Function safe_arch::shuffle_m256

source · 
pub fn shuffle_m256<const IMM: i32>(a: m256, b: m256) -> m256
Available with target feature avx only.
Expand description

Shuffle the f32 lanes from a and b together using an immediate control value.

This works like shuffle_abi_f32_all_m128, but with the low 128 bits and high 128 bits each doing a shuffle at the same time. Each index (0..=3) only refers to a lane within a given 128 bit portion of the 256 bit inputs. You cannot cross data between the two 128 bit halves.