An Algorithm for Hardware/Software Partitioning Using Constraint-Driven and Elimination of Slack Time

One of the key problems in hardware/software co-design is hardware/software partitioning. The problems in the presented algorithms of hardware/software partitioning are analyzed and a new hardware/software partitioning algorithm is presented, based on the combination of the constraint-driven and elimination of slack time. First, an approach used to get area-time（A-T） curve of nodes is put forward. The nodes are mapped to hardware or software by comparing the value of time constraint criticality with threshold. Hardware/software executive areas are decided in terms of hardware/software areas constraints criticality, and then the executive time of nodes is obtained through A-T curve of nodes. Finally, the performance optimization is done further by eliminating slack time existing among nodes.