviour

Fixes #8287

avcodec/apedec: Check remaining space in decode_array_0000()

Fixes: Timeout (? -> 2sec)
Fixes: 17886/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_APE_fuzzer-5728165124636672
Fixes: 18131/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_APE_fuzzer-5710803432374272
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/apedec: Check error flag after entropy_decode*

Fixes: 17886/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_APE_fuzzer-5728165124636672
Fixes: 18131/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_APE_fuzzer-5710803432374272
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/f_reverse: fix memory leaks

Fixes #8283

avfilter: add arnndn filter

avfilter/vf_atadenoise: compensate for small overall brightness loss

This is very hard to spot.

avfilter/vsrc_testsrc: rgbtestsrc: fix undefined behaviour

Fixes #8238

avfilter/vf_gblur: fix heap-buffer overflow

Fixes #8282

avformat/chromaprint: Fix writing raw fingerprint

The pointer fp after the call to chromaprint_get_raw_fingerpoint() points to an array
of uint32_t whereas the current code assumed just a char stream. Thus when writing the
raw fingerprint, the output would be truncated by a factor of 4.

For reference the declaration of the function from chromaprint.h is:
int chromaprint_get_raw_fingerprint(ChromaprintContext *ctx, uint32_t **fingerprint, int *size);

avfilter/af_afade: make sure that in is available

avfilter/dnn: unify the layer load function in native mode

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>

avfilter/dnn: unify the layer execution function in native mode

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>

avfilter/dnn: add DLT prefix for enum DNNLayerType to avoid potential conflicts

and also change CONV to DLT_CONV2D for better description

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>

avformat/latmenc: abort if no extradata is available

Fixes ticket #8273.

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>

avfilter/vf_bm3d: make sure nb_jobs is at least 1

tools/target_dec_fuzzer: Also fuzz keyframe & disposal flags

This should improve coverage

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavc/libxavs2: fix parameter setting result determination

Signed-off-by: hwren <hwrenx@126.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/af_afade: fix heap-buffer overflow

Fixes #8276

avfilter/vf_edgedetect: fix heap-buffer overflow

Fixes #8275

avfilter/vf_avgblur: fix heap-buffer overflow

Fixes #8274

avcodec/av1_parse: Use av_fast_realloc() for OBU array

Based on commits 22bec0d33f4231487547581a1f77e2e8e6eade88 and
cebb446911fdc6c42d5a480b441b025c399e4a88.

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

avcodec/av1_parse: simplify memset call

Removed (new_size - pkt->nals_allocated) because this value is always 1
during the call.
Based on commit 78b86c30d3860135042505dd4a9cbd95c4e6257d.

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

avcodec/h2645_parse: Reset nal_buffer_size on uninit

Without reseting nal_buffer_size, av_fast_realloc will fail if
ff_h2645_packet_split is called with the unitialized pkt as argument.

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

ffprobe: fix output of packets_and_frames section in JSON format

The "type" entry was hardcoded with an trailing comma, even if it was
the only entry in the section.

Fixes ticket #8228.

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

avfilter: fix typo in comments

Signed-off-by: mypopy@gmail.com

tools/target_dec_fuzzer: Also fuzz codec_tag

This should improve coverage

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

tools/target_dec_fuzzer: Also fuzz FF_COMPLIANCE_EXPERIMENTAL

This should improve coverage

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

tools/target_dec_fuzzer: Also fuzz block_align

This should improve coverage

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/vf_lenscorrection: make width/height int

Somehow previous correct fix broke usage.

doc/APIchanges: add missing entry for the new runtime param AVOption flag

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

avfilter/vf_atadenoise: rewrite using macro

doc/filters: document colorchannelmixer commands

doc/filters: document amplify commands

doc/filters: document scroll commands

avfilter/vf_bwdif: fix heap-buffer overflow

Fixes #8261

avfilter/af_atempo: use ff_filter_process_command()

avfilter/af_biquads: use ff_filter_process_command()

avfilter/af_rubberband: use ff_filter_process_command()

avfilter/vf_drawbox: use ff_filter_process_command()

avfilter/vf_colorchannelmixer: add support for commands

avfilter/vf_amplify: add support for commands

avfilter/vf_scroll: add support for commands

avfilter/af_anlmdn: switch to ff_filter_process_command()

avfilter/vf_avgblur: switch to ff_filter_process_command()

avfilter/vf_gblur: switch to ff_filter_process_command()

avfilter/af_anlms: switch to ff_filter_process_command()

doc/filters: document new feature

avutil/opt: print runtime flag too

avfilter/avfilter: add ff_filter_process_command()

avutil/opt: add AV_OPT_FLAG_RUNTIME_PARAM flag

avfilter/vf_geq: reindent

avfilter/vf_geq: allow user to set interpolation method

avfilter/vf_edgedetect: fix coverity issue

avformat/chromaprint: improve logging message

Setting silence_threshold requires that -algorithm is set to 3.

avformat/chromaprint: Fix fp_format option

The fp_format option was incorrectly declared,
so it could not be set via string constants.

avfilter/vf_geq: Use av_clipd() instead of av_clipf()

With floats we cannot represent all 32bit integer dimensions

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

avformat/Makefile: Fix aiffdec replaygain dependency

Forgotten in 6390f52a.

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

avcodec/mips: Fixed four warnings in vc1dsp

Change the stride argument to ptrdiff_t in the following functions:
ff_put_no_rnd_vc1_chroma_mc8_mmi, ff_put_no_rnd_vc1_chroma_mc4_mmi,
ff_avg_no_rnd_vc1_chroma_mc8_mmi, ff_avg_no_rnd_vc1_chroma_mc4_mmi.

Reviewed-by: Shiyou Yin <yinshiyou-hf@loongson.cn>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/wmaprodec: Check that the streams channels do not exceed the overall channels

Fixes: NULL pointer dereference
Fixes: 18075/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_XMA1_fuzzer-5708262036471808
Fixes: 18087/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_XMA1_fuzzer-5740627634946048
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_libvmaf: fix filtering of >8 bit data

This is what reference does.

avfilter/vf_lenscorrection: fix division by zero

Fixes #8265

avfilter/vf_fieldorder: fix heap-buffer overflow

Fixes #8264

avfilter/vf_convolution: fix undefined behaviour

Fixes #8263

avfilter/vf_bm3d: fix possible infinite loop

avfilter/vf_bm3d: fix heap-buffer overflows

Fixes #8262

avfilter/vf_edgedetect: check if height is big enough

Fixes #8260

avfilter/vf_waveform: fix typos which caused crash

avfilter/vf_colorconstancy: fix overreads in gauss array

Fixes #8250