-
Notifications
You must be signed in to change notification settings - Fork 95
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
Feature/gitlab ci #2252
base: next-hypre-outerloop-cuda
Are you sure you want to change the base?
Feature/gitlab ci #2252
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Most of the code seems to be able to run only on a very specific host.
I would thus prefer if you could hide it, I think /.gitlab/lassen/
would be a possible place ...
Also, it would be nice if the job would bail out without an error if it not run on a system that has the required features. I tried to run it on IPP's gitlab and the job is stuck, so it seems I have to cancel it, which implies it will be reported as a failed build.
@dschwoerer A bit of background for this: this is essentially for testing the GPU branch on LLNL's machines. It won't be running automatically, but will required a member of the LLNL Github organisation to post It is a good point about this being quite specific to that setup. @adrienbernede is the file required to be called |
Ah thanks, I assumed it would be just pulled to a gitlab instance, where the tests would be automatically triggered. If the setup is more complicated then a different name might be nicer, but the current setup is not nice, as I do understand that such a setup needs to be quite specific - and I am perfectly happy with the files being in the repo, I just don't like them being in the tests folder, as I think it is confusing to mix CI instructions with the general tests. So I think there are 2 issues:
|
@dschwoerer @ZedThree , thank you for reviewing this. I knew little about the context, and your comments help a lot. Since there was no CI setup for Gitlab, I was acting as if it would only be used on LLNL Gitlab instance. Now, it is possible to make it safe to run on any other instance, it only requires some specific changes. There are 2 ways to handle a "Gitlab agnostic configuration":
After writing the this, I tend to go for 1. although 2. was may initial preference. |
Question for lassen users: What do you think of the change from I duplicated the logic to be free to make it more generic and modular. |
This PR adds a basic CI implementation to test BOUT-dev on Gitlab LC GPU platform.
This is a basic implementation, some things would need to be improved. In particular:
config-bout-gpu.sh
and modified it so that it is more generic, and extracted the CMake configurtion in a.cmake
file that is invoked on command line.