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

Allow LDCache to be used to discover libcuda.so in CDI spec generation #735

Closed
elezar opened this issue Oct 11, 2024 · 0 comments · Fixed by #755
Closed

Allow LDCache to be used to discover libcuda.so in CDI spec generation #735

elezar opened this issue Oct 11, 2024 · 0 comments · Fixed by #755
Assignees

Comments

@elezar
Copy link
Member

elezar commented Oct 11, 2024

On systems where driver libraries are not installed at "standard" library paths the nvidia-ctk cdi generate --mode=nvml command fails with output similar to:

# nvidia-ctk cdi generate --output=/etc/cdi/nvidia.yaml
WARN[0000] Failed to locate nvidia-cdi-hook: pattern nvidia-cdi-hook not found
INFO[0000] Using /var/lib/nvidia/lib64/libnvidia-ml.so.535.183.06
INFO[0000] Auto-detected mode as 'nvml'                
INFO[0000] Selecting /dev/nvidia0 as /dev/nvidia0      
INFO[0000] Selecting /dev/dri/card0 as /dev/dri/card0  
INFO[0000] Selecting /dev/dri/renderD128 as /dev/dri/renderD128
WARN[0000] failed to locate libcuda.so: pattern libcuda.so.*.* not found
64-bit library libcuda.so.*.*: not found; using *.*
INFO[0000] Using driver version 535.183.06              
ERRO[0000] failed to generate CDI spec: failed to create edits common for entities: failed to create discoverer for common entities: failed to create discoverer for driver files: failed to create discoverer for driver libraries: failed to get libraries for driver version: failed to locate libcuda.so.535.183.06: pattern libcuda.so.535.183.06 not found
64-bit library libcuda.so.535.183.06: not found

This can be rectified by running:

nvidia-ctk cdi generate --mode=nvml --library-search-path=/var/lib/nvidia/lib64

In the past the LDCache could be used on such systems allowing libcuda.so to be discovered successfully.

This was reported in v1.16.2

@elezar elezar self-assigned this Oct 11, 2024
@elezar elezar linked a pull request Oct 21, 2024 that will close this issue
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

Successfully merging a pull request may close this issue.

1 participant