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

nvidia-docker is deprecated #8

Open
SirYwell opened this issue Apr 27, 2024 · 2 comments
Open

nvidia-docker is deprecated #8

SirYwell opened this issue Apr 27, 2024 · 2 comments

Comments

@SirYwell
Copy link

SirYwell commented Apr 27, 2024

https://github.com/NVIDIA/nvidia-docker is an archived repository. The replacement seems to be https://github.com/NVIDIA/nvidia-container-toolkit, but I'm not sure what needs to be changed to stay up-to-date here.

Edit: I just noticed that the README here is updated already, so I assume it is just outdated on https://tornadovm.readthedocs.io/en/latest/docker.html.

@jjfumero
Copy link
Member

Hi @SirYwell . Yes, thanks for the report. Indeed, not long ago I installed on Fedora using the NVIDIA guidelines:

https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html

We will update the documentation.

@jjfumero
Copy link
Member

jjfumero commented Apr 29, 2024

Hi @SirYwell . thanks for the report. From my understanding of the OpenCL, since this is a pragma emitted for FP16, if there is no support, then the OpenCL JIT compiler will ignore that pragma.

Of course, if you also use FP16, the it will fail.

Otherwise the problem might be another thing. Which version of the matrix multiply are you running? Can you share the Java code?

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

2 participants