Value-based constraint control flow integrity

Abstract

Control ow integrity (CFI) is a generic technique that prevents a control ow hijackingattacks by verifying the legitimacy of indirect branches against a predened set of targets. State-of-theartCFI solutions focus on reducing the number of targets using the context of a program such as the pathto the indirect branch and the origin of the code pointer. However, these solutions work with an impracticalassumption that the attacker only compromises control data

non-control data such as condition data that canalso be abused by attackers are not considered. To overcome these limitations, in this paper, we proposevalue-based constraint CFI (vCFI) to improve the effectiveness of CFI by retrieving and protecting alldata that can potentially be manipulated for control ow hijacking. We rst perform static analysis suchas dependency, condition, and data analyses to derive all control ow-related data. Then, vCFI protectsthese data during runtime by instrumenting a program to be hardened. We implemented vCFI as a compilerextension and evaluated its performance using SPEC CPU2006. The performance degradation caused byadopting vCFI was reasonable, and the average overhead was 13.6%.