Christophe Gisquet 2aef3d66c9 SBR DSP x86: implement SSE sbr_hf_gen ...

Start and end index are multiple of 2, therefore guaranteeing aligned access.
Also, this allows to generate 4 floats per loop, keeping the alignment all
along.

Timing:
- 32 bits: 326c -> 172c
- 64 bits: 323c -> 156c

Signed-off-by: Diego Biurrun <diego@biurrun.de>

2012-12-07 11:04:26 +01:00

..

ac3dsp_init.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

ac3dsp.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

cabac.h

x86: cabac: allow building with suncc

2012-08-13 14:51:52 +01:00

cavsdsp.c

x86: mmx2 ---> mmxext in function names

2012-10-31 17:53:57 +01:00

dct32.asm

build: Drop AVX assembly ifdefs

2012-11-11 20:43:28 +01:00

deinterlace.asm

x86: yasm: Use complete source path for macro helper %includes

2012-10-31 00:37:42 +01:00

dnxhdenc.c

x86: more specific checks for availability of required assembly capabilities

2012-09-07 18:16:04 +02:00

dsputil_avg_template.c

x86: h264: Convert 8-bit QPEL inline assembly to YASM

2012-11-25 20:38:35 +01:00

dsputil_mmx.c

x86: fix build without inline asm

2012-11-26 01:50:47 +01:00

dsputil_mmx.h

x86: mmx2 ---> mmxext in function names

2012-10-31 17:53:57 +01:00

dsputil_qns_template.c

x86: avcodec: Drop silly "_mmx" suffix from dsputil template names

2012-09-07 13:50:52 +02:00

dsputil_rnd_template.c

x86: avcodec: Drop silly "_mmx" suffix from dsputil template names

2012-09-07 13:50:52 +02:00

dsputil.asm

x86: h264: Convert 8-bit QPEL inline assembly to YASM

2012-11-25 20:38:35 +01:00

dsputilenc_mmx.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

dsputilenc.asm

x86: dsputilenc: port to cpuflags

2012-11-28 16:05:44 +01:00

fdct.c

x86: mmx2 ---> mmxext in function names

2012-10-31 17:53:57 +01:00

fft_init.c

x86: call most of the x86 dsp init functions under if (ARCH_X86)

2012-10-08 11:54:05 +02:00

fft.asm

build: Drop AVX assembly ifdefs

2012-11-11 20:43:28 +01:00

fft.h

x86: Use consistent 3dnowext function and macro name suffixes

2012-08-03 14:00:47 +02:00

fmtconvert_init.c

x86: Replace checks for CPU extensions and flags by convenience macros

2012-09-08 18:18:34 +02:00

fmtconvert.asm

x86: SPLATD: port to cpuflags

2012-11-18 18:34:05 +01:00

h264_chromamc_10bit.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

h264_chromamc.asm

x86: h264_chromamc: port to cpuflags

2012-11-25 17:25:10 +01:00

h264_deblock_10bit.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

h264_deblock.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

h264_i386.h

x86: cabac: allow building with suncc

2012-08-13 14:51:52 +01:00

h264_idct_10bit.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

h264_idct.asm

x86: h264_idct: port to cpuflags

2012-11-28 00:28:09 +01:00

h264_intrapred_10bit.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

h264_intrapred_init.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

h264_intrapred.asm

x86: h264_intrapred: Fix C function names in comments

2012-11-18 18:34:05 +01:00

h264_qpel_8bit.asm

x86: h264 qpel: use the correct number of utilized xmm regs in cglobal

2012-11-25 18:48:43 -05:00

h264_qpel_10bit.asm

x86: h264_qpel_10bit: drop unused parameter from MC10/MC20/MC30 macros

2012-11-10 14:49:09 +01:00

h264_qpel.c

x86: h264: Convert 8-bit QPEL inline assembly to YASM

2012-11-25 20:38:35 +01:00

h264_weight_10bit.asm

x86: h264_weight_10bit: port to cpuflags

2012-11-13 19:07:09 +01:00

h264_weight.asm

x86: h264_weight: port to cpuflags

2012-11-27 21:10:38 +01:00

h264dsp_init.c

x86: h264dsp: Fix linking with yasm and optimizations disabled

2012-11-28 14:45:28 +01:00

idct_mmx_xvid.c

x86: mmx2 ---> mmxext in function names

2012-10-31 17:53:57 +01:00

idct_sse2_xvid.c

Don't include common.h from avutil.h

2012-08-15 22:32:06 +03:00

idct_xvid.h

x86: mmx2 ---> mmxext in function names

2012-10-31 17:53:57 +01:00

imdct36.asm

x86: include x86inc.asm in x86util.asm

2012-10-31 00:37:42 +01:00

lpc.c

x86: more specific checks for availability of required assembly capabilities

2012-09-07 18:16:04 +02:00

Makefile

x86: h264: Convert 8-bit QPEL inline assembly to YASM

2012-11-25 20:38:35 +01:00

mathops.h

Move MASK_ABS macro to libavcodec/mathops.h

2012-08-09 00:58:20 +01:00

mlpdsp.c

dsputil: split out mlp dsp function

2012-10-11 12:01:08 +02:00

motion_est.c

x86: mmx2 ---> mmxext in function names

2012-10-31 17:53:57 +01:00

mpegaudiodec.c

x86: call most of the x86 dsp init functions under if (ARCH_X86)

2012-10-08 11:54:05 +02:00

mpegvideo.c

avcodec: Drop silly and/or broken printf debug output

2012-10-01 10:24:28 +02:00

mpegvideoenc_template.c

x86: mpegvideoenc: Do not abuse HAVE_ variables for template instantiation

2012-08-29 00:14:52 +02:00

mpegvideoenc.c

x86: mmx2 ---> mmxext in function names

2012-10-31 17:53:57 +01:00

pngdsp_init.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

pngdsp.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

proresdsp_init.c

x86: Replace checks for CPU extensions and flags by convenience macros

2012-09-08 18:18:34 +02:00

proresdsp.asm

x86: yasm: Use complete source path for macro helper %includes

2012-10-31 00:37:42 +01:00

rv34dsp_init.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

rv34dsp.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

rv40dsp_init.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

rv40dsp.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

sbrdsp_init.c

SBR DSP x86: implement SSE sbr_hf_gen

2012-12-07 11:04:26 +01:00

sbrdsp.asm

SBR DSP x86: implement SSE sbr_hf_gen

2012-12-07 11:04:26 +01:00

simple_idct.c

x86: avcodec: Drop silly "_mmx" suffixes from filenames

2012-08-28 18:37:34 +02:00

snowdsp.c

x86: avcodec: Drop silly "_mmx" suffixes from filenames

2012-08-28 18:37:34 +02:00

vc1dsp_init.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

vc1dsp_mmx.c

x86: mmx2 ---> mmxext in function names

2012-10-31 17:53:57 +01:00

vc1dsp.asm

x86: PABSW: port to cpuflags

2012-11-05 14:51:10 +01:00

vc1dsp.h

x86: vc1: call ff_vc1dsp_init_x86() under if (ARCH_X86)

2012-10-08 11:54:05 +02:00

vp3dsp_init.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

vp3dsp.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

vp8dsp_init.c

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

vp8dsp.asm

x86: mmx2 ---> mmxext in asm constructs

2012-11-14 00:58:51 +01:00

vp56_arith.h

x86: vp56: cmov version of vp56_rac_get_prob requires inline asm

2012-09-05 21:30:46 +02:00

vp56dsp_init.c

x86: Replace checks for CPU extensions and flags by convenience macros

2012-09-08 18:18:34 +02:00

vp56dsp.asm

x86: yasm: Use complete source path for macro helper %includes

2012-10-31 00:37:42 +01:00

w64xmmtest.c

win64: add a XMM clobber test configure option.

2012-02-02 12:00:48 -08:00