Remove cuda driver calls from core - #1247
Draft
andrea-cimatoribus-pix4d wants to merge 1 commit into
Draft
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Tentative proposal to remove direct calls to the CUDA driver. This allows loading a library compiled with MatX when a driver is not available. The patch largely originates from an LLM, as I have very limited experience with both MatX and the low-level CUDA APIs involved.
The code builds locally on Ubuntu 26 with CUDA 13.2. I see some test failures, but I see them on main as well, and I'm not sure how to interpret them. I don't know how to run the CI either.