* commit '17a63ff0':
  h264: update flag name in ff_h264_decode_ref_pic_list_reordering()
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'ee6280ca':
  h264: parse two additional constraint flags

Conflicts:
	libavcodec/h264_ps.c

See: 9091ba9dMerged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'e70ab7c1':
  h264: add MVCD to the list of High profiles in SPS
Merged-by: Michael Niedermayer <michaelni@gmx.at>

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

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

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

The new function has the ability to allocate the structure, allowing it to grow
without needing major bumps
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

This structure is used in the interface between libs and thus cannot have
fields added in the middle without major bump
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

This is neccessary for maintaining ABI compatibility with FFmpeg 2.1
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  latm: Always reconfigure if no extradata was set previously

See: d039b607Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '6e4009d4':
  arm: dcadsp: implement decode_hf as external NEON asm
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '4cb69642':
  dcadec: simplify decoding of VQ high frequencies

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

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

* commit '7686afd0':
  dca: factorize scaling in inverse ADPCM
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '08e3ea60':
  x86: synth filter float: implement SSE2 version

Conflicts:
	libavcodec/x86/dcadsp.asm
	libavcodec/x86/dcadsp_init.c

See: 2cdbcc00Merged-by: Michael Niedermayer <michaelni@gmx.at>

Timings for Arrandale:
          C    SSE
win32:  2108   334
win64:  1152   322

Factorizing the inner loop with a call/jmp is a >15 cycles cost, even with
the jmp destination being aligned.

Unrolling for ARCH_X86_64 is a 20 cycles gain.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* commit '57b1eb9f':
  dcadsp: scan coefficients linearly in dca_lfe_fir

Conflicts:
	libavcodec/dcadsp.c

See: 9ae8e231Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'ad507d79':
  x86: dcadsp: implement SSE lfe_dir

Conflicts:
	libavcodec/x86/dcadsp.asm

See: 16924311Merged-by: Michael Niedermayer <michaelni@gmx.at>

Results for Arrandale/Windows:
32: 1670 -> 316
64:  728 -> 298
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

This is the name used in the specifications.

Also comment all previous profiles.

* commit '87ec849f':
  dcadec: remove scaling in lfe_interpolation_fir

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

* commit 'a55546f4':
  proresenc: Reuse proper dsputil infrastructure for FDCT
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '92e598a5':
  prores: Drop DSP infrastructure for prores encoder bits

Conflicts:
	libavcodec/Makefile
	libavcodec/proresdsp.c
	libavcodec/proresenc_kostya.c

Note, these changes only affect one of the 2 prores encoders we have
If someone wants to add optimizations to the affected encoder, or needs/wants
this infrastructure, then iam happy to revert this
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'd6acefe0':
  proresenc: Drop unnecessary DCT permutation bits

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

* commit 'b2365049':
  prores: Use consistent names for DSP arch initialization functions

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

AAC LOAS can have new audio config objects in the stream itself.

Make sure the decoder reconfigures itself when the first one arrives
midstream.

Bug-Id: 644
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>

Fixes Ticket3212
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

The vector dequantization has a test in a loop preventing effective SIMD
implementation. By moving it out of the loop, this loop can be DSPized.

Therefore, modify the current DSP implementation. In particular, the
DSP implementation no longer has to handle null loop sizes.

The decode_hf implementations have following timings:

For x86 Arrandale:
        C  SSE SSE2 SSE4
win32: 260 162  119  104
win64: 242 N/A   89   72

The arm NEON optimizations follow in a later patch as external asm. The
now unused check for the y modifier in arm inline asm is removed from
configure.

Based on a patch from Christophe Gisquet.

Unrolling of the m == 0 case avoids a possible use of the uninitilized
value sum when s->predictor_history is not set. I failed to find a
sample for it. It also reduced the cycle count from 220 to 150 on
sandy bridge, x86_64 linux, gcc 4.8.2 compared to his patch.

Timings for Arrandale:
          C    SSE
win32:  2108   334
win64:  1152   322

Factorizing the inner loop with a call/jmp is a >15 cycles cost, even with
the jmp destination being aligned.

Unrolling for ARCH_X86_64 is a 20 cycles gain.
Signed-off-by: Janne Grunau <janne-libav@jannau.net>

This change is inspired by x86 asm where it frees a register.
Signed-off-by: Janne Grunau <janne-libav@jannau.net>

Results for Arrandale/Windows:
32: 1670 -> 316
64:  728 -> 298
Signed-off-by: Janne Grunau <janne-libav@jannau.net>

The scaling factor is constant so it is faster to scale the
FIR coefficients in the tables during compilation.
Signed-off-by: Janne Grunau <janne-libav@jannau.net>

None of the encoder bits are arch-optimized.

No permutation is necessary for the FDCT.