Changelog: list the new aptX features

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

aptx: add raw muxer and demuxer for aptX

aptx: implement the aptX bluetooth codec

The encoder was reverse engineered from binary library and from
EP0398973B1 patent (long expired).
The decoder was simply deduced from the encoder.

decode: add a per-frame private data for hwaccel use

This will be useful in the CUVID hwaccel. It should also eventually
replace current decoder-specific mechanisms used by various other
hwaccels.

Merges Libav commit 704311b2946d74a80f65906961cd9baaa18683a3.

decode: add a mechanism for performing delayed processing on the decoded frames

This will be useful in the CUVID hwaccel.

Merges Libav commit badf0951f54c1332e77455dc40398f3512540c1b.

decode: add a method for attaching lavc-internal data to frames

Use the AVFrame.private_ref field.

This new struct will be useful in the following commits.

Merges Libav commit 359a8a3e2d1194b52b6c386f94fd0929567dfb67.

avutil/frame: Add private_ref to AVFrame

This gives FFmpeg libs a field that they can freely and safely use.
Avoiding the need of wrapping of a users opaque_ref field and its issues.

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

lavc/libx265: switch to ff_alloc_packet2

ff_alloc_packet have been deprecated, switch to use
ff_alloc_packet2.

Signed-off-by: Jun Zhao <jun.zhao@intel.com>
Reviewed-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavc/libkvazaar: switch to ff_alloc_packet2.

ff_alloc_packet have been deprecated, switch to use the
ff_alloc_packet2.

Signed-off-by: Jun Zhao <jun.zhao@intel.com>
Reviewed-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavu/timecode: clarify error msg for timecode_rate

The user-supplied value for timecode_rate in drawtext is rounded
to nearest integer. So, a supplied value of 0.49 or lower is rounded to 0.
This throws a misleading error message which says "Timecode frame rate must be
specified". Changed message to account for values under one.

Also noted supported framerates for drop TC.

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

avformat/dashdec: use the current DASHContext for the rep_dest

fix ticket id: #6820

Signed-off-by: Steven Liu <lq@onvideo.cn>
Tested-by: mrskman

lavf/dashdec: Fix several memleaks.

Signed-off-by: Carl Eugen Hoyos <ceffmpeg@gmail.com>
Tested-by: Steven Liu <lq@chinaffmpeg.org>

avcodec/cngdec: Fix integer clipping

Fixes: runtime error: value -36211.7 is outside the range of representable values of type 'short'
Fixes: 2992/clusterfuzz-testcase-6649611793989632
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

configure: fix the non pkg-config libmfx check on Windows

Based on a patch by Aaron Levinson.

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

avdevice/decklink_dec: make some function static

Reviewed-by: Aaron Levinson <alevinsn_dev@levland.net>
Signed-off-by: James Almer <jamrial@gmail.com>

lavc/pngdec: fix av_bprint_finalize() usage.

aacenc: mark the preset 5.0/5.1 layouts correctly with back speakers

The spec is correct, it does list these layouts as having rear speakers.
Questionable how many decoders correctly interpret those correctly since
side is way more popular.
Also fixes fate-aac-yoraw-encode.

Reported-by: pkviet <pkv.stream@gmail.com>
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

aacenc: use the PCE comment field for encoder ID

Also handle extradata of variable size (for bitexact/if PCEs aren't used).

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

aacenc: support extended channel layouts using PCEs

This commit implements support for PCE (Program Configuration Elements) in the
AAC encoder, and as such allows for encoding of channel layouts not present
in the presets defined by the spec (which only lists the 8 most common ones).

This has been a highly requested feature and is also the first open source encoder
to support this many layouts.

Many thanks to pkviet <pkv.stream@gmail.com> who implemented support for and
verified all channel layouts.

pngdec: expose gAMA and cHRM chunks as side/meta data

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

pngdec: check for bprint finalization sucess on icc data parsing

It can fail in case of OOM.

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

vc2enc_dwt: pad the temporary buffer by the slice size

Since non-Haar wavelets need to look into pixels outside the frame, we
need to pad the buffer. The old factor of two seemed to be a workaround
that fact and only padded to the left and bottom. This correctly pads
by the slice size and as such reduces memory usage and potential
exploits.
Reported by Liu Bingchang.

Ideally, there should be no temporary buffer but the encoder is designed
to deinterleave the coefficients into the classical wavelet structure
with the lower frequency values in the top left corner.

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

configure: add missing avutil deps for hwcontext modules

Fixes ticket #6812

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

lavfi/scale2ref: Set output frame rate to main input frame rate.

Fixes ticket #6817.

avformat/ty: fix format specifiers in debug log messages

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

avcodec/mips: Improve hevc non-uni hv mc msa functions

Use mask buffer.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/mips: Improve hevc uni weighted 4 tap vt mc msa functions

Use global mask buffer for appropriate mask load.
Use immediate unsigned saturation for clip to max saving one vector register.
Remove unused macro.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/mips: Improve hevc uni 4 tap hv mc msa functions

Use global mask buffer for appropriate mask load.
Remove unused macro and table.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/mips: Improve hevc bi wgt 4 tap hv mc msa functions

Use global mask buffer for appropriate mask load.
Use immediate unsigned saturation for clip to max saving one vector register.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/qsvenc_jpeg: fix copyright header

Fixes fate-source after 1926f13a206bfbd99abf536e7424e2b321f2e2ad.

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

checkasm/llviddsp: fix warnings about mixed declaration and code

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

Merge commit '70946e605924e2108c39f96faa369c220177f301'

* commit '70946e605924e2108c39f96faa369c220177f301':
  h264dec: Fix mix of lossless and lossy MBs decoding
  h264_cabac: Fix CABAC+8x8dct in 4:4:4
  h264dec: fix Lossless Decoding (Profile 244) for 8x8 Intra Prediction

This commit is a noop, see
6f7ca1f55be1270e0d7c33409de4473e4dd00add
840b41b2a643fc8f0617c0370125a19c02c6b586
cf231b68da1150c100114f2c5671b7ed740f917a

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

Merge commit '45c4bf3df03ef53ae61fa1473424d4ae024f22e4'

* commit '45c4bf3df03ef53ae61fa1473424d4ae024f22e4':
  h264dec: track the last seen value of x264_build

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

Merge commit '45df7adc1d9b7e8fbae5af9328baa6ab3562002b'

* commit '45df7adc1d9b7e8fbae5af9328baa6ab3562002b':
  imgutils: add function to clear an image to black
  lavc, lavu: move frame cropping to a convenience function
  rmdec: add missing brackets to compound statement

This commit is a noop, see
65b83ce01b94eae944290c11257799a9288f2d4c
caa12027baf1180453846c58da08fc87accc0ff6
463b81de2b252691d75417643597c42684bf830d

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

Merge commit '550cb6a85d0f7211289f7a789527d48cb29460ff'

* commit '550cb6a85d0f7211289f7a789527d48cb29460ff':
  lavf/vf_deinterlace_qsv: Enable the qsv deinterlace vpp

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

Merge commit '2fd6e7d077f590e4d7195356f9baeb271f8b9ae2'

* commit '2fd6e7d077f590e4d7195356f9baeb271f8b9ae2':
  libavcodec/mjpeg_qsv: Add QSV MJPEG encoder

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

Merge commit '2b1324bd167553f49736e4eaa94f96da9982925e'

* commit '2b1324bd167553f49736e4eaa94f96da9982925e':
  lavf: allow avformat_close_input() with NULL
  dxva: DXVA2_ModeHEVC_VLD_Main10 does not support Main
  caf: add an Opus tag
  hevc: Make sure to update the current frame transfer characteristic

This commit is a noop, see
dc9735eb67ca696d926922735a9bf01776d5c328
b4093e60c51af493a6dad7819264ef769736227f
5a3b602acda68fe5ca09082dc753179450a97a13
c64da19bbc1d3b405703ef3b705dba39a450504f
36cd017acd9cac0e6695124c052a59fb1fc13145

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

lavc/dnxhddata: Do not print frame rates with supported profiles.

The list is unmaintained, misleading and the frame rates are
ignored by all compatible players.

avformat: move priv_pts from AVStream to an internal struct

It has no reason to be in a public header, even if defined as private.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>

avutil/hwcontext_d3d11va: Fix leak when wrapping texture in AVD3D11FrameDescriptor

Reviewed-by: Aaron Levinson <alevinsn_dev@levland.net>
Signed-off-by: James Almer <jamrial@gmail.com>

Merge commit '95ce02b35d3d1bb16111031df1d82a6e4d894d36'

* commit '95ce02b35d3d1bb16111031df1d82a6e4d894d36':
  rmdec: don't ignore the return value of av_get_packet()

See 65b83ce01b94eae944290c11257799a9288f2d4c

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

Merge commit '825e463a170c7004c63030dc484b2b2de869227b'

* commit '825e463a170c7004c63030dc484b2b2de869227b':
  build: Add feature test macros for glibc 2.19+

This commit is a noop, see
26e8895b7395ed180dcb7e30465058c45050b5e2
af1818276ef271af98e2e2bbabb4dc875b4fa7d8

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

Merge commit '0f5ad12ba2b538cb329c507ecc914e06bfa70194'

* commit '0f5ad12ba2b538cb329c507ecc914e06bfa70194':
  flac: Use a local cache for decode_residual()

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

Merge commit '15f1cc09a406cf6296818d475a256902235eefc4'

* commit '15f1cc09a406cf6296818d475a256902235eefc4':
  flac: Postpone unlikely condition checks

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

Merge commit '9c1e111406bd4cbb04d8e8611b71bebf203bec5e'

* commit '9c1e111406bd4cbb04d8e8611b71bebf203bec5e':
  flac: Convert to the new bitstream reader
  bitstream: Avoid undefined behavior in bitstream_skip()

This commit is a noop, see
https://ffmpeg.org/pipermail/ffmpeg-devel/2017-April/209609.html

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

Merge commit 'fd92dafaff8844b5fedf94679b93d953939a7f7b'

* commit 'fd92dafaff8844b5fedf94679b93d953939a7f7b':
  bink: Split read_dct_coeffs()

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

Merge commit '5be9939b46a6a4d3860a1b1a872aa50897095970'

* commit '5be9939b46a6a4d3860a1b1a872aa50897095970':
  fate: Add cinepak encoder vsynth tests

See d7f62f033c214f3863acfd8e0672407b0bf91a3b

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

Merge commit 'ed3b2a999d189771c9b0ada9a9200117595fa474'

* commit 'ed3b2a999d189771c9b0ada9a9200117595fa474':
  cinepakenc: misc small changes
  cinepakenc: K&R formatting cosmetics
  cinepakenc: Stop using AVPicture
  cinepakenc: Drop broken debug code
  cinepakenc: add option handling for flexibility
  cinepakenc: fixes and improvements
  Add Cinepak encoder

See
59dbc36f49db5cfd9d2ad4b00ef2e3336173ee8d
762c4dc082764896e27f3c53c14a65871d8fe910
bf23642dcc86890fa2e32697c3b930b3f1ac0d88
0ab25dac2f0b170ecc0e37530cf15f83b9f9e56d

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

compat/w32dlfcn: add missing config.h include

Fixes warnings about undefined HAVE_WINRT

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

Merge commit '4d330da006fe48178a4c8047f06270925eaedf63'

* commit '4d330da006fe48178a4c8047f06270925eaedf63':
  os_support: Use HAVE_UWP instead of manually checking WINAPI_FAMILY
  d3d11va: Check WINAPI_FAMILY instead of HAVE_LOADLIBRARY
  lavf: Remove codec_tag from dashenc and smoothstreamingenc
  hevc: Add support for alternative transfer characterics SEI

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

configure: libvmaf is not GPLv2-compatible.

It is licensed under Apache 2.0.

avcodec/mips: Improve hevc bi 4 tap hv mc msa functions

Use global mask buffer for appropriate mask load.
Use immediate unsigned saturation for clip to max saving one vector register.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/mips: Improve avc avg mc 10, 30, 01 and 03 msa functions

Align the mask buffer to 64 bytes.
Load the specific destination bytes instead of MSA load and pack.
Remove unused macros and functions.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

libavcodec/mepgvideo : remove warning introduce by previous patch

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

libavcodec/lossless_video_dsp : cosmetic add better separator for each function, in order to make reading of the asm file easier

libavcodec/lossless_videodsp : add add_bytes avx2 version

checkasm/llviddsp : add test for other dsp func

add_median_pred
add_left_pred : add two func one with acc 0, and one with random acc
add_left_pred16

fate/utvideo : add test for decoding and encoding

libavcodec/utvideo : simplify decode_plane

the func is only call with step = 1
no need to pass it in the func

libavcodec/blockdsp : fix comment. clear_block need 32 aligned now

avutil/pixdesc: remove reference to avcodec_get_chroma_sub_sample

Don't recommend to use deprecated functions.

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

avcodec: remove remaining uses of avcodec_get_chroma_sub_sample

Replace them with av_pix_fmt_get_chroma_sub_sample.

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

Merge commit 'a594f17f83a1ffdc1eec18818208fe39487dd5d7'

* commit 'a594f17f83a1ffdc1eec18818208fe39487dd5d7':
  dvbsubdec: Free subrect memory on allocation error
  dvbsubdec: Fixed segfault when decoding subtitles

See
fbb59a3bf4c8668197521b3db5f26d0e93aed1a6
39dfe6801a3f67c2964a829b65e8e38e3cb22217

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

Merge commit '3fdf50f9e864c88da2139cf066832944de81acaa'

* commit '3fdf50f9e864c88da2139cf066832944de81acaa':
  hwframe: Allow hwaccel frame allocators to align surface sizes
  dxva: verbose-log decoder GUID list
  hwcontext_d3d11va: add option to enable debug mode
  dxva: support DXGI_FORMAT_420_OPAQUE decoding
  hwcontext_d3d11va: allocate staging texture lazily
  hwcontext_d3d11va: fix crash on frames_init failure
  dxva: fix some warnings
  dxva: add declarative profile checks

This commit is a noop, see
e2afcc33e0bcba92ab6c767f09f17a67911a4928
39f201a0ec7913f7509a01fb0fa6634a24e52203
1509d739a036b9838e12f28dac9f09ac37bc3928
6f5ff3269b12776a32b12887128ba034ad86e08b
8d7fdba7b8670db9dd5023dde6fe3a06674933c4
289d387330d8a0a07ff98c9e09215627a70ace2d
f0bcedaf37ed4fdb082c88826b8bd3a5dd26382a
d14179e3d49eb0f2533db16151e01abb0018b165

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

Merge commit '0e83e95c60892747f2f1bd8382f915f2397f99a5'

* commit '0e83e95c60892747f2f1bd8382f915f2397f99a5':
  configure: Reset X86ASM_DEP(FLAGS) when probing for the assembler program

See f66086adac5eb9b964c89ea0480487f48875f690

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

Merge commit '7cb053e4ddf258e2dbf52ccc586548680742d758'

* commit '7cb053e4ddf258e2dbf52ccc586548680742d758':
  nvenc: Minimize the surface allocation
  nvenc: Use a fifo to manage the free surface pool
  nvenc: Add an explicit auto alias

This commit is a noop, see
7b0689c55f1f8ee3abca58e1e3522f1c33beb60b
8de3458a07376b0a96772e586b6dba5e93432f52
de2faec2faf34ee948d400101299d4ccdc1f3a50

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

Merge commit '857e26b655a769e5a56bada1a0d9adb44cc176b7'

* commit '857e26b655a769e5a56bada1a0d9adb44cc176b7':
  build: Add an option for passing linker flags to the shared library build

See e34a3468f294f21878875e0f6ad5564cd52af70e.

--extra-ldlibflags is left in place for compatibility reasons.

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

Merge commit '333a34da3a6d13155de480e7d16ec5fec7f4cd6c'

* commit '333a34da3a6d13155de480e7d16ec5fec7f4cd6c':
  configure: Automatically add -isysroot for darwin if --sysroot is specified

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

Merge commit 'f519e131c101ae3948dabf1f7f598efbff5fb536'

* commit 'f519e131c101ae3948dabf1f7f598efbff5fb536':
  af_bs2b: Add missing casts to complex pointer assignments

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

avformat/ty: do not overread chunk

Signed-off-by: Paul B Mahol <onemda@gmail.com>

avformat/ty: try harder to find stream signature

Search all chunks, not just first one.

Signed-off-by: Paul B Mahol <onemda@gmail.com>

avformat/rtpdec_asf: free the buffer pointed by the AVIOContext

Don't free the buffer allocated in ff_wms_parse_sdp_a_line() after
calling avformat_open_input(), as it may free it and replace it with
another one.

Should fix ticket #6808

Tested-by: Carl Eugen Hoyos <ceffmpeg@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>

avformat: add TiVo ty demuxer

Signed-off-by: Paul B Mahol <onemda@gmail.com>

avfilter:vf_libvmaf: improve docs.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>

avfilter/vf_libvmaf: fix errors while running with psnr=1.

This patch fixes the seg fault which ocuured while running libvmaf filter
with option psnr=1.

Signed-off-by: ashk43712 <ashk43712@gmail.com>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>

lavf/amr: Do not set codec_tag for raw amr_nb and raw amr_wb.

ffplay: use SDL2 audio API

It allows us to specify what kind of audio parameter changes are allowed.

Should fix ticket #6721.

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

Maintainers : add myself for exr

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

avcodec/sbrdsp_fixed: Fix integer overflow in shift in sbr_hf_g_filt_c()

Fixes: runtime error: shift exponent 66 is too large for 64-bit type 'long long'
Fixes: 3642/clusterfuzz-testcase-minimized-5443853801750528
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/aacsbr_fixed: Fix division by zero in sbr_gain_calc()

Fixes: 3642/clusterfuzz-testcase-minimized-5443853801750528
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avutil/softfloat: Add FLOAT_MIN

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

avcodec/h264idct_template: Fix integer overflows in ff_h264_idct8_add()

Fixes: runtime error: signed integer overflow: -503316480 + -2013265038 cannot be represented in type 'int'
Fixes: 3805/clusterfuzz-testcase-minimized-6578427831255040
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavf/amr: Add amrnb and amrwb demuxers.

Fixes ticket #6678.

lavc/v4l2_context: Change the type of the ioctl cmd to uint32_t.

Fixes a warning on recent Linux:
libavcodec/v4l2_context.c: In function 'ff_v4l2_context_set_status':
libavcodec/v4l2_context.c:496:26: warning: comparison is always false due to limited range of data type

avcodec/qdrw: support 16bpp files with bppcnt == 2 && bpp == 8

avformat/mvdec: check for EOF

Fixes #6805.

Signed-off-by: Paul B Mahol <onemda@gmail.com>

ffmpeg.c: fix code style in seek_to_start

Signed-off-by: Peter Große <pegro@friiks.de>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

ffmpeg.c: fix calculation of input file duration in seek_to_start()

Fixes looping files without audio or when using stream_copy, where
ist->nb_samples is not set since no decoding is done.

This fixes ticket #5719 and also fixes an endless loop with the sample
in ticket #6139.

Signed-off-by: Peter Große <pegro@friiks.de>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/mips: Improve hevc uni weighted 4 tap hz mc msa functions

Use global mask buffer for appropriate mask load.
Use immediate unsigned saturation for clip to max saving one vector register.
Remove unused macro.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/mips: Improve hevc uni 4 tap hz and vt mc msa functions

Use global mask buffer for appropriate mask load.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/xan: Check for bitstream end in xan_huffman_decode()

Fixes: Timeout
Fixes: 3707/clusterfuzz-testcase-6465922706440192
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/exr: fix undefined shift in pxr24_uncompress()

Fixes: runtime error: left shift of 255 by 24 places cannot be represented in type 'int'
Fixes: 3787/clusterfuzz-testcase-minimized-5728764920070144
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>

Revert "vc2enc: Calculate average slice quantiser correctly"

This reverts commit 626d200e76e15487cff20992256a55919432b0b3.

ffplay: only use hardware accelerated SDL texture formats

Typically only a small subset of the SDL texture formats are supported directly
by the SDL renderer drivers, the rest is software emulated. It's better if
libswscale does the format conversion to a hardware-accelerated texture format
instead of SDL.

This should fix video render slowdowns with some texture formats after
3bd2228d05a05eab5f91ac00b01efac9cb07649b.

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

ffplay: create the window and the renderer before starting playback

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

vc2enc: Calculate average slice quantiser correctly

h264: Make Truncating SPS/PPS message debug.

h2645_parse: Allocate a single buffer per packet

Drastically reduces memory usage on pathological streams.
Fixes ticket #6789

lavc/alac: Avoid allocating huge memory blocks for malicious alac input.

avcodec/mips: Improve hevc bi wgt 4 tap hz and vt mc msa functions

Use global mask buffer for appropriate mask load.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>