如何分析 OpenMP 瓶颈

How to profile OpenMP bottlenecks(如何分析 OpenMP 瓶颈)

问题描述

我有一个由 OpenMP 并行化的循环，但由于任务的性质，有 4 个 critical 子句.

I have a loop that has been parallelized by OpenMP, but due to the nature of the task, there are 4 critical clauses.

分析加速并找出哪个关键子句(或者可能是非关键子句(！))在循环中占用最多时间的最佳方法是什么?

What would be the best way to profile the speed up and find out which of the critical clauses (or maybe non-critical(!) ) take up the most time inside the loop?

我使用 Ubuntu 10.04 和 g++ 4.4.3

I use Ubuntu 10.04 with g++ 4.4.3

推荐答案

OpenMP 包括用于测量时序性能的函数 omp_get_wtime() 和 omp_get_wtick() (此处的文档)，我建议使用这些.

OpenMP includes the functions omp_get_wtime() and omp_get_wtick() for measuring timing performance (docs here), I would recommend using these.

否则尝试分析器.我更喜欢可以在此处找到的 google CPU 分析器.

Otherwise try a profiler. I prefer the google CPU profiler which can be found here.

还有这个 答案.

这篇关于如何分析 OpenMP 瓶颈的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持编程学习网！