LockedDown: Exploiting Contention on Host-GPU PCIe Bus for Fun and Profit
LockedDown: Exploiting Contention on Host-GPU PCIe Bus for Fun and Profit
复制标题
DOI：
10.1109/eurosp53844.2022.00025
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Mert Side;F. Yao;Zhenkai Zhang
中科院分区：
文献类型：
--
作者：
Mert Side;F. Yao;Zhenkai Zhang
现代图形处理单元 (GPU) 的部署在传统计算和云计算中都迅速增长。然而，这种广泛部署带来的潜在安全问题尚未得到彻底调查。在本文中，我们披露了一个新的可利用的侧通道漏洞，该漏洞普遍存在于配备现代 GPU 的系统中。此漏洞是由于主机 GPU PCIe 总线上引起的可测量争用造成的。为了证明此漏洞的可利用性，我们进行了两个案例研究。在第一个案例研究中，我们利用该漏洞构建了一个在虚拟化 NVIDIA GPU 上运行的跨虚拟机隐蔽通道。据我们所知，这是第一个探讨虚拟化 GPU 环境下隐蔽通道攻击的工作。隐蔽通道的速度可达 90 kbps，且错误率相当低。在第二个案例研究中，我们利用该漏洞发起网站指纹攻击，可以准确推断用户浏览了哪些网页。该攻击在 Windows 和 Linux 上针对 Chrome 和 Firefox 等流行浏览器进行了评估，结果表明这种指纹识别方法可以实现高达 95.2% 的准确率。此外，针对Tor浏览器对该攻击进行了评估，准确率高达90.6%。
The deployment of modern graphics processing units (GPUs) has grown rapidly in both traditional and cloud computing. Nevertheless, the potential security issues brought forward by this extensive deployment have not been thoroughly investigated. In this paper, we disclose a new exploitable side-channel vulnerability that ubiquitously exists in systems equipped with modern GPUs. This vulnerability is due to measurable contention caused on the host-GPU PCIe bus. To demonstrate the exploitability of this vulnerability, we conduct two case studies. In the first case study, we exploit the vulnerability to build a cross-VM covert channel that works on virtualized NVIDIA GPUs. To the best of our knowledge, this is the first work that explores covert channel attacks under the circumstances of virtualized GPUs. The covert channel can reach a speed up to 90 kbps with a considerably low error rate. In the second case study, we exploit the vulnerability to mount a website fingerprinting attack that can accurately infer which web pages are browsed by a user. The attack is evaluated against popular browsers like Chrome and Firefox on both Windows and Linux, and the results show that this fingerprinting method can achieve up to 95.2% accuracy. In addition, the attack is evaluated against Tor browser, and up to 90.6% accuracy can be achieved.