Differential computation analysis: Hiding your white-box designs is not enough

Abstract

Although all current scientific white-box approaches of standardized cryptographic primitives are broken, there is still a large number of companies which sell “secure” white-box products. In this paper, we present a new approach to assess the security of white-box implementations which requires neither knowledge about the look-up tables used nor any reverse engineering effort. This differential computation analysis (DCA) attack is the software counterpart of the differential power analysis attack as applied by the cryptographic hardware community. We developed plugins to widely available dynamic binary instrumentation frameworks to produce software execution traces which contain information about the memory addresses being accessed. To illustrate its effectiveness, we show how DCA can extract the secret key from numerous publicly (non-commercial) available white-box programs implementing standardized cryptography by analyzing these traces to identify secret-key dependent correlations. This approach allows one to extract the secret key material from white-box implementations significantly faster and without specific knowledge of the white-box design in an automated manner.