I. Introduction
Graphics Processing Units (GPUs) are becoming an inevitable part of every computing system because of their ability to provide fast and energy-efficient computation. Given such ability, GPUs are also now being used to accelerate a variety of cryptographic algorithms. For example, the popular Advanced Encryption Standard (AES) algorithm [21] is known to achieve significant speedups on GPUs compared to CPUs [6], [9], [17], [23] as the AES algorithm exposes abundant thread-level parallelism to leverage high bandwidth and compute throughput of GPUs. With such increasing popularity of GPUs to accelerate security-sensitive applications, it is imperative to keep GPUs secure against a variety of side-channel attacks and other security vulnerabilities.