Function _mm512_insertf32x8

pub fn _mm512_insertf32x8<const IMM8: i32>(a: __m512, b: __m256) -> __m512

Copy a to dst, then insert 256 bits (composed of 8 packed single-precision (32-bit) floating-point elements) from b into dst at the location specified by IMM8.

Intel’s Documentation