Abstract
This paper proposes a methodology to access data and manage the content of distributed memories in FPGA designs through the configuration bitstream. Thanks to the methods proposed, it is possible to read and write the data content of registers without using the in/out ports of registers in a straightforward fashion. Hence, it offers the possibility of performing several operations, such as, to load, copy or compare the information stored in registers without the necessity of physical interconnections. This work includes two flows that simplify the designing process when using the proposed approach: while the first enables the protection or unprotection of writing on different partial regions through the bitstream, the second permits homogeneous instances of a design implemented in different reconfigurable regions to be obtained without losing efficiency. The approach is based and has been physically validated on the ZYNQ from Xilinx, and when using partially reconfigurable designs, it does not affect the hardware overhead nor the maximum operating frequency of the design.
Funder
Department of Education
Ministerio de Ciencia Innovación of Spain
Subject
Electrical and Electronic Engineering,Computer Networks and Communications,Hardware and Architecture,Signal Processing,Control and Systems Engineering
Reference31 articles.
1. Xilinx Corp. (2022, November 18). 7 Series FPGAs Memory Resources UG473 (v1.14). Xilinx Documentation. Available online: http://www.xilinx.com.
2. Gong, L., and Diessel, O. (2015). Functional Verification of Dynamically Reconfigurable FPGA-Based Systems, Springer.
3. Yu, H., Lee, H., Lee, S., Kim, Y., and Lee, H.M. (2018). Recent Advances in FPGA Reverse Engineering. Electronics, 7.
4. Poverelli, J., and Brewer, F. (2019, January 19–21). Direct Bitstream Processing for High Performance Feedback Control. Proceedings of the IEEE Conference on Control Technology and Applications (CCTA), Hong Kong, China.
5. Asadi, G., and Tahoori, M. (2005, January 20–22). Soft Error Rate Estimation and Mitigation for SRAM-Based FPGAs. Proceedings of the ACM/SIGDA International Symposium on Field-Programmable Gate Arrays, Monterey, CA, USA.