Tue, 15 Sep 2009 16:18:16 +0000 |
vitor |
Add two more sizes to ff_sine_windows[] and also pad it with NULLs so
libavcodec
|
Fri, 11 Sep 2009 02:24:22 +0000 |
mru |
ARM: NEON optimised MDCT
libavcodec
|
Fri, 11 Sep 2009 02:24:19 +0000 |
mru |
Prepare for optimised forward MDCT implementations
libavcodec
|
Thu, 10 Sep 2009 08:50:03 +0000 |
mru |
ARM: NEON optimised FFT and MDCT
libavcodec
|
Sun, 06 Sep 2009 08:56:10 +0000 |
reimar |
ff_sine_windows table contains only constant data, too.
libavcodec
|
Sun, 06 Sep 2009 08:53:14 +0000 |
reimar |
The ff_cos_tabs table itself is constant, too, so mark it as such.
libavcodec
|
Sat, 29 Aug 2009 16:55:55 +0000 |
mru |
One declaration of mm_flags is enough
libavcodec
|
Sat, 29 Aug 2009 16:55:50 +0000 |
mru |
Move declarations of some mmx functions to dsputil_mmx.h
libavcodec
|
Sat, 29 Aug 2009 16:55:46 +0000 |
mru |
Move DECLARE_ALIGNED_8 definition next to DECLARE_ALIGNED_16
libavcodec
|
Fri, 28 Aug 2009 23:39:02 +0000 |
mru |
Make DECLARE_ALIGNED_8 align to 8 bytes, never to 16
libavcodec
|
Thu, 27 Aug 2009 15:38:59 +0000 |
vitor |
Mark parameter src of vector_clipf() as const
libavcodec
|
Thu, 27 Aug 2009 14:49:36 +0000 |
vitor |
SSE optimized vector_clipf(). 10% faster TwinVQ decoding.
libavcodec
|
Tue, 16 Jun 2009 09:00:55 +0000 |
darkshikari |
idct_dc for VC-1/WMV3 decoder; ~11% faster decoding overall.
libavcodec
|
Sat, 16 May 2009 14:17:08 +0000 |
serge |
Support for getting (i)MDCT output multiplied by a constant scaling factor.
libavcodec
|
Fri, 15 May 2009 15:34:22 +0000 |
ramiro |
mlp: Simplify adressing of state and coeffs arrays for both filters by making
libavcodec
|