This came up while debugging a problem with mpv:
https://github.com/mpv-player/mpv/issues/4155Signed-off-by: wm4 <nfxjfg@googlemail.com>

* commit 'e3f941cb':
  checkasm: add a test for HEVC IDCT
Merged-by: Clément Bœsch <u@pkh.me>

* commit '9b2ccafb':
  aarch64: Add missing sign extension in ff_h264_idct8_add_neon
Merged-by: Clément Bœsch <u@pkh.me>

* commit 'cbd84b8a':
  nvenc: Fix error log

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

* commit 'da284837':
  nvenc: Force high_444 profile for 444 input

This commit is a noop, see 20abda6bMerged-by: Clément Bœsch <u@pkh.me>

* commit 'e4128c08':
  Revert "hevc: x86: Refactor IDCT macro declarations"

So apparently this was technically correct be reverted due to
authorship. Reverted as well in FFmpeg for now...

See http://lists.libav.org/pipermail/libav-devel/2016-October/079560.htmlMerged-by: Clément Bœsch <u@pkh.me>

* commit '5801f9ed':
  h264_intrapred: x86: Update comments left behind in 95c89da3Merged-by: Clément Bœsch <u@pkh.me>

* commit '20abcaa2':
  configure: #include stdint.h as part of libxavs test

This commit is a noop, see 20c4fb2eMerged-by: Clément Bœsch <u@pkh.me>

* commit 'd9dccc03':
  hevc: x86: Refactor IDCT macro declarations
Merged-by: Clément Bœsch <u@pkh.me>

* commit 'be630b1e':
  d3d11va: Use the proper decoding slice index

This commit is a noop, see 9b462a0bMerged-by: Clément Bœsch <u@pkh.me>

* commit '715f139c': (23 commits)
  vp9lpf/x86: make filter_16_h work on 32-bit.
  vp9lpf/x86: make filter_48/84/88_h work on 32-bit.
  vp9lpf/x86: make filter_44_h work on 32-bit.
  vp9lpf/x86: make filter_16_v work on 32-bit.
  vp9lpf/x86: make filter_48/84_v work on 32-bit.
  vp9lpf/x86: make filter_88_v work on 32-bit.
  vp9lpf/x86: make filter_44_v work on 32-bit.
  vp9lpf/x86: save one register in SIGN_ADD/SUB.
  vp9lpf/x86: store unpacked intermediates for filter6/14 on stack.
  vp9lpf/x86: move variable assigned inside macro branch.
  vp9lpf/x86: simplify ABSSUM_CMP by inverting the comparison meaning.
  vp9lpf/x86: remove unused register from ABSSUB_CMP macro.
  vp9lpf/x86: slightly simplify 44/48/84/88 h stores.
  vp9lpf/x86: make cglobal statement more conservative in register allocation.
  vp9lpf/x86: save one register in loopfilter surface coverage.
  vp9lpf/x86: add ff_vp9_loop_filter_[vh]_44_16_{sse2,ssse3,avx}.
  vp9lpf/x86: add ff_vp9_loop_filter_h_{48,84}_16_{sse2,ssse3,avx}().
  vp9lpf/x86: add an SSE2 version of vp9_loop_filter_[vh]_88_16
  vp9lpf/x86: add ff_vp9_loop_filter_[vh]_88_16_{ssse3,avx}.
  vp9lpf/x86: add ff_vp9_loop_filter_[vh]_16_16_sse2().
  ...

All these commits are cherry-picks from FFmpeg. Maybe some slight
differences sneaked in but the Libav codebase still differs too much
with our own to make a proper diff. This merge is a noop.
Merged-by: Clément Bœsch <u@pkh.me>

* commit 'dc3fe45f':
  fate: Add test for rscc palette
Merged-by: Clément Bœsch <u@pkh.me>

* commit 'c19830aa':
  rscc: Support palette format

See 11777eb8Merged-by: Clément Bœsch <u@pkh.me>

* commit 'b8d5070d':
  avcodec: Document AV_PKT_DATA_PALETTE side data type
Merged-by: Clément Bœsch <u@pkh.me>

* commit '497c0879':
  avidec: Set palette alpha as fully opaque

This commit is a noop, see 64cafe34Merged-by: Clément Bœsch <u@pkh.me>

* commit 'bad4aad4':
  avidec: Do not special case palette on big-endian

This commit is a noop, see 64cafe34Merged-by: Clément Bœsch <u@pkh.me>

* commit '310c55f1':
  pixfmt: Document alternative names for smpte 431 and 432
Merged-by: Clément Bœsch <u@pkh.me>

* commit '5a5df90d':
  vaapi_h265: Add main 10 encode support

This commit is a noop, see b9514756Merged-by: Clément Bœsch <u@pkh.me>

* commit 'eaaaabf6':
  hwcontext_vaapi: Enable P010 support

This commit is a noop, see 7e0623b7Merged-by: Clément Bœsch <u@pkh.me>

* commit 'b8cac1e8':
  vaapi_h265: Fix buffering parameters
  vaapi_h265: Fix slice header writing
  vaapi_h264: Write bitstream restriction fields
  vaapi_h264: Fix CFR mode with frame_rate set in AVCodecContext
  vaapi_encode: Decide on GOP setup before initialising sequence parameters

This merge is a noop, see:
  30ebabca vaapi_h265: Fix buffering parameters
  ae0230cc vaapi_h265: Fix slice header writing
  6796e6ea vaapi_h264: Write bitstream restriction fields
  658c5afa vaapi_h264: Fix CFR mode with frame_rate set in AVCodecContext
  ded1859d vaapi_encode: Decide on GOP setup before initialising sequence parameters
Merged-by: Clément Bœsch <u@pkh.me>

* commit '5cc0057f':
  lavu: remove the custom atomic API

This commit is a noop. The removal is postponed until all usages in
FFmpeg are dropped as well. A patchset is on discussion on the
mailing-list:
http://ffmpeg.org/pipermail/ffmpeg-devel/2017-March/209003.htmlMerged-by: Clément Bœsch <u@pkh.me>

This supports retrieving the device from a provided hw_frames_ctx, and
automatically creating a hw_frames_ctx if hw_device_ctx is set.

The old API is not deprecated yet. The user can still use
av_vdpau_bind_context() (with or without setting hw_frames_ctx), or use
the API before that by allocating and setting hwaccel_context manually.

Cherry-picked from Libav commit 1a7ddba5.
(Adds missing APIchanges entry to the Libav version.)
Reviewed-by: Mark Thompson <sw@jkqxz.net>

This "reuses" the flags introduced for the av_vdpau_bind_context() API
function, and makes them available to all hwaccels. This does not affect
the current vdpau API, as av_vdpau_bind_context() should obviously
override the AVCodecContext.hwaccel_flags flags for the sake of
compatibility.

Cherry-picked from Libav commit 16a163b5.
Reviewed-by: Mark Thompson <sw@jkqxz.net>

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

In this case, the user only supplies a device and the frame context
is allocated internally by lavc.

(cherry picked from commit 5dd9a4b8)

libavcodec/vaapi.h:58:1: warning: attribute 'deprecated' is ignored, place it after "struct" to apply attribute to type declaration [-Wignored-attributes]

(cherry picked from commit ed6a891c)
Signed-off-by: Mark Thompson <sw@jkqxz.net>

Should fix Chromium issue 704064.
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

* commit '59c70227':
  pthread_frame: use atomics for frame progress

This commit is a noop, see b6587421Merged-by: Clément Bœsch <u@pkh.me>

* commit '64a31b28':
  pthread_frame: use atomics for PerThreadContext.state

This commit is a noop, see 74926269Merged-by: Clément Bœsch <u@pkh.me>

* commit 'db273325':
  pthread_frame: use a thread-safe way for signalling threads to die

This commit is a noop, see 4845f072Merged-by: Clément Bœsch <u@pkh.me>

* commit '8385ba53':
  mmaldec: convert to stdatomic
Merged-by: Clément Bœsch <u@pkh.me>

* commit '3a165c18':
  v4l2: convert to stdatomic
Merged-by: Clément Bœsch <u@pkh.me>

* commit '27079a42':
  buffer: convert to stdatomic
Merged-by: Clément Bœsch <u@pkh.me>

USE_ATOMICS is only set if there is no thread implementation enabled, in
which case you can't expect any lock mechanism from FFmpeg.

This is also conflicting with the incoming use of stdatomic.

* commit 'eb34d403':
  Add a compat dummy stdatomic.h used when threading is disabled
  Add a compat stdatomic.h implementation based on pthreads
  Add a compat stdatomic.h implementation based on suncc atomics
  Add a compat stdatomic.h implementation based on windows atomics
  Add a compat stdatomic.h implementation based on GCC atomics

This merge is a noop, see:
  41e891e8 Add a compat dummy stdatomic.h used when threading is disabled
  74b5f108 Add a compat stdatomic.h implementation based on pthreads
  70faadc8 Add a compat stdatomic.h implementation based on suncc atomics
  c91e72ed Add a compat stdatomic.h implementation based on windows atomics
  3359eede Add a compat stdatomic.h implementation based on GCC atomics
Merged-by: Clément Bœsch <u@pkh.me>