pub unsafe fn _mm_cvtps_epi32(a: __m128) -> __m128i
Converts packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers.
a
Intel’s documentation