This paper studies and evaluates the extent to which automated compiler techniques can defend against timing-based side channel attacks on modern x86 processors. We study how modern x86 processors can leak timing information through side channels that relate to data flow. We study the efficiency, effectiveness, portability, predictability and sensitivity of several mitigating code transformations that eliminate or minimize key-dependent execution time variations. Furthermore, we discuss the extent to which compiler backends are a suitable tool to provide automated support for the proposed mitigations.

Existing compiler techniques can transform code to make its timing behavior independent of sensitive values to prevent information leakage through time side channels. Those techniques are hampered, however, by their static nature and dependence on details of the processor targeted during the compilation. This paper presents a dynamic compiler approach based on offline profiles and JIT compiler strategies. This approach reduces overhead significantly and enables a trade-off between provided protection and overhead. Furthermore, it supports adaptive policies in which the protection adapts to run-time changes in the requirements. A prototype implementation in the Jikes Research VM is evaluated on RSA encryption, HMAC key verification, and IDEA encryption.