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

Build CUDA 12.0 on linux-aarch64 #98

Open
xhochy opened this issue Jan 15, 2024 · 2 comments
Open

Build CUDA 12.0 on linux-aarch64 #98

xhochy opened this issue Jan 15, 2024 · 2 comments

Comments

@xhochy
Copy link
Member

xhochy commented Jan 15, 2024

This was skipped in #97 due to glibc+cudnn dependency issues.

@jakirkham
Copy link
Member

The GLIBC issues with cuDNN should now be addressed: conda-forge/cudnn-feedstock#96

This will require using GLIBC 2.28 for Linux ARM builds, which is now possible as conda-forge supports newer GLIBC's: conda-forge/conda-forge-pinning-feedstock#6626

Here are the relevant docs on using a newer GLIBC

Please reach out if you have any questions

@h-vetinari
Copy link
Member

This will require using GLIBC 2.28 for Linux ARM builds

None of the CUDNN-consuming feedstocks I've seen actually requires this. What's important is that the glibc in the image is new enough, but that's now always the case (unless you explicitly override the alma9 default). IOW, the c_stdlib_version can very, very likely stay at 2.17

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

No branches or pull requests

3 participants