COMRace: Detecting Data Race Vulnerabilities in COM Objects

Authors: 

Fangming Gu and Qingli Guo, Institute of Information Engineering, Chinese Academy of Sciences and School of Cyber Security, University of Chinese Academy of Sciences; Lian Li, Institute of Computing Technology, Chinese Academy of Sciences and School of Computer Science and Technology, University of Chinese Academy of Sciences; Zhiniang Peng, Sangfor Technologies Inc and Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences; Wei Lin, Xiaobo Yang, and Xiaorui Gong, Institute of Information Engineering, Chinese Academy of Sciences and School of Cyber Security, University of Chinese Academy of Sciences

Abstract: 

The Microsoft Component Object Model (COM) is the foundation for many key Microsoft technologies and we develop COMRace, the first data race vulnerability detection tool for commercial off-the-shelf COM objects. COMRace targets a severe but previously overlooked flaw in the COM threading model, which makes COM objects prone to data race attacks. In COMRace, we apply static binary analyses to identify thread-unsafe interface methods in off-the-shelf COM binaries, then further verify binary analyses results with automatically synthesized proof-of-concept exploits (PoC). We have applied COMRace to 10,420 registered COM objects on the windows platform and the tool reports 186 vulnerable interface methods. COMRace automatically synthesizes 234 PoCs for 256 selected method pairs (82 unsafe methods) with conflict accesses, and there are 194 PoCs triggering race conditions. Furthermore, 145 PoCs lead to critical memory corruptions, exposing 26 vulnerabilities confirmed by the Common Vulnerabilities and Exposures (CVE) database.

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@inproceedings {281348,
author = {Fangming Gu and Qingli Guo and Lian Li and Zhiniang Peng and Wei Lin and Xiaobo Yang and Xiaorui Gong},
title = {{COMRace}: Detecting Data Race Vulnerabilities in {COM} Objects},
booktitle = {31st USENIX Security Symposium (USENIX Security 22)},
year = {2022},
isbn = {978-1-939133-31-1},
address = {Boston, MA},
pages = {3019--3036},
url = {https://www.usenix.org/conference/usenixsecurity22/presentation/gu-fangming},
publisher = {USENIX Association},
month = aug,
}

Presentation Video