* commit '7fa00653a550c0d24b3951c0f9fed6350ecf5ce4':
  h264: add a parameter to the FIELD_PICTURE macro.
  h264: add a parameter to the FRAME_MBAFF macro.

Conflicts:
	libavcodec/h264.c
	libavcodec/h264_loopfilter.c
	libavcodec/h264_refs.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'da6be8fcec16a94d8084bda8bb8a0a411a96bcf7':
  h264: add a parameter to the MB_FIELD macro.
  h264: add a parameter to the MB_MBAFF macro.

Conflicts:
	libavcodec/h264.c
	libavcodec/h264_cabac.c
	libavcodec/h264_cavlc.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '48d0fd2d62a476e1db9298163f1fc0abae26cc67':
  h264: merge common_init() into ff_h264_decode_init.

Conflicts:
	libavcodec/h264.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '25408b2a0660c1e6c8555559c4ed71dff2ede31e':
  h264: make ff_h264_frame_start static.

Conflicts:
	libavcodec/h264.c
	libavcodec/h264.h

Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '1c4073efd24164ac6eaa52c544f5cdb0e5f6aee5':
  fate: add tests for h264 decoder reinit
  h264: fix bit depth changes with frame threading

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

This way it does not look like a constant.

This way it does not look like a constant.

This way it does not look like a constant.

This way it does not look like a constant.

There is no point in keeping those separate.

It is not called from outside h264.c

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

The versions of the headers are the same as those provided with x264
for consistency's sake.

Dynamically loads the library itself, rather than through VfW.
Supports AvxSynth on Linux and OS X.

Supports the new colorspaces added in AviSynth 2.6 when used
with AviSynth 2.6.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

The metadata must be set before saving the frame to avci->to_free,
otherwise it will leak.

Signed-off-by: Hendrik Leppkes <h.leppkes@gmail.com>
Signed-off-by: Clément Bœsch <ubitux@gmail.com>

This should fix compilation on openbsd

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* cigaes/master:
  lavfi/vf_yadif: use standard options parsing.
  lavfi/vf_unsharp: use standard options parsing.
  lavfi/vf_transpose: use standard options parsing.
  lavfi/vf_pad: use standard options parsing.
  lavfi/vf_fps: use standard options parsing.
  lavfi/vf_fade: use standard options parsing.
  lavi/vf_drawbox: use standard options parsing.
  lavfi/vf_delogo: use standard options parsing.
  lavfi/vf_decimate: use standard options parsing.
  lavfi/vf_crop: use standard options parsing.
  lavfi/af_volume: use standard options parsing.
  lavfi/vf_tile: use standard options parsing.
  lavfi/avf_concat: use standard options parsing.
  lavfi: add common code to handle options parsing.
  lavf/vobsub: free index pseudo-packet.
  ffmpeg: fix freeing of sub2video frame.
  lavfi: add sine audio source.
  lavu/opt: add AV_OPT_TYPE_DURATION.
  lavfi/concat: fix silence duration computation.
  lavf/concatdec: support seeking.

Merged-by: Michael Niedermayer <michaelni@gmx.at>

This patch adds the enums for the ContentEncryption elements.
This patch also adds support for parsing the ContentEncKeyID. The
ContentEncKeyID is then base64 encoded and stored in the stream's
metadata.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Inside send_silence(), delta_ts was used inconsistently:
sometimes as the new value, sometimes as the old value.

Consistently use it as the new value, and add an argument
to know the last segment duration.

Fixes: otonajoshi_avformat_tries_to_load_whole.ts
Reported-by: JEEB (on IRC)
Thanks-to: nevcairiel
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* cigaes/master:
  lavu/frame: use channels rather than channel_layout.
  lavf: avformat_seek_file(): validate stream_index.
  lavf/concatdec: fix possible leak in case of malloc failure.
  lavfi/buffersink: check av_frame_ref() failure.

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Floating point FFT (nips optimized) breaks when hard coded tables are
not enabled because MIPS optimization of floating point FFT uses only
ff_init_ff_cos_tabs(16) which is not enabled by default in that case.

This patch is fixing it.

Signed-off-by: Nedeljko Babic <nbabic@mips.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fix CID 747737.