pub unsafe fn _mm256_cvtps_epi32(a: __m256) -> __m256i
Converts packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers.
a
Intel’s documentation