search

NVlabs / nvdiffrast

Nvdiffrast - Modular Primitives for High-Performance Differentiable Rendering
Other
1.35k stars 144 forks source link

nvidia driver version #42

Closed haoxurt closed 3 years ago

haoxurt commented 3 years ago

Hi, I have ran the demo with nvidia driver 450.102.04. However, I failed when I used other environment with nvidia driver 450.80.02. Could you tell me your nvidia driver version?

s-laine commented 3 years ago

I seem to have 450.80.02 in our local cluster and it has been working fine.

haoxurt commented 3 years ago

I built a docker following Dockerfile and it occur the error [F glutil.cpp:338] eglInitialize() failed

s-laine commented 3 years ago

That is quite surprising, and I don't know how to help, as we don't see the issue in our end. Did you reboot after installing the driver? Did you build the Docker container with the same driver version installed as you had when running it? I don't know if that matters, but it might.

nurpax commented 3 years ago

I tried with driver version 460.91.03 using the latest code from this repo and the sample seems to work for me with the following commands:

bash ./run_sample.sh --build-container
bash ./run_sample.sh ./samples/torch/cube.py --resolution 32

Can you confirm that the above still fails for you, even after a reboot of your machine?

Note that you need to let Docker use gpus with the --gpus all switch as is done in the run_sample.sh script.

If the problem persists, make sure that you can run some other GPU commands in your container, like nvidia-smi.

haoxurt commented 3 years ago

That is quite surprising, and I don't know how to help, as we don't see the issue in our end. Did you reboot after installing the driver? Did you build the Docker container with the same driver version installed as you had when running it? I don't know if that matters, but it might.

I have not solved the problem, it coud be related to basic system environment of server.