An Improved Method On Static Binary Analysis To Enhance The Context-Sensitive Cfi

Qintao Shen, Lei Luo, Jun Ma, Jie Yu, Qingbo Wu, Yongqi Ma, Zhengji LiuZenodo 2017 · DOI

Abstract

Control Flow Integrity (CFI) is one of the most<br> promising technique to defend Code-Reuse Attacks (CRAs).<br> Traditional CFI Systems and recent Context-Sensitive CFI use coarse<br> control flow graphs (CFGs) to analyze whether the control flow<br> hijack occurs, left vast space for attackers at indirect call-sites. Coarse<br> CFGs make it difficult to decide which target to execute at indirect<br> control-flow transfers, and weaken the existing CFI systems actually.<br> It is an unsolved problem to extract CFGs precisely and perfectly<br> from binaries now. In this paper, we present an algorithm to get a<br> more precise CFG from binaries. Parameters are analyzed at indirect<br> call-sites and functions firstly. By comparing counts of parameters<br> prepared before call-sites and consumed by functions, targets of<br> indirect calls are reduced. Then the control flow would be more<br> constrained at indirect call-sites in runtime. Combined with CCFI,<br> we implement our policy. Experimental results on some popular<br> programs show that our approach is efficient. Further analysis show<br> that it can mitigate COOP and other advanced attacks.
BibTeX
@inproceedings{shen2017improved,
  author = {Qintao Shen and Lei Luo and Jun Ma and Jie Yu and Qingbo Wu and Yongqi Ma and Zhengji Liu},
  title = {{An Improved Method On Static Binary Analysis To Enhance The Context-Sensitive Cfi}},
  booktitle = {Zenodo},
  year = {2017},
  doi = {10.5281/zenodo.1131424},
}

Loading related content...

Figure