An improved transformational approach to the scheduling problem in high-level synthesis is described. Based on an existing approach called SALSA, it uses an extended move set and lower bounds on resource costs to quickly find high-quality schedules of data-oriented control-data flow graphs. Results show the ability to find high-quality schedules for difficult scheduling problems in small amounts of CPU time. Results show that in contrast to other approaches, execution times can actually decrease as schedule length increases
Published in:
Circuits and Systems, 1993., ISCAS '93, 1993 IEEE International Symposium on
Date of Conference: 3-6 May 1993