Function _mm256_rsqrt14_ps

#[target_feature(enable = "avx512f", enable = "avx512vl")]
pub fn _mm256_rsqrt14_ps(a: __m256) -> __m256

Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.

Intel's documentation