Qintao Shen and Lei Luo and Jun Ma and Jie Yu and Qingbo Wu and Yongqi Ma and Zhengji Liu
An Improved Method on Static Binary Analysis to Enhance the ContextSensitive CFI
848 - 854
2017
11
7
International Journal of Computer and Information Engineering
https://publications.waset.org/pdf/10007533
https://publications.waset.org/vol/127
World Academy of Science, Engineering and Technology
Control Flow Integrity (CFI) is one of the most
promising technique to defend CodeReuse Attacks (CRAs).
Traditional CFI Systems and recent ContextSensitive CFI use coarse
control flow graphs (CFGs) to analyze whether the control flow
hijack occurs, left vast space for attackers at indirect callsites. Coarse
CFGs make it difficult to decide which target to execute at indirect
controlflow transfers, and weaken the existing CFI systems actually.
It is an unsolved problem to extract CFGs precisely and perfectly
from binaries now. In this paper, we present an algorithm to get a
more precise CFG from binaries. Parameters are analyzed at indirect
callsites and functions firstly. By comparing counts of parameters
prepared before callsites and consumed by functions, targets of
indirect calls are reduced. Then the control flow would be more
constrained at indirect callsites in runtime. Combined with CCFI,
we implement our policy. Experimental results on some popular
programs show that our approach is efficient. Further analysis show
that it can mitigate COOP and other advanced attacks.
Open Science Index 127, 2017