Merge commit '1e56173515826aa4d680d3b216d80a3879ed1c68'

* commit '1e56173515826aa4d680d3b216d80a3879ed1c68':
  rtsp: add pkt_size option

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

avcodec/wavpackenc: use ff_clz()

avcodec/cinepak: Add missing const

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

avfilter/vf_tonemap: add slice threading

avfilter/avf_abitscope: switch to activate

avfilter/af_biquads: add timeline support

avfilter/af_stereowiden: add timeline support

avfilter/af_extrastereo: add timeline support

avfilter/af_crossfeed: add timeline support

avcodec/agm: Check output size before allocation

Fixes: OOM
Fixes: 14198/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AGM_fuzzer-5723579234123776
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/truemotion2: Fix 2 integer overflows in tm2_update_block()

Fixes: signed integer overflow: -2147483648 + -1 cannot be represented in type 'int'
Fixes: 14107/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_TRUEMOTION2_fuzzer-5694078680825856
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/cinepak: Check slice_size before allocating image

Fixes: Timeout (16sec -> 125msec)
Fixes: 14283/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CINEPAK_fuzzer-5742851457024000
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Tomas Härdin <tjoppen@acc.umu.se>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter/avf_showspectrum: convert some doubles to floats

swscale/ppc: VSX-optimize hscale_fast

./ffmpeg -f lavfi -i yuvtestsrc=duration=1:size=1200x1440 -sws_flags fast_bilinear \
        -s 2400x720 -f rawvideo -vframes 5 -pix_fmt abgr -nostats test.raw

4.27 speedup for hyscale_fast:
  24796 UNITS in hyscale_fast,    4096 runs,      0 skips
   5797 UNITS in hyscale_fast,    4096 runs,      0 skips

4.48 speedup for hcscale_fast:
  19911 UNITS in hcscale_fast,    4095 runs,      1 skips
   4437 UNITS in hcscale_fast,    4096 runs,      0 skips

lavu/hwcontext_qsv: Fix the realign check for hwupload

Fix the aligned check in hwupload, input surface should be 16 aligned
too.

Partly fix #7830.

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

avfilter/af_asoftclip: add timeline support

avfilter/af_dynaudnorm: add timeline support

avfilter/af_dynaudnorm: always update delay when flushing

avfilter/vf_normalize: add timeline support

avfilter/af_adeclick: switch to activate

avfilter/af_rubberband: make use of initial input timestamp

avfilter/avf_showwaves: Add draw mode also to showwavespic filter

See commit a8c2d375ca68b7f001564ced14d8ac0757f53a29

avfilter/f_ebur128: use correct type for chl

avfilter/af_dynaudnorm: switch to activate

avfilter/af_rubberband: switch to activate

avfilter/af_rubberband: check if rbs is valid

avfilter/af_sofalizer: switch to activate

avfilter/avf_avectorscope: switch to activate

avfilter/avf_ahistogram: switch to activate

avfilter/avf_ahistogram: properly name pads

avfilter/avf_showspectrum: add log scale for frequency plot

avfilter/af_astats: fix msvc compile error

MSVC requires an explicit cast from void * to void when applying the
ternary conditional operator to switch between methods that return
void.

avcodec/cbs_h2645: use the fixed() macro for forbidden_zero_bit

This follows the spec definition, and removes a field from the relevant
structs.

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

avcodec/cbs_h2645: add macros to read and write fields with no custom range of values

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

avcodec/cbs_av1: add missing value range constrains to timecode Metadata OBU

Also infer the value time_offset_length as 0 when it's not present.

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

avformat/mxfdec: fix and enhance RIP KLV length checks

KLV length is BER encoded (variable size), but the code assumed the encoding to
always use 4 bytes.

Fixes parsing Random Index Pack in samples/MXF/issue2160/PW0805A0V01.4C5B5636.EFA330.mxf.

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

avformat/mxfdec: take into account run-in in find_partition_by_offset

Also rename the function to find_partition_by_absolute_offset to make it clear
offset is absolute.

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

avformat/mxfdec: guess wrapping of tracks by other tracks with the same body sid

This affects the following samples:

samples/ffmpeg-bugs/roundup/issue1775/av_seek_frame_failure.mxf
samples/ffmpeg-bugs/trac/ticket1957/16ch.mxf
samples/ffmpeg-bugs/trac/ticket5016/r0.mxf
samples/ffmpeg-bugs/trac/ticket5016/r1.mxf
samples/ffmpeg-bugs/trac/ticket5316/hq.MXF
samples/ffmpeg-bugs/trac/ticket5316/hqx.MXF

Some AVPacket->pos values are changed because for frame wrapped tracks we point
to the KLV offset and not the data.

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

avformat/mxfdec: rework mxf_essence_container_end

We find the last essence container much faster if we go through the partitions
backwards...

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

avcodec/scpr: avoid using uninitialized value

Fixes #7872.

avfilter/vf_lut3d: increase MAX_LEVEL

avfilter/vf_lut3d: fix range domain processing for .cube format

The ranges are for input, not for output.

avcodec/pnm_parser: Remember the length already scanned for ascii images

Fixes: speed regression with xmap_samsung_gear_2560x1280.pgm
Found-by: Michael Koch
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/pnm_parser: Remember the size of the image and do not reparse the header

Fixes: Timeout (11sec -> 60ms)
Fixes: 14270/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_PAM_fuzzer-5734809634078720
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>

avcodec/scpr3: add missing check for decode_value3() return value

Fixes ticket #7866.

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

avformat/microdvddec: skip empty lines

avformat/subtitles: ignore extra '\r' at line endings

avcodec/nvenc: fix indentation

avcodec/nvenc: only unregister input resources when absolutely needed

This reverts nvenc to old behaviour, which in some super rare edge cases
performs better.
The implication of this is that any potential API user who relies on
nvenc cleaning up every frames device resources after it's done using
them will have to change their usage pattern.

That should not be a problem, since pretty much every normal usage
pattern automatically implies that surfaces are reused from a common
pool, since constant re-allocation is also very expensive.

lavf/hls: Remove HLSContext.strict_std_compliance field

After the commit 9f61abc8111, we can use AVFormatContext.strict_std_compliance
instead of HLSContext.strict_std_compliance to avoid the code redundancy.

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>

x86/opusdsp: replace loads with shuffles

Has a slight speedup.
Can't be carried over to aarch64, since it has no shufps-like instruction.

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

avfilter: add audio soft clip filter

avfilter/vf_stack: use time_base from framesync

Fixes non-monotonous timestamps.

avcodec/gdv: Check input palette size before rescale()

Fixes: Timeout (22sec -> 11sec)
Fixes: 13576/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_GDV_fuzzer-5681024577568768
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/af_astats: count number of NaNs/Infs/denormals for floating-point audio too

lavfi/opencl: add more opencl helper macro

Signed-off-by: Ruiling Song <ruiling.song@intel.com>

avcodec/jpeg2000dec: Fix return type of get_plt()

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

avcodec/jpeg2000dec: Check PLT data somewhat

Fixes: Timeout (21sec -> 0.6sec)
Fixes: 14134/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_JPEG2000_fuzzer-5768371078955008
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/jpeg2000dec: Replace the step_x/y assert by a check in the CPRL case as with the PCRL case

Fixes: assertion failure
Fixes: 14246/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_JPEG2000_fuzzer-5758393601490944
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/jpeg2000: Check stepsize before using it

Fixes: value 1.87633e+10 is outside the range of representable values of type 'int'
Fixes: Undefined behavior
Fixes: 14246/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_JPEG2000_fuzzer-5758393601490944
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/qtrle: Check how much of the chunk is available before decoding

Fixes: Timeout (10sec -> 2sec)
Fixes: 13979/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_QTRLE_fuzzer-5635157718990848
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_fixed: Fix undefined shift in noise_scale()

Fixes: 13655/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AAC_FIXED_fuzzer-5120559430500352
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/bintext: Add error message when resolution is too small for font.

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

avcodec/zmbv: optimize motion compensation with memcpy()

Fixes: Timeout (16 sec - 7 sec)
Fixes: 14237/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ZMBV_fuzzer-5693453897302016
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Tomas Härdin <tjoppen@acc.umu.se>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavfi/frei0r: Fix a union member type and remove an unneeded cast.

avcodec/h264_ps: use get_se_golomb_long() to parse some sps fields

All three may be up to 32 bits wide.

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

avcodec/h264_ps: fix storage size for offset_for_ref_frame

The spec defines the valid range of values to be INT32_MIN + 1 to INT32_MAX, inclusive.

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

avfilter/af_surround: improve rear channels separation

avfilter/af_surround: check for invalid magnitude and phase difference

avfilter/af_surround: add angle option

avfilter/af_surround: fix typo

avcodec/arbc: Try to correct keyframe/frame type

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

avcodec/arbc: Skip unchanged frames

Fixes: Timeout (16sec -> 5sec)
Fixes: 14128/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ARBC_fuzzer-5767365721063424
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avutil/avstring: Fix bug and undefined behavior in av_strncasecmp()

The function in case of n=0 would read more bytes than 0.
The end pointer could be beyond the allocated space, which
is undefined.

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

avformat/mov: Skip stsd adjustment without chunks

Fixes: Assertion failure
Fixes: clusterfuzz-testcase-minimized-media_pipeline_integration_fuzzer-5683096400822272
Found-by: Clusterfuzz
Reported-by: Dan Sanders <sandersd@google.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avformat/aadec: Check for scanf() failure

Fixes: use of uninitialized variables
Fixes: blank.aa
Found-by: Chamal De Silva <chamal.desilva@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/dxtory: Check slice_size against minimum in dxtory_decode_v2()

Fixes: Timeout (33sec -> 16 milli sec)
Fixes: 14181/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DXTORY_fuzzer-5681840708386816
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/dsicinvideo: check the amount decoded by cin_decode_huffman()

Fixes: Timeout (158sec -> 36sec)
Fixes: 14214/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DSICINVIDEO_fuzzer-5633569034076160
This is untested with valid cin files as none of the files i found cover the changed
codepath

Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/agm: Check that there is available input in read_code()

Fixes: Timeout (46sec -> 7ms)
Fixes: 14030/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AGM_fuzzer-5721258760601600
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

tests/fate-run: New variable hostexecsuf for local fate tools.

Allows running fate in-tree on wsl with msvc.

compat/windows/makedef: Allow building shared libs with MSVC under WSL

A similar patch was posted by Gilles Khouzam in his blog.

avdevice/alsa: fix indefinite stop on closing PCM capture

Fixes: https://bugs.archlinux.org/task/58619
Found-by: Elias (Bleuzen) https://bugs.archlinux.org/user/26956
Signed-off-by: Nicolas George <george@nsup.org>

libavformat/mov: limit nb_frames_for_fps to INT_MAX

It's this or add overflow detection in mov_read_header().

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

avfilter/af_surround: export more channel's in/out gains

avfilter/af_surround: switch to activate

avfilter/af_surround: expose window size to user

avfilter/af_surround: add 6.1/6.0 upmix from stereo

avcodec/cuviddec: improve progressive frame detection

There are 2 types of problems when using adaptive deinterlace with cuvid:

1. Sometimes, in the middle of transcoding, cuvid outputs frames with visible horizontal lines (as though weave deinterlace method was chosen);
2. Occasionally, on scene changes, cuvid outputs a wrong frame, which should have been shown several seconds before (as if the frame was assigned some wrong PTS value).

The reason is that sometimes CUVIDPARSERDISPINFO has property progressive_frame equal to 1 with interlaced videos.
In order to fix the problem we should check if the video is interlaced or progressive in the beginning of a video sequence (cuvid_handle_video_sequence).
And then we just use this information instead of the property progressive_frame in CUVIDPARSERDISPINFO (which is unreliable).

Signed-off-by: Timo Rothenpieler <timo@rothenpieler.org>

examples/avio_reading: Use avio_context_free() to free AVIOContext

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>

avfilter/af_surround: make channel spread from stereo image user configurable

avfilter/af_surround: add lfe_mode option

avfilter/af_surround: avoid divisions with very small numbers

avformat/mpegenc - reject unsupported audio streams

Only MP1, MP2, MP3, 16-bit PCM_DVD, PCM S16BE,
AC3 and DTS audio codecs are supported by the muxer.

lavf/oggparsevorbis: Fix change the case of metadata keys issue

The spec in https://xiph.org/vorbis/doc/v-comment.html states that
the metadata keys are case-insensitive, so don't change the case
and update the fate test case.

Fix #7784

Reviewed-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>

avformat/dashenc: Fix a bug with writing "final" manifest

This bug was introduced in the commit 951561b64ee6c11f01daedd9dcf73276cc1e765b

avformat/dashdec: add ProgramInformation parser

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

avcodec/agm: remove ;;

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

avcodec/agm: Check for too many too short codes in make_new_tree()

Fixes: SEGV on unknown address
Fixes: 14198/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AGM_fuzzer-5723579234123776
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>

avcodec/agm: Do not crash on invalid codes

I do not know if such vlc trees are allowed in agm, I have no specification
So i do not know if these should be treated as error, or not.
But the code does contain a check for idx < 0 already ...

Fixes: Stack-buffer-overflow in get_tree_codes
Fixes: 14189/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AGM_fuzzer-5745747003179008
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>

avcodec/ccaption_dec: Add a blank like at the end to avoid rollup reading from outside

Fixes: index 20 out of bounds for type 'const char *[4][128]'
Fixes: 14367/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CCAPTION_fuzzer-5718819672162304
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>