I found the optimisation of 2 samples per iteration obscured the
underlying algorithm.  I had to write it out on paper and translate into
a mathematical sum to see that the two samples are unconnected.  I hope
that if anyone else is struggling to understand the code that this will
be useful.
Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

This change is inspired by x86 asm, where this frees a register.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

vector_fmul and vector_fmac_scalar are guaranteed that they can process in
batch of 16 elements, but their SSE versions only does 8 at a time.

Therefore, unroll them a bit.
299 to 261c for 256 elements in vector_fmac_scalar on Arrandale/Win64.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes artifacts
Fixes use of freed memory
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

fix integer overflows
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>

This makes them similar to the other checks
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

Signed-off-by: Lukasz Marek <lukasz.m.luki@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

Fixes fate issues on mingw64
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  lavfi doxy: add buffer{src,sink}.h to the main lavfi doxy group

Conflicts:
	libavfilter/buffersink.h
	libavfilter/buffersrc.h
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'f758ea6e':
  buffersink: document special error codes returned from av_buffersink_get_frame
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Peter Ross <pross@xvid.org>
Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Lukasz Marek <lukasz.m.luki@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Lukasz Marek <lukasz.m.luki@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Both are required by the VP7 decoder.
Signed-off-by: Peter Ross <pross@xvid.org>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  ppc: reduce overreads when loading 8 pixels in altivec dsp functions
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '3fbad007':
  utvideoenc: Enable support for multiple slices and use them

Conflicts:
	libavcodec/utvideoenc.c
	tests/fate/utvideo.mak

See: efec857cMerged-by: Michael Niedermayer <michaelni@gmx.at>

The official Ut Video decoder only threads with slices, thus until
now any files encoded by the libavcodec encoder have only been
decodable with a single thread. The default slice count is now
set to subsampled_height / 120.

Also sets slices to 1 for the Ut Video encoder tests to keep them
green.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* commit '4127e6ae':
  libx265: Remove redundant default param call

See: 25bc8390Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '41836c4e':
  libx265: Fix use of uninitialized input picture

See: 955544e4Merged-by: Michael Niedermayer <michaelni@gmx.at>

This extends the exception from H264 to H264 + HEVC

Fixes Ticket3385
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

No testcase known, this is a theoretical improvment
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Altivec can only load naturally aligned vectors. To handle possibly
unaligned data a second vector is loaded from an offset of the original
location and the data is recovered through a vector permutation.
Overreads are minimal if the offset for second load points to the last
element of data. This is 7 for loading eight 8-bit pixels and overreads
are reduced from 16 bytes to 8 bytes if the pixels are 64-bit aligned.
For unaligned pixels the overread is reduced from 23 bytes to 15 bytes
in the worst case.

Signed-off-by: Reynaldo H. Verdejo Pinochet <r.verdejo@sisa.samsung.com>

They were basically the same.
Signed-off-by: Reynaldo H. Verdejo Pinochet <r.verdejo@sisa.samsung.com>

Signed-off-by: Reynaldo H. Verdejo Pinochet <r.verdejo@sisa.samsung.com>

Signed-off-by: Reynaldo H. Verdejo Pinochet <r.verdejo@sisa.samsung.com>

Signed-off-by: Reynaldo H. Verdejo Pinochet <r.verdejo@sisa.samsung.com>

Signed-off-by: Reynaldo H. Verdejo Pinochet <r.verdejo@sisa.samsung.com>

There were a few grammar and spelling mistakes in this filter
description. I went through it and corrected them without changing
the meaning.

The official Ut Video decoder only threads with slices, thus until
now any files encoded by the libavcodec encoder have only been
decodable with a single thread. The default slice count is now
set to subsampled_height / 120.

Also sets slices to 1 for the Ut Video encoder tests to keep them
green.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

* qatar/master:
  doc/examples: misc Doxygen markup improvements

Conflicts:
	doc/examples/muxing.c
	doc/examples/transcode_aac.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>