Search for: All records

Pre-silicon tools for hardening hardware against side-channel and fault injection attacks have become popular recently. However, the security of the system is still threatened by sophisticated physical attacks, which exploit the physical layer characteristics of the computing system beyond the integrated circuits (ICs) and, therefore, bypass the conventional countermeasures. Further, environmental conditions for the hardware can also impact side-channel leakage and fault vulnerability in unexpected ways that are challenging to model in pre-silicon. Thus, attacks cannot be addressed solely by conventional countermeasures at higher layers of the compute stack due to the lack of awareness about the events occurring at the physical layer during runtime. In this paper, we first discuss why the current pre-silicon security and verification tools might fail to achieve security against physical threats in the post-silicon phase. Afterward, we provide insights from the fields of power/signal integrity (PI/SI), and failure analysis (FA) to understand the fundamental issue with the failed current practices. We argue that hardware-based moving target defenses (MTDs) to randomize the physical fabric’s characteristics of the system can mitigate such unaccounted post-silicon threats. We show the effectiveness of such an approach by presenting the results of two case studies in which we perform powerful attacks, such as impedance analysis and laser voltage probing. Finally, we review the overhead of our proposed approach and show that the imposed overhead by MTD solutions can be addressed by making them active only when a threat is detected. 

Design-for-test/debug (DfT/D) introduces scan chain testing to increase testability and fault coverage by inserting scan flip-flops. However, these scan chains are also known to be a liability for security primitives. In previous research, the dynamically obfuscated scan chain (DOSC) was introduced to protect logic-locking keys from scan-based attacks by obscuring test patterns and responses. In this paper, we present DOSCrack, an oracle-guided attack to de-obfuscate DOSC using symbolic execution and binary clustering, which significantly reduces the candidate seed space to a manageable quantity. Our symbolic execution engine employs scan mode simulation and satisfiability modulo theories (SMT) solvers to reduce the possible seed space, while obfuscation key clustering allows us to effectively rule out a group of seeds that share similarities. An integral component of our approach is the use of sequential equivalence checking (SEC), which aids in identifying distinct simulation patterns to differentiate between potential obfuscation keys. We experimentally applied our DOSCrack framework on four different sizes of DOSC benchmarks and compared their runtime and complexity. Finally, we propose a low-cost countermeasure to DOSCrack which incorporates a nonlinear feedback shift register (NLFSR) to increase the effort of symbolic execution modeling and serves as an effective defense against our DOSCrack framework. Our research effectively addresses a critical vulnerability in scan-chain obfuscation methodologies, offering insights into DfT/D and logic locking for both academic research and industrial applications. Our framework highlights the need to craft robust and adaptable defense mechanisms to counter evolving scan-based attacks.