Merge pull request #63 from intel/release-26q1-1

Release 26q1-1

Author: byako

charts/intel-gaudi-resource-driver/Chart.yaml

@@ -3,20 +3,20 @@ name: intel-gaudi-resource-driver-chart
 description: A Helm chart for a Dynamic Resource Allocation (DRA) Intel Gaudi Resource Driver
 
 type: application
-version: 0.7.0
-appVersion: v0.7.0
+version: 0.7.1
+appVersion: v0.7.1
 home: https://github.com/intel/intel-resource-drivers-for-kubernetes/charts
 
 dependencies:
   - name: node-feature-discovery
     alias: nfd
-    version: "0.17.4"
+    version: "0.18.3"
     condition: nfd.enabled
     repository: https://kubernetes-sigs.github.io/node-feature-discovery/charts
 
 annotations:
   org.opencontainers.image.url: "https://github.com/intel/intel-resource-drivers-for-kubernetes"
   org.opencontainers.image.source: "https://github.com/intel/intel-resource-drivers-for-kubernetes"
-  org.opencontainers.image.version: "0.7.0"
+  org.opencontainers.image.version: "0.7.1"
   org.opencontainers.image.title: "Intel Gaudi Resource Driver"
   org.opencontainers.image.description: "This chart installs the Intel Gaudi resource driver on Kubernetes."

charts/intel-gaudi-resource-driver/README.md

@@ -51,7 +51,7 @@ You may also run `helm show values` on this chart's dependencies for additional
 | image.repository | string | `intel` |
 | image.name | string | `"intel-gaudi-resource-driver"` |
 | image.pullPolicy | string | `"IfNotPresent"` |
-| image.tag | string | `"v0.7.0"` |
+| image.tag | string | `"v0.7.1"` |
 
 > [!Note]
 > If you change the image tag to be used in Helm chart deployment, ensure that the version of the container image is consistent with deployment YAMLs - they might change between releases.

charts/intel-gaudi-resource-driver/templates/resource-driver.yaml

@@ -38,6 +38,8 @@ spec:
           mountPath: /var/lib/kubelet/plugins
         - name: cdi
           mountPath: /etc/cdi
+        - name: etc-habanalabs
+          mountPath: /etc/habanalabs
         - name: varruncdi
           mountPath: /var/run/cdi
         # when using fake sysfs - mount at the same place as on host
@@ -58,6 +60,9 @@ spec:
       - name: plugins
         hostPath:
           path: /var/lib/kubelet/plugins
+      - name: etc-habanalabs
+        hostPath:
+          path: /etc/habanalabs
       - name: cdi
         hostPath:
           path: {{ .Values.cdi.staticPath }}

charts/intel-gaudi-resource-driver/values.yaml

@@ -13,7 +13,7 @@ image:
   repository: ghcr.io/intel/intel-resource-drivers-for-kubernetes
   name: intel-gaudi-resource-driver
   pullPolicy: IfNotPresent
-  tag: "v0.7.0"
+  tag: "v0.7.1"
 
 serviceAccount:
   create: true

charts/intel-gpu-resource-driver/Chart.yaml

@@ -3,20 +3,20 @@ name: intel-gpu-resource-driver-chart
 description: A Helm chart for a Dynamic Resource Allocation (DRA) Intel GPU Resource Driver
 
 type: application
-version: 0.10.0
-appVersion: v0.10.0
+version: 0.10.1
+appVersion: v0.10.1
 home: https://github.com/intel/intel-resource-drivers-for-kubernetes/charts
 
 dependencies:
   - name: node-feature-discovery
     alias: nfd
-    version: "0.17.4"
+    version: "0.18.3"
     condition: nfd.enabled
     repository: https://kubernetes-sigs.github.io/node-feature-discovery/charts
 
 annotations:
   org.opencontainers.image.url: "https://github.com/intel/intel-resource-drivers-for-kubernetes"
   org.opencontainers.image.source: "https://github.com/intel/intel-resource-drivers-for-kubernetes"
-  org.opencontainers.image.version: "0.10.0"
+  org.opencontainers.image.version: "0.10.1"
   org.opencontainers.image.title: "Intel GPU Resource Driver"
   org.opencontainers.image.description: "This chart installs the Intel GPU resource driver on Kubernetes."

charts/intel-gpu-resource-driver/README.md

@@ -16,6 +16,11 @@ helm install \
     intel-gpu-resource-driver oci://ghcr.io/intel/intel-resource-drivers-for-kubernetes/intel-gpu-resource-driver-chart
 ```
 
+> [!NOTE]
+> Starting v0.10.0, [XPUM Daemon](https://github.com/intel/xpumanager/tree/v2.x/xpumd) is used for health monitoring and devices' details discovery.
+> It is not currently part of this chart, and needs to be installed separately.
+
+
 > [!NOTE]
 > For Kubernetes clusters using [Pod Security Standards](https://kubernetes.io/docs/concepts/security/pod-security-standards/),
 > pre-create the namespace with the respective label allowing to use HostPath Volumes.
@@ -46,12 +51,14 @@ helm show values oci://ghcr.io/intel/intel-resource-drivers-for-kubernetes/intel
 
 You may also run `helm show values` on this chart's dependencies for additional options.
 
-| Key | Type | Default |
-|-----|------|---------|
-| image.repository | string | `intel` |
-| image.name | string | `"intel-gpu-resource-driver"` |
-| image.pullPolicy | string | `"IfNotPresent"` |
-| image.tag | string | `"v0.10.0"` |
+| Key | Type | Default | Comment |
+|-----|------|---------|---------|
+| image.repository | string | `intel` ||
+| image.name | string | `"intel-gpu-resource-driver"` ||
+| image.pullPolicy | string | `"IfNotPresent"` ||
+| image.tag | string | `"v0.10.1"` ||
+| kubeletPlugin.healthMonitoring.enabled | bool | true | Enable (default) GPU details discovery method. Also, [health monitoring](../../doc/gpu/USAGE.md#health-monitoring-support). Requires [xpumd](https://github.com/intel/xpumanager/tree/v2.x/xpumd) |
+| kubeletPlugin.privileged | bool | false | Enable alternative method for discovering GPU details when health monitoring is disabled |
 
 ## Deploying to RedHat OpenShift Container Platform
 

charts/intel-gpu-resource-driver/templates/node-feature-rules.yaml

@@ -21,7 +21,7 @@ spec:
         - matchFeatures:
           - feature: kernel.enabledmodule
             matchExpressions:
-              i915: {op: Exists}
+              xe: {op: Exists}
 ---
 apiVersion: nfd.k8s-sigs.io/v1alpha1
 kind: NodeFeatureRule

charts/intel-gpu-resource-driver/templates/resource-driver.yaml

@@ -58,8 +58,6 @@ spec:
               fieldPath: metadata.namespace
         - name: SYSFS_ROOT
           value: "/sysfs"
-        - name: ZES_ENABLE_SYSMAN
-          value: "1"
         volumeMounts:
         - name: plugins-registry
           mountPath: /var/lib/kubelet/plugins_registry

charts/intel-gpu-resource-driver/values.yaml

@@ -13,7 +13,7 @@ image:
   repository: ghcr.io/intel/intel-resource-drivers-for-kubernetes
   name: intel-gpu-resource-driver
   pullPolicy: IfNotPresent
-  tag: "v0.10.0"
+  tag: "v0.10.1"
 
 serviceAccount:
   create: true

charts/intel-qat-resource-driver/Chart.yaml

@@ -3,20 +3,20 @@ name: intel-qat-resource-driver-chart
 description: A Helm chart for a Dynamic Resource Allocation (DRA) Intel QAT Resource Driver
 
 type: application
-version: 0.4.2
-appVersion: v0.4.2
+version: 0.4.3
+appVersion: v0.4.3
 home: https://github.com/intel/intel-resource-drivers-for-kubernetes/charts
 
 dependencies:
   - name: node-feature-discovery
     alias: nfd
-    version: "0.17.4"
+    version: "0.18.3"
     condition: nfd.enabled
     repository: https://kubernetes-sigs.github.io/node-feature-discovery/charts
 
 annotations:
   org.opencontainers.image.url: "https://github.com/intel/intel-resource-drivers-for-kubernetes"
   org.opencontainers.image.source: "https://github.com/intel/intel-resource-drivers-for-kubernetes"
-  org.opencontainers.image.version: "0.4.2"
+  org.opencontainers.image.version: "0.4.3"
   org.opencontainers.image.title: "Intel QAT Resource Driver"
   org.opencontainers.image.description: "This chart installs the Intel QAT resource driver on Kubernetes."