ScanSAT: Unlocking Static and Dynamic Scan Obfuscation

Abstract

While financially advantageous, outsourcing key steps, such as testing, to potentially untrusted Outsourced Assembly and Test (OSAT) companies may pose a risk of compromising on-chip assets. Obfuscation of scan chains is a technique that hides the actual scan data from the untrusted testers; logic inserted between the scan cells, driven by a secret key, hides the transformation functions that map the scan-in stimulus (scan-out response) and the delivered scan pattern (captured response). While static scan obfuscation utilizes the same secret key, and thus, the same secret transformation functions throughout the lifetime of the chip, dynamic scan obfuscation updates the key periodically. In this paper, we propose ScanSAT: an attack that transforms a scan obfuscated circuit to its logic-locked version and applies the Boolean satisfiability (SAT) based attack, thereby extracting the secret key. We implement our attack, apply on representative scan obfuscation techniques, and show that ScanSAT can break both static and dynamic scan obfuscation schemes with 100% success rate. Moreover, ScanSAT is effective even for large key sizes and in the presence of scan compression.

A Secure Scan Controller for Protecting Logic Locking

“…Attackers who own the reverse-engineered netlist can resort to ScanSAT attack [22]. Based on a SAT solver, it targets scan protection methods that corrupt scan data in the absence of a secret key.…”

“…However, as is the case with logic locking, the obfuscated scan output reflects the inversion effect of key-gates and it can be used to trace back to the secret key. ScanSAT attack is indeed capable of modelling the locked scan chains as a combination [22] Shift & Leak [33] Encrypt Flip-Flop [26] × Dynamic Scan Obfuscation [32] × Secure Cell [25] × Proposed denotes resilience against the attack × denotes susceptibility to the attack of key-gates inserted at PPIs and PPOs. Using the same attack model as SAT attack, the secret key is deduced by the SAT solver.…”

DynUnlock: Unlocking Scan Chains Obfuscated using Dynamic Keys

“…We modify the code-base to dump a conjunctive normal form (CNF) after each iteration, which may reveal some of the seed bits. Unlike [16], we can carry out our attack for just one capture cycle. To recover more bits, we restart the LFSR circuit and obtain a new DIP and its corresponding output pattern from the SAT tool, and recover more seed bits.…”

“…The defense in [13] can be viewed as the most rigorous, and thus, the most secure, dynamic scan locking defense. Indeed, the defense in [12] was broken recently even for its most rigorous version where the key is updated for every pattern (p=1) [16]. Yet, the defense in [13] remains unbroken.…”

Fortifying RTL Locking Against Oracle-Less (Untrusted Foundry) and Oracle-Guided Attacks

“…While DisORC demonstrates resilience against oracle-guided attacks, it is only applied at the gate-level for a locked netlist. Scan locking techniques [18], [19], [20] which insert key-driven logic on scan paths are also shown to be vulnerable to modeling-based attacks [21], [22].…”