SSE loading ints into __m128

Question

What are the gcc's intrinsic for loading 4 ints into __m128 and 8 ints into __m256 (aligned/unaligned)? What about unsigned ints?

Clarification: You're looking GCC's intrinsics, and not Intel's intrinsics correct? — Mysticial, Feb 24 '12 at 20:31
Don't know precisely, but intrinsic usable from gcc 4.6+ "out-of-box". — Cartesius00, Feb 24 '12 at 20:32
Use https://software.intel.com/sites/landingpage/IntrinsicsGuide/ to find all you need to know. — user276648, May 28 '18 at 09:32

score 9 · Accepted Answer · edited May 28 '18 at 09:31

9

Using Intel's SSE intrnisics, the ones you're looking for are:

Documentation:

There's no distinction between signed or unsigned. You'll need to cast the pointer to __m128i* or __m256i*.

Note that these are Intel's SSE intrinsics and will work in GCC, Clang, MSVC, and ICC.
The GCC intrinsics work only in, well, GCC AFAIK of.

edited May 28 '18 at 09:31

user276648

answered Feb 24 '12 at 20:36

Mysticial

Nice, thank you, very much. I have on other follow-up question: http://stackoverflow.com/questions/9437860/sse-ints-vs-floats-practice – Cartesius00 Feb 24 '12 at 20:55

1 Answers1