Four types of experiments are executed combined with global memory accesses:
Single precision Flops (multiply-additions)
Double precision Flops (multiply-additions)
Half precision Flops (multiply-additions)
Integer multiply-addition operations
Building is based now on CMake files. Each implementation resides in a separate folder:
CUDA implementation: mixbench-cuda
OpenCL implementation: mixbench-opencl
HIP implementation: mixbench-hip
SYCL implementation: mixbench-sycl
CUDA Study Steps
NOTE
CUDA gpu 编程学习,基于 《CUDA 编程——基础与实践》(樊哲勇)。
包含章节:
CUDA 官方文档
CUDA c++编程指南
CUDA c++最佳实践指南
CUDA 运行时API手册
CUDA 数学函数库API手册
CUDA 编程案例
CUDA Samples
CUDA 性能测试
CUDA Bechmarks
Four types of experiments are executed combined with global memory accesses: Single precision Flops (multiply-additions) Double precision Flops (multiply-additions) Half precision Flops (multiply-additions) Integer multiply-addition operations
Building is based now on CMake files. Each implementation resides in a separate folder: CUDA implementation: mixbench-cuda OpenCL implementation: mixbench-opencl HIP implementation: mixbench-hip SYCL implementation: mixbench-sycl
生成的测试结果类似: