A Processor with Dynamically Reconfigurable Circuit for Floating-Point Arithmetic

This paper describes about dynamic reconfiguration to miniaturize arithmetic circuits in general-purpose processor. Dynamic reconfiguration is a technique to realize required functions by changing hardware construction during operation. The proposed arithmetic circuit performs floating-point arithmetic which is frequently used in science and technology. The data format is floating-point based on IEEE754. The proposed circuit is designed using VHDL, and verified the correct operation by simulations and experiments.

Authors:

• Yukinari Minagi
• Akinori Kanasugi

Keywords:

• dynamic reconfiguration
• floating-point arithmetic
• double precision
• FPGA

References:

[1] T. J. Todman, et al, "Reconfigurable computing: architectures and design
methods", IEE Proc.-Computers & Digital Techniques, vol. 152, no. 2,
pp. 193 - 207, 2005.
[2] T. Sato, H. Watanabe, K. Shiba, "Implementation of dynamically
reconfigurable processor DAPDNA-2", VLSI Design, Automation and
Test, 2005 IEEE VLSI-TSA International Symposium, pp. 323-324, 2005.
[3] M J. Myjak, J. G. Delgado-Frias, "A Medium-Grain Reconfigurable
Architecture for DSP: VLSI Design, Benchmark Mapping, and
Performance", IEEE Trans. on VLSI Systems, vol.16, no.1, pp.14-23, Jan
2008.
[4] H. Shimada, Y. Hayakawa and A. Kanasugi, "An Architecture of
Dynamically Reconfigurable Arithmetic Circuit", Proc. of 2009 Int. Conf.
on Electronics Packaging, pp.963-966, 2009.
[5] A. Kanasugi, A. Tsukahara, "A Processor for Genetic Algorithm using
Dynamically Reconfigurable Memory", Journal of Convergence
Information Technology, vol.2, no.1, pp. 4-15, 2007.
[6] A. Tsukahara, A. Kanasugi, "Genetic Algorithm with Dynamic Variable
Number of Individuals and Accuracy", International Journal of Control,
Automation, and Systems, vol. 7, no. 1, pp.1-6, 2009.
[7] MIPS Technologies, Inc., MIPS32 Architecture for Programmers Volume
I: Introduction to the MIPS32 Architecture, July 1, 2005.
[8] IEEE Std. 754 - 1985, IEEE Standard for Binary Floating-Point
Arithmetic, IEEE, 1985.
[9] Xilinx, Inc., ML401/ML402/ML403 Evaluation Platform User Guide
UG080 (v2.5), May 2006