0

I'm invoking nvmlDeviceGetCount() on a system with 2 GPUs, and it returns a device count of 0 GPUs - with no error. Why would it say that?

Additional information:

  • CUDA version: 12.6.68 (and bundled driver)
  • ... in which a macro defines nvmlDeviceGetCount to be nvmlDeviceGetCount_v2.
  • nvidia-smi lists these two GPUs, and I am able to run CUDA kernels on them
  • OS: Devuan GNU/Linux Excalibur
  • Kernel: Linux 6.6.13
Improve this question
2
  • Which version of the function are you using? The documentation describes something completely consistent with your output if it turns out you are using the v1 version. Commented Dec 9, 2024 at 23:29
  • @talonmies: nvmlDeviceGetCount_v2(). Commented Dec 10, 2024 at 7:21
Related questions
187
How do I select which GPU to run a job on?
4
Using Theano with GPU on Ubuntu 14.04 on AWS g2
16
Error: NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver
Load 4 more related questions Show fewer related questions

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.