ffmpeg/libavutil/x86
Clément Bœsch 45c7f3997e avutil/pixelutils: faster pixelutils_sad_[au]_16x16
~560 → ~500 decicycles

This is following the comments from Michael in
https://ffmpeg.org/pipermail/ffmpeg-devel/2014-August/160599.html

Using 2 registers for accumulator didn't help. On the other hand,
some re-ordering between the movs and psadbw allowed going ~538 to ~500.
2014-08-23 10:18:53 +02:00
..
asm.h x86/vp9: inital AVX2 intra_pred 2014-06-08 02:37:20 +02:00
bswap.h
cpu.c x86/cpu: check for OS support before enabling AVX2 2014-03-25 17:56:43 +01:00
cpu.h x86: add detection for FMA3 instruction set 2014-02-23 15:29:36 +01:00
cpuid.asm Merge commit '79793f833784121d574454af4871866576c0749d' 2014-07-01 15:43:40 +02:00
emms.asm
emms.h
float_dsp_init.c x86/float_dsp: port vector_fmul_window to yasm 2014-06-08 12:41:32 +02:00
float_dsp.asm x86/float_dsp: add missing femms 2014-06-08 20:06:28 +02:00
intreadwrite.h
lls_init.c drop LLS1, rename LLS2 to LLS 2014-08-09 23:20:31 +02:00
lls.asm drop LLS1, rename LLS2 to LLS 2014-08-09 23:20:31 +02:00
Makefile avutil: add pixelutils API 2014-08-05 21:05:52 +02:00
pixelutils_init.c avutil: add pixelutils API 2014-08-05 21:05:52 +02:00
pixelutils.asm avutil/pixelutils: faster pixelutils_sad_[au]_16x16 2014-08-23 10:18:53 +02:00
pixelutils.h avutil: add pixelutils API 2014-08-05 21:05:52 +02:00
timer.h avutil/timer: Fix units for x86 after c708b5403346255ea5adc776645616cc7c61f078 2014-03-09 15:22:02 +01:00
w64xmmtest.h
x86inc.asm Merge commit '79793f833784121d574454af4871866576c0749d' 2014-07-01 15:43:40 +02:00
x86util.asm x86/hevc_deblock: improve 8bit transpose store macros 2014-08-03 04:24:15 +02:00