Add per-node stack selection between device-plugin and DRA

Signed-off-by: Karthik Vetrivel <kvetrivel@nvidia.com>

Author: karthikvetrivel

assets/gpu-feature-discovery/0500_daemonset.yaml

@@ -19,6 +19,7 @@ spec:
     spec:
       nodeSelector:
         nvidia.com/gpu.deploy.gpu-feature-discovery: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       tolerations:
         - key: nvidia.com/gpu
           operator: Exists

assets/state-cc-manager/0500_daemonset.yaml

@@ -20,6 +20,7 @@ spec:
           effect: NoSchedule
       nodeSelector:
         nvidia.com/gpu.deploy.cc-manager: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       priorityClassName: system-node-critical
       serviceAccountName: nvidia-cc-manager
       containers:

assets/state-container-toolkit/0500_daemonset.yaml

@@ -18,6 +18,7 @@ spec:
     spec:
       nodeSelector:
         nvidia.com/gpu.deploy.container-toolkit: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       tolerations:
         - key: nvidia.com/gpu
           operator: Exists

assets/state-dcgm-exporter/0800_daemonset.yaml

@@ -18,6 +18,7 @@ spec:
     spec:
       nodeSelector:
         nvidia.com/gpu.deploy.dcgm-exporter: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       tolerations:
         - key: nvidia.com/gpu
           operator: Exists

assets/state-dcgm/0400_dcgm.yml

@@ -18,6 +18,7 @@ spec:
     spec:
       nodeSelector:
         nvidia.com/gpu.deploy.dcgm: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       tolerations:
         - key: nvidia.com/gpu
           operator: Exists

assets/state-device-plugin/0500_daemonset.yaml

@@ -18,6 +18,7 @@ spec:
     spec:
       nodeSelector:
         nvidia.com/gpu.deploy.device-plugin: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       tolerations:
         - key: nvidia.com/gpu
           operator: Exists

assets/state-driver/0500_daemonset.yaml

@@ -26,6 +26,7 @@ spec:
     spec:
       nodeSelector:
         nvidia.com/gpu.deploy.driver: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       tolerations:
         - key: nvidia.com/gpu
           operator: Exists

assets/state-kata-device-plugin/0500_daemonset.yaml

@@ -18,6 +18,7 @@ spec:
     spec:
       nodeSelector:
         nvidia.com/gpu.deploy.kata-sandbox-device-plugin: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       tolerations:
         - key: nvidia.com/gpu
           operator: Exists

assets/state-kata-manager/0600_daemonset.yaml

@@ -20,6 +20,7 @@ spec:
           effect: NoSchedule
       nodeSelector:
         nvidia.com/gpu.deploy.kata-manager: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       priorityClassName: system-node-critical
       serviceAccountName: nvidia-kata-manager
       hostPID: true

assets/state-mig-manager/0600_daemonset.yaml

@@ -16,6 +16,7 @@ spec:
     spec:
       nodeSelector:
         nvidia.com/gpu.deploy.mig-manager: "true"
+        nvidia.com/gpu-operator.resource-allocation.mode: "device-plugin"
       tolerations:
         - key: nvidia.com/gpu
           operator: Exists