VCVTNEPS2BF16 - ConVerT No Exception Packed Single to BFloat16
float -> bfloat16
VCVTNEPS2BF16 xmm1{k1}{z}, xmm2/m128/m32bcst (V5+BF16+VL
__m128bh _mm_cvtne2ps_pbh(__m128 a)
__m128bh _mm_mask_cvtne2ps_pbh(__m128bh s, __mmask8 k, __m128 a)
__m128bh _mm_maskz_cvtne2ps_pbh(__mmask8 k, __m128 a)
VCVTNEPS2BF16 xmm1{k1}{z}, ymm2/m256/m32bcst (V5+BF16+VL
__m128bh _mm256_cvtne2ps_pbh(__m256 a)
__m128bh _mm256_mask_cvtne2ps_pbh(__m128bh s, __mmask8 k, __m256 a)
__m128bh _mm256_maskz_cvtne2ps_pbh(__mmask8 k, __m256 a)
VCVTNEPS2BF16 ymm1{k1}{z}, zmm2/m512/m32bcst (V5+BF16
__m256bh _mm512_cvtne2ps_pbh(__m512 a)
__m256bh _mm512_mask_cvtne2ps_pbh(__m256bh s, __mmask16 k, __m512 a)
__m256bh _mm512_maskz_cvtne2ps_pbh(__mmask16 k, __m512 a)
x86/x64 SIMD Instruction List
Feedback