An efficient algorithm for performance-optimal FPGA technology mapping with retiming

Jason Cong, Chang Wu

Abstract: It is known that most field programmable gate array (FPGA) mapping algorithms consider only combinational circuits. Pan and Liu [1996] recently proposed a novel algorithm, named SeqMapII, of technology mapping with retiming for clock period minimization. Their algorithm, however, requires O(K/sup 3/n/sup 5/log(Kn/sup 2/)logn) run time and O(K/sup 2/n/sup 2/) space for sequential circuits with n gates. In practice, these requirements are too high for targeting K-lookup-table-based FPGA's implementing medium or large designs. In this paper, we present three strategies to improve the performance of the SeqMapII algorithm significantly. Our algorithm works in O(K/sup 2/nln|P/sub v/|logn) run time and O(K|P/sub v/|) space, where n/sub l/ is the number of labeling iterations and |P/sub v/| is the size of the partial flow network. In practice, both n/sub l/ and |P/sub v/| are less than n. Area minimization is also considered in our algorithm based on efficient low-cost K-cut computation.