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

Update Score-P / Scalasca to new versions #1981

Closed
Thyre opened this issue May 15, 2024 · 1 comment
Closed

Update Score-P / Scalasca to new versions #1981

Thyre opened this issue May 15, 2024 · 1 comment

Comments

@Thyre
Copy link
Contributor

Thyre commented May 15, 2024

Hi there,

I'm one of the developers working on the performance measurement infrastructure Score-P, mainly on the OpenMP and accelerator support.

After updating Score-P and OPARI2 to newer versions in the Fedora repositories, @adrianreber made me aware of Score-P and Scalasca being in the OpenHPC repositories and also mentioned that these are a bit outdated.


After discussing with colleagues for a bit, we're interested in providing newer versions of Score-P and Scalasca and try to keep them updated as well. We're doing the same thing for EasyBuild and spack already and also maintain a Ubuntu PPA for this.

Based on the current state of the SPEC files, I would suggest the following approach:

  • First, split the vendor packages delivered in Score-P into its own packages. Score-P normally includes the latest and greatest versions when it releases, but sometimes new versions of our dependent libraries release before a new version of Score-P. This is especially important with the idea to also integrate Cube GUI into OpenHPC (see CubeGUI -- A Qt-based graphical explorer to view performance profiles submissions#49 ).
  • Next, update Score-P for the latest versions. We are able to test things for Intel compilers and GCC and have a extensive internal testing suite for it. For Intel compilers (if the LLVM-based compilers are used), oneAPI 2024.0 or newer will be required as older versions will not give any information about called functions (unfortunately).
  • Then, update Scalasca. This should be straight-forward.

I'll probably start working on this in the coming days / weeks, depending on how much other work there is.

@adrianreber
Copy link
Member

@Thyre Thanks a lot for your help. I agree with your approach. Let's do it.

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

2 participants