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

exago+python fails to build with Spack on Ascent #19

Closed
nkoukpaizan opened this issue Sep 26, 2023 · 1 comment · Fixed by #20
Closed

exago+python fails to build with Spack on Ascent #19

nkoukpaizan opened this issue Sep 26, 2023 · 1 comment · Fixed by #20
Assignees
Milestone

Comments

@nkoukpaizan
Copy link
Collaborator

I'm in the process of upgrading the Ascent build system to use a Spack submodule (see diff), and I am getting build errors for ExaGO:

Error sample
>> 327    CMakeFiles/test_pflow_functionality.dir/selfcheck.cpp.o: In function 'fmt_row(std::ostream&, int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)':

>> 328    selfcheck.cpp:(.text+0x824): undefined reference to 'std::__cxx11::basic_stringstream<char, std::char_traits<char>, std::allocator<char> >::basic_stringstream()'

I narrowed it down to +python variant, as the build works for ~python, though it could be something else that happens to be triggered with +python. The old modules still work on Ascent with the develop branch (Python included), and the Spack build seems to work on Summit. I am not sure what could be causing the current issue on Ascent.

CC: @cameronrutherford

@nkoukpaizan nkoukpaizan self-assigned this Sep 26, 2023
@cameronrutherford
Copy link
Contributor

Can you make a draft PR with these changes so that I can test this out and help debug? The diff is great but I think that would serve code review a little better.

I imagine this is some compiler error that we are having that hasn't come up on any other platforms before. Other ideas are to do with logging enabled/disabled affecting our python wrapper, and perhaps the PETSc/other versions that you are pinning.

It's also unclear from that backtrace what actual line is causing the error. It is also possible that shared/static libs is part of the problem here, but hard to say. I also notice that you are a fan or merge commits as opposed to rebasing, and so just wondering if that woudl affect things.

We also are moving away from using fmt through a TPL and instead favoring a newer C++ standard (cc @jaelynlitz on this since we were 90% of the way there iirc), so hopefully this gets less problematic.

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

Successfully merging a pull request may close this issue.

2 participants