Added _mm_rsqrt_ps, plus minor fixes

jratcliff63367 / sse2neon

Automatically exported from code.google.com/p/sse2neon

284 stars 130 forks source link

Added _mm_rsqrt_ps, plus minor fixes #4

Closed digitizdat closed 4 years ago

digitizdat commented 9 years ago

1) Added the _mm_rsqrt_ps (reciprocal square root) macro using @mtklein's technique for improving accuracy. 2) Addressed issues #1 and #3. 3) Updated the call to the obsolete ::memalign in SSE2NEONTEST.cpp to the C11 ::aligned_alloc.

jserv commented 4 years ago

Drop as DLTcollab/sse2neon implemented the similar work.