Runtime Profiling of OpenCL Workloads Using LLVM-based Code Instrumentation

  • Yu, Yongseung.
  • Kang, Seokwon
  • Park, Yongjun
Citations

WEB OF SCIENCE

1
Citations

SCOPUS

1

초록

GPUs, which are widely used high-performance hardware accelerators in heterogeneous computing, and programming models for architectures such as OpenCL and CUDA, have recently been developed to achieve high productivity. LLVM is an open-source compiler infrastructure that enables low-level optimization through LLVM intermediate representation (LLVM IR) in various programming language environments. In this paper, we propose a fully-automatic Dynamic Profiling framework which performs instruction-level analysis through IR-level code instrumentation for typical OpenCL workload kernels.

키워드

Dynamic ProfilingGPULLVMOpenCLGraphics processing unitProgram compilersCode instrumentationDynamic ProfilingHeterogeneous computingHigh-performance hardwareIntermediate representationsLanguage environmentLLVMOpenCLOpen source software
제목
Runtime Profiling of OpenCL Workloads Using LLVM-based Code Instrumentation
저자
Yu, Yongseung.Kang, SeokwonPark, Yongjun
DOI
10.1109/TENCON.2018.8650390
발행일
2019-10
유형
Conference Paper
저널명
IEEE Region 10 Annual International Conference, Proceedings/TENCON
2018-October
페이지
1520 ~ 1524