-
Notifications
You must be signed in to change notification settings - Fork 6
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
Problem with master branch #147
Comments
@abaillod I just tried with the latest state of the
|
I just noted that you are using 12 OpenMP threads when running the latest master branch:
Can you maybe retry with only one thread by exporting |
Thanks, I didn't realize I had set OMP_NUM_THREADS to 12. Changing to only a single thread solved the problem. The final force |f| is slightly different on my machine than on yours, but I guess this is expected since we have different architectures.
Maybe we should add a flag or something in preset.f90 that stops SPEC if the number of threads is not one? Or does anyone use SPEC with multiple threads? This can cause unnecessary headaches... |
That would defeat the purpose of the OpenMP parallelization ;-) The real problem is (I think) that SPEC cannot deal so well with being assigned more computing power than it can make use of. |
Using the current master version 6d2693c
on Ubuntu, compiled with OMP_NUM_THREADS=12
12 MPI ranks
Seems like the issue has been resolved then? |
Hello,
I have a problem running the master branch. When running the latest version of the master branch on this simple, circular tokamak input file: Input_0.txt, I get the following output
While when I use the version 3.0 of SPEC, I get good convergence, i.e.
Let me also mention that this is a comparison between version 3.0 and the latest version of the master branch - I did
The same problem appears when running with the branch
update_makefile
.I don't know what is causing the problem. This might be due to some compilation issue, since I have been struggling with compilation on my machine for a few weeks. Could somebody try to replicate the problem?
Thanks!
The text was updated successfully, but these errors were encountered: