-
Notifications
You must be signed in to change notification settings - Fork 30
Add support for (limited) Python dependencies: nvidia-cutlass-dsl and einops #302
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
Conversation
This option can be used to specify a limited set of Python dependencies.
MekkCyber
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Awesome! I just have a qq
| libcufile | ||
| ] | ||
| ++ lib.optionals (cudaAtLeast "13.0") [ cuda_crt ]; | ||
|
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
is there any reference for why we need cuda_crt only for >=13.0 ?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
They moved some headers from the cuda_nvcc package to the new cuda_crt package in CUDA 13.
This change adds support for kernel dependencies. Dependencies are specified in
build.toml, e.g.:The set of dependencies is limited to avoid bringing back the problem of resolving Python dependencies. Currently only
nvidia-cutlass-dslandeinopsare supported.This PR also adds the derivations for the
nvidia-cutlass-dslpackage.