Toward Taming the Overhead Monster for Data-Flow Integrity

TL;DR

This paper introduces a hardware-assisted parallel approach to significantly reduce the performance overhead of Data-Flow Integrity (DFI), enabling its practical deployment without compromising security standards.

Contribution

A novel hardware-assisted parallel method is proposed to lower DFI overhead, maintaining original security guarantees while achieving a 4x performance improvement.

Findings

Complete enforcement of original DFI criteria achieved

Average 4x reduction in performance overhead

Validated on SPEC CPU 2006 benchmarks

Abstract

Data-Flow Integrity (DFI) is a well-known approach to effectively detecting a wide range of software attacks. However, its real-world application has been quite limited so far because of the prohibitive performance overhead it incurs. Moreover, the overhead is enormously difficult to overcome without substantially lowering the DFI criterion. In this work, an analysis is performed to understand the main factors contributing to the overhead. Accordingly, a hardware-assisted parallel approach is proposed to tackle the overhead challenge. Simulations on SPEC CPU 2006 benchmark show that the proposed approach can completely enforce the DFI defined in the original seminal work while reducing performance overhead by 4x, on average.