Function safe_arch::shuffle_m256d

source ยท 
pub fn shuffle_m256d<const IMM: i32>(a: m256d, b: m256d) -> m256d
Available with target feature avx only.
Expand description

Shuffle the f64 lanes from a and b together using an immediate control value.

The control value uses the lowest 4 bits only.

  • bit 0 picks between lanes 0 or 1 from A.
  • bit 1 picks between lanes 0 or 1 from B.
  • bit 2 picks between lanes 2 or 3 from A.
  • bit 3 picks between lanes 2 or 3 from B.

Note that this shuffle cannot move data between the lower half of the lanes and the upper half of the lanes.