When coding lossless jpeg the priv context will be pointing to LJpegEncContext
rather than MpegEncContext, which the function expects.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

* commit '2575b191':
  build: Add separate flags for checkheaders targets
Merged-by: Clément Bœsch <u@pkh.me>

* commit '984e50e7':
  build: Add -Wpedantic and -Wextra to extra warning flags
Merged-by: Clément Bœsch <u@pkh.me>

* commit '33a2b73b':
  mpeg4audio: correctly propagate meaningful error values

This commit is a noop, see 50b14539Merged-by: Clément Bœsch <u@pkh.me>

* commit 'd82d5379':
  mmaldec: initialize refcount using atomic_init()
Merged-by: Clément Bœsch <u@pkh.me>

* commit '518a2247':
  configure: Be more helpful when requiring options set
Merged-by: Clément Bœsch <u@pkh.me>

* commit 'a91f1023':
  examples: fix a typo in an error message

This commit is a noop, see 3aa1ff30Merged-by: Clément Bœsch <u@pkh.me>

* commit '706af922':
  lavu: Document the color properties enumeration values origin
Merged-by: Clément Bœsch <u@pkh.me>

* commit '5168026a':
  options_table: Do not rely on enum size as option bound
Merged-by: Clément Bœsch <u@pkh.me>

* commit 'ff9db5cf':
  lavc: Use a stricter check for the color properties values
Merged-by: Clément Bœsch <u@pkh.me>

* commit '0a35f128':
  cabac: x86: Give optimizations header a more meaningful name
Merged-by: Clément Bœsch <u@pkh.me>

* commit 'cad42fad':
  aarch64: vp9itxfm: Skip empty slices in the first pass of idct_idct 16x16 and 32x32

This commit is a noop, see 8b11a89cMerged-by: Clément Bœsch <u@pkh.me>

* commit '9c8bc74c':
  arm: vp9itxfm: Skip empty slices in the first pass of idct_idct 16x16 and 32x32

This commit is a noop, see 388f6e67Merged-by: Clément Bœsch <u@pkh.me>

* commit '3c87039a':
  arm: vp9itxfm: Only reload the idct coeffs for the iadst_idct combination

This commit is a noop, see ecd343aaMerged-by: Clément Bœsch <u@pkh.me>

* commit 'c4c5f538':
  vp9dsp: add DC only versions for idct/idct.

This commit is a noop, see 64821f5aMerged-by: Clément Bœsch <u@pkh.me>

* commit 'e4382a4a':
  hevc: Eliminate pointless variable indirection

This commit is a noop, the code is different in FFmpeg.
Merged-by: Clément Bœsch <u@pkh.me>

* commit '5c890225':
  hevc: Drop pointless av_unused attribute
Merged-by: Clément Bœsch <u@pkh.me>

* commit '0983f911':
  metasound: Drop unused tables

This commit is mostly a noop, see
276a8666
e8319f60Merged-by: Clément Bœsch <u@pkh.me>

* commit 'c21d78a9':
  configure: Integrate X11 checks into vaapi/vdpau checks
Merged-by: Clément Bœsch <u@pkh.me>

* commit '8b56dbe7':
  configure: Do not add newlines in filter()/filter_out() functions
Merged-by: Clément Bœsch <u@pkh.me>

* commit '9254344e':
  configure: Move hardware-accelerated codec deps out of hwaccel section
Merged-by: Clément Bœsch <u@pkh.me>

* commit 'd4f2a681':
  configure: MMAL-related decoders should depend on, not select, mmal
Merged-by: Clément Bœsch <u@pkh.me>

Saves around 5kbps.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

Apparently its only use is to enable comfort noise/error recovery.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

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

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

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

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

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

or if x/y go beyond padded area.

This is mostly useful when paired with the aspect option.
Defaults aren't changed.

Idea for this was taken from mpv's soon-to-be-removed expand vf.
Reviewed-by: Paul B Mahol <onemda@gmail.com>

It was addressed in b8f26779

While they shouldn't be present, they are harmless if they are.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>

This merges commits 8e2ea691 and
096a8eff by Anton Khirnov, with the
following change:

- extract_extradata_check() is added to know if the codec is supported
by the bsf before trying to initialize it. This behaviour is similar to
the old AVCodecParser.split checks.

The FATE reference changes are due to the filtered out NAL units that
the old AVCodecParser.split implementation left alone.
Decoding is unchanged as the functions that parse extradata simply
ignored said unnecessary NAL units.
Signed-off-by: James Almer <jamrial@gmail.com>

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

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

The av_log() is done outside the lock, but this way the accesses to the
field (reads and writes) are always protected by a mutex. The av_log()
is not run inside the lock context because it may involve user callbacks
and doing that in performance-sensitive code is probably not a good idea.

This should fix occasional tsan warnings when running fate-h264, like:

WARNING: ThreadSanitizer: data race (pid=10916)
  Write of size 4 at 0x7d64000174fc by main thread (mutexes: write M2313):
    #0 update_context_from_user src/libavcodec/pthread_frame.c:335 (ffmpeg+0x000000df7b06)
[..]
  Previous read of size 4 at 0x7d64000174fc by thread T1 (mutexes: write M2311):
    #0 ff_thread_await_progress src/libavcodec/pthread_frame.c:592 (ffmpeg+0x000000df8b3e)

I'm hoping that this will address the remaining tsan fate-h264 issues:

WARNING: ThreadSanitizer: data race (pid=24478)
  Read of size 8 at 0x7dbc0001c828 by main thread (mutexes: write M3243):
    #0 ff_h264_ref_picture src/libavcodec/h264_picture.c:107 (ffmpeg+0x0000013b78d8)
[..]
  Previous write of size 1 at 0x7dbc0001c82e by thread T2 (mutexes: write M3245):
    #0 ff_h264_direct_ref_list_init src/libavcodec/h264_direct.c:137 (ffmpeg+0x000001382c93)

But I'm not sure because I haven't been able to reproduce locally.

Fix decoding frame.jpg from ticket #267

Regression since 9c7ee374 / 212c6a1d