avformat/svag: extend format long description

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

avcodec: extend long decription for adpcm psx codec

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

avformat/msf: extend format long description

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

Add myself as maintainer for Hap

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Revert "configure: add -Wstrict-prototypes when available"

This reverts commit e6a93e59adeea1c140f2ef63e4055a15823bfeda,
Wstrict-prototypes is already enabled.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>

configure: add -Wstrict-prototypes when available

GCC (and Clang) have this useful warning that is not enabled by -Wall or
-Wextra. This will ensure that issues like those fixed in
4da52e3630343e8d3a79aef2cafcb6bf0b71e8da
will trigger warnings.

Reviewed-by: Hendrik Leppkes <h.leppkes@gmail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>

avutil/qsort: use the do while form for AV_QSORT, AV_MSORT

Reviewed-by: Clément Bœsch <u@pkh.me>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>

vf_psnr/ssim: don't crash if stats_file is NULL.

avcodec/hap: set bits_per_coded_sample

fixes issue where alpha is ignored in some players

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter: add shuffleframes filter

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

avformat/mov: Autodetect mp3s which need parsing

mp3 packets all have the same duration and number of samples
if their duration indicated in the container varies then thats an
indication that they are not 1 mp3 packet each.
If this autodetection fails for some case then please contact us
and provide a testcase.

Fixes Ticket4938

avdevice/pulse_audio_common: add av_warn_unused_result

This does not trigger any warnings, but adds robustness.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>

vf_ssim: print per-channel dB values.

vf_psnr: remove %0.2f format specifiers for stream summary line.

This makes output equally precise as vf_ssim.

avcodec/libzvbi-teletextdec: Add variable to fix build

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter,swresample,swscale: use fabs, fabsf instead of FFABS

It is well known that fabs and fabsf are at least as fast and sometimes
faster than the FFABS macro, at least on the gcc+glibc combination.
For instance, see the reference:
http://patchwork.sourceware.org/patch/6735/.
This was a patch to glibc in order to remove their usages of a macro.

The reason essentially boils down to fabs using the __builtin_fabs of
the compiler, while FFABS needs to infer to not use a branch and to
simply change the sign bit. Usually the inference works, but sometimes
it does not. This may be easily checked by looking at the asm.

This also has the added benefit of reducing macro usage, which has
problems with side-effects.

Note that avcodec is not handled here, as it is huge and
most things there are integer arithmetic anyway.

Tested with FATE.

Reviewed-by: Clément Bœsch <u@pkh.me>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>

Merge commit '3ee2c60cc296eee3f63d7b5fee9b4332eeeac9fa'

* commit '3ee2c60cc296eee3f63d7b5fee9b4332eeeac9fa':
  utils: Use data buffers directly instead of an AVPicture

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'ff7956fcbf8e59b21654b95038de3ed88a850a9e'

* commit 'ff7956fcbf8e59b21654b95038de3ed88a850a9e':
  avplay: Replace avpicture functions with imgutils

avplay and ffplay have diverged quite a bit, and ffplay should be updated
independently.

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

avcodec/libzvbi: Update for AVSubtitleRect changes

Merge commit 'a17a7661906ba295d67afd80ac0770422e1b02b3'

* commit 'a17a7661906ba295d67afd80ac0770422e1b02b3':
  lavc: Add data and linesize to AVSubtitleRect

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

avcodec/nuv: Fix 'libavcodec/nuv.c:83:19: warning: passing argument 3 of av_image_copy from incompatible pointer type'

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec/x86/vc1dsp: Remove unused macro

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Merge commit 'f890677d05bc4e8b494a73373ab4cc19791bf884'

* commit 'f890677d05bc4e8b494a73373ab4cc19791bf884':
  Replace any remaining avpicture function with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

avfilter/vf_zscale: fix typo

Fixes #4958 as found by nicol.

Signed-off-by: Lou Logan <lou@lrcd.com>

Merge commit '13bddab7de10aebf6efb98aa6d7ff0c51bb0e364'

* commit '13bddab7de10aebf6efb98aa6d7ff0c51bb0e364':
  nuv: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '48c06386831604921bdaf4fb77ea02766cd615f4'

* commit '48c06386831604921bdaf4fb77ea02766cd615f4':
  dpx: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

roqvideodec: use av_frame_copy

Merge commit 'f0a106578d759de6183eea3c75f8373b6d3153c1'

* commit 'f0a106578d759de6183eea3c75f8373b6d3153c1':
  roqvideodec: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'ef3a3519c10620c4206738595bf03fc0bed71802'

* commit 'ef3a3519c10620c4206738595bf03fc0bed71802':
  rawdec: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'fcc1280acb6e6f682b34c2101b075b82f83d71ba'

* commit 'fcc1280acb6e6f682b34c2101b075b82f83d71ba':
  rawenc: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '3496cec43304ac040d6d05f9d500a6f93cc049e7'

* commit '3496cec43304ac040d6d05f9d500a6f93cc049e7':
  msrle: Use AVFrame instead of AVPicture

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'ac981d16415e7fd99683e10297781c7d9ec1a8cd'

* commit 'ac981d16415e7fd99683e10297781c7d9ec1a8cd':
  APIchanges: Fill in missing dates and hashes

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

ffmpeg: add abort_on option to allow aborting on empty output

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>

ffmpeg: fix ffmpeg.h trailing whitespace

How this passed through the commit hook?

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>

qsvdec: fix get_format and hwaccel_context handling

This enables the qsv transcoder to actually get activated

Merge commit 'fb472e1a11a4e0caed2c3c91da01ea8e35d9e3f8'

* commit 'fb472e1a11a4e0caed2c3c91da01ea8e35d9e3f8':
  avconv: add support for Intel QSV-accelerated transcoding

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'dc923bc23b3efd949d0bf67ff1abdb95059e5843'

* commit 'dc923bc23b3efd949d0bf67ff1abdb95059e5843':
  qsvenc: add an API for allocating opaque surfaces

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '2ec96b6bd5bf7b22978711bcf2cee702bee89c6f'

* commit '2ec96b6bd5bf7b22978711bcf2cee702bee89c6f':
  qsvenc: cosmetics, reindent

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit 'f6f32fc93d39caf329869c1bff8ad024ccab1d42'

* commit 'f6f32fc93d39caf329869c1bff8ad024ccab1d42':
  qsvenc: set the timestamp for PIX_FMT_QSV frames as well

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

Merge commit '772c87c5a658f36d7c0612f5da583fc2bfa54f79'

* commit '772c87c5a658f36d7c0612f5da583fc2bfa54f79':
  qsvenc: support passing arbitrary external buffers to the encoder

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>

avutil/intmath: fix undefined behavior in ff_ctzll_c()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

utils: Use data buffers directly instead of an AVPicture

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

libavformat/electronicarts: also demux mpeg audio layer 2

Signed-off-by: Peter Ross <pross@xvid.org>
http://wiki.multimedia.cx/index.php?title=Electronic_Arts_SCxl
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

lavc/x86/vc1dsp_init: Fix compilation with --disable-yasm.

lavf/mpjpegdec: Return 0 if an allocation inside the probe function fails.

avformat/electronicarts: fix demuxing of certain eam files

Such files have gaps between header chunks.

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

avformat/electronicarts: support ADPCM PSX

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

mmaldec: Add mpeg2 decoding support

Register mmaldec as mpeg2 decoder. Supporting mpeg2 in mmaldec is just a
matter of setting the correct MMAL_ENCODING on the input port. To ease the
addition of further supported mmal codecs a macro is introduced to generate
the decoder and decoder class structs.

Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: wm4 <nfxjfg@googlemail.com>

x86/Makefile: move decoder/encoder objects out of the subsystems section

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

ffmpeg: exit on corrupt packets or decoded frames if exit_on_error flag is present

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>

ffmpeg: factorize checking decoder result

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>

ffmpeg: exit on av_write_trailer failure if exit_on_error is set

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>

ffmpeg: log failed av_write_trailer

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>

vc1dsp: Port ff_vc1_put_ver_16b_shift2_mmx to yasm

This function is only used within other inline asm functions, hence the
HAVE_MMX_INLINE guard. Per recent discussions, we should not worry about
the performance of inline asm-only builds.

libavformat/mxfdec.c: Report dark metadata keys only when they match no parser at all, not everytime they fail to match one.

Reviewed-by: Tomas Härdin <tomas.hardin@codemill.se>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

libavformat/mxfdec.c: cosmetics: Add missing space after '?' in log message.

Reviewed-by: Tomas Härdin <tomas.hardin@codemill.se>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avcodec: rename avpriv_color_frame to ff_color_frame

It is only used inside libavcodec.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>

avcodec: drop 2 suffix from avpriv_ac3_parse_header2

avpriv_ac3_parse_header was removed in commit 3dfb643.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>

avfilter/vf_ssim: Add support for writing stats to stdout

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

huffyuvencdsp: Cherry pick changes left out in the last commit

Oops.

huffyuvencdsp: Add ff_diff_bytes_{sse2,avx2}

SSE2 version 4%-35% faster than MMX depending on the width.
AVX2 version 1%-13% faster than SSE2 depending on the width.

mmaldec: Fix avpriv_atomic_get usage

There is no avpriv_atomic_get, instead avpriv_atomic_int_get is to be used for
integers. This fixes building mmaldec.

Signed-off-by: Julian Scheel <julian@jusst.de>
Reviewed-by: wm4 <nfxjfg@googlemail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

aacdec: fix strict prototype warning

Function protoypes without arguments require a void argument in C,
instead of an empty list.

avplay: Replace avpicture functions with imgutils

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

lavc: Add data and linesize to AVSubtitleRect

Use the new fields directly instead of the ones from AVPicture.
This removes a layer of indirection which serves no pratical purpose
whatsoever, and will help in removing AVPicture structure completely
later.

Every subtitle encoder/decoder seamlessly points to the new arrays,
so it is possible to deprecate AVSubtitleRect.pict.

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

Replace any remaining avpicture function with imgutils

avpicture_get_size() -> av_image_get_buffer_size()

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

nuv: Replace avpicture functions with imgutils

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

dpx: Replace avpicture functions with imgutils

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

roqvideodec: Replace avpicture functions with imgutils

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

rawdec: Replace avpicture functions with imgutils

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

rawenc: Replace avpicture functions with imgutils

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

msrle: Use AVFrame instead of AVPicture

Callers always use a frame and cast it to AVPicture, change
ff_msrle_decode() to accept an AVFrame directly instead.

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

lavf: add oggparsedaala and bump micro

This commit introduces a parser for the current bitstream produced by
Daala. It currently bears a large similarity with Theora, another
codec produced by Xiph. While likely to change in the future, its basic
format of packet parsing should remain fairly identical with its current
structure.
Once the bitstream freezes, there are a few probable simplifications
that could be made. Also, the current version (major, minor and micro)
is stuck at zero so it's unusable as a way to warn about possible
incompatibilities. This will change once the bitstream freezes,
however until then this file is strictly targeting the current git
master of the reference encoder, libdaala.

This file was developed independently at the same time by both myself
and Vittorio Giovara, who used libav as a starting point. For fairness,
and to prevent confusion and allegations, his name has been added to the
copyright in the license header as well, and vice versa.

lavc: add AV_CODEC_ID_DAALA to the list of codec IDs and bump minor

This commit shall introduce the first step of adding support for the
Daala next generation video codec to FFmpeg. Although still in
development, the codec is showing good progress and exchanging work
through IETF drafts. The companies behind Daala are also participating
in the Alliance for Open Media, so it's likely that whatever the result
any of these collaborations produce it's probable that elements from
Daala could be used in them, or perhaps this codec itself could be the
result.

libvpxenc: remove some unused ctrl id mappings

VP8E_UPD_ENTROPY, VP8E_UPD_REFERENCE, VP8E_USE_REFERENCE were removed
from libvpx and the remaining values were never used here

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Zern <jzern@google.com>

huffyuvencdsp: Convert ff_diff_bytes_mmx to yasm

Heavily based upon ff_add_bytes by Christophe Gisquet.

Reviewed-by: James Almer <jamrial@gmail.com>
Signed-off-by: Timothy Gu <timothygu99@gmail.com>

avfilter/vf_psnr: Add support for writing stats to stdout

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avutil: merge avpriv_float_dsp_init into avpriv_float_dsp_alloc

Also replace the last two usages of avpriv_float_dsp_init with
avpriv_float_dsp_alloc.

Reviewed-by: James Almer <jamrial@gmail.com>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>

avcodec: remove unused avpriv_ac3_parse_header

It was replaced by avpriv_ac3_parse_header2.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>

avutil: install des.h, rc4.h and tree.h as public headers

These headers contain functions supposed to be public.

libavutil/des.h:
 av_des_alloc
 av_des_crypt
 av_des_init
 av_des_mac
libavutil/rc4.h:
 av_rc4_alloc
 av_rc4_crypt
 av_rc4_init
libavutil/tree.h
 av_tree_destroy
 av_tree_enumerate
 av_tree_find
 av_tree_insert
 av_tree_node_alloc
 av_tree_node_size

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>

dnxhdenc: Optimize get_pixels_8x4_sym for 10-bit

This reverts commit 628e6d0164febc8e69b0f10dfa487e8a2dd1a28a and uses
a better fix.

Before:
4483 decicycles in get_pixels_8x4_sym,  131032 runs,     40 skips

After:
2569 decicycles in get_pixels_8x4_sym,  131054 runs,     18 skips

dca_parser: don't overwrite the sample rate, it may not be correct

The parser only reads the dca core sample rate, which is limited to a
maximum of 48000 Hz, while X96 and HD extensions can increase the sample
rate up to 192000 Hz.

This change prevents the parser and decoder fighting over the sample rate,
potentially confusing user applications. This also fixes sample rate
display of >48000Hz files with ffmpeg/ffprobe when using libdcadec.

Fixes ticket #4397

avformat/mp3dec: improve junk skipping heuristic

Commit 2b3e9bbfb529e6bde238aeb511b55ebe461664c8 caused problems for a
certain API user:

https://code.google.com/p/chromium/issues/detail?id=537725
https://code.google.com/p/chromium/issues/detail?id=542032

The problem seems rather arbitrary, because if there's junk, anything
can happen. In this case, the imperfect junk skipping just caused it to
read different junk, from what I can see.

We can improve the accuracy of junk detection by a lot by checking if 2
consecutive frames use the same configuration. While in theory it might
be completely fine for the 1st frame to have a different format than the
2nd frame, it's exceedingly unlikely, and I can't think of a legitimate
use-case.

This is approximately the same mpg123 does for junk skipping. The
set of compared header bits is the same as the libavcodec mp3 parser
uses for similar purposes.

avformat: add wve demuxer

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

avformat: add msf demuxer

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

compat/solaris/make_sunver.pl: Use /usr/bin/env perl instead of /usr/bin/perl

This is how the other perl scripts in git call perl

Reviewed-by: Ganesh Ajjanagadde <gajjanag@mit.edu>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

huffyuvencdsp: Use intptr_t for width

It is done this way in huffyuvdsp as well.

x86: vc1dsp_mmx: Move yasm initiation steps to vc1dsp_init

That's where all yasm initiation steps are. Also removes the overlap
between the two files.

x86: fpel: Remove erroneous ff_put_pixels8_mmxext prototype

This function does not exist.

x86: fpel: Move prototypes for 4-px block functions

avcodec/mpegaudio_parser: Update comment to match code

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

vp9_parser: fix endless loop w/0-sized frame

treat this the same as an over-sized superframe packet to break out of
the parser loop and allow the decoder to fail.

Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: James Zern <jzern@google.com>

APIchanges: Fill in missing dates and hashes

lavu/intmath.h: Move x86 only msvc/icl functions to x86 specific header.

Signed-off-by: Matt Oliver <protogonoi@gmail.com>

lavu/intmath.h: Add msvc/icl ctzll optimisations.

Signed-off-by: Matt Oliver <protogonoi@gmail.com>

avutil/opt: display a better default value for int/int64 options

Example:

% ./ffmpeg -h encoder=aac
  -aac_coder         <int>        E...A... Coding algorithm (from -1 to 3) (default twoloop)
     faac                         E...A... FAAC-inspired method
     anmr                         E...A... ANMR method
     twoloop                      E...A... Two loop searching method
     fast                         E...A... Constant quantizer
[...]

avfilter/af_flanger: free frame on ENOMEM

Signed-off-by: Kyle Swanson <k@ylo.ph>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avutil: use EINVAL instead of -1 for the return code of crypto related init functions

These functions return an error typically when the key size is an
incorrect number. AVERROR(EINVAL) is more specific than -1.

Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>

avformat/vag: Remove unused variable pos

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

avfilter: add zscale filter

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