Ignore vendored docs in sanity

.markdownlint-cli2.yaml

@@ -0,0 +1,2 @@
+ignores:
+  - "vendor/**/*.md"

docs/.markdownlint.yaml → .markdownlint.yaml

@@ -1,6 +1,3 @@
-ignores:
-  - CHANGELOG.md
-  - "vendor/**"
 default: true
 MD013: false
 MD024:
@@ -12,3 +9,5 @@ MD029:
 MD033: false
 MD034: false
 MD041: false
+ignores:
+  - "vendor/**/*.md"

.spellcheck.yml → .spellcheck.local.yaml

@@ -33,9 +33,9 @@
 matrix:
 - name: Markdown
   sources:
-  - ['docs/**/*.md', '!docs/doxygen/mainpage.md', '!docs/contributing/documentation-standards.md']
-  - ['tools/autotag/templates/**/*.md', '!tools/autotag/templates/**/5*.md', '!tools/autotag/templates/**/6.0*.md', '!tools/autotag/templates/**/6.1*.md']
-  expect_match: false
+  - '!vendor/**|docs/**/*.md|!docs/doxygen/mainpage.md|!docs/contributing/documentation-standards.md'
+  expect_match: true
+  jobs: 4
   aspell:
     lang: en
   dictionary:
@@ -115,8 +115,9 @@ matrix:
   - pyspelling.filters.url:
 - name: reST
   sources:
-  - 'docs/**/*.rst'
-  expect_match: false
+  - '!vendor/**|docs/**/*.rst'
+  expect_match: true
+  jobs: 4
   aspell:
     lang: en
   dictionary:

api/v1alpha1/deviceconfig_types.go

@@ -117,7 +117,7 @@ type DriverSpec struct {
 	// example tag is coreos-416.94-5.14.0-427.28.1.el9_4.x86_64-6.2.2 and ubuntu-22.04-5.15.0-94-generic-6.1.3
 	//+operator-sdk:csv:customresourcedefinitions:type=spec,displayName="Image",xDescriptors={"urn:alm:descriptor:com.amd.deviceconfigs:image"}
 	// +optional
-	// +kubebuilder:validation:Pattern=`^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$`
+	// +kubebuilder:validation:Pattern=`^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[$a-zA-Z0-9_]+(?:[._-][$a-zA-Z0-9_]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$`
 	Image string `json:"image,omitempty"`
 
 	// driver image registry TLS setting for the container image
@@ -251,12 +251,11 @@ type DevicePluginSpec struct {
 	// +optional
 	DevicePluginTolerations []v1.Toleration `json:"devicePluginTolerations,omitempty"`
 
-	// resource naming strategy for device plugin
-	//+operator-sdk:csv:customresourcedefinitions:type=spec,displayName="ResourceNamingStrategy",xDescriptors={"urn:alm:descriptor:com.amd.deviceconfigs:ResourceNamingStrategy"}
-	// +kubebuilder:validation:Enum=single;mixed
-	// +kubebuilder:default:="single"
+	// device plugin arguments is used to pass supported flags and their values while starting device plugin daemonset
+	// supported flag values: {"resource_naming_strategy": {"single", "mixed"}}
+	//+operator-sdk:csv:customresourcedefinitions:type=spec,displayName="DevicePluginArguments",xDescriptors={"urn:alm:descriptor:com.amd.deviceconfigs:devicePluginArguments"}
 	// +optional
-	ResourceNamingStrategy string `json:"resourceNamingStrategy,omitempty"`
+	DevicePluginArguments map[string]string `json:"devicePluginArguments,omitempty"`
 
 	// node labeller image
 	//+operator-sdk:csv:customresourcedefinitions:type=spec,displayName="NodeLabellerImage",xDescriptors={"urn:alm:descriptor:com.amd.deviceconfigs:nodeLabellerImage"}

bundle/manifests/amd-gpu-operator.clusterserviceversion.yaml

@@ -30,7 +30,7 @@ metadata:
         }
       ]
     capabilities: Basic Install
-    createdAt: "2025-03-20T06:06:57Z"
+    createdAt: "2025-03-25T06:19:27Z"
     operatorframework.io/suggested-namespace: openshift-amd-gpu
     operators.operatorframework.io/builder: operator-sdk-v1.32.0
     operators.operatorframework.io/project_layout: go.kubebuilder.io/v3
@@ -152,6 +152,13 @@ spec:
         path: devicePlugin
         x-descriptors:
         - urn:alm:descriptor:com.amd.deviceconfigs:devicePlugin
+      - description: 'device plugin arguments is used to pass supported flags and
+          their values while starting device plugin daemonset supported flag values:
+          {"resource_naming_strategy": {"single", "mixed"}}'
+        displayName: DevicePluginArguments
+        path: devicePlugin.devicePluginArguments
+        x-descriptors:
+        - urn:alm:descriptor:com.amd.deviceconfigs:devicePluginArguments
       - description: device plugin image
         displayName: DevicePluginImage
         path: devicePlugin.devicePluginImage
@@ -192,11 +199,6 @@ spec:
         path: devicePlugin.nodeLabellerTolerations
         x-descriptors:
         - urn:alm:descriptor:com.amd.deviceconfigs:nodeLabellerTolerations
-      - description: resource naming strategy for device plugin
-        displayName: ResourceNamingStrategy
-        path: devicePlugin.resourceNamingStrategy
-        x-descriptors:
-        - urn:alm:descriptor:com.amd.deviceconfigs:ResourceNamingStrategy
       - description: upgrade policy for device plugin and node labeller daemons
         displayName: UpgradePolicy
         path: devicePlugin.upgradePolicy

bundle/manifests/amd.com_deviceconfigs.yaml

@@ -190,6 +190,13 @@ spec:
               devicePlugin:
                 description: device plugin
                 properties:
+                  devicePluginArguments:
+                    additionalProperties:
+                      type: string
+                    description: |-
+                      device plugin arguments is used to pass supported flags and their values while starting device plugin daemonset
+                      supported flag values: {"resource_naming_strategy": {"single", "mixed"}}
+                    type: object
                   devicePluginImage:
                     description: device plugin image
                     pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
@@ -306,13 +313,6 @@ spec:
                           type: string
                       type: object
                     type: array
-                  resourceNamingStrategy:
-                    default: single
-                    description: resource naming strategy for device plugin
-                    enum:
-                    - single
-                    - mixed
-                    type: string
                   upgradePolicy:
                     description: upgrade policy for device plugin and node labeller
                       daemons
@@ -357,7 +357,7 @@ spec:
                       for OpenShift the default value is image-registry.openshift-image-registry.svc:5000/$MOD_NAMESPACE/amdgpu_kmod
                       image tag will be in the format of <linux distro>-<release version>-<kernel version>-<driver version>
                       example tag is coreos-416.94-5.14.0-427.28.1.el9_4.x86_64-6.2.2 and ubuntu-22.04-5.15.0-94-generic-6.1.3
-                    pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
+                    pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[$a-zA-Z0-9_]+(?:[._-][$a-zA-Z0-9_]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
                     type: string
                   imageRegistrySecret:
                     description: secrets used for pull/push images from/to private

cmd/main.go

@@ -35,9 +35,6 @@ package main
 import (
 	"flag"
 
-	"github.com/ROCm/gpu-operator/internal/configmanager"
-	"github.com/ROCm/gpu-operator/internal/metricsexporter"
-	"github.com/ROCm/gpu-operator/internal/testrunner"
 	kmmv1beta1 "github.com/rh-ecosystem-edge/kernel-module-management/api/v1beta1"
 	"k8s.io/apimachinery/pkg/runtime"
 	utilruntime "k8s.io/apimachinery/pkg/util/runtime"
@@ -51,11 +48,15 @@ import (
 	_ "k8s.io/client-go/plugin/pkg/client/auth"
 
 	gpuev1alpha1 "github.com/ROCm/gpu-operator/api/v1alpha1"
+	utils "github.com/ROCm/gpu-operator/internal"
 	"github.com/ROCm/gpu-operator/internal/cmd"
 	"github.com/ROCm/gpu-operator/internal/config"
+	"github.com/ROCm/gpu-operator/internal/configmanager"
 	"github.com/ROCm/gpu-operator/internal/controllers"
 	"github.com/ROCm/gpu-operator/internal/kmmmodule"
+	"github.com/ROCm/gpu-operator/internal/metricsexporter"
 	"github.com/ROCm/gpu-operator/internal/nodelabeller"
+	"github.com/ROCm/gpu-operator/internal/testrunner"
 	//+kubebuilder:scaffold:imports
 )
 
@@ -107,8 +108,9 @@ func main() {
 	}
 
 	client := mgr.GetClient()
-	kmmHandler := kmmmodule.NewKMMModule(client, scheme)
-	nlHandler := nodelabeller.NewNodeLabeller(scheme)
+	isOpenShift := utils.IsOpenShift(setupLogger)
+	kmmHandler := kmmmodule.NewKMMModule(client, scheme, isOpenShift)
+	nlHandler := nodelabeller.NewNodeLabeller(scheme, isOpenShift)
 	metricsHandler := metricsexporter.NewMetricsExporter(scheme)
 	testrunnerHandler := testrunner.NewTestRunner(scheme)
 	configmanagerHandler := configmanager.NewConfigManager(scheme)

config/crd/bases/amd.com_deviceconfigs.yaml

@@ -186,6 +186,13 @@ spec:
               devicePlugin:
                 description: device plugin
                 properties:
+                  devicePluginArguments:
+                    additionalProperties:
+                      type: string
+                    description: |-
+                      device plugin arguments is used to pass supported flags and their values while starting device plugin daemonset
+                      supported flag values: {"resource_naming_strategy": {"single", "mixed"}}
+                    type: object
                   devicePluginImage:
                     description: device plugin image
                     pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
@@ -302,13 +309,6 @@ spec:
                           type: string
                       type: object
                     type: array
-                  resourceNamingStrategy:
-                    default: single
-                    description: resource naming strategy for device plugin
-                    enum:
-                    - single
-                    - mixed
-                    type: string
                   upgradePolicy:
                     description: upgrade policy for device plugin and node labeller
                       daemons
@@ -353,7 +353,7 @@ spec:
                       for OpenShift the default value is image-registry.openshift-image-registry.svc:5000/$MOD_NAMESPACE/amdgpu_kmod
                       image tag will be in the format of <linux distro>-<release version>-<kernel version>-<driver version>
                       example tag is coreos-416.94-5.14.0-427.28.1.el9_4.x86_64-6.2.2 and ubuntu-22.04-5.15.0-94-generic-6.1.3
-                    pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
+                    pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[$a-zA-Z0-9_]+(?:[._-][$a-zA-Z0-9_]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
                     type: string
                   imageRegistrySecret:
                     description: secrets used for pull/push images from/to private

config/manifests/bases/amd-gpu-operator.clusterserviceversion.yaml

@@ -123,6 +123,13 @@ spec:
         path: devicePlugin
         x-descriptors:
         - urn:alm:descriptor:com.amd.deviceconfigs:devicePlugin
+      - description: 'device plugin arguments is used to pass supported flags and
+          their values while starting device plugin daemonset supported flag values:
+          {"resource_naming_strategy": {"single", "mixed"}}'
+        displayName: DevicePluginArguments
+        path: devicePlugin.devicePluginArguments
+        x-descriptors:
+        - urn:alm:descriptor:com.amd.deviceconfigs:devicePluginArguments
       - description: device plugin image
         displayName: DevicePluginImage
         path: devicePlugin.devicePluginImage
@@ -163,11 +170,6 @@ spec:
         path: devicePlugin.nodeLabellerTolerations
         x-descriptors:
         - urn:alm:descriptor:com.amd.deviceconfigs:nodeLabellerTolerations
-      - description: resource naming strategy for device plugin
-        displayName: ResourceNamingStrategy
-        path: devicePlugin.resourceNamingStrategy
-        x-descriptors:
-        - urn:alm:descriptor:com.amd.deviceconfigs:ResourceNamingStrategy
       - description: upgrade policy for device plugin and node labeller daemons
         displayName: UpgradePolicy
         path: devicePlugin.upgradePolicy

docs/installation/openshift-olm.md

@@ -204,6 +204,14 @@ spec:
     "feature.node.kubernetes.io/amd-gpu": "true"
 ```
 
+Things to note:
+
+1. By default, there is no need to specify the image field in CR for Openshift. Default will be used which is: image-registry.openshift-image-registry.svc:5000/$MOD_NAMESPACE/amdgpu_kmod
+
+2. If users specify image, $MOD_NAMESPACE can be a place holder , KMM Operator can automatically translate it to the namespace
+
+3. Openshift internal registry has image url restriction, OpenShift users cannot use image like `<registry URL>/<repo name>` , it requires the image URL to be `<registry URL>/<project name or namespace>/<repo name>`. However, if any other registry is being used by the user, the image URL can be of either form.
+
 The operator will:
 
 1. Collect worker node system specifications

helm-charts-k8s/Chart.lock

@@ -6,4 +6,4 @@ dependencies:
   repository: file://./charts/kmm
   version: v1.0.0
 digest: sha256:f9a315dd2ce3d515ebf28c8e9a6a82158b493ca2686439ec381487761261b597
-generated: "2025-03-20T06:06:33.9562362Z"
+generated: "2025-03-25T06:19:17.248998622Z"

helm-charts-k8s/crds/deviceconfig-crd.yaml

@@ -194,6 +194,13 @@ spec:
               devicePlugin:
                 description: device plugin
                 properties:
+                  devicePluginArguments:
+                    additionalProperties:
+                      type: string
+                    description: |-
+                      device plugin arguments is used to pass supported flags and their values while starting device plugin daemonset
+                      supported flag values: {"resource_naming_strategy": {"single", "mixed"}}
+                    type: object
                   devicePluginImage:
                     description: device plugin image
                     pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
@@ -310,13 +317,6 @@ spec:
                           type: string
                       type: object
                     type: array
-                  resourceNamingStrategy:
-                    default: single
-                    description: resource naming strategy for device plugin
-                    enum:
-                    - single
-                    - mixed
-                    type: string
                   upgradePolicy:
                     description: upgrade policy for device plugin and node labeller
                       daemons
@@ -361,7 +361,7 @@ spec:
                       for OpenShift the default value is image-registry.openshift-image-registry.svc:5000/$MOD_NAMESPACE/amdgpu_kmod
                       image tag will be in the format of <linux distro>-<release version>-<kernel version>-<driver version>
                       example tag is coreos-416.94-5.14.0-427.28.1.el9_4.x86_64-6.2.2 and ubuntu-22.04-5.15.0-94-generic-6.1.3
-                    pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
+                    pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[$a-zA-Z0-9_]+(?:[._-][$a-zA-Z0-9_]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
                     type: string
                   imageRegistrySecret:
                     description: secrets used for pull/push images from/to private registry

helm-charts-openshift/Chart.lock

@@ -6,4 +6,4 @@ dependencies:
   repository: file://./charts/kmm
   version: v1.0.0
 digest: sha256:25200c34a5cc846a1275e5bf3fc637b19e909dc68de938189c5278d77d03f5ac
-generated: "2025-03-20T06:06:55.80187139Z"
+generated: "2025-03-25T06:19:26.060856628Z"

helm-charts-openshift/crds/deviceconfig-crd.yaml

@@ -194,6 +194,13 @@ spec:
               devicePlugin:
                 description: device plugin
                 properties:
+                  devicePluginArguments:
+                    additionalProperties:
+                      type: string
+                    description: |-
+                      device plugin arguments is used to pass supported flags and their values while starting device plugin daemonset
+                      supported flag values: {"resource_naming_strategy": {"single", "mixed"}}
+                    type: object
                   devicePluginImage:
                     description: device plugin image
                     pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
@@ -310,13 +317,6 @@ spec:
                           type: string
                       type: object
                     type: array
-                  resourceNamingStrategy:
-                    default: single
-                    description: resource naming strategy for device plugin
-                    enum:
-                    - single
-                    - mixed
-                    type: string
                   upgradePolicy:
                     description: upgrade policy for device plugin and node labeller
                       daemons
@@ -361,7 +361,7 @@ spec:
                       for OpenShift the default value is image-registry.openshift-image-registry.svc:5000/$MOD_NAMESPACE/amdgpu_kmod
                       image tag will be in the format of <linux distro>-<release version>-<kernel version>-<driver version>
                       example tag is coreos-416.94-5.14.0-427.28.1.el9_4.x86_64-6.2.2 and ubuntu-22.04-5.15.0-94-generic-6.1.3
-                    pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[a-z0-9]+(?:[._-][a-z0-9]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
+                    pattern: ^([a-z0-9]+(?:[._-][a-z0-9]+)*(:[0-9]+)?)(/[$a-zA-Z0-9_]+(?:[._-][$a-zA-Z0-9_]+)*)*(?::[a-z0-9._-]+)?(?:@[a-zA-Z0-9]+:[a-f0-9]+)?$
                     type: string
                   imageRegistrySecret:
                     description: secrets used for pull/push images from/to private registry

internal/controllers/upgrademgr.go

@@ -151,11 +151,6 @@ func (n *upgradeMgr) HandleUpgrade(ctx context.Context, deviceConfig *amdv1alpha
 		// 1. Set init status for unprocessed nodes
 		n.helper.handleInitStatus(ctx, &nodeList.Items[i])
 
-		if !n.helper.isNodeReadyForUpgrade(ctx, &nodeList.Items[i]) {
-			res = ctrl.Result{Requeue: true, RequeueAfter: time.Second * 20}
-			continue
-		}
-
 		// 2. Handle failed nodes
 		if n.helper.isNodeStateUpgradeFailed(ctx, &nodeList.Items[i], deviceConfig) {
 			n.helper.clearUpgradeStartTime(nodeList.Items[i].Name)
@@ -193,6 +188,11 @@ func (n *upgradeMgr) HandleUpgrade(ctx context.Context, deviceConfig *amdv1alpha
 			continue
 		}
 
+		if !n.helper.isNodeReadyForUpgrade(ctx, &nodeList.Items[i]) {
+			res = ctrl.Result{Requeue: true, RequeueAfter: time.Second * 20}
+			continue
+		}
+
 		//This node is a candidate for selection
 		candidateNodes = append(candidateNodes, nodeList.Items[i])
 	}