* qatar/master:
  dsputil: x86: Fix compile error
  dsputil: x86: Convert h263 loop filter to yasm
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '12b54a1f':
  mpegvideo: remove an unused function parameter
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '605b047b':
  rv10: improve buffer size check.
Merged-by: Michael Niedermayer <michaelni@gmx.at>

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

* commit 'c5fcdb44':
  error_resilience: remove a useless if() and FIXME
  h264: remove silly macros
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'ca1fe6c0':
  h263: remove an unused parameter from ff_h263_decode_init_vlc
  flac: add channel layout masks for streams with 7 or 8 channels.
  flac: don't check the number of channels before setting the channel layout.

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

* commit 'a846dccb':
  h264chroma: x86: Fix building with yasm disabled
  rv34: Drop now unnecessary dsputil dependencies

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

* commit '79dad2a9':
  dsputil: Separate h264chroma

Conflicts:
	libavcodec/dsputil_template.c
	libavcodec/ppc/dsputil_ppc.c
	libavcodec/vc1dec.c
	libavcodec/vc1dsp.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

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

This is unnecessary, as ff_flac_set_channel_layout can handle any number of channels.

Accidentally prefixed ff_ with cextern.
Signed-off-by: Martin Storsjö <martin@martin.st>

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

Non-ascii chars are not correctly rendered in both MAN and HTML output or
may even cause compilation failure. Thus avoid to use them until we find
a satisfying way to deal with them.

Fix trac ticket #2243.

The DTS needs to be resynched against the segment start PTS, or the
resulting DTS may result < PTS.
Reported-by: Owen Jones <riots6@gmail.com>

See thread:
Subject: [FFmpeg-user] pts/dts error using reset_timestamps while splitting a DVD
Date: Sat, 19 Jan 2013 08:58:27 +0000

Fixes ticket #1764.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Check slice count and input buffer size before constructing a possibly
invalid pointer, not after.

pp_time is never set for h264

They serve no useful purpose and wreak all kind of havoc when h264.h is
included elsewhere.

They were added to the latest FLAC specification:
https://git.xiph.org/?p=flac-website.git;a=commit;h=65c199a2Signed-off-by: Anton Khirnov <anton@khirnov.net>

This is unnecessary, as ff_flac_set_channel_layout can handle any number of channels.
Signed-off-by: Anton Khirnov <anton@khirnov.net>

* rdp/combined2:
  dshow: allow for more codec ID's
  dshow: better error reporting
  dshow: fix log message
Merged-by: Michael Niedermayer <michaelni@gmx.at>

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

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

Modified the fate test crc generator to print the side_data's
crc if side_data is present.
Signed-off-by: Vignesh Venkatasubramanian <vigneshv@google.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  mpegvideo: initialize dummy reference frames.

Conflicts:
	libavcodec/mpegvideo.c
	tests/ref/fate/filter-yadif-mode1

The luma init is left at 0x80 as it was previously, and tested and
working since a long time. Its also mandated to be 0x80 by at least
one specification. Cases that need different values already had
code in place to handle them.
Merged-by: Michael Niedermayer <michaelni@gmx.at>

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

* commit 'dff6197d':
  nuv: do not rely on get_buffer() initializing the frame.
  yop: initialize palette to 0

Conflicts:
	libavcodec/nuv.c
	tests/ref/fate/nuv-rtjpeg-fh
	tests/ref/fate/yop
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '8136f234':
  yop: check for input overreads.
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Martin Storsjö <martin@martin.st>

* commit '06cf597c':
  yop: check that extradata is large enough.

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

* commit '89f11f49':
  qtrle: fix the topmost line for 1bit

Conflicts:
	tests/ref/fate/qtrle-1bit

See: 1af91978Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '685e6f2e':
  xxan: properly handle odd heights.
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '2cd40680':
  fraps: fix off-by one bug for version 1.

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

* commit 'da7baaaa':
  aasc: fix output for msrle compression.
  msrledec: check bounds before constructing a possibly invalid pointer,
  cmdutils: remove ansi from the list of broken codecs.

Conflicts:
	tests/ref/fate/aasc

Also See: e7117f1cMerged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '238614de':
  cdgraphics: do not rely on get_buffer() initializing the frame.
  svq1: replace struct svq1_frame_size with an array.
  vf_yadif: silence a warning.

Conflicts:
	libavcodec/svq1dec.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '48a4ffa7':
  asf: K&R formatting cosmetics
  vc1dec: use codec_id instead of codec_tag for VC1IMAGE
  sh4: drop unused functions

Conflicts:
	libavformat/asf.c
	libavformat/asfdec.c
	libavformat/asfenc.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>