NG: llamacpp on the ai node (Qwen3.8-27B) with the AMD GPU device plugin

Co-Authored-By: Claude <noreply@anthropic.com>

k8s-ng/core/gpu/kustomization.yaml

@@ -0,0 +1,11 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+helmCharts:
+  - name: amd-gpu
+    repo: https://rocm.github.io/k8s-device-plugin/
+    version: 0.22.0
+    releaseName: amd-gpu-device-plugin
+    namespace: gpu-system
+    valuesFile: values.yaml
+    includeCRDs: true

k8s-ng/core/gpu/values.yaml

@@ -0,0 +1,35 @@
+# AMD GPU device plugin: exposes the Strix Halo iGPU of the "ai" node as the
+# amd.com/gpu resource (and injects /dev/kfd and /dev/dri into the pod).
+nfd:
+  enabled: false
+
+labeller:
+  enabled: false
+
+dp:
+  image:
+    repository: docker.io/rocm/k8s-device-plugin
+    tag: "1.31.0.9"
+  updateStrategy:
+    type: RollingUpdate
+    rollingUpdate:
+      maxUnavailable: 1
+
+securityContext:
+  allowPrivilegeEscalation: false
+  capabilities:
+    drop:
+      - ALL
+
+# The ai node carries the label workload/ai=true and the taint workload=ai:NoSchedule
+# (hosts/ai in the nix repo).
+tolerations:
+  - key: workload
+    operator: Equal
+    value: ai
+    effect: NoSchedule
+
+node_selector_enabled: true
+node_selector:
+  kubernetes.io/arch: amd64
+  workload/ai: "true"