Programmatic method for reducing cost of control in parallel processes

Abstract

A parallel compiler exploits temporal recursion to reduce the cost of control code generated in transforming a sequential nested loop program into a set of parallel processes mapped to an array of processors. A parallel compiler process transforms a nested loop program into a set of single loops, where each single loop is assigned to execute on a processor element in a parallel processor array. The parallel compiler obtains a mapping of iterations of the nested loop to processor elements in the array and a schedule of start times for initiating execution of the iterations on corresponding processor elements in the array. Based on this mapping and iteration schedule, the parallel compiler generates code to compute iteration coordinates on a processor element for an iteration of the single loop from iteration coordinates computed on the same processor element for a previous iteration of the single loop. The parallel compiler uses this method to generate code to compute loop indices, memory addresses, and tests of loop bounds efficiently based on values from a previous iteration.