Thu 21 Jun 2018 15:15 - 15:40 at Grand Ballroom AB - Concurrency Debugging Chair(s): Tony Hosking

As GPUs have become an integral part of nearly every pro- cessor, GPU programming has become increasingly popular. GPU programming requires a combination of extreme levels of parallelism and low-level programming, making it easy for concurrency bugs such as data races to arise. These concurrency bugs can be extremely subtle and di cult to debug due to the massive numbers of threads running concurrently on a modern GPU. While some tools exist to detect data races in GPU pro- grams, they are often prohibitively slow or focused only on a small class of data races in shared memory. Compared to prior work, our race detector, CURD, can detect data races precisely on both shared and global memory, selects an appropriate race detection algorithm based on the synchronization used in a program, and utilizes efficient compiler instrumentation to reduce performance overheads. Across 53 benchmarks, we find that using CURD incurs an aver- age slowdown of just 2.88x over native execution. CURD is 2.1x faster than Nvidia’s CUDA-Racecheck race detector, despite detecting a much broader class of races. CURD finds 35 races across our benchmarks, including bugs in established benchmark suites and in sample programs from Nvidia.

Thu 21 Jun

pldi-2018-papers
14:00 - 15:40: PLDI Research Papers - Concurrency Debugging at Grand Ballroom AB
Chair(s): Tony HoskingAustralian National University / Data61
pldi-2018-papers14:00 - 14:25
Talk
Hongyu LiuUniversity of Texas at San Antonio, USA, Sam SilvestroUniversity of Texas at San Antonio, USA, Wei WangUniversity of Texas at San Antonio, USA, Chen TianHuawei Lab, USA, Tongping Liu
Media Attached
pldi-2018-papers14:25 - 14:50
Talk
Bozhen LiuTexas A&M University, USA, Jeff HuangTexas A&M University
Media Attached
pldi-2018-papers14:50 - 15:15
Talk
Jake RoemerOhio State University, Kaan GençOhio State University, USA, Michael BondOhio State University
Media Attached
pldi-2018-papers15:15 - 15:40
Talk
Yuanfeng PengUniversity of Pennsylvania, Vinod Grover, Joseph DeviettiUniversity of Pennsylvania
Media Attached