Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Nvprof是比较旧的工具了,逐渐被官方弃用了,建议更新为nsight system #10

Open
reed-lau opened this issue Mar 8, 2023 · 3 comments

Comments

@reed-lau
Copy link

reed-lau commented Mar 8, 2023

### NVIDIA `nvprof` 命令

如题目所述。

@roastduck
Copy link
Collaborator

nsight 虽然比较新,但是因为各个功能分布在不同的子命令里,似乎反而不是那么容易上手。例如要测各个 kernel 的执行时间要用一种子命令;要分析 GPU 上的硬件指标又要用另一种子命令,并且还要用另外的命令打开分析结果。如果有比较简洁的 nsight 教程,欢迎分享。@xxcclong 有何建议?

@HuangCongQing
Copy link

NVIDIA 计算能力7.5及以上的GPU设备不再支持nvprof工具进行性能剖析,提示使用Nsight Compute作为替代品,如下图所示
image

link: https://blog.csdn.net/yan31415/article/details/109643517

@roastduck
Copy link
Collaborator

实验平台用的 GPU 是 P100,Compute Capability 是 6.0,可以用 nvprof。我个人觉得本课程的目标不是教如何使用具体的软件:如果 nsight 确实更好用,可以改教 nsight;但如果用 nsight 只是为了兼容特定硬件,就不太有必要了。如果你要用 T4 或者 A100 的话,可以参照 nsight 的文档使用,其中的原理是一样的,只是命令不一样。另外, A100 甚至没法在中国买到了。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants