VCVTPD2PH - ConVerT Packed Double to Packed Half
double -> FP16
VCVTPD2PH xmm1{k1}{z}, xmm2/m128/m64bcst (V5+FP16+VL
__m128h _mm_cvtpd_ph(__m128d a)
__m128h _mm_mask_cvtpd_ph(__m128h s, __mmask8 k, __m128d a)
__m128h _mm_maskz_cvtpd_ph(__mmask8 k, __m128d a)
VCVTPD2PH xmm1{k1}{z}, ymm2/m256/m64bcst (V5+FP16+VL
__m128h _mm256_cvtpd_ph(__m256d a)
__m128h _mm256_mask_cvtpd_ph(__m128h s, __mmask8 k, __m256d a)
__m128h _mm256_maskz_cvtpd_ph(__mmask8 k, __m256d a)
VCVTPD2PH xmm1{k1}{z}, zmm2/m512/m64bcst{er} (V5+FP16
__m128h _mm512_cvtpd_ph(__m512d a)
__m128h _mm512_mask_cvtpd_ph(__m128h s, __mmask8 k, __m512d a)
__m128h _mm512_maskz_cvtpd_ph(__mmask8 k, __m512d a)
__m128h _mm512_cvt_roundpd_ph(__m512d a, int r)
__m128h _mm512_mask_cvt_roundpd_ph(__m128h s, __mmask8 k, __m512d a, int r)
__m128h _mm512_maskz_cvt_roundpd_ph(__mmask8 k, __m512d a, int r)
x86/x64 SIMD Instruction List
Feedback