Parallel execution of multiple sequential instruction streams
Matthew Chang, Y. K. Chan · 2002
Most of the existing loop partitioning schemes require the dependence distances between the iterations of the loops to be constants which means that these schemes will not be applicable when the dependence distance is variable. When each iteration of a loop is treated as a sequential instruction stream, parallel execution of these loops is possible if the data dependency between each iteration can be resolved during run-time. We propose a parallel architecture which is able to resolve these kind of undetermined data dependencies. A notable feature of the proposed architecture is that it will initiate executing multiple instruction streams before the data dependency between the instruction streams, if any, has been resolved and will restart the execution if necessary.>