VPOPCNTB - POPulation CouNT Byte

VPOPCNTB xmm1{k1}{z}, xmm2/m128    (V5+BITALG+VL
__m128i _mm_popcnt_epi8(__m128i a)
__m128i _mm_mask_popcnt_epi8(__m128i s, __mmask16 k, __m128i a)
__m128i _mm_maskz_popcnt_epi8(__mmask16 k, __m128i a)

For each BYTE, count number of bits set to one in (1) and store the result in (2).
VPOPCNTB ymm1{k1}{z}, ymm2/m256    (V5+BITALG+VL
__m256i _mm256_popcnt_epi8(__m256i a)
__m256i _mm256_mask_popcnt_epi8(__m256i s, __mmask32 k, __m256i a)
__m256i _mm256_maskz_popcnt_epi8(__mmask32 k, __m256i a)

For each BYTE, count number of bits set to one in (1) and store the result in (2).
VPOPCNTB zmm1{k1}{z}, zmm2/m512    (V5+BITALG
__m512i _mm512_popcnt_epi8(__m512i a)
__m512i _mm512_mask_popcnt_epi8(__m512i s, __mmask64 k, __m512i a)
__m512i _mm512_maskz_popcnt_epi8(__mmask64 k, __m512i a)

For each BYTE, count number of bits set to one in (1) and store the result in (2).
x86/x64 SIMD Instruction List  Feedback