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

caliper cuda modifier #194

Merged
merged 2 commits into from
Apr 16, 2024
Merged

caliper cuda modifier #194

merged 2 commits into from
Apr 16, 2024

Conversation

rfhaque
Copy link
Collaborator

@rfhaque rfhaque commented Apr 3, 2024

No description provided.

@rfhaque rfhaque requested a review from pearce8 April 3, 2024 01:12
@github-actions github-actions bot added the configs New or modified system config label Apr 3, 2024
@rfhaque rfhaque changed the title caliper cuda modifier implementation caliper cuda modifier Apr 3, 2024
@pearce8
Copy link
Collaborator

pearce8 commented Apr 4, 2024

@rfhaque Can you please test on all cuda machines on-site we have defined in benchpark?

@rfhaque
Copy link
Collaborator Author

rfhaque commented Apr 4, 2024

@pearce8 Its already tested on pascal and lassen.

@pearce8
Copy link
Collaborator

pearce8 commented Apr 5, 2024

We need to impose a constraint on the spack spec as this should only be used on NVIDIA GPUs (where CUDA is available). If a user tries to use this modifier on a different architecture, we should make sure they get an error indicating that the modifier does not work on this architecture. Doug provided this example: https://github.com/GoogleCloudPlatform/ramble/blob/develop/lib/ramble/ramble/language/modifier_language.py#L285

@pearce8
Copy link
Collaborator

pearce8 commented Apr 5, 2024

@rfhaque Also, how would one turn on this modifier? @slabasan needs to know for the docs.

@rfhaque
Copy link
Collaborator Author

rfhaque commented Apr 5, 2024

in benchpark setup, we specify --modifier=caliper-cuda

@rfhaque
Copy link
Collaborator Author

rfhaque commented Apr 5, 2024

We need to impose a constraint on the spack spec as this should only be used on NVIDIA GPUs (where CUDA is available). If a user tries to use this modifier on a different architecture, we should make sure they get an error indicating that the modifier does not work on this architecture. Doug provided this example: https://github.com/GoogleCloudPlatform/ramble/blob/develop/lib/ramble/ramble/language/modifier_language.py#L285

I'll take a look at this

@pearce8 pearce8 merged commit cc7e69e into develop Apr 16, 2024
6 checks passed
@pearce8 pearce8 deleted the feature_caliper_cuda_modifier branch April 16, 2024 01:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
configs New or modified system config
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants