Merge commit '0d4d43513786f1df4d561e1fac924fb0722c6700'

* commit '0d4d43513786f1df4d561e1fac924fb0722c6700':
  hevc: Add NEON add_residual for bitdepth 8

See 03cecf45c134ebbaecb62505fe444ade423ea7dc

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

avcodec/mips: Improve avc chroma avg hv mc msa functions

Replace generic with block size specific function.
Load the specific destination bytes instead of MSA load and pack.

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 22, 11, 31, 13 and 33 msa functions

Remove loops and unroll as block sizes are known.
Load the specific destination bytes instead of MSA load and pack.
Remove unused macro 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>

avcodec/aacdec_fixed: Fix integer overflow in apply_dependent_coupling_fixed()

Fixes: runtime error: signed integer overflow: 623487 * 536870912 cannot be represented in type 'int'
Fixes: 3594/clusterfuzz-testcase-minimized-4650622935629824
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 integer overflow in predict()

Fixes: runtime error: signed integer overflow: -2110708110 + -82837504 cannot be represented in type 'int'
Fixes: 3547/clusterfuzz-testcase-minimized-6009386439802880
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

ffmpeg: Fix stored encoder metadata with -bitexact

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

Merge commit '970c76f32283bddf3a5afd24fe52db7a96186244'

* commit '970c76f32283bddf3a5afd24fe52db7a96186244':
  mlp_parser: Drop in-parser downmix functionality

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

Merge commit 'dd3aa85b68c017c419acb0c39ff6aa890ce89e87'

* commit 'dd3aa85b68c017c419acb0c39ff6aa890ce89e87':
  aac_ac3_parser: Drop in-parser downmix functionality

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

Merge commit 'c7bf98d4372e9217c8d75d2ca1f298385909e4da'

* commit 'c7bf98d4372e9217c8d75d2ca1f298385909e4da':
  avprobe: use av_spherical_projection_name() to print spherical projection names
  dump: use av_spherical_projection_name() to print spherical projection names
  spherical: add functions to retrieve and request projection names

This commit is a noop, see
9033e8723c86ed31872b22bd576602d48e2b9d0e
2efb70c37992b16b589e7405cd36e2f15a34c8ec
2a2854f57842e843ae0d4bd0ebe617135b2f63a0

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

Merge commit '7acb90333a187b0e847b66f9d3511245423dc0ce'

* commit '7acb90333a187b0e847b66f9d3511245423dc0ce':
  vaapi: Add external control of allow-profile-mismatch
  lavc: Add flag to allow profile mismatch with hardware decoding
  vaapi_encode: Use gop_size consistently in RC parameters

This commit is a noop, see
b658b5399e5d75cec44b09d79b00f59fa83cca43
49ae8a5e87f99b15514f5ef075d801e9ea8374e9
38820631746f86a2252c71ed4a588eb679dd9f40

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

Merge commit '85c2bf9d2cc7dad3d5c6dc79f802a74efa8d8fce'

* commit '85c2bf9d2cc7dad3d5c6dc79f802a74efa8d8fce':
  avfoundation: update to use AVCodecParameters

This commit is a noop, see f1560dbb2a0089d63ca61a324673f7e2e30fb517

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

Merge commit '94c54d97e7f4fe90570c323803f2bdf6246c1010'

* commit '94c54d97e7f4fe90570c323803f2bdf6246c1010':
  mlp: Factor out channel layout subset checks

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

Merge commit 'cea7fd9afb8488a6c48f7d7ee38602e1fd3dd425'

* commit 'cea7fd9afb8488a6c48f7d7ee38602e1fd3dd425':
  mlp: Drop ff_ prefix from a static function

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

Merge commit '1f6e5dd4d4dd9c02371ad211049d27edb6d95286'

* commit '1f6e5dd4d4dd9c02371ad211049d27edb6d95286':
  mlp: Rename the channel layout variable

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

Merge commit 'bee89ed3ec61a4a0e72255ae6f4a4274d25b5ccd'

* commit 'bee89ed3ec61a4a0e72255ae6f4a4274d25b5ccd':
  matroskaenc: don't warn about unknown spherical metadata when there isn't any

This commit is a noop.

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

Merge commit '04b0f0e371ff81b682274b574fb465ba4395c09f'

* commit '04b0f0e371ff81b682274b574fb465ba4395c09f':
  mem: uninline av_malloc(z)_array()

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

Merge commit '3d69dd65c6771c28d3bf4e8e53a905aa8cd01fd9'

* commit '3d69dd65c6771c28d3bf4e8e53a905aa8cd01fd9':
  hevc: Add support for bitdepth 10 for IDCT DC

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

Merge commit '358adef0305618219522858e471edf7e0cb4043e'

* commit '358adef0305618219522858e471edf7e0cb4043e':
  hevc: Add NEON IDCT DC functions for bitdepth 8

See 03cecf45c134ebbaecb62505fe444ade423ea7dc

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

Merge commit 'd0ce0634e09403eecaea4283cf82a2c5cea7d1e1'

* commit 'd0ce0634e09403eecaea4283cf82a2c5cea7d1e1':
  clearvideo: Convert to the new bitstream reader

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 '189157c3fc8eeb691e3684b09d971eb5ddb47d5b'

* commit '189157c3fc8eeb691e3684b09d971eb5ddb47d5b':
  Add ClearVideo decoder

See a63496cc882428aefafc85d2f60e0908b020bffe

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

Merge commit 'c55f2ae2b8d4de473f945aebd6a7a89b6b83e87a'

* commit 'c55f2ae2b8d4de473f945aebd6a7a89b6b83e87a':
  avprobe: Handle unknown values for the color description

This commit is a noop.

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

avcodec/avcodec.h: remove doxy from the old bsf API functions

Make it clear that these are deprecated and the new API should be
used instead.

As a side effect, this slightly reduces differences with libav.

Reviewed-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>

Merge commit 'b812db66af160bfac2f64aef2ade4aed7eb76b12'

* commit 'b812db66af160bfac2f64aef2ade4aed7eb76b12':
  utvideodec: Fix gradient prediction when stride does not match width

This commit is a noop, see 49255370044cf4a58c81a88cb8206aee62086346

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

Merge commit '7c25523cc8e618e77dc84d960e41e9644eaf8c33'

* commit '7c25523cc8e618e77dc84d960e41e9644eaf8c33':
  utvideodec: Fix decoding odd sizes with interlaced video with some formats

See 9ef21a897c64417a0575cbc6fad6222f3163d103

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

Merge commit '8c0cadd17e98547d84e82111550caca4fb40ff8d'

* commit '8c0cadd17e98547d84e82111550caca4fb40ff8d':
  avplay: Do not try to allocate new frames when the player is closing

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

Merge commit '41262498ff89ef48e5ce4823b2fc2dc514a3eb0c'

* commit '41262498ff89ef48e5ce4823b2fc2dc514a3eb0c':
  configure: Fix sem_timedwait probe

This commit is a noop. see abb5efca263daff670ebb0a07af4fc902e457ade

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

Merge commit '44aafa3325ec09a120bdb81b8579b8dfd65b1693'

* commit '44aafa3325ec09a120bdb81b8579b8dfd65b1693':
  configure: Properly escape arguments in filter/filter_out helper functions

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

Merge commit '3889dfde9c0284b008c5cfcbc921c1d137fb81e2'

* commit '3889dfde9c0284b008c5cfcbc921c1d137fb81e2':
  dca: Move the downmix request check outside the loop

This commit is a noop.

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

Merge commit '2e1ab6a67dd5d9a30558d008aa6e541c68d76221'

* commit '2e1ab6a67dd5d9a30558d008aa6e541c68d76221':
  vc1dec: raise an error if sprite picture data is missing

This commit is a noop, see 30f680ee0a2707af9a649a0aa3fd951d18a25c05

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

Merge commit 'ecc5c4db2dd3a0f328d95df89daa59f78b4b2810'

* commit 'ecc5c4db2dd3a0f328d95df89daa59f78b4b2810':
  doc/examples/output: Cast pointer to the right (const) type

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

Merge commit 'f25609ff06b093d82548c92fc2031cd2e66c20d3'

* commit 'f25609ff06b093d82548c92fc2031cd2e66c20d3':
  doc/examples/filter_audio: const correctness

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

Merge commit '0fd0d4fd0a518e30ff23972828ad7cf7f35cfb9d'

* commit '0fd0d4fd0a518e30ff23972828ad7cf7f35cfb9d':
  swscale-test: const correctness

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

avdevice/libndi_newtek_dec: remove extra bracket

Fixes compilation broken in 2245476e5c451552f8f32a4a881f6b4968898a47

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

avdevice/decklink_dec: remove extra bracket

Fixes compilation broken in 2245476e5c451552f8f32a4a881f6b4968898a47

Fixes ticket #6791

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

x86/bswapdsp: add missing preprocessor wrappers for AVX2 functions

Fixes build with old nasm/yasm.

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

decode: avoid leaks on failure in ff_get_buffer()

If the get_buffer() call fails, the frame might have some side data
already set. Make sure it gets freed.

Merges Libav commit de77671438c24ffea93398c8dc885d4dd04477de.

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

lavc/avrndec: remove AV_CODEC_CAP_DR1, as it's broken

In the is_mjpeg case, the user's get_buffer2 callback is not called,
thus completely breaking the API.

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

libavformat/mov.c: use calculated dts offset when seeking in streams

Subtract the calculated dts offset from the requested timestamp before
seeking. This fixes an error "Error while filtering: Operation not
permitted" observed with a short file which contains only one key frame
and starts with negative timestamps.

Then, av_index_search_timestamp() returns a valid negative timestamp,
but mov_seek_stream bails out with AVERROR_INVALIDDATA.

Fixes ticket #6139.

Signed-off-by: Jonas Licht <jonas.licht@fem.tu-ilmenau.de>
Signed-off-by: Peter Große <pegro@friiks.de>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

movenc: add m4b to ipod format extensions

m4b is commonly used as extension for m4a audiobook files.
The format is exactly the same. The only thing that differs
is the extension.

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

fate: Add bitexact flag to the invalid-elst test.

Fixes fate-mov-invalid-elst-entry-count on a ppc system.

fate: Fix some of the fitsdec tests on big-endian.

avdevice: remove usage of deprecated setter and getter functions

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

ffmpeg: remove usage of deprecated getter functions

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

ffprobe: remove usage of deprecated getter functions

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

lavf/aviobuf: return EINVAL when reading from a write-only context.

Signed-off-by: Nicolas George <george@nsup.org>

lavf/avio: temporarily accept 0 as EOF.

Print a warning to let applicatios fix their use.
After a deprecation period, check with a low-level assert.
Also make the constraint explicit in the doxygen comment.

Signed-off-by: Nicolas George <george@nsup.org>

examples/avio_reading: return AVERROR_EOF at EOF.

Signed-off-by: Nicolas George <george@nsup.org>

avcodec/jpeglsdec: Check for end of bitstream in ls_decode_line()

Fixes: 1773/clusterfuzz-testcase-minimized-4832523987189760
Fixes: Timeout
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/jpeglsdec: Check ilv for being a supported value

Fixes: 1773/clusterfuzz-testcase-minimized-4832523987189760
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

dashenc: copy stream frame rate to output stream

Leads to setting of DefaultDuration in Matroska muxer based on frame rate instead of timebase.
Fixes playback in Chrome.

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

dashenc: Don't output frameRates at both AS and Representation-level. Only output maxFrameRate at AS-level.

Fix conformance regarding section "3.2.4. Presence of Attributes and
Element" of the "Guidelines for Implementation: DASH-IF Interoperability
Points V4.1" (http://dashif.org/guidelines/)

Signed-off-by: Anton Schubert <ischluff@mailbox.org>
Signed-off-by: Peter Große <pegro@friiks.de>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

dashenc: allow AdaptationSets to have different average frame rates

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

dashenc: don't write header data before the first packet arrives

Fixes: 1b8ef01f04ab ("dashenc: add webm support")
Signed-off-by: Peter Große <pegro@friiks.de>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

dashenc: set DASH related options for the subsequent matroska muxer when using webm

This patch is inspired by the ffmpeg webm_chunk muxer and fixes that all resulting
tracks have the same track number.

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

dashenc: fix bitrate estimation with correct scaling

Signed-off-by: Anton Schubert <ischluff@mailbox.org>
Signed-off-by: Peter Große <pegro@friiks.de>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavf/rtpenc: Add support for 24 bit pcm encoding as defined by RFC 3190.

Fixes ticket #6770.

libavcodec/bswapdsp : add AVX2 func for bswap_buf (swap uint32_t)

lsws/yuv2rgb: Fix yuva2rgb32 on big endian hardware.

avformat/avformat.h: fix pairs typo

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

avformat: deprecate getters and setters for AVFormatContext and AVStream fields

The fields can be accessed directly, so these are not needed anymore.

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

avformat: move public AVStream fields up in the struct

Remove the silly second notice.

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

avutil/frame: deprecate getters and setters for AVFrame fields

The fields can be accessed directly, so these are not needed anymore.

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

avcodec/magicyuvenc: improve compression

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

libavcodec/magicyuv : remove unneed variable assignment

avformat/hlsenc: reindent hlsenc code

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

avformat/hlsenc: fix missing first segment bug in fmp4 mode

fix ticket id: #6776
fix code logic error, need not check first segment.

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

avformat/hlsenc: fix base_output_dirname is null when basename_size is 0 bug

fix ticket id: #6777
when use argument hls_segment_filename, the basename_size will be 0

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

avformat: remove use of deprecated AVFMT_FLAG_KEEP_SIDE_DATA flag

It has no effect whatsoever since the major bump.
Replace the flag's documentation to reflect this as well.

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

lavfi/palettegen: Fix compilation after 1da8c4e.

Found-by: James Almer

lavfi/palettegen: Allow setting the background colour.

Merge commit '8fd8f91e47f33cd82371a97ac81afc476144964f'

* commit '8fd8f91e47f33cd82371a97ac81afc476144964f':
  qsvenc: Make sure the interlaced encoding works
  qsvenc: Use MFXVideoENCODE_Query() to update the parameters

This commit is a noop, see
dd8319767e1674d030b5d803c43029e67db5efb8
ae5b67ee64bec00c52881ff8426fddaf477341c1

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

Merge commit 'e00db9f78bb475ed5103364f61892f4e75ef89ba'

* commit 'e00db9f78bb475ed5103364f61892f4e75ef89ba':
  checkasm: hevc: Add a hevc_ prefix to the add_residual functions

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

Merge commit '1731c3530bffb876deb9e00dfffdf9841a8412cd'

* commit '1731c3530bffb876deb9e00dfffdf9841a8412cd':
  mm: Skip unexpected audio packets

This commit is a noop. The generic code already drops packets from
non-existant streams.

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

Merge commit 'f2c469b73f8ac3dd6a98d38281f61b68ea6dd336'

* commit 'f2c469b73f8ac3dd6a98d38281f61b68ea6dd336':
  mjpeg: Report non-3 component rgb lossless as not supported

[15:56:27] <jamrial> michaelni: can't reproduce the failure f2c469b73f is trying to "fix"
[15:56:30] <jamrial> do we support what that commit claims is unsupported?
[16:56:54] <michaelni> jamrial, yes, we support lossless rgba

This commit is a noop.

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

fate: change fate-ffmpeg-attached_pics to encode to pcm_s16le

Previously alac encoder was used, from a first glance I thought it is bitexact,
but it turns out it is using floating point arithmetic as well, so probably it
is not. Fixes fate failures on mingw32/64.

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

Merge commit '279e3aaa14daba6b7a37d75f3fb6e29c732d123f'

* commit '279e3aaa14daba6b7a37d75f3fb6e29c732d123f':
  flv: Validate the packet size

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

avformat/ac3dec: Check buf2 before adding 16 in ac3_eac3_probe()

This is needed since e0250cf3651e6417e0117486a7816b45fb2d34cd as that uses end-buf2
Note, there are more than 16 bytes allocated beyond "end"

Fixes: regression (segfault) with probetest
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavf/mov.c: Fix parsing of edit list atoms with invalid elst entry count.

Signed-off-by: Sasi Inguva <isasi@google.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

dashenc: move UTCTiming element to the end of the manifest

Required by comformance XSD [1].

[1] https://github.com/Dash-Industry-Forum/Conformance-and-reference-source/blob/master/conformance/MPDValidator/schemas/DASH-MPD.xsd

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

dashenc: use av_dict_set_int instead of static function

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

Merge commit '4809781586d1c956005f72946a2aab5915eab350'

* commit '4809781586d1c956005f72946a2aab5915eab350':
  travis: Exclude gcc for the macOS target

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

Merge commit '5352802da81f2083e65d466612e639a4e6e5530e'

* commit '5352802da81f2083e65d466612e639a4e6e5530e':
  dca: Account for lfe when checking for the channel count

This commit is a noop.

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

Merge commit '37f573543c4fd7f44339e04d8d15b95118493ddd'

* commit '37f573543c4fd7f44339e04d8d15b95118493ddd':
  swscale: Convert the check check_image_pointers helper to a macro

[14:52:50] <jamrial> michaelni, ubitux: 37f573543c yay or nay? we don't have the warnings mentioned there because we cast the argument
[14:54:10] <@ubitux> i don't really see the improvement, no opinion
[15:04:14] <michaelni> jamrial, functions are better/cleaner than macros
[15:04:20] <michaelni> IMHO that is

This commit is a noop.

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

Merge commit 'f56fa95cd13f627891a1bfb66bf61b971b9e0238'

* commit 'f56fa95cd13f627891a1bfb66bf61b971b9e0238':
  swscale: Do not shift negative values directly

14:30:21] <jamrial> michaelni: do you prefer how f56fa95cd1 (libav) handles left shift of negative value, or how we're currently doing it?
[14:30:31] <jamrial> ours is cleaner looking, IMO
[14:30:57] <@ubitux> i agree
[14:31:13] <@ubitux> if we didn't forget any, i think this commit is better noop'ed
[14:31:57] <jamrial> ubitux: we even have a bunch of other cases this commit doesn't handle that i'd have to manually change, so yeah, i'm leaning towards nooping it
[14:34:04] <michaelni> jamrial, +1, ours as its cleaner

This commit is a noop.

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

avfilter/zscale: fix memory leak

avfilter/zscale: fix segfault on library error

Merge commit '378460fef1e563704bc5a307ba748a2b819cc09d'

* commit '378460fef1e563704bc5a307ba748a2b819cc09d':
  utvideodec: Support for gradient prediction
  utvideodec: Reuse the huffyuv add_left
  utvideodec: Support ULY4 and ULH4
  utvideodec: Support UQRA and UQRG
  swscale: Add gbrap10 output
  swscale: Add input support for gbrap10 pixel format
  pixfmt: Add gbrap10 pixel format
  utvideodec: Support UQY2

This commit is a noop, see
115e63c8d6791ebdc921caf40c3def024690af2e
8100426fe4ba41b74404cbbd83184afa3239f64b
9d30690f20343ed3057fe4abf82009598d320ce5
b828788d58cb6f55e0bffa4682d75ddf84b4214b
84efdabc94251f1f6f483eb737e95d53376ae921
c62cb9bf5a50df416532afd8d5f79ebade174ac7
ea93052db3594f93f2d10be085a770184da0513d
faa94a576f5f3de10fc7016e0d94229faa1c2159

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

Merge commit '9e4a5eb51b9f3b2bff0ef08e0074b7fe4893075d'

* commit '9e4a5eb51b9f3b2bff0ef08e0074b7fe4893075d':
  avformat: Free the internal codec context at the end

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

Merge commit '19c5fa9eea47d45b5eedce3bff4edeaf83fb3942'

* commit '19c5fa9eea47d45b5eedce3bff4edeaf83fb3942':
  golomb: Simplify get_ur_golomb_jpegls

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

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

lavfi/paletteuse: fix debug_kdtree after aba926e7d

configure: add missing llvidencdsp dependency to MagicYUV encoder

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

avcodec: bump minor after addition of MagicYUV encoder

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

lavfi/paletteuse: fix debug_mean_error after aba926e7d

lavfi/paletteuse: fix debug_accuracy after aba926e7d

avcodec: add MagicYUV encoder

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

lavfi/paletteuse: simplify color get function prototypes

lavfi/paletteuse: move "new" option before debugging options

lavfi/paletteuse: fix to support transparency

This patch enables paletteuse to identify the transparency in incoming
video and tag transparent pixels on outgoing video with the correct
index from the palette.

This requires tracking the transparency index in the palette,
establishing an alpha threshold below which a pixel is considered
transparent and above which the pixel is considered opaque, and
additional changes to track the alpha value throughout the conversion
process.

This change is a partial fix for https://trac.ffmpeg.org/ticket/4443
However, animated GIFs are still output incorrectly due to a bug
in gif optimization which does not correctly handle transparency.

Signed-off-by: Clément Bœsch <u@pkh.me>

lswr/swresample: Mention the actually supported formats when erroring out.

Fixes ticket #6779.

Merge commit '1fe858136b315796dd5349f3b4448a29d1bd6fa1'

* commit '1fe858136b315796dd5349f3b4448a29d1bd6fa1':
  utvideodec: Prevent possible signed overflow

This commit is a noop, see e86444b19d0b63c098298243fb20fd577f34cf34

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