Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Bundle libdevice.10.bc with jaxlib wheels. #6581

Merged
merged 1 commit into from
Apr 29, 2021

Conversation

hawkinsp
Copy link
Collaborator

libdevice.10.bc is a redistributable part of the CUDA SDK.

This avoids problems trying to locate a copy of libdevice inside the user's CUDA installation.

@hawkinsp hawkinsp requested a review from skye April 28, 2021 15:45
@google-cla google-cla bot added the cla: yes label Apr 28, 2021
@hawkinsp hawkinsp added the pull ready Ready for copybara import and testing label Apr 28, 2021
Copy link
Member

@skye skye left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just to clarify, this works with all cuda versions?

Also does this significantly increase the jaxlib binary size?

build/BUILD.bazel Outdated Show resolved Hide resolved
@skye
Copy link
Member

skye commented Apr 28, 2021

This is awesome btw

libdevice.10.bc is a redistributable part of the CUDA SDK.

This avoids problems trying to locate a copy of libdevice inside the user's CUDA installation.
@hawkinsp
Copy link
Collaborator Author

Just to clarify, this works with all cuda versions?

Also does this significantly increase the jaxlib binary size?

This should work across CUDA 11 minor releases. For other versions, we are using the same version that came with the CUDA release the user is using.

The file is ~450k uncompressed, so yes it makes jaxlib bigger but not that much bigger.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cla: yes pull ready Ready for copybara import and testing
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants