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.

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

Disable CUDA for build #1625

Open
meerfrau opened this issue May 5, 2023 · 1 comment
Open

Disable CUDA for build #1625

meerfrau opened this issue May 5, 2023 · 1 comment
Labels
enhancement

Comments

@meerfrau
Copy link

meerfrau commented May 5, 2023

I don't a Nvidia GPU and therefore no CUDA. The Python build automatically includes CUDA libs.

This could lead to failure:
strip: ./usr/lib/python3.10/site-packages/triton/third_party/cuda/bin/ptxas.jL2mdh: warning: allocated section .dynsym' not in segment`

How do I disable it at build time?
@ptillet
Copy link
Collaborator

ptillet commented May 6, 2023

I don't think including cuda libs should be a problem as long as it's not used? Seems like the issue you're seeing is more an arm64 vs x86 problem?

@SomeoneSerge SomeoneSerge mentioned this issue Oct 25, 2023
Merged
ZzEeKkAa pushed a commit to ZzEeKkAa/triton that referenced this issue Aug 5, 2024
@whitneywhtsang
Use minimum subgroup size to determine DPAS capability (triton-lang#1625
fea510c 
)

Signed-off-by: Whitney Tsang <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@ptillet @meerfrau