The vGPU Device Manager is a tool designed for system administrators to make working with vGPU devices easier. It allows administrators to declaratively define a set of possible vGPU device configurations they would like applied to all GPUs on a node.
This will generate a binary called xgv-vgpu-dm.
make cmds-xgv-vgpu-dm- Create a specific vGPU device from a configuration file.
sudo ./xgv-vgpu-dm apply -f examples/config-vgpu.yaml -c PANGU-A0-1G-1-CORE- Create a specific vGPU device with debug out.
sudo ./xgv-vgpu-dm -v apply -f examples/config-vgpu.yaml -c PANGU-A0-128M-1-CORE- Build image
make -f deployments/operator/Makefile build-dev- Create the configmap file of vgpu-config-file
kubectl apply -f deplyments/operator/configmap.yaml- Deploy vgpu-device-manager
kubectl apply -f deplyments/operator/xdxct-vgpu-device-manager.yaml- Check the vgpu-device-manager
kubectl label node <node-name> --overwrite xdxct.com/vgpu-config=PANGU-A0-1G-1-CORE
$ cat /sys/class/mdev_bus/<pci-id>/<vgpu-uuid>/mdev_type/name
Type ID: 8; Type Name: XGV_V0_128M_1_CORE