Polygonal Iteration Space Partitioning

This work presents a new set of loop transformations to expose and maximize data locality in loop-nests with non-uniform reuse patterns. The proposed set of transformations use the norms of the Polyhedral Model to represent loop-nests and then leverages such a representation to partition the iteration space into polygonally shaped partitions with maximum locality. However, the partitioning algorithm tends to produce partitions with complex geometry (shape) and with progressively smaller number of iterations, which, in practice, introduces much run-time overhead. This work also focuses on containing the number of partitions and properly manage their geometry at run-time, to contain unnecessary overhead. The proposed transformations also exposes loop level parallelism, by grouping together independent iterations, thus improving performance of both serial and parallel execution. In parallel execution a selective mapping of partitions to threads based on the type of reuse these partitions exhibit is proposed. The proposed transformations show a consistent performance speedup on serial execution (up to 1.2x over Polly) and parallel execution (up to 3.17x over PLuTo) of some loop-nests.