PADDB - Packed ADD Byte

PADDB xmm1, xmm2/m128    (S2
__m128i _mm_add_epi8(__m128i a, __m128i b)

For each BYTE, calculate (1) + (2) and set the result to (3).
VPADDB xmm1, xmm2, xmm3/m128    (V1
__m128i _mm_add_epi8(__m128i a, __m128i b)
VPADDB xmm1{k1}{z}, xmm2, xmm3/m128    (V5+BW+VL
__m128i _mm_mask_add_epi8(__m128i s, __mmask16 k, __m128i a, __m128i b)
__m128i _mm_maskz_add_epi8(__mmask16 k, __m128i a, __m128i b)

For each BYTE, calculate (1) + (2) and set the result to (3).
VPADDB ymm1, ymm2, ymm3/m256    (V2
__m256i _mm256_add_epi8(__m256i a, __m256i b)
VPADDB ymm1{k1}{z}, ymm2, ymm3/m256    (V5+BW+VL
__m256i _mm256_mask_add_epi8(__m256i s, __mmask32 k, __m256i a, __m256i b)
__m256i _mm256_maskz_add_epi8(__mmask32 k, __m256i a, __m256i b)

For each BYTE, calculate (1) + (2) and set the result to (3).
VPADDB zmm1{k1}{z}, zmm2, zmm3/m512    (V5+BW
__m512i _mm512_add_epi8(__m512i a, __m512i b)
__m512i _mm512_mask_add_epi8(__m512i s, __mmask64 k, __m512i a, __m512i b)
__m512i _mm512_maskz_add_epi8(__mmask64 k, __m512i a, __m512i b)

For each BYTE, calculate (1) + (2) and set the result to (3).

x86/x64 SIMD Instruction List  Feedback