Use CUDA math wheels #5966
Merged
Use CUDA math wheels #5966
Conversation
This file contains 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
KyleFromNVIDIA
added
5 - DO NOT MERGE
bdice
reviewed
Jul 22, 2024
bdice
approved these changes
Jul 22, 2024
|
Given that 24.08 is now in burndown, I'm not sure it makes sense to make a large sweeping change like this so late, but I'm open to counter-arguments. |
I agree. I think we are going to need to push wheel improvements to 24.10 |
bdice
reviewed
Aug 12, 2024
vyasr
approved these changes
Aug 14, 2024
KyleFromNVIDIA
changed the title
KyleFromNVIDIA
removed
the
5 - DO NOT MERGE
bdice
approved these changes
Aug 15, 2024
There was a problem hiding this comment.
@KyleFromNVIDIA This is pretty clean now. Can you do a little writeup on the effect this PR has on wheel sizes, wheel file contents, and build times?
|
Comparing against 990db71 with Python 3.11, CUDA 12, x86_64... OldBuild Time: 10:41 NewBuild Time: 6:53 |
|
/merge |
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Outdated suggestions cannot be applied.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.
Use CUDA math wheels to reduce wheel size by not statically linking CUDA math libraries.
Contributes to rapidsai/build-planning#35