Geek Answers Handbook

Permanent SIMD floats

I'm trying my best to optimize the code that I have using Microsoft built-in tools. One of the biggest problems when optimizing my code is the LHS, which occurs when I want to use a constant. It seems that there is some information about the generation of certain constants ( here and here - section 13.4 ), but all its assemblies (which I would prefer to avoid).

The problem is that I am trying to implement the same thing with intrinsics, msvc complains about incompatible types, etc. Does anyone know of any equivalent tricks using intrinsics?

Example - Create {1.0,1.0,1.0,1.0}

//pcmpeqw xmm0,xmm0 __m128 t = _mm_cmpeq_epi16( t, t ); //pslld xmm0,25 _mm_slli_epi32(t, 25); //psrld xmm0,2 return _mm_srli_epi32(t, 2);

This generates a bunch of errors about an incompatible type (__m128 vs _m128i). I'm new to this, so I'm pretty sure I'm missing something obvious. Can anyone help?

TL; DR . How can I generate a __m128 vec filled with single point constant floats with ms internal functions?

Thanks for reading:)

+7
source share
2 answers

Just enter __m128i into __m128 using _ mm_castsi128_ps . In addition, the second line should be

 t = _mm_slli_epi32(t, 25)
+3
source

Try _mm_set_ps , _mm_set_ps1 or _mm_set1_ps .

+4
source

More articles:


All Articles