Flexibility Inlining into Arithmetic Data-paths Exploiting A Regular Interconnection Scheme

This paper presents a design technique for coarse grained reconfigurable cores targeting mostly DSP applications. The proposed technique inlines flexibility into custom carry-save-arithmetic (CSA) datapaths exploiting a stable and canonical interconnection scheme. The canonical interconnection is revealed by a uniformity transformation imposed on the basic architectures of CSA multipliers and CSA chain-adders/subtractors. The design flow for the implementation of the core is analyzed in detail, and a novel reconfigurable architecture prototype is presented. The paper concludes with the experimental results showing that our architecture performs an average latency reduction of 32.63%, compared with datapaths of primitive computational resources, with a tolerable overhead in hardware utilization.