A polynomial time method for optimal software pipelining

A Unified Framework for Instruction Scheduling and Mapping for Function Units with Structural Hazards

“…As stated in [25], software pipelining methods vary in several aspects: (1) whether or not they consider finite resources, (2) whether they model simple or complex resource usage, (3) whether the algorithm is one-pass, iterative, incremental, or enumerative searchbased. In particular, in terms of resource constraints, the work reported in [1,22,24,33] does not consider any resource constraint while the methods reported in [2,4,11,16,21,26,[34][35][36] deal with function unit constraints but with simple resource usage. Both function unit and register resource constraints are considered in [7,18].…”

Scientific Foundations

“…It amounts to scheduling techniques as found in [17], [34], [22], but their application in the domain of SoC design raises new opportunities in optimization, such as those based on low-power dissipation criteria.…”

FEADS: A Framework for Exploring the Application Design Space on Network Processors

“…We use the concept of r-periodic scheduling (20) to determine the optimal number of application instances to be scheduled. In this scheme, the schedule of two consecutive iterations of a loop may not be the same, but will repeat every r iterations.…”

“…This is similar to the r-periodic schedule discussed in Ref. 20. Hence the number of instances required to arrive at the optimal throughput needs to be determined.…”

“…This is similar to r-periodic scheduling. (20) We propose a simple heuristic to estimate the number of instances required. To obtain the dynamic performance metrics, we use the Petri net (PN) model developed in Ref.…”