[−][src]Macro safe_arch::blend_imm_m256
This is supported with target feature
avx
only.Blends the f32
lanes according to the immediate mask.
Each bit 0 though 7 controls lane 0 through 7. Use 0 for the $a
value and
1 for the $b
value.
let a = m256::from_array([10.0, 20.0, 30.0, 40.0, 50.0, 60.0, 70.0, 80.0]); let b = m256::from_array([100.0, 200.0, 300.0, 400.0, 500.0, 600.0, 700.0, 800.0]); // let c = blend_imm_m256!(a, b, 0b0011_0110).to_array(); assert_eq!(c, [10.0, 200.0, 300.0, 40.0, 500.0, 600.0, 70.0, 80.0]);