Function _mm_maskz_rcp_ph 

pub fn _mm_maskz_rcp_ph(k: __mmask8, a: __m128h) -> __m128h

Available on x86 and target feature avx512fp16 and target feature avx512vl only.

Compute the approximate reciprocal of packed 16-bit floating-point elements in a and stores the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 1.5*2^-12.

Intel’s documentation