Skip to content

opal/cuda: Handle CUDA VMM pointers in accelerator check_addr function #12757

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 1 commit into from
Aug 27, 2024
Merged

opal/cuda: Handle CUDA VMM pointers in accelerator check_addr function #12757

merged 1 commit into from
Aug 27, 2024

Conversation

@Akshay-Venkatesh
Copy link
Contributor

@Akshay-Venkatesh Akshay-Venkatesh commented Aug 14, 2024 •
edited by janjust
Loading

Support CUDA VMM pointers
@jsquyres
Copy link
Member

jsquyres commented Aug 14, 2024

@Akshay-Venkatesh Why did you put bot:notacherrypick on this PR?
@Akshay-Venkatesh Akshay-Venkatesh force-pushed the topic/main/detect-host-numa-as-device-mem branch from deda336 to 036c997 Compare August 14, 2024 21:22
@Akshay-Venkatesh
Copy link
Contributor Author

Akshay-Venkatesh commented Aug 14, 2024

@Akshay-Venkatesh Why did you put bot:notacherrypick on this PR?

@jsquyres I've addressed this now
@jsquyres jsquyres requested a review from bosilca August 15, 2024 10:58
@jsquyres
Copy link
Member

jsquyres commented Aug 15, 2024

@Akshay-Venkatesh Why did you put bot:notacherrypick on this PR?

@jsquyres I've addressed this now

I still see bot:notacherrypick in the description of this PR.

You can find a description of what bot:notacherrypick is supposed to be used for in https://docs.open-mpi.org/en/v5.0.x/developers/git-github.html#git-branch-scheme. Please use it approrpiately.
bosilca
bosilca requested changes Aug 15, 2024
View reviewed changes
Copy link
Member

@bosilca bosilca left a comment

Choose a reason for hiding this comment

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

I left comments on the PR for the 4.1. Please address them here as well
@Akshay-Venkatesh
Copy link
Contributor Author

Akshay-Venkatesh commented Aug 23, 2024

I left comments on the PR for the 4.1. Please address them here as well

@bosilca Thanks for catching the leaked handle. I've made relevant fixes to handle VMM pointers in general. Please review when you get a chance and thanks in advance.
@Akshay-Venkatesh Akshay-Venkatesh changed the title opal/cuda: avoid direct access to cumem host numa memory opal/cuda: Handle CUDA VMM pointers in accelerator check_addr function Aug 23, 2024
@Akshay-Venkatesh Akshay-Venkatesh force-pushed the topic/main/detect-host-numa-as-device-mem branch from 845a9d2 to 86b9876 Compare August 26, 2024 20:47
@janjust
Copy link
Contributor

janjust commented Aug 27, 2024

@Akshay-Venkatesh please squash commits and force push, I'll cherry-pick to v5.0.x
@Akshay-Venkatesh Akshay-Venkatesh force-pushed the topic/main/detect-host-numa-as-device-mem branch from 86b9876 to b0bc58f Compare August 27, 2024 19:39
@Akshay-Venkatesh
opal/cuda: Handle VMM pointers in cuda_check_addr
87d05de 
Signed-off-by: Akshay Venkatesh <akvenkatesh@nvidia.com>
@Akshay-Venkatesh Akshay-Venkatesh force-pushed the topic/main/detect-host-numa-as-device-mem branch from a872e54 to 87d05de Compare August 27, 2024 19:41
@janjust janjust merged commit 8369c1c into open-mpi:main Aug 27, 2024
@Akshay-Venkatesh Akshay-Venkatesh mentioned this pull request Sep 30, 2024
Merged
@github-actions github-actions bot mentioned this pull request May 8, 2025
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Target: main

4 participants

@Akshay-Venkatesh @jsquyres @janjust @bosilca