Allow Numba NVRTC Binding Search Additional Paths#254
Merged
gmarkall merged 11 commits intoNVIDIA:mainfrom May 20, 2025
Merged
Conversation
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
Contributor
Author
|
/ok to test |
|
Auto-sync is disabled for ready for review pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
Contributor
Author
|
/ok to test |
Contributor
Author
|
/ok to test |
gmarkall
requested changes
May 19, 2025
Contributor
gmarkall
left a comment
gmarkall
left a comment
There was a problem hiding this comment.
I think the feature and implementation looks good. I have a couple of items on the diff:
- Can we use "NVRTC" rather than "RTC" in the config variable name to avoid ambiguity? I have tried to make this all as suggestions that can be accepted to avoid manually going and editing the code
- I suggested some clarifications for readability in the docs - let me know if you prefer another form of any of these.
- The
override_configcontext manager is normally used to handle changing configuration during a test, so that the original configuration is restored if the test does not run to completion.
Following resolution of these I think this is good to merge.
Co-authored-by: Graham Markall <535640+gmarkall@users.noreply.github.com>
Co-authored-by: Graham Markall <535640+gmarkall@users.noreply.github.com>
…isVoid/numba-cuda into fea-additional-rtc-search-paths
Contributor
Author
|
/ok to test |
gmarkall
approved these changes
May 20, 2025
gmarkall
added
5 - Ready to merge
Merged
isVoid
added a commit
that referenced
this pull request
May 21, 2025
- Allow External Code to Use Cooperative Group (#240) - Improve debug info for kernel arguments (#242) - Allow Numba NVRTC Binding Search Additional Paths (#254) - Add Bfloat16 High Level API, Documentation (#245) - add a test to use bf16 bindings inside device functions (#244) - Change CI to only be manually triggered to save on CI runs (#252) - Simplify the CI build and test matrix (#249)
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
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.
Currently, the nvrtc binding searches a few hardcoded internal paths when looking for headers. This limits the foreign function usage to only depend on standard CUDA libraries. This PR adds
CUDA_RTC_EXTRA_SEARCH_PATHSentry, a colon separated path list, which defines additional search paths when compiling external functions.Note that these search paths are placed after the standard cudatookit and numba-cuda internal search paths.
closes #46