Lock-based synchronization for GPU architectures
Yunlong Xu, Lan Gao, Rui Wang, Zhongzhi Luan, Weiguo Wu, Depei Qian · 2016
Modern GPUs have shown promising results in accelerating compute-intensive and numerical workloads with limited data sharing. However, emerging GPU applications manifest ample amount of data sharing among concurrently executing threads. Often data sharing requires mutual exclusion mechanism to ensure data integrity in multithreaded environment. Although modern GPUs provide atomic primitives that can be leveraged to construct fine-grained locks, the existing GPU lock implementations either incur frequent concurrency bugs, or lead to extremely low hardware utilization due to the Single Instruction Multiple Threads (SIMT) execution paradigm of GPUs.