This paper proposes a function-generation based area-aware configurable logic block (CLB) architecture and an associated packing technique, for SRAM-based FPGAs. The new CLB architecture provides the same logic functionality, but occupies 38% less area, consumes 38.31% less power and requires 50% less configuration-bits per CLB when compared to the standard 4-LUT CLB architecture. The proposed
... [Show full abstract] packing technique is timing-driven and is shown to produce designs with almost same routing cost and performance overhead as that produced by the T-VPack algorithm on standard benchmark circuits.