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

[REVIEW] Jit context cache #5219

Merged
merged 3 commits into from
May 19, 2020
Merged

Conversation

devavret
Copy link
Contributor

Partially takes care of #5133, specifically the first requirement:

cudf::binary_operation currently ignores the CUDA context of the caller thread, which makes the JIT compiled PTX loaded on a wrong device.

@devavret devavret requested a review from a team as a code owner May 18, 2020 20:08
@devavret devavret requested review from trxcllnt and nvdbaranec May 18, 2020 20:08
@devavret devavret added 3 - Ready for Review Ready for review by team 4 - Needs Review Waiting for reviewer to review or respond labels May 18, 2020
@GPUtester
Copy link
Collaborator

Please update the changelog in order to start CI tests.

View the gpuCI docs here.

Copy link
Contributor

@trxcllnt trxcllnt left a comment

Choose a reason for hiding this comment

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

Looks good to me. Only question is should we use the CUDA runtime API instead of the driver API?

@devavret
Copy link
Contributor Author

Looks good to me. Only question is should we use the CUDA runtime API instead of the driver API?

I tried to, but couldn't find the runtime equivalent of cuCtxGetCurrent.

@harrism
Copy link
Member

harrism commented May 19, 2020

rerun tests

@devavret devavret merged commit f1488b3 into rapidsai:branch-0.14 May 19, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
3 - Ready for Review Ready for review by team 4 - Needs Review Waiting for reviewer to review or respond
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants