Intel Intrinsic suffixes
MMX _pi16 4 x packed short int _pi32 2 x packed int _pi8 8 x packed signed char _pu16 4 x packed unsigned short int _pu8 8 x packed unsigned char _si64 single 64-bit binary (logical)
SSE _ps 4 x packed float _ss single scalar float _si32 single 32-bit int _si64 single 64-bit long int
SSE2 _epi16 8 x packed short int _epi32 4 x packed int _epi64 2 x packed long int _epi8 16 x packed signed char _epu16 8 x packed unsigned short int _epu32 4 x packed unsigned int _epu8 16 x packed unsigned char _pd 2 x packed double _sd single scalar double _pi64 single long int _si128 single 128-bit binary (logical)
AVX/AVX2 __m256_* _ps 8 x packed float _pd 4 x packed double _epi16 16 x packed short int _epi32 8 x packed int _epi64 4 x packed long int _epi8 32 x packed signed char _epu16 16 x packed unsigned short int _epu32 8 x packed unsigned int _epu8 32 x packed unsigned char _ss single scalar float (broadcast/splat) _sd single scalar double _si256 single 256-bit binary (logical) _pd256 cast / zero extend _ps256 cast / zero extend _pd128 cast _ps128 cast
AVX512 __m512_* _ps 16 x packed float _pd 8 x packed double _epi16 32 x packed short int _epi32 16 x packed int _epi64 8 x packed long int _epi8 64 x packed signed char _epu16 32 x packed unsigned short int _epu32 16 x packed unsigned int _epu64 8 x packed unsigned long int _epu8 64 x packed unsigned char _ss single scalar float _sd single scalar double _si512 single 512-bit binary (logical) _pd512 cast / zero extend _ps512 cast / zero extend