CUDA程序并发错误的自动检测

AUTOMATIC DETECTION OF SYNCHRONIZATION BUGS IN CUDA PROGRAMS

吴明远

11849319

硕士

计算机技术

张煜群

2020-05-30

2020-07-08

哈尔滨工业大学

深圳

在当前的工业界，GPU程序的开发和维护已经成为了重要的新平台。机器学习的任务在GPU平台上能够获得更加优秀的效果。CUDA（Computer Unified Device Architecture）是nVidia开发的一个GPU计算平台。虽然CUDA已成为通用GPU计算的主要并行计算平台和编程模型，但CUDA引发的错误模式尚未有成熟的自动化解决方案。在此篇论文中，我们将提出一个轻量级的CUDA并发错误检测工具Simulee，通过自动生成触发内存访问的输入，追踪并分析实际被执行的CUDA程序语句序列，并基于该序列，收集不同线程的内存访问信息并做出是否存在并发错误的判断。为了从效能和效率两个维度评价Simulee的性能，本文还从Github上选择了7个流行的CUDA相关项目，来作为测试的对象。实验结果证明，Simulee可以从24个手动分析的并发bug中成功检测21个出来，同时还检测出了24个尚未发现的并发错误bug，其中的10个已经被开发者确认。与此同时，Simulee在检测中体现的性能都显著强于已知的CUDA 并发错误检测的方案。

While CUDA(Computer Unified Device Architecture) has become a mainstream parallel computing platform and programming model for general-purpose GPU computing, how to effectively and efficiently detect CUDA synchronization bugs remains a challenging open problem. In this paper, we propose the first lightweight CUDA synchronization bug detection framework, namely Simulee, to model CUDA program execution by interpreting the corresponding LLVM bytecode and collecting the memory-access information for automatically detecting general CUDA synchronization bugs. To evaluate the effectiveness and efficiency of Simulee, we construct a benchmark with 7 popular CUDA-related projects from Github, upon which we conduct an extensive set of experiments. The experimental results suggest that Simulee can detect 21 out of the 24 manually identified bugs in our preliminary study and also 24 previously unknown bugs among all projects, 10 of which have already been confirmed by the developers. Furthermore, Simulee significantly outperforms state-of-the-art approaches for CUDA synchronization bug detection.

CUDA 并发错误 输入优化生成 动态语句序列分析 内存访问模型

CUDA synchronization bugs input generation dynamic execution path memory-access modeling

中文

联合培养

南方科技大学

吴明远. CUDA程序并发错误的自动检测[D]. 深圳. 哈尔滨工业大学,2020.