A novel method for improving the operation autonomy of SIMD processing elements

A novel method for improving the operation autonomy of the processing elements (PE) of SIMD-like machines is presented and discussed in this paper. The paper shows that it is feasible to avoid most branches and it is also possible to emulate nested if-then-else sentences on the processing elements by combining guarded instructions and pseudo branches. This prevents unnecessary intervention by the array control unit in many data-dependant computations, particularly those with short branch sections. The paper also shows that the simplicity of the method allows it to be implemented both in fine-grain and coarse-grain SIMD/ASIMD architectures because it does not require significant additional silicon area. Finally, it is shown that pseudo branches can be used to control the power saving of those processing elements that have instructions ified.