Jason Garrett-Glaser 8a467b2d44 VP8: 30% faster idct_mb ... Take shortcuts based on statistically common situations. Add 4-at-a-time idct_dc function (mmx and sse2) since rows of 4 DC-only DCT blocks are common. TODO: tie this more directly into the MB mode, since the DC-level transform is only used for non-splitmv blocks? Originally committed as revision 24452 to svn://svn.ffmpeg.org/ffmpeg/trunk		2010-07-23 02:58:27 +00:00
..
cavsdsp_mmx.c
cpuid.c	Remove FF_MM_SSE2/3 flags for CPUs where this is generally not faster than	2010-07-19 22:38:23 +00:00
dct32_sse.c
dnxhd_mmx.c
dsputil_h264_template_mmx.c
dsputil_h264_template_ssse3.c
dsputil_mmx_avg_template.c
dsputil_mmx_qns_template.c
dsputil_mmx_rnd_template.c
dsputil_mmx.c	relicense h264 deblock sse2 to lgpl	2010-07-22 00:39:49 +00:00
dsputil_mmx.h	Add header declarations for mmx/sse constants missing them	2010-07-21 10:02:07 +00:00
dsputil_yasm.asm
dsputilenc_mmx.c	Remove FF_MM_SSE2/3 flags for CPUs where this is generally not faster than	2010-07-19 22:38:23 +00:00
fdct_mmx.c
fft_3dn2.c
fft_3dn.c
fft_mmx.asm	more credits to D. J. Bernstein for fft	2010-07-18 20:06:42 +00:00
fft_sse.c
fft.c
fft.h
h264_deblock_sse2.asm	relicense h264 deblock sse2 to lgpl	2010-07-22 00:39:49 +00:00
h264_i386.h
h264_idct_sse2.asm
h264_intrapred.asm
h264dsp_mmx.c
idct_mmx_xvid.c
idct_mmx.c
idct_sse2_xvid.c
idct_xvid.h
lpc_mmx.c
Makefile	relicense h264 deblock sse2 to lgpl	2010-07-22 00:39:49 +00:00
mathops.h
mlpdsp.c
mmx.h
motion_est_mmx.c
mpegaudiodec_mmx.c
mpegvideo_mmx_template.c
mpegvideo_mmx.c
rv40dsp_mmx.c
simple_idct_mmx.c
snowdsp_mmx.c
vc1dsp_mmx.c	Move ff_pw_* from vc1dsp_mmx.c to dsputil_mmx.c	2010-07-21 10:02:03 +00:00
vc1dsp_yasm.asm
vp3dsp_mmx.c
vp3dsp_mmx.h
vp3dsp_sse2.c
vp3dsp_sse2.h
vp6dsp_mmx.c
vp6dsp_mmx.h
vp6dsp_sse2.c
vp6dsp_sse2.h
vp8dsp-init.c	VP8: 30% faster idct_mb	2010-07-23 02:58:27 +00:00
vp8dsp.asm	VP8: 30% faster idct_mb	2010-07-23 02:58:27 +00:00
x86inc.asm	sync yasm macros from x264	2010-07-21 22:45:16 +00:00
x86util.asm