Container: fix broken libibverbs-dev install on CUDA 13 base image #889
+54
−25
Conversation
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
Signed-off-by: Ovidiu Mara <[email protected]>
Signed-off-by: Ovidiu Mara <[email protected]>
Signed-off-by: Ovidiu Mara <[email protected]>
|
👋 Hi ovidiusm! Thank you for contributing to ai-dynamo/nixl. Your PR reviewers will review your contribution then trigger the CI to test your changes. 🚀 |
|
/build |
Signed-off-by: Ovidiu Mara <[email protected]>
|
/build |
|
/build |
Signed-off-by: Ovidiu Mara <[email protected]>
ovidiusm
force-pushed
the
nixl-fix-build-cuda-13
branch
from
85292aa to
40648a6
Compare
|
/build |
Signed-off-by: Ovidiu Mara <[email protected]>
Signed-off-by: Ovidiu Mara <[email protected]>
Signed-off-by: Ovidiu Mara <[email protected]>
|
/build |
aranadive
approved these changes
Oct 13, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
What?
ibverbs dev packages are broken in CUDA base images (missing symlinks).
Direct reinstallation from the base image is not possible because the deb files have been deleted (probably to shrink the image).
Direct reinstallation from the Ubuntu repository is not always possible/desired because the Ubuntu versions are older.
DOCA repository contains the deb packages, but it needs to be added before reinstallation is attempted, not after. Current scripts are adding it too late.
Why?
Seeing recurrent issues where ibverbs installation is not functional. This has caused issues in both internal and external projects.
How?
Updated procedure to prepare the environment:
The same procedure should be used on CUDA 12.x and 13.x
Also added a version check for GPUNETIO plugin since it only supports CUDA 12.8.x and 12.9.x. A newer plugin version will be added separately with support for 13.x. The check is needed also to avoid building on CUDA 12.6 image in one of the CI tests.