Extracting Both Affine and Non-linear Synchronization-Free Slices in Program Loops

An approach is presented permitting for extracting both affine and non-linear synchronization-free slices in program loops. It requires an exact dependence analysis. To describe and implement the approach, the dependence analysis by Pugh and Wonnacott was chosen where dependences are found in the form of tuple relations. The approach is based on operations on integer tuple relations and sets and it has been implemented and verified by means of the Omega project software. Results of experiments with the UTDSP benchmark suite are discussed. Speed-up and efficiency of parallel code produced by means of the approach is studied.