CAD for Assurance of Electronic Systems
 

ObfusGEM

By: Michael Zuzak (University of Maryland, College Park) and Ankur Srivastava (University of Maryland, College Park)

Stage: RTL, HDL

Summary

ObfusGEM is a comprehensive logic locking simulator designed to evaluate the supply-chain security of locked processor models. It implements an error injection framework inspired by the architectural error resilience community to “close the loop” between module-level logic locking or hardware security techniques and their application and architecture-level impact. We provide a library of existing hardware security techniques and configurations along with a simulation framework in hopes of providing a tool enabling the design and evaluation of hardware security configurations for specific processor architectures or devices.

Contact

Michael Zuzak

Input/Output Interface

  • Input: GEM5 Processor Model, Disk Image, Processor Applications, Logic Locking Configuration
  • Output: Failure Rate of Application, Time to Failure for Application, Failure State for Application

Dependencies

  • g++ version 4.8 or newer or clang version 3.1 or newer
  • Python, version 2.6 – 2.7
  • SCons, version 0.98.1 or newer
  • zlib, any recent version
  • m4, the macro processor
  • protobuf, version 2.1 or newer
  • pydot
  • Perl

Licensing Info

BSD-3-Clause License

References

Zuzak, Michael; Srivastava, Ankur

ObfusGEM: Enhancing Processor Design Obfuscation Through Security-Aware On-Chip Memory and Data Path Design Proceedings Article

In: International Symposium on Memory Systems (MEMSYS), 2020.

BibTeX

Binkert, Nathan; Beckmann, Bradford; Black, Gabriel; Reinhardt, Steven K; Saidi, Ali; Basu, Arkaprava; Hestness, Joel; Hower, Derek R; Krishna, Tushar; Sardashti, Somayeh; Sen, Rathijit; Sewell, Korey; Shoaib, Muhammad; Vaish, Nilay; Hill, Mark D; Wood, David A

The Gem5 Simulator Journal Article

In: SIGARCH Comput. Archit. News, vol. 39, no. 2, pp. 1–7, 2011, ISSN: 0163-5964.

Abstract | Links | BibTeX

Acknowledgments

  • This work was supported by the ARCS Foundation, the National Science Foundation (NSF) Grant 1642424, and the Air Force Office of Scientific Research Grant FA9550-14-1-0351

User Guide Link can be found at this link