Sample timings on ped1080p.webm (of the ssse3 functions):
iadst_idct:  4672 -> 1175 cycles
idct_iadst:  4736 -> 1263 cycles
iadst_iadst: 4924 -> 1438 cycles
Total decoding time changed from 6.565s to 6.413s.

Also fix trac ticket #3296.
Signed-off-by: Georg Martius <martius@mis.mpg.de>
Signed-off-by: Stefano Sabatini <stefasab@gmail.com>

In particular:
* set default value for accuracy to 15 (max)
* add zoomspeed for dynamic zoom (optzoom=2)
* make camera path optimization algorithm configurable
* update optzoom documenation in code to use new localmotions calculation function commandline arguments
* add debug option in vidstabtransform
Signed-off-by: Georg Martius <martius@mis.mpg.de>
Signed-off-by: Stefano Sabatini <stefasab@gmail.com>

Players that conform to the specification can only playback such
files correctly on big-endian hardware.

Fixes ticket #3311.

This allows decoding broken mkv files containing G.726 audio.

The value is wrong and leads to broken files.

Fixes: use of uninitialized memeory
Fixes: msan_uninit-mem_7fbf26b5fefe_5981_intro.jv
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes: use of uninitialized memeory
Fixes: msan_uninit-mem_7fa421d0e222_1765_Girl_With_The_Dragon_Tattoo_2_23_56.mkv
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  x86: dsputil: Simplify xvmc deprecation conditional

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

* commit 'dc4d726b':
  build: Add separate hidden config option for the intrax8 code

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

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

* qatar/master:
  aarch64: port neon clobber test from arm
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'f896bca0':
  aarch64: h264 (bi)weight NEON optimizations
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '36e3b1f2':
  aarch64: h264 loop filter NEON optimizations
Merged-by: Michael Niedermayer <michaelni@gmx.at>

This allows for slightly cleaner dependency specification.

* commit 'c65d67ef':
  aarch64: hpeldsp NEON optimizations

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

* commit 'd5dd8c7b':
  aarch64: h264 qpel NEON optimizations
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '8438b3f0':
  aarch64: h264 idct NEON assembler optimizations
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '71617884':
  aarch64: h264 chroma motion compensation NEON optimizations
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'b7b17ed6':
  aarch64: add cpuflags support for NEON and VFP
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '1e9265cd':
  dct-test: test ff_simple_idct_neon only on ARM for now
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '2620df13':
  mov: Free an earlier allocated array if allocating a new one
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'd51f0996':
  mov: Free intermediate arrays in the normal cleanup function

Conflicts:
	libavformat/mov.c

See: a6b3e6d0Merged-by: Michael Niedermayer <michaelni@gmx.at>

4412 decicycles in ff_vp9_loop_filter_h_16_16_ssse3, 4193462 runs, 842 skips
3600 decicycles in ff_vp9_loop_filter_h_16_16_avx, 4193621 runs, 683 skips

3010 decicycles in ff_vp9_loop_filter_v_16_16_ssse3, 4193528 runs, 776 skips
2678 decicycles in ff_vp9_loop_filter_v_16_16_avx, 4193742 runs, 562 skips

23025 decicycles in ff_vp9_idct_idct_32x32_add_ssse3, 2096871 runs, 281 skips
19943 decicycles in ff_vp9_idct_idct_32x32_add_avx, 2096815 runs, 337 skips

4675 decicycles in ff_vp9_idct_idct_16x16_add_ssse3, 4194018 runs, 286 skips
3980 decicycles in ff_vp9_idct_idct_16x16_add_avx, 4194022 runs, 282 skips

967 decicycles in ff_vp9_idct_idct_8x8_add_ssse3, 16776972 runs, 244 skips
887 decicycles in ff_vp9_idct_idct_8x8_add_avx, 16777002 runs, 214 skips

Ported from ARMv7 NEON.

Ported from ARMv7 NEON.

Ported from ARMv7 NEON.

Ported from ARMv7 NEON.

Ported from ARMv7 NEON.

Since RV40 and VC-1 use almost the same algorithm so optimizations for
those two decoders are easy to do and included.

NEON and VFP are currently mandatory for all ARMv8 profiles. Both are
handled as extensions as far as cpuflags are concerned. This is
consistent with handling x86_64 which always has SSE2, but still
handles it as an extension.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>

It could probably also be considered an error if the pointer isn't
null at this point, but then we might risk rejecting some
slightly broken files that we might have handled so far.

Sample-Id: 00000496-google
Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>

These arrays are normally freed at the end of mov_read_trak,
but make sure they're freed in case mov_read_trak returned
early (due to errors) or in case the atoms that allocate arrays
are encountered at some other point than within a trak (which
we don't have checks against).

Sample-Id: 00000496-google
Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>