* commit '63bac48f':
  x86: dsputil: Move rv40-specific functions where they belong

Conflicts:
	libavcodec/x86/dsputil_mmx.c
	libavcodec/x86/dsputil_mmx.h
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '92f8e06e':
  x86: dsputil hpeldsp: Move shared template functions into separate object

Conflicts:
	libavcodec/x86/Makefile
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '7edaf4ed':
  x86: rnd_template: Eliminate pointless OP_AVG macro indirection
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'db9aee6c':
  oma: properly forward errors in oma_read_packet
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '6f98508e':
  oma: return meaningful errors
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'e55e8b2c':
  lavc: do not leak frames in avcodec_decode_audio4

Conflicts:
	libavcodec/utils.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'e0c53c34':
  nut: use meaningful error values
  FATE: use a less ambiguous end time for filter-trim-time test

Conflicts:
	libavformat/nutdec.c
	tests/fate/filter-video.mak
Merged-by: Michael Niedermayer <michaelni@gmx.at>

This is in line with the new name for the libavfilter variable.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* commit '9779f930':
  configure: add dependencies to lavfi .pc file

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

* commit 'bce70f81':
  configure: make avconv select (a)trim filters

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

* commit '481575d2':
  avconv: remove -deinterlace

Conflicts:
	Changelog
	doc/ffmpeg.texi
	ffmpeg.c
	ffmpeg_opt.c

The merge removes the documentation for the deprecated -deinterlace
the code itself has already been replaced a while ago by code
that inserts yadif, that insertion code is left in place.
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'a97d8cc1':
  indeo3: use unaligned reads on reference blocks.
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '70f96615':
  jpeg2000dec: initialize static data just once
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '45a1694f':
  jpeg2000dec: cleanup properly after decoding errors

Conflicts:
	libavcodec/jpeg2000dec.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

MSVC complains about the 32bits addressing, while mingw/gcc does not.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Also extend documentation accordingly.

In particular, mention the fact that the amortized element addition cost
is constant.

Fixes a few green artifacts at the top
Fixes rest of Ticket 2535
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes green trash
Fixes part of Ticket2535
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* cigaes/master:
  lavfi: add comments to explain the negotiation loop.
  lavfi: fix filter format negotiation loop.
  ffmpeg: move a local variable definition later.
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Fixes division by zero

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes integer overflow and out of array accesses

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes out of array accesses

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes out of array reads

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes out of array reads

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes array overread

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

query_formats() returning EAGAIN is not considered a progress
in the format negotiation.

If the filter returns EAGAIN but did set some of its formats
lists, it could be considered a partial success and counted
as progress in the negotiation. Not counting it is not a
problem because it currently only happens in the first round,
where there will always be some progress on the other filters.

Fix a segfault when using the -map option without input files.

This fixes out of array accesses.

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Padding data is not supposed to be copied.

Prevent spurios EIO on EOF.

CC:libav-stable@libav.org

Notwithstanding the return value the frame should be freed if
is not going to be used.
Reported-by: Matthew Wolenetz <wolenetz@chromium.org>