PyXEL: Exploring Bitstream Analysis to Assess and Enhance the Robustness of Designs on FPGAs

Abstract

Commercial hardware-reconfigurable systems-on-chip are highly attractive for mission-critical applications in the space and automotive industries. However, their vulnerability to soft errors is a major concern, and analyzing the robustness of these systems is a complex task due to the lack of dedicated tools, information, and methodologies available. PyXEL is a tool designed to address these issues, providing the methodology for automating reliability analysis based on radiation and fault injection campaigns and facilitating the development of mitigation solutions based on customized place-and-route. Furthermore, PyXEL offers the methodology for visualizing, decoding, and analyzing the configuration data of programmable hardware devices, enabling more precise and efficient evaluation and analysis of the robustness of systems implemented on programmable hardware devices.