The Preprocessed Doacross Loop

Abstract : Dependencies between loop iterations cannot always be characterized during program compilation. Doacross loops typically make use of a-priori knowledge of inter-iteration dependencies to carry out required synchronizations. We propose a type of doacross loop that allows us to schedule iterations of a loop among processors without advance knowledge of inter-iteration dependencies. The method proposed for loop iterations requires us to carry out parallelizable preprocessing and postprocessing steps during program execution. Keywords: Shared memory, Runtime.