Available on x86-64 and target feature
avx
only.Expand description
Loads 256-bits of integer data from unaligned memory into result.
This intrinsic may perform better than _mm256_loadu_si256
when the
data crosses a cache line boundary.