x86/float_dsp: unroll loop in vector_fmac_scalar
authorJames Almer <jamrial@gmail.com>
Wed, 16 Apr 2014 05:09:36 +0000 (02:09 -0300)
committerMichael Niedermayer <michaelni@gmx.at>
Wed, 16 Apr 2014 16:36:52 +0000 (18:36 +0200)
commit11b36b1ee051e53c815114f8c258e23166ea9ed1
treeb83aee02b5cba539bce45a351c1fbd810e38706a
parent27f184ef40e8da09d9dfc8f71a8d1893729381f6
x86/float_dsp: unroll loop in vector_fmac_scalar

~6% faster SSE2 performance. AVX/FMA3 are unaffected.

Signed-off-by: James Almer <jamrial@gmail.com>
Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
libavutil/x86/float_dsp.asm