There is no need to duplicate the input reference, since a filter should
not pass along a reference which is later modified. If this happens the
filter passing the reference should be fixed.

Also improve consistency with start_frame() of settb, allowing a pending
factorization.

The documentation was erroneously removed in 7cf9aadd.

Fixes Ticket1586
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes Ticket1502
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

* qatar/master:
  avformat: Drop pointless "format" from container long names
  swscale: bury one more piece of inline asm under HAVE_INLINE_ASM.
  wv: K&R formatting cosmetics
  configure: Add missing descriptions to help output
  h264_ps: declare array of colorspace strings on its own line.
  fate: amix: specify f32 sample format for comparison
  tiny_psnr: support 32-bit float samples
  eamad/eatgq/eatqi: call special EA IDCT directly
  eamad: remove use of MpegEncContext
  mpegvideo: remove unnecessary inclusions of faandct.h
  af_asyncts: avoid overflow in out_size with large delta values
  af_asyncts: add first_pts option

Conflicts:
	configure
	libavcodec/eamad.c
	libavcodec/h264_ps.c
	libavformat/crcenc.c
	libavformat/ffmdec.c
	libavformat/ffmenc.c
	libavformat/framecrcenc.c
	libavformat/md5enc.c
	libavformat/nutdec.c
	libavformat/rawenc.c
	libavformat/yuv4mpeg.c
	tests/tiny_psnr.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

When start_frame has returned, the buffer reference's PTS
may have been modified by another filter.

Based on code by: carl
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

broken by: cbb2c95e
i had fixed that commit but pushed from an old branch
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

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

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

Also bump minor version and add APIchanges entry.

While here refactor code.
Signed-off-by: Paul B Mahol <onemda@gmail.com>

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

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

* qatar/master:
  flvdec: remove spurious use of stream id
  lavf: deprecate r_frame_rate.
  lavf: round estimated average fps to a "standard" fps.

Conflicts:
	ffmpeg.c
	ffprobe.c
	libavformat/avformat.h
	libavformat/electronicarts.c
	libavformat/flvdec.c
	libavformat/rawdec.c
	libavformat/utils.c
	tests/ref/fate/iv8-demux
Merged-by: Michael Niedermayer <michaelni@gmx.at>

The new name is more meaningful in that context.

Keeping the set options in the context when the component is still used
can be useful for debugging purposes.

Avoid to fill the same picture again and again with the same content.

Optimize computation, and provides an example for the use of the
AV_PERM_REUSE permission flag.

Signed-off-by: Mans Rullgard <mans@mansr.com>

Signed-off-by: Mans Rullgard <mans@mansr.com>

* commit 'fe1c1198':
  lavf: use dts difference instead of AVPacket.duration in find_stream_info()
  avf: introduce nobuffer option
  fate: make yadif tests consistent across systems
  vf_hqdn3d: support 9 and 10bit colordepth
  vf_hqdn3d: reduce intermediate precision
  vf_hqdn3d: simplify and optimize
  factor identical ff_inplace_start_frame out of two filters
  vf_hqdn3d: cosmetics
  avprobe/avconv: fix tentative declaration compile errors on MSVS.

Conflicts:
	doc/APIchanges
	ffmpeg.c
	ffprobe.c
	libavformat/avformat.h
	libavformat/options_table.h
	libavformat/utils.c
	libavformat/version.h
	tests/fate/filter.mak
	tests/ref/fate/filter-yadif-mode0
	tests/ref/fate/filter-yadif-mode1
Merged-by: Michael Niedermayer <michaelni@gmx.at>

These decoders use a special non-MPEG2 IDCT.  Call it directly
instead of going through dsputil.  There is never any reason
to use a regular IDCT with these decoders or to use the EA IDCT
with other codecs.

This also fixes the bizarre situation of eamad and eatqi decoding
incorrectly if eatgq is disabled.
Signed-off-by: Mans Rullgard <mans@mansr.com>

There is no sense in pulling in this monster struct just for
a handful of fields.  The code does not call any functions
expecting an MpegEncContext.
Signed-off-by: Mans Rullgard <mans@mansr.com>

Signed-off-by: Mans Rullgard <mans@mansr.com>

Improves subjective quality

Formula and testing by: kamedo2 <fujisakihir90@yahoo.co.jp>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

This allows for padding/trimming at the start of stream. By default, no
assumption is made about the first frame's expected pts, so no padding or
trimming is done.

We match streams by codec id now.