Update NVIDIA driver symlink script#158
Merged
bedroge merged 6 commits intoEESSI:mainfrom Feb 18, 2026
Merged
Conversation
Contributor
Author
|
Although we don't have the symlinks yet, I can actually already test this in the container - it will just create the symlinks in What I did: Works as intended. After implementing the variant symlinks, we should retest, try to use the |
Contributor
|
Tested in the container using EESSI 2025.06 and without having configured the variant symlinks: With the variant symlink reconfigured as Wiping that dir and doing it again using Also checked the symlinks, and the pointed to the expected locations. |
bedroge
reviewed
Feb 17, 2026
bedroge
reviewed
Feb 17, 2026
bedroge
reviewed
Feb 17, 2026
bedroge
reviewed
Feb 17, 2026
bedroge
reviewed
Feb 17, 2026
| # Do some checks on existence of links and that we don't end up at /dev/null (the default), so we can print some informative information | ||
| # One downside is that we can't explicitely check if something is a variant symlink, so we'll just assume that if it's a link AND it | ||
| # lives in our CVMFS repository, it must be a variant symlink | ||
| nvidia_trusted_dir="${EESSI_EPREFIX}/lib/nvidia" |
Contributor
There was a problem hiding this comment.
Does this mean that the script will no longer work for 2023.06?
Contributor
Author
There was a problem hiding this comment.
Hm, yeah, that's annoying, this script is in an unversioned prefix. I mean, if we deploy this only for 2025.06, we keep the old version for 2023.06. But then if we want to update that, we have to revert all changes, etc. Maybe we should just duplicate the script? I.e. create something like scripts/gpu_support/nvidia/2023.06/link_nvidia_host_libraries.sh? Or should it be at higher level 2023.06/scripts/gpu_support...?
Contributor
Author
There was a problem hiding this comment.
Ok, I did solve it differently in the end. Only the symlink_mode function changed. I now just duplicated it, and call the correct one based on the value of EESSI_VERSION (note that check_eessi_initialized is called before symlink_mode, so we can be sure that EESSI is initialized and that EESSI_VERSION has been set by the time it gets to that part of the script).
The duplication means there is some code duplication - but actually not to much: the function has changed substantially, so I think the small amount of duplication that is still left is acceptable - especially since it's unlikely we'll still change anything on the 2023.06 side.
Co-authored-by: Bob Dröge <b.e.droge@rug.nl>
Co-authored-by: Bob Dröge <b.e.droge@rug.nl>
…yer-scripts into link_nvidia_drivers
Contributor
Author
|
Tested using: Gives: As expected. Then, modify the CVMFS config: And run again: The result looks fine: (of course, we don't need to symlink this in a container, as singularity sets Similarly, if I set And the result looks as expected: |
Contributor
Author
|
Also tested Looks as expected! |
Contributor
|
bot: build repo:eessi.io-2023.06-software instance:eessi-bot-deucalion for:arch=aarch64/a64fx |
|
New job on instance
|
|
New job on instance
|
Contributor
|
Staging PR merged. |
Contributor
|
@casparvl Do we need to update docs accordingly @ https://eessi.io/docs/site_specific_config/gpu/#exposing-nvidia-gpu-drivers ? |
Contributor
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
We'll need the following variant symlinks to be in place before this script can work as intended:
And then:
This can then be quite easily tested from within the container:
This should error out stating that the variant symlink resolves to
/dev/null. Then, you can change/etc/cvmfs/default.localto set e.g.EESSI_NVIDIA_OVERRIDE_DEFAULT(e.g. to/opt/eessi/nvidia) and run the linking script again - this should the install the symlinks.