[−][src]Function safe_arch::shr_each_u64_m256i
#[must_use]pub fn shr_each_u64_m256i(a: m256i, count: m256i) -> m256i
This is supported with target feature
avx2
only.Lanewise u64
shift right by the matching i64
lane in count
.
let a = m256i::from([0_u64, 1111, 2999, 13888]); let count = m256i::from([1_u64, 2, 3, 4]); let b: [u64; 4] = shr_each_u64_m256i(a, count).into(); assert_eq!(b, [0, 277, 374, 868]);
- Intrinsic:
_mm256_srlv_epi64
- Assembly:
vpsrlvq ymm, ymm, ymm