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

[FEA] Remove Pascal support #9692

Closed
2 of 3 tasks
sameerz opened this issue Nov 14, 2023 · 7 comments
Closed
2 of 3 tasks

[FEA] Remove Pascal support #9692

sameerz opened this issue Nov 14, 2023 · 7 comments
Assignees
Labels
feature request New feature or request

Comments

@sameerz
Copy link
Collaborator

sameerz commented Nov 14, 2023

Is your feature request related to a problem? Please describe.
RAPIDS will drop support for the Pascal architecture GPU in 24.02. The RAPIDS Spark team will drop support for Pascal as well.

Describe the solution you'd like

  • Remove any Pascal specific tests
  • Remove the Pascal build pipeline
  • Update documentation to note support for Pascal only in 23.12 and prior.

Describe alternatives you've considered
N/A

Additional context
https://docs.rapids.ai/notices/rsn0034/

Tasks

@sameerz sameerz added feature request New feature or request ? - Needs Triage Need team to review and classify labels Nov 14, 2023
@mattahrens mattahrens removed the ? - Needs Triage Need team to review and classify label Nov 14, 2023
@GaryShen2008
Copy link
Collaborator

I searched in the code and didn't find any code commented with Pascal.
The only place in GpuColumnarToRowExec. isAcceleratedTransposeSupported has already checked as Cuda.getComputeCapabilityMajor > 6, which means above Pascal.
So, it may be no necessary to change any code.

@pxLi
Copy link
Collaborator

pxLi commented Dec 1, 2023

internal pascal pipeline for 24.02 has been stopped

@nvliyuan
Copy link
Collaborator

nvliyuan commented Dec 1, 2023

CC @SurajAralihalli

@sameerz
Copy link
Collaborator Author

sameerz commented Dec 1, 2023

The only place in GpuColumnarToRowExec. isAcceleratedTransposeSupported has already checked as Cuda.getComputeCapabilityMajor > 6, which means above Pascal.
So, it may be no necessary to change any code.

Are there any other places in the code where we check whether the user is running on a supported GPU?

@pxLi
Copy link
Collaborator

pxLi commented Dec 1, 2023

related tickets:
#8336
#9528

@GaryShen2008
Copy link
Collaborator

Remove myself since it only needs to update doc.

@GaryShen2008 GaryShen2008 removed their assignment Jan 25, 2024
@nvliyuan
Copy link
Collaborator

close this issue since pr merged

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
feature request New feature or request
Projects
None yet
Development

No branches or pull requests

5 participants