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

How to explicitly provide CUDA install location? #375

Open
aterenin opened this Issue Apr 7, 2019 · 2 comments

Comments

Projects
None yet
3 participants
@aterenin
Copy link

commented Apr 7, 2019

I'm attempting to build CUDAnative on a shared system with multiple CUDA installations. I am NOT an administrator on this system.

I'd like to use CUDA 10. This is installed in /usr/local/cuda-10.0/. There is also a CUDA 9 installation on the system.

I get the error Available CUDA toolchain does not provide libdevice..

I'm unable to find any information via Google on how to go about resolving this error.

@vchuravy

This comment has been minimized.

Copy link
Member

commented Apr 7, 2019

Setting one of the environment variables CUDA_PATH, CUDA_HOME, or CUDA_ROOT should do the trick https://github.com/JuliaGPU/CUDAapi.jl/blob/master/src/discovery.jl#L205

@aterenin

This comment has been minimized.

Copy link
Author

commented Apr 7, 2019

Thanks, setting CUDA_PATH was exactly what was needed. It'd be good to document this, though at minimum someone who is trying to figure this out can now at least find this GitHub issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.