refactor(logging): enhance log messages for device resource counting …

…and processing

fix: improve UUID parsing and error handling in ExtractMigTemplatesFromUUID function
refactor: update test cases for ExtractMigTemplatesFromUUID with comprehensive scenarios
chore: clean up imports and enhance test readability across various test files

Signed-off-by: haitwang-cloud <[email protected]>

fix: nvidia-device-plugin no version info (Project-HAMi#779)

Signed-off-by: chaunceyjiang <[email protected]>
Signed-off-by: haitwang-cloud <[email protected]>

Implement initial E2E test suite setup (Project-HAMi#775)

Signed-off-by: wen.rui <[email protected]>
Signed-off-by: haitwang-cloud <[email protected]>

Draft E2E test design (Project-HAMi#633)

Signed-off-by: Rei1010 <[email protected]>
Signed-off-by: wen.rui <[email protected]>
Signed-off-by: haitwang-cloud <[email protected]>

.github/workflows/auto-release.yaml

@@ -134,6 +134,7 @@ jobs:
     uses: ./.github/workflows/call-e2e.yaml
     with:
       ref: ${{ needs.ensure-tag.outputs.tag }}
+      type: "release"
 
   # excute a compatibility test when hami release
   release-e2e-upgrade:

.github/workflows/call-e2e.yaml

@@ -4,15 +4,75 @@ on:
   workflow_call:
     inputs:
       ref:
+        description: 'Reference id to run tests'
         required: true
         type: string
-permissions: write-all
+      type:
+        description: 'E2E type'
+        required: true
+        type: string
+        default: pullrequest
 
 jobs:
-  e2e:
-    runs-on: ubuntu-latest
+  e2e-test:
+    strategy:
+      matrix:
+        include:
+          - device: nvidia
+            type: tesla-p4
+#          - device: nvidia
+#            type: rtx-4090
+#          - device: huawei
+#            type: ascend-910b
+    runs-on: [ "${{ matrix.device }}", "${{ matrix.type }}" ]
+    environment: ${{ matrix.device }}
+    env:
+      E2E_TYPE: ${{ inputs.type }}
     steps:
+      - name: checkout code
+        uses: actions/checkout@v4
+
+      - name: install Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: "1.21"
+
+      - name: setup e2e env
+        run: |
+          make e2e-env-setup
+
+      - name: download hami helm
+        if: inputs.type == 'pullrequest'
+        uses: actions/download-artifact@v4
+        with:
+          name: chart_package_artifact
+          path: charts/
+
+      - name: download hami image
+        if: inputs.type == 'pullrequest'
+        uses: actions/download-artifact@v4
+        with:
+          name: hami-image
+          path: ./image
+
+      - name: load e2e image
+        if: inputs.type == 'pullrequest'
+        run: |
+          echo "Loading Docker image from image.tar..."
+          if [ -z "${VSPHERE_GPU_VM_IP}" ]; then
+            echo "Error: VSPHERE_GPU_VM_IP is not defined!"
+            exit 1
+          fi
+          scp ./image/image.tar root@$VSPHERE_GPU_VM_IP:/home/
+          ssh root@$VSPHERE_GPU_VM_IP "nerdctl load -i /home/image.tar"
+          ssh root@$VSPHERE_GPU_VM_IP "nerdctl image ls | grep hami"
+
+      - name: deploy hami helm
+        env:
+          HAMI_VERSION: ${{ inputs.ref }}
+        run: |
+          make helm-deploy
+
       - name: e2e test
-        # https://github.com/actions/virtual-environments/issues/709
         run: |
-          echo "Need to add e2e test"
+          make e2e-test

.github/workflows/ci.yaml

@@ -9,7 +9,7 @@ on:
     branches-ignore:
       - 'dependabot/**'
 permissions:
-  contents: read  # for actions/checkout to fetch code  
+  contents: read  # for actions/checkout to fetch code
 
 env:
   REGISTRY: docker.io
@@ -38,10 +38,11 @@ jobs:
         run: make lint
       - name: import alias
         run: hack/verify-import-aliases.sh
+
   test:
     name: Unit test
     needs: lint # rely on lint successful completion
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -71,11 +72,29 @@ jobs:
           fail_ci_if_error: false
           verbose: true
 
+  get_version:
+    name: get_version
+    runs-on: ubuntu-22.04
+    outputs:
+      version: ${{ steps.parse_version.outputs.version }}
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Parse the version
+        id: parse_version
+        run: |
+          value=$(make -f Makefile.defs print-version)
+          suffix="$(git rev-parse --short HEAD)"
+          tag="${value}-${suffix}"
+          echo "Generated tag: ${tag}"
+          echo "version=${tag}" >> $GITHUB_OUTPUT
 
   build:
-    name: compile
-    runs-on: ubuntu-latest
-    needs: test # rely on test successful completion
+    name: Compile
+    runs-on: ubuntu-22.04
+    environment: nvidia
+    needs: [ test, get_version ]
     steps:
       - uses: actions/checkout@master
       - name: Free disk space
@@ -88,22 +107,16 @@ jobs:
           echo "=========after clean up, the left CI disk space"
           df -h
 
-      - name: Get the version
-        id: get_version
-        run: |
-          tag="$(git rev-parse --short HEAD)"
-          echo ::set-output name=VERSION::${tag}
-
       - name: Checkout code
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
-      
+
       - name: Checkout submodule
         uses: Mushus/[email protected]
         with:
           basePath: # optional, default is .
-          submodulePath: libvgpu 
+          submodulePath: libvgpu
 
       - name: Set up QEMU
         uses: docker/setup-qemu-action@v3
@@ -124,13 +137,88 @@ jobs:
         with:
           context: .
           file: ${{ env.IMAGE_ROOT_PATH }}/Dockerfile
-          labels: ${{ steps.meta.outputs.labels }}
-          platforms: ${{ env.BUILD_PLATFORM }}
+          labels: ${{ needs.get_version.outputs.version }}
           build-args: |
-            VERSION=${{ steps.get_version.outputs.VERSION }}
+            VERSION=${{ needs.get_version.outputs.version }}
             GOLANG_IMAGE=golang:1.22.5-bullseye
             NVIDIA_IMAGE=nvidia/cuda:12.2.0-devel-ubuntu20.04
             DEST_DIR=/usr/local
-          tags: ${{ steps.meta.outputs.tags }}
+          tags: ${{ env.REGISTRY }}/${{ env.IMAGE_REPO }}:${{ needs.get_version.outputs.version }}
           push: false
+          load: true
           github-token: ${{ env.REGISTER_PASSWORD }}
+
+      - name: List images
+        run: |
+          docker images
+
+      - name: Save Docker image to image.tar
+        run: |
+          docker save ${{ env.REGISTRY }}/${{ env.IMAGE_REPO }}:${{ needs.get_version.outputs.version }} -o image.tar
+
+      - name: Upload image.tar as artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: hami-image
+          path: image.tar
+          retention-days: 5
+          if-no-files-found: error
+
+  # packages tgz from /charts of original branch, deploy to /charts of target branch
+  package_chart:
+    runs-on: ubuntu-22.04
+    needs: [ test, get_version ]
+    env:
+      HELM_VERSION: v3.8.1
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          ref: ${{ needs.get_ref.outputs.ref }}
+
+      - name: Configure Git
+        run: |
+          git config user.name "$GITHUB_ACTOR"
+          git config user.email "[email protected]"
+
+      - name: Install Helm
+        uses: azure/setup-helm@v4
+        with:
+          version: ${{ env.HELM_VERSION }}
+
+      - name: Lint helm chart
+        run: |
+          make lint_chart
+
+      - name: Package Chart
+        continue-on-error: false
+        env:
+          VERSION: ${{ needs.get_version.outputs.version }}
+        run: |
+          cd charts
+          make clean
+          make
+          if ! ls *.tgz &>/dev/null ; then
+            echo "failed to generate chart"
+            exit 1
+          fi
+          cd ..
+          mkdir -p tmp
+          mv charts/*.tgz tmp
+
+      - name: Upload Artifact
+        uses: actions/[email protected]
+        with:
+          name: chart_package_artifact
+          path: tmp/*
+          retention-days: 5
+          if-no-files-found: error
+
+  # execute a full e2e test when hami code merge
+  e2e_test:
+    uses: ./.github/workflows/call-e2e.yaml
+    needs: [ package_chart, get_version, build ]
+    with:
+      ref: ${{ needs.get_version.outputs.version }}
+      type: "pullrequest"

.gitignore

@@ -8,4 +8,6 @@ libvgpu.so
 .idea
 vendor
 license
-vgpuvalidator
+vgpuvalidator
+_output/coverage/coverage_pkg.txt
+.DS_Store

Makefile

@@ -9,6 +9,7 @@ docker:
 	--build-arg TARGET_ARCH=${TARGET_ARCH} \
 	--build-arg NVIDIA_IMAGE=${NVIDIA_IMAGE} \
 	--build-arg DEST_DIR=${DEST_DIR} \
+	--build-arg VERSION=${VERSION} \
 	--build-arg GOPROXY=https://goproxy.cn,direct \
 	. -f=docker/Dockerfile -t ${IMG_TAG}
 
@@ -19,6 +20,7 @@ dockerwithlib:
 	--build-arg TARGET_ARCH=${TARGET_ARCH} \
 	--build-arg NVIDIA_IMAGE=${NVIDIA_IMAGE} \
 	--build-arg DEST_DIR=${DEST_DIR} \
+	--build-arg VERSION=${VERSION} \
 	--build-arg GOPROXY=https://goproxy.cn,direct \
 	. -f=docker/Dockerfile.withlib -t ${IMG_TAG}
 
@@ -73,3 +75,15 @@ lint_chart:
           aquasec/trivy:$(TRIVY_VERSION) config --exit-code 1  --severity $(LINT_TRIVY_SEVERITY_LEVEL) /tmp/src/charts  ; \
       (($$?==0)) || { echo "error, failed to check chart trivy" && exit 1 ; } ; \
       echo "chart trivy check: pass"
+
+.PHONY: e2e-env-setup
+e2e-env-setup:
+	./hack/e2e-test-setup.sh
+
+.PHONY: helm-deploy
+helm-deploy:
+	./hack/deploy-helm.sh "${E2E_TYPE}" "${KUBE_CONF}" "${HAMI_VERSION}"
+
+.PHONY: e2e-test
+e2e-test:
+	./hack/e2e-test.sh "${E2E_TYPE}" "${KUBE_CONF}"

Makefile.defs

@@ -13,9 +13,16 @@ TARGETARCH ?= amd64
 DESTDIR_BIN ?= $(ROOT_DIR)/output/$(TARGETARCH)/bin
 DESTDIR_BASH_COMPLETION ?= $(ROOT_DIR)/output/$(TARGETARCH)/bash-completion
 
-VERSION = $(shell cat $(dir $(lastword $(MAKEFILE_LIST)))/VERSION)
+VERSION?=""
+ifeq ($(VERSION), "")
+    VERSION=$(shell cat $(dir $(lastword $(MAKEFILE_LIST)))/VERSION)
+endif
 
 ECHO_GEN=echo "  GEN   $(RELATIVE_DIR)/"
 
 LINT_TRIVY_SEVERITY_LEVEL ?= CRITICAL
 TRIVY_VERSION=0.36.0
+
+.PHONY: print-version
+print-version:
+	@echo $(VERSION)

charts/Makefile

@@ -1,4 +1,5 @@
 # get VERSION
+.DEFAULT_GOAL := all
 include  ../Makefile.defs
 
 VERSION_REGEX := '[vV]*[0-9]\+\.[0-9]\+\.[0-9]\+.*'
@@ -11,6 +12,7 @@ all: update-versions lint package
 #update version in chart
 update-versions:
 	$(ECHO_GEN) " Updating Chart version to $(VERSION)"
+	echo "VERSION=$(VERSION)"
 	echo "VERSION_MAJOR=$(VERSION_MAJOR)"
 	echo "GIT_VERSION=$(GIT_VERSION)"
 	echo "FULL_BUILD_VERSION=$(FULL_BUILD_VERSION)"
@@ -22,10 +24,11 @@ update-versions:
 	sed -i 's/version: "*'$(VERSION_REGEX)'"*/version: "'$$hami_version'"/g' $(VALUES_FILE)
 
 lint: update-versions
-	helm lint --with-subcharts --values ./hami/values.yaml ./hami
+	helm lint --with-subcharts --values ./hami/values.yaml ./hami --debug
 
 package: lint
-	helm package ./hami
+	helm package ./hami --debug
 
 clean:
 	rm -f *.tgz
+

docker/Dockerfile

@@ -4,10 +4,11 @@ ARG NVIDIA_IMAGE=nvidia/cuda:12.2.0-devel-ubuntu20.04
 FROM $GOLANG_IMAGE AS build
 FROM $GOLANG_IMAGE AS gobuild
 ARG GOPROXY
+ARG VERSION
 ADD . /k8s-vgpu
 #RUN --mount=type=cache,target=/go/pkg/mod \
 #    cd /k8s-vgpu && make all
-RUN cd /k8s-vgpu && make all
+RUN cd /k8s-vgpu && make all VERSION=$VERSION
 RUN go install github.com/NVIDIA/mig-parted/cmd/nvidia-mig-parted@latest
 
 FROM $NVIDIA_IMAGE AS nvbuild

docker/Dockerfile.withlib

@@ -5,8 +5,9 @@ FROM $GOLANG_IMAGE AS build
 FROM $GOLANG_IMAGE AS GOBUILD
 ADD . /k8s-vgpu
 ARG GOPROXY=https://goproxy.cn,direct
+ARG VERSION
 RUN go env -w GO111MODULE=on
-RUN cd /k8s-vgpu && make all
+RUN cd /k8s-vgpu && make all VERSION=$VERSION
 RUN go install github.com/NVIDIA/mig-parted/cmd/nvidia-mig-parted@latest
 
 #FROM ubuntu:24.04