-
Notifications
You must be signed in to change notification settings - Fork 214
dev: add pixi setup for reproducible environments [no-ci] #1066
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
Auto-sync is disabled for ready for review pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
test_nvidia_wheels_cu12 = [ | ||
test = ["pytest>=6.2.4"] | ||
cu12 = [ | ||
"cuda-toolkit[nvcc,cublas,nvrtc,cudart,cufft,curand,cusolver,cusparse,npp,nvfatbin,nvjitlink,nvjpeg,cccl]==12.*", |
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.
Does pixi handle the pypi --> conda mapping here where if using pixi does it install the pip packages or map to conda packages?
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.
For the pyproject.toml integration it maps only to pypi packages.
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.
For now, I think that's preferable to avoid having to maintain two lists of packages.
I don't think this needs to run CI, so I'll try out the new skipper functionality. |
/ok to test |
@kkraus14 I think this needs an approval again, because I rebased. |
"nvidia-nccl-cu12; sys_platform != 'win32'", | ||
"nvidia-nvshmem-cu12; sys_platform != 'win32'", | ||
] | ||
test_nvidia_wheels_cu13 = [ |
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.
test_nvidia_wheels_cu13
and test_nvidia_wheels_cu12
are referenced here:
:~/forked/cuda-python $ git grep 'test_nvidia_wheels_cu'
.github/workflows/test-wheel-linux.yml: pip install --only-binary=:all: -v . --group "test_nvidia_wheels_cu${TEST_CUDA_MAJOR}" --group test_nvidia_wheels_host
.github/workflows/test-wheel-windows.yml: pip install --only-binary=:all: -v . --group "test_nvidia_wheels_cu${TEST_CUDA_MAJOR}" --group test_nvidia_wheels_host
cuda_pathfinder/pyproject.toml:test_nvidia_wheels_cu12 = [
cuda_pathfinder/pyproject.toml:test_nvidia_wheels_cu13 = [
toolshed/collect_site_packages_dll_files.ps1:pip install --only-binary=:all: -e '.[test,test_nvidia_wheels_cu12,test_nvidia_wheels_host]'
toolshed/collect_site_packages_dll_files.ps1:pip install --only-binary=:all: -e '.[test,test_nvidia_wheels_cu13,test_nvidia_wheels_host]'
toolshed/collect_site_packages_so_files.sh:pip install --only-binary=:all: -e . --group test --group test_nvidia_wheels_cu12 --group test_nvidia_wheels_host
toolshed/collect_site_packages_so_files.sh:pip install --only-binary=:all: -e . --group test --group test_nvidia_wheels_cu13 --group test_nvidia_wheels_host
Do those need to be updated?
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.
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.
This was an oversight on my part in reviewing. I saw the green checkmark and thought it ran through CI before being retitled to [no-ci]
but I obviously misread things. Will be more careful moving forward.
|
…IDIA#1066)" This reverts commit 868fa53.
Reverted in #1072. |
This PR adds a pixi setup for
cuda_pathfinder
leveraging most of the existingpyproject.toml
setup.