I. Introduction
Recently, the reconfigurable logic (RL) circuits have attracted much attention due to their expanding applications: artificial intelligence, data analytics, video transcoding, and cybersecurity. The most well-known RL chip is a field-programmable gate array (FPGA) which consists of arrays of simple logic functions and memories. For the past 30 years, FPGA chips have achieved higher density and higher speed by reducing the size of devices [1], [2]. However, even if the FPGA market is burgeoning, the applications of RL circuits are still limited because they show lower density, higher power consumption, and lower speed than application-specific integrated circuits. These limitations are originated from the fact that the conventional RL circuits only consist of CMOS devices on a silicon substrate for their logic and routing parts.