Christophe Gisquet 133b34207c x86: float dsp: unroll SSE versions ...

vector_fmul and vector_fmac_scalar are guaranteed that they can process in
batch of 16 elements, but their SSE versions only does 8 at a time.

Therefore, unroll them a bit.
299 to 261c for 256 elements in vector_fmac_scalar on Arrandale/Win64.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

2014-02-15 18:54:21 +01:00

..

asm.h

…

bswap.h

…

cpu.c

Merge commit '4d6ee0725553a43ba88d6f8327ebcf8f1c5ae8d4'

2013-10-26 02:36:36 +02:00

cpu.h

x86: add missing XOP checks and macros

2014-02-11 03:46:52 +01:00

cpuid.asm

…

emms.asm

…

emms.h

…

float_dsp_init.c

Merge remote-tracking branch 'qatar/master'

2013-07-18 09:53:47 +02:00

float_dsp.asm

x86: float dsp: unroll SSE versions

2014-02-15 18:54:21 +01:00

intreadwrite.h

…

lls_init.c

rename new lls code to lls2 to avoid conflict with the old which has a different ABI

2013-11-17 16:41:08 +01:00

lls.asm

rename new lls code to lls2 to avoid conflict with the old which has a different ABI

2013-11-17 16:41:08 +01:00

Makefile

…

timer.h

Merge remote-tracking branch 'qatar/master'

2013-09-22 11:51:38 +02:00

w64xmmtest.h

Reinstate proper FFmpeg license for all files.

2013-08-30 15:47:38 +00:00

x86inc.asm

x86inc: Extend FMA_INSTR functionality

2014-02-13 22:14:24 +01:00

x86util.asm

Merge commit 'c6908d6b4b377a04a5d055ba874bdbcf06c80497'

2013-10-14 16:06:22 +02:00