A Survey of Techniques for Improving Security of GPUs

Abstract

Graphics processing unit (GPU), although a powerful performance-booster, also has many security vulnerabilities. Due to these, the GPU can act as a safe-haven for stealthy malware and the weakest 'link' in the security 'chain'. In this paper, we present a survey of techniques for analyzing and improving GPU security. We classify the works on key attributes to highlight their similarities and differences. More than informing users and researchers about GPU security techniques, this survey aims to increase their awareness about GPU security vulnerabilities and potential countermeasures.

Virtually secure: A taxonomic assessment of cybersecurity challenges in virtual reality environments

“…VR depends on GPU devices to process rendered images. As such, when exploited, GPU vulnerabilities can have direct impact on VR experience [21]. Odeleye et al have developed frame rate manipulation attacks that exploit GPU vulnerabilities to cause missed and dropped frames in frame processing and can cause considerable discomfort to the users [15].…”

GPU Devices for Safety-Critical Systems: A Survey

“…These issues have been a matter of different categories of attacks (see survey [129]), such as information extraction using binary leaking or covert-side channel attacks that take advantage of cache information, memory access patterns timing information, electromagnetic signals, or power. This set of techniques has been able to extract input data, architecture and parameters of the model.…”

Securing GPU via region-based bounds checking

“…By running a malicious gadget code on an altered control flow path, attackers can collect sensitive data or escalate the attacker's privilege to the system administrator. Recent work has shown that GPUs are also vulnerable to buffer overflow attacks [12,14,47]. For example, research by Miele [45] describes GPU attack scenarios in which the attacker manipulates the function pointer by using a buffer overflow and successfully forces the GPU kernel to execute the malicious function.…”