The docker run "--gpus" flag fails to limit the number of GPUs in a single machine to a specific container.
If I wanted to run 2 instances of rocm-docker with "--gpus 0" and "--gpus 1", "rocm-smi --showhw" still seems to detect both gpus in all running containers on my system.
Is this expected behavior? Is there another flag that should be used? Should this be a feature request?
Hello,
The docker run "--gpus" flag fails to limit the number of GPUs in a single machine to a specific container.
If I wanted to run 2 instances of rocm-docker with "--gpus 0" and "--gpus 1", "rocm-smi --showhw" still seems to detect both gpus in all running containers on my system.
Is this expected behavior? Is there another flag that should be used? Should this be a feature request?
Kind regards,
Ben