Graham Booker f4a02f6e9b Use a faster way to compute 255-val: Instead of creating a vector of ...

all 255s, and then doing the subtraction, nor of the vector with itself: saves
one instruction and a register.
Patch by Graham Booker % gbooker A tamu P edu%

Originally committed as revision 9340 to svn://svn.ffmpeg.org/ffmpeg/trunk

2007-06-16 18:59:10 +00:00

..

dsputil_altivec.c

use macro Use DECLARE_ALIGNED_16 to align stack-allocated variables

2007-05-17 14:14:53 +00:00

dsputil_altivec.h

Altivec version of h264_(h|v)_loop_filter_luma

2007-06-09 19:13:34 +00:00

dsputil_ppc.c

…

dsputil_ppc.h

…

fdct_altivec.c

…

fft_altivec.c

…

float_altivec.c

…

gcc_fixes.h

…

gmc_altivec.c

use macro Use DECLARE_ALIGNED_16 to align stack-allocated variables

2007-05-17 14:14:53 +00:00

h264_altivec.c

Use a faster way to compute 255-val: Instead of creating a vector of

2007-06-16 18:59:10 +00:00

h264_template_altivec.c

use macro Use DECLARE_ALIGNED_16 to align stack-allocated variables

2007-05-17 14:14:53 +00:00

idct_altivec.c

…

int_altivec.c

…

mathops.h

…

mpegvideo_altivec.c

use macro Use DECLARE_ALIGNED_16 to align stack-allocated variables

2007-05-17 14:14:53 +00:00

mpegvideo_ppc.c

…

snow_altivec.c

…

types_altivec.h

…

vc1dsp_altivec.c

…