VEXP2PS - Exponential 2 Packed Single
VPEXP2PS zmm1{k1}{z}, zmm2/m512/m32bcst{sae} (V5+ER
__m512 _mm512_exp2a23_round_ps(__m512 a, int sae)
__m512 _mm512_mask_exp2a23_round_ps(__m512 s, __mmask16 k, __m512 a, int sae)
__m512 _mm512_maskz_exp2a23_round_ps(__mmask16 k, __m512 a, int sae)
For each float, calculate approximate 2x, where x=(1), and set the result to (2). (relative error < 2-23)
x86/x64 SIMD Instruction List
Feedback