Refactor CUDA verification to use nvidia-smi and fix silent GPU allocation failures. #255

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.

Sign up for GitHub

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

Jump to bottom

Merged

ehsaniara merged 2 commits into main from cuda

Jan 12, 2026

Owner

ehsaniara commented Jan 12, 2026

Changes:

CUDA Verifier Refactoring

Add CUDAVerifierInterface with counterfeiter mock generation
Use external test package (gpu_test) to avoid import cycles
Simplify tests to focus on logic, not JSON parsing

GPU Service Bug Fix

Jobs requiring GPUs now fail with clear error if GPU support is disabled
Previously: jobs silently ran without GPUs (potential runtime failures)
Add warning log when GPU allocation returns nil


          Jobs requiring GPUs now fail with clear error if GPU support is disabled

316c4a4

ehsaniara self-assigned this


          wip: cuda implementation

36a6a8c

ehsaniara merged commit 5677684 into main

10 checks passed

ehsaniara deleted the cuda branch

January 12, 2026 20:58

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet