pub fn _mm_cvtsd_i64(a: __m128d) -> i64
Convert the lower double-precision (64-bit) floating-point element in a to a 64-bit integer, and store the result in dst.
Intel’s documentation