Intrinsics for Packed Move with Extend Operations
- _mm256_cvtepi8_epi16/32/64
Performs packed move with sign-extend on 8-bit signed integers to 16/32/64-bit integers. The corresponding Intel® AVX2 instruction is VPMOVSXBW,VPMOVSXBD, or VPMOVSXBQ.
- _mm256_cvtepi16_epi32/64
Performs packed move with sign-extend on 16-bit signed integers to 32/64-bit integers. The corresponding Intel® AVX2 instruction is VPMOVSXWD orVPMOVSXWQ.
- _mm256_cvtepi32_epi64
Performs packed move with sign-extend on 32-bit signed integers to 64-bit integers. The corresponding Intel® AVX2 instruction is VPMOVSXDQ.
- _mm256_cvtepu8_epi16/32/64
Performs packed move with zero-extend on 8-bit unsigned integers to 16/32/64-bit integers. The corresponding Intel® AVX2 instruction is VPMOVZXBW,VPMOVZXBD, or VPMOVZXBQ.
- _mm256_cvtepu16_epi32/64
Performs packed move with zero-extend on 16-bit unsigned integers to 32/64-bit integers. The corresponding Intel® AVX2 instruction is VPMOVZXWD or VPMOVZXWQ.
- _mm256_cvtepu32_epi64
Performs packed move with zero-extend on 32-bit unsigned integers to 64-bit integers. The corresponding Intel® AVX2 instruction is VPMOVZXDQ.