versions: bump containerd active version to 2.2

SSIA

Signed-off-by: Fabiano Fidêncio <ffidencio@nvidia.com>

.github/actionlint.yaml

@@ -10,11 +10,6 @@ self-hosted-runner:
     - amd64-nvidia-a100
     - amd64-nvidia-h100-snp
     - arm64-k8s
-    - containerd-v1.7-overlayfs
-    - containerd-v2.0-overlayfs
-    - containerd-v2.1-overlayfs
-    - containerd-v2.2
-    - containerd-v2.2-overlayfs
     - garm-ubuntu-2004
     - garm-ubuntu-2004-smaller
     - garm-ubuntu-2204
@@ -25,6 +20,7 @@ self-hosted-runner:
     - ppc64le-k8s
     - ppc64le-small
     - ubuntu-24.04-ppc64le
+    - ubuntu-24.04-s390x
     - metrics
     - riscv-builder
     - sev-snp

.github/dependabot.yml

@@ -12,7 +12,6 @@ updates:
       - "/src/tools/agent-ctl"
       - "/src/tools/genpolicy"
       - "/src/tools/kata-ctl"
-      - "/src/tools/runk"
       - "/src/tools/trace-forwarder"
     schedule:
       interval: "daily"

.github/workflows/basic-ci-amd64.yaml

@@ -26,8 +26,6 @@ jobs:
       matrix:
         containerd_version: ['active']
         vmm: ['dragonball', 'cloud-hypervisor', 'qemu-runtime-rs']
-    # TODO: enable me when https://github.com/containerd/containerd/issues/11640 is fixed
-    if: false
     runs-on: ubuntu-22.04
     env:
       CONTAINERD_VERSION: ${{ matrix.containerd_version }}
@@ -147,49 +145,22 @@ jobs:
           name: kata-static-tarball-amd64${{ inputs.tarball-suffix }}
           path: kata-artifacts
 
+      - name: get-kata-tools-tarball
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        with:
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
+
       - name: Install kata
         run: bash tests/integration/nydus/gha-run.sh install-kata kata-artifacts
 
+      - name: Install kata-tools
+        run: bash tests/integration/nydus/gha-run.sh install-kata-tools kata-tools-artifacts
+
       - name: Run nydus tests
         timeout-minutes: 10
         run: bash tests/integration/nydus/gha-run.sh run
 
-  run-runk:
-    name: run-runk
-    # Skip runk tests as we have no maintainers. TODO: Decide when to remove altogether
-    if: false
-    runs-on: ubuntu-22.04
-    env:
-      CONTAINERD_VERSION: lts
-    steps:
-      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-        with:
-          ref: ${{ inputs.commit-hash }}
-          fetch-depth: 0
-          persist-credentials: false
-
-      - name: Rebase atop of the latest target branch
-        run: |
-          ./tests/git-helper.sh "rebase-atop-of-the-latest-target-branch"
-        env:
-          TARGET_BRANCH: ${{ inputs.target-branch }}
-
-      - name: Install dependencies
-        run: bash tests/integration/runk/gha-run.sh install-dependencies
-
-      - name: get-kata-tarball
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
-        with:
-          name: kata-static-tarball-amd64${{ inputs.tarball-suffix }}
-          path: kata-artifacts
-
-      - name: Install kata
-        run: bash tests/integration/runk/gha-run.sh install-kata kata-artifacts
-
-      - name: Run runk tests
-        timeout-minutes: 10
-        run: bash tests/integration/runk/gha-run.sh run
-
   run-tracing:
     name: run-tracing
     strategy:
@@ -367,8 +338,16 @@ jobs:
           name: kata-static-tarball-amd64${{ inputs.tarball-suffix }}
           path: kata-artifacts
 
-      - name: Install kata
-        run: bash tests/functional/kata-agent-apis/gha-run.sh install-kata kata-artifacts
+      - name: get-kata-tools-tarball
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        with:
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
+
+      - name: Install kata & kata-tools
+        run: | 
+          bash tests/functional/kata-agent-apis/gha-run.sh install-kata kata-artifacts
+          bash tests/functional/kata-agent-apis/gha-run.sh install-kata-tools kata-tools-artifacts
 
       - name: Run kata agent api tests with agent-ctl
         run: bash tests/functional/kata-agent-apis/gha-run.sh run

.github/workflows/basic-ci-s390x.yaml

@@ -26,8 +26,6 @@ jobs:
       matrix:
         containerd_version: ['active']
         vmm: ['qemu-runtime-rs']
-    # TODO: enable me when https://github.com/containerd/containerd/issues/11640 is fixed
-    if: false
     runs-on: s390x-large
     env:
       CONTAINERD_VERSION: ${{ matrix.containerd_version }}
@@ -48,7 +46,7 @@ jobs:
           TARGET_BRANCH: ${{ inputs.target-branch }}
 
       - name: Install dependencies
-        run: bash tests/integration/cri-containerd/gha-run.sh
+        run: bash tests/integration/cri-containerd/gha-run.sh install-dependencies
         env:
           GH_TOKEN: ${{ github.token }}
 

.github/workflows/build-checks.yaml

@@ -12,7 +12,12 @@ name: Build checks
 jobs:
   check:
     name: check
-    runs-on: ${{ matrix.runner || inputs.instance }}
+    runs-on: >-
+      ${{
+        ( contains(inputs.instance, 's390x') && matrix.component.name == 'runtime' ) && 's390x' ||
+        ( contains(inputs.instance, 'ppc64le') && (matrix.component.name == 'runtime' || matrix.component.name == 'agent') ) && 'ppc64le' ||
+        inputs.instance
+      }}
     strategy:
       fail-fast: false
       matrix:
@@ -70,36 +75,6 @@ jobs:
               - protobuf-compiler
         instance:
           - ${{ inputs.instance }} 
-        include:
-          - component:
-              name: runtime
-              path: src/runtime
-              needs:
-                - golang
-                - XDG_RUNTIME_DIR
-            instance: ubuntu-24.04-s390x
-            runner: s390x
-          - component:
-              name: runtime
-              path: src/runtime
-              needs:
-                - golang
-                - XDG_RUNTIME_DIR
-            instance: ubuntu-24.04-ppc64le
-            runner: ppc64le
-          - component:
-              name: agent
-              path: src/agent
-              needs:
-                - rust
-                - libdevmapper
-                - libseccomp
-                - protobuf-compiler
-                - clang
-            instance: ubuntu-24.04-ppc64le
-            runner: ppc64le
-
-             
 
     steps:
       - name: Adjust a permission for repo

.github/workflows/build-kata-static-tarball-amd64.yaml

@@ -41,16 +41,11 @@ jobs:
       matrix:
         asset:
           - agent
-          - agent-ctl
           - busybox
           - cloud-hypervisor
           - cloud-hypervisor-glibc
           - coco-guest-components
-          - csi-kata-directvolume
           - firecracker
-          - genpolicy
-          - kata-ctl
-          - kata-manager
           - kernel
           - kernel-confidential
           - kernel-dragonball-experimental
@@ -59,11 +54,11 @@ jobs:
           - nydus
           - ovmf
           - ovmf-sev
+          - ovmf-tdx
           - pause-image
           - qemu
           - qemu-snp-experimental
           - qemu-tdx-experimental
-          - trace-forwarder
           - virtiofsd
         stage:
           - ${{ inputs.stage }}
@@ -153,8 +148,8 @@ jobs:
         if: ${{ startsWith(matrix.asset, 'kernel-nvidia-gpu') }}
         uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
         with:
-          name: kata-artifacts-amd64-${{ matrix.asset }}-headers${{ inputs.tarball-suffix }}
-          path: kata-build/kata-static-${{ matrix.asset }}-headers.tar.zst
+          name: kata-artifacts-amd64-${{ matrix.asset }}-modules${{ inputs.tarball-suffix }}
+          path: kata-build/kata-static-${{ matrix.asset }}-modules.tar.zst
           retention-days: 15
           if-no-files-found: error
 
@@ -171,6 +166,8 @@ jobs:
           - rootfs-image
           - rootfs-image-confidential
           - rootfs-image-mariner
+          - rootfs-image-nvidia-gpu
+          - rootfs-image-nvidia-gpu-confidential
           - rootfs-initrd
           - rootfs-initrd-confidential
           - rootfs-initrd-nvidia-gpu
@@ -240,8 +237,8 @@ jobs:
         asset:
           - busybox
           - coco-guest-components
-          - kernel-nvidia-gpu-headers
-          - kernel-nvidia-gpu-confidential-headers
+          - kernel-nvidia-gpu-modules
+          - kernel-nvidia-gpu-confidential-modules
           - pause-image
     steps:
       - uses: geekyeggo/delete-artifact@f275313e70c08f6120db482d7a6b98377786765b # v5.1.0
@@ -362,3 +359,104 @@ jobs:
           path: kata-static.tar.zst
           retention-days: 15
           if-no-files-found: error
+
+  build-tools-asset:
+    name: build-tools-asset
+    runs-on: ubuntu-22.04
+    permissions:
+      contents: read
+      packages: write
+    strategy:
+      matrix:
+        asset:
+          - agent-ctl
+          - csi-kata-directvolume
+          - genpolicy
+          - kata-ctl
+          - kata-manager
+          - trace-forwarder
+        stage:
+          - ${{ inputs.stage }}
+    steps:
+      - name: Login to Kata Containers quay.io
+        if: ${{ inputs.push-to-registry == 'yes' }}
+        uses: docker/login-action@74a5d142397b4f367a81961eba4e8cd7edddf772 # v3.4.0
+        with:
+          registry: quay.io
+          username: ${{ vars.QUAY_DEPLOYER_USERNAME }}
+          password: ${{ secrets.QUAY_DEPLOYER_PASSWORD }}
+
+      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+        with:
+          ref: ${{ inputs.commit-hash }}
+          fetch-depth: 0 # This is needed in order to keep the commit ids history
+          persist-credentials: false
+
+      - name: Rebase atop of the latest target branch
+        run: |
+          ./tests/git-helper.sh "rebase-atop-of-the-latest-target-branch"
+        env:
+          TARGET_BRANCH: ${{ inputs.target-branch }}
+
+      - name: Build ${{ matrix.asset }}
+        id: build
+        run: |
+          make "${KATA_ASSET}-tarball"
+          build_dir=$(readlink -f build)
+          # store-artifact does not work with symlink
+          mkdir -p kata-tools-build && cp "${build_dir}"/kata-static-"${KATA_ASSET}"*.tar.* kata-tools-build/.
+        env:
+          KATA_ASSET: ${{ matrix.asset }}
+          TAR_OUTPUT: ${{ matrix.asset }}.tar.gz
+          PUSH_TO_REGISTRY: ${{ inputs.push-to-registry }}
+          ARTEFACT_REGISTRY: ghcr.io
+          ARTEFACT_REGISTRY_USERNAME: ${{ github.actor }}
+          ARTEFACT_REGISTRY_PASSWORD: ${{ secrets.GITHUB_TOKEN }}
+          TARGET_BRANCH: ${{ inputs.target-branch }}
+          RELEASE: ${{ inputs.stage == 'release' && 'yes' || 'no' }}
+
+      - name: store-artifact ${{ matrix.asset }}
+        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
+        with:
+          name: kata-tools-artifacts-amd64-${{ matrix.asset }}${{ inputs.tarball-suffix }}
+          path: kata-tools-build/kata-static-${{ matrix.asset }}.tar.zst
+          retention-days: 15
+          if-no-files-found: error
+
+  create-kata-tools-tarball:
+    name: create-kata-tools-tarball
+    runs-on: ubuntu-22.04
+    needs: [build-tools-asset]
+    permissions:
+      contents: read
+      packages: write
+    steps:
+      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+        with:
+          ref: ${{ inputs.commit-hash }}
+          fetch-depth: 0
+          fetch-tags: true
+          persist-credentials: false
+      - name: Rebase atop of the latest target branch
+        run: |
+          ./tests/git-helper.sh "rebase-atop-of-the-latest-target-branch"
+        env:
+          TARGET_BRANCH: ${{ inputs.target-branch }}
+      - name: get-artifacts
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        with:
+          pattern: kata-tools-artifacts-amd64-*${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
+          merge-multiple: true
+      - name: merge-artifacts
+        run: |
+          ./tools/packaging/kata-deploy/local-build/kata-deploy-merge-builds.sh kata-tools-artifacts versions.yaml kata-tools-static.tar.zst
+        env:
+          RELEASE: ${{ inputs.stage == 'release' && 'yes' || 'no' }}
+      - name: store-artifacts
+        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
+        with:
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-static.tar.zst
+          retention-days: 15
+          if-no-files-found: error

.github/workflows/build-kata-static-tarball-arm64.yaml

@@ -134,8 +134,8 @@ jobs:
         if: ${{ startsWith(matrix.asset, 'kernel-nvidia-gpu') }}
         uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
         with:
-          name: kata-artifacts-arm64-${{ matrix.asset }}-headers${{ inputs.tarball-suffix }}
-          path: kata-build/kata-static-${{ matrix.asset }}-headers.tar.zst
+          name: kata-artifacts-arm64-${{ matrix.asset }}-modules${{ inputs.tarball-suffix }}
+          path: kata-build/kata-static-${{ matrix.asset }}-modules.tar.zst
           retention-days: 15
           if-no-files-found: error
 
@@ -150,6 +150,7 @@ jobs:
       matrix:
         asset:
           - rootfs-image
+          - rootfs-image-nvidia-gpu
           - rootfs-initrd
           - rootfs-initrd-nvidia-gpu
     steps:
@@ -215,7 +216,7 @@ jobs:
       matrix:
         asset:
           - busybox
-          - kernel-nvidia-gpu-headers
+          - kernel-nvidia-gpu-modules
     steps:
       - uses: geekyeggo/delete-artifact@f275313e70c08f6120db482d7a6b98377786765b # v5.1.0
         with:

.github/workflows/build-kata-static-tarball-ppc64le.yaml

@@ -32,7 +32,7 @@ jobs:
     permissions:
       contents: read
       packages: write
-    runs-on: ppc64le-small
+    runs-on: ubuntu-24.04-ppc64le
     strategy:
       matrix:
         asset:
@@ -89,7 +89,7 @@ jobs:
 
   build-asset-rootfs:
     name: build-asset-rootfs
-    runs-on: ppc64le-small
+    runs-on: ubuntu-24.04-ppc64le
     needs: build-asset
     permissions:
       contents: read
@@ -170,7 +170,7 @@ jobs:
 
   build-asset-shim-v2:
     name: build-asset-shim-v2
-    runs-on: ppc64le-small
+    runs-on: ubuntu-24.04-ppc64le
     needs: [build-asset, build-asset-rootfs, remove-rootfs-binary-artifacts]
     permissions:
       contents: read
@@ -230,7 +230,7 @@ jobs:
 
   create-kata-tarball:
     name: create-kata-tarball
-    runs-on: ppc64le-small
+    runs-on: ubuntu-24.04-ppc64le
     needs: [build-asset, build-asset-rootfs, build-asset-shim-v2]
     permissions:
       contents: read

.github/workflows/build-kata-static-tarball-s390x.yaml

@@ -32,7 +32,7 @@ permissions: {}
 jobs:
   build-asset:
     name: build-asset
-    runs-on: s390x
+    runs-on: ubuntu-24.04-s390x
     permissions:
       contents: read
       packages: write
@@ -257,7 +257,7 @@ jobs:
 
   build-asset-shim-v2:
     name: build-asset-shim-v2
-    runs-on: s390x
+    runs-on: ubuntu-24.04-s390x
     needs: [build-asset, build-asset-rootfs, remove-rootfs-binary-artifacts]
     permissions:
       contents: read
@@ -319,7 +319,7 @@ jobs:
 
   create-kata-tarball:
     name: create-kata-tarball
-    runs-on: s390x
+    runs-on: ubuntu-24.04-s390x
     needs:
       - build-asset
       - build-asset-rootfs

.github/workflows/build-kubectl-image.yaml

@@ -0,0 +1,75 @@
+name: Build kubectl multi-arch image
+
+on:
+  schedule:
+    # Run every Sunday at 00:00 UTC
+    - cron: '0 0 * * 0'
+  workflow_dispatch:
+    # Allow manual triggering
+  push:
+    branches:
+      - main
+    paths:
+      - 'tools/packaging/kubectl/Dockerfile'
+      - '.github/workflows/build-kubectl-image.yaml'
+
+permissions: {}
+
+env:
+  REGISTRY: quay.io
+  IMAGE_NAME: kata-containers/kubectl
+
+jobs:
+  build-and-push:
+    name: Build and push multi-arch image
+    runs-on: ubuntu-24.04
+    permissions:
+      contents: read
+      packages: write
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+        with:
+          persist-credentials: false
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@29109295f81e9208d7d86ff1c6c12d2833863392 # v3.6.0
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@b5ca514318bd6ebac0fb2aedd5d36ec1b5c232a2 # v3.10.0
+
+      - name: Login to Quay.io
+        uses: docker/login-action@74a5d142397b4f367a81961eba4e8cd7edddf772 # v3.4.0
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ vars.QUAY_DEPLOYER_USERNAME }}
+          password: ${{ secrets.QUAY_DEPLOYER_PASSWORD }}
+
+      - name: Get kubectl version
+        id: kubectl-version
+        run: |
+          KUBECTL_VERSION=$(curl -L -s https://dl.k8s.io/release/stable.txt)
+          echo "version=${KUBECTL_VERSION}" >> "$GITHUB_OUTPUT"
+
+      - name: Generate image metadata
+        id: meta
+        uses: docker/metadata-action@902fa8ec7d6ecbf8d84d538b9b233a880e428804 # v5.7.0
+        with:
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+          tags: |
+            type=raw,value=latest
+            type=raw,value={{date 'YYYYMMDD'}}
+            type=raw,value=${{ steps.kubectl-version.outputs.version }}
+            type=sha,prefix=
+
+      - name: Build and push multi-arch image
+        uses: docker/build-push-action@ca052bb54ab0790a636c9b5f226502c73d547a25 # v5.4.0
+        with:
+          context: tools/packaging/kubectl/
+          file: tools/packaging/kubectl/Dockerfile
+          platforms: linux/amd64,linux/arm64,linux/s390x,linux/ppc64le
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max

.github/workflows/ci.yaml

@@ -147,7 +147,7 @@ jobs:
       tag: ${{ inputs.tag }}-s390x
       commit-hash: ${{ inputs.commit-hash }}
       target-branch: ${{ inputs.target-branch }}
-      runner: s390x
+      runner: ubuntu-24.04-s390x
       arch: s390x
     secrets:
       QUAY_DEPLOYER_PASSWORD: ${{ secrets.QUAY_DEPLOYER_PASSWORD }}
@@ -165,7 +165,7 @@ jobs:
       tag: ${{ inputs.tag }}-ppc64le
       commit-hash: ${{ inputs.commit-hash }}
       target-branch: ${{ inputs.target-branch }}
-      runner: ppc64le-small
+      runner: ubuntu-24.04-ppc64le
       arch: ppc64le
     secrets:
       QUAY_DEPLOYER_PASSWORD: ${{ secrets.QUAY_DEPLOYER_PASSWORD }}
@@ -233,14 +233,14 @@ jobs:
         env:
           TARGET_BRANCH: ${{ inputs.target-branch }}
 
-      - name: get-kata-tarball
+      - name: get-kata-tools-tarball
         uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
         with:
-          name: kata-static-tarball-amd64-${{ inputs.tag }}
-          path: kata-artifacts
+          name: kata-tools-static-tarball-amd64-${{ inputs.tag }}
+          path: kata-tools-artifacts
 
-      - name: Install tools
-        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-artifacts
+      - name: Install kata-tools
+        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-tools-artifacts
 
       - name: Copy binary into Docker context
         run: |
@@ -314,6 +314,7 @@ jobs:
     needs: publish-kata-deploy-payload-amd64
     uses: ./.github/workflows/run-k8s-tests-on-nvidia-gpu.yaml
     with:
+      tarball-suffix: -${{ inputs.tag }}
       registry: ghcr.io
       repo: ${{ github.repository_owner }}/kata-deploy-ci
       tag: ${{ inputs.tag }}-amd64
@@ -473,7 +474,7 @@ jobs:
       vmm: ${{ matrix.params.vmm }}
 
   run-cri-containerd-tests-arm64:
-    if: ${{ inputs.skip-test != 'yes' }}
+    if: false
     needs: build-kata-static-tarball-arm64
     strategy:
       fail-fast: false

.github/workflows/docs.yaml

@@ -0,0 +1,32 @@
+name: Documentation
+on:
+  push:
+    branches:
+      - main
+permissions: {}
+jobs:
+  deploy-docs:
+    name: deploy-docs
+    permissions:
+      contents: read
+      pages: write
+      id-token: write
+    environment:
+      name: github-pages
+      url: ${{ steps.deployment.outputs.page_url }}
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/configure-pages@v5
+      - uses: actions/checkout@v5
+        with:
+          persist-credentials: false
+      - uses: actions/setup-python@v5
+        with:
+          python-version: 3.x
+      - run: pip install zensical
+      - run: zensical build --clean
+      - uses: actions/upload-pages-artifact@v4
+        with:
+          path: site
+      - uses: actions/deploy-pages@v4
+        id: deployment

.github/workflows/gatekeeper.yaml

@@ -10,7 +10,9 @@ on:
       - opened
       - synchronize
       - reopened
+      - edited
       - labeled
+      - unlabeled
 
 permissions: {}
 

.github/workflows/kata-runtime-classes-sync.yaml

@@ -1,43 +0,0 @@
-name: kata-runtime-classes-sync
-
-on:
-  pull_request:
-    types:
-      - opened
-      - edited
-      - reopened
-      - synchronize
-
-permissions: {}
-
-concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
-  cancel-in-progress: true
-
-jobs:
-  kata-deploy-runtime-classes-check:
-    name: kata-deploy-runtime-classes-check
-    runs-on: ubuntu-22.04
-    steps:
-    - name: Checkout code
-      uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-      with:
-        persist-credentials: false
-    - name: Ensure the split out runtime classes match the all-in-one file
-      run: |
-        pushd tools/packaging/kata-deploy/runtimeclasses/
-        echo "::group::Combine runtime classes"
-        for runtimeClass in $(find . -type f \( -name "*.yaml" -and -not -name "kata-runtimeClasses.yaml" \) | sort); do
-            echo "Adding ${runtimeClass} to the resultingRuntimeClasses.yaml"
-            cat "${runtimeClass}" >> resultingRuntimeClasses.yaml;
-        done
-        echo "::endgroup::"
-        echo "::group::Displaying the content of resultingRuntimeClasses.yaml"
-        cat resultingRuntimeClasses.yaml
-        echo "::endgroup::"
-        echo ""
-        echo "::group::Displaying the content of kata-runtimeClasses.yaml"
-        cat kata-runtimeClasses.yaml
-        echo "::endgroup::"
-        echo ""
-        diff resultingRuntimeClasses.yaml kata-runtimeClasses.yaml

.github/workflows/payload-after-push.yaml

@@ -131,7 +131,7 @@ jobs:
       repo: kata-containers/kata-deploy-ci
       tag: kata-containers-latest-ppc64le
       target-branch: ${{ github.ref_name }}
-      runner: ppc64le-small
+      runner: ubuntu-24.04-ppc64le
       arch: ppc64le
     secrets:
       QUAY_DEPLOYER_PASSWORD: ${{ secrets.QUAY_DEPLOYER_PASSWORD }}

.github/workflows/publish-kata-deploy-payload.yaml

@@ -50,6 +50,24 @@ jobs:
           fetch-depth: 0
           persist-credentials: false
 
+      - name: Remove unnecessary directories to free up space
+        run: |
+          sudo rm -rf /usr/local/.ghcup
+          sudo rm -rf /opt/hostedtoolcache/CodeQL
+          sudo rm -rf /usr/local/lib/android
+          sudo rm -rf /usr/share/dotnet
+          sudo rm -rf /opt/ghc
+          sudo rm -rf /usr/local/share/boost
+          sudo rm -rf /usr/lib/jvm
+          sudo rm -rf /usr/share/swift
+          sudo rm -rf /usr/local/share/powershell
+          sudo rm -rf /usr/local/julia*
+          sudo rm -rf /opt/az
+          sudo rm -rf /usr/local/share/chromium
+          sudo rm -rf /opt/microsoft
+          sudo rm -rf /opt/google
+          sudo rm -rf /usr/lib/firefox
+
       - name: Rebase atop of the latest target branch
         run: |
           ./tests/git-helper.sh "rebase-atop-of-the-latest-target-branch"

.github/workflows/release-ppc64le.yaml

@@ -31,7 +31,7 @@ jobs:
     permissions:
       contents: read
       packages: write
-    runs-on: ppc64le-small
+    runs-on: ubuntu-24.04-ppc64le
     steps:
       - name: Login to Kata Containers ghcr.io
         uses: docker/login-action@74a5d142397b4f367a81961eba4e8cd7edddf772 # v3.4.0

.github/workflows/release-s390x.yaml

@@ -35,7 +35,7 @@ jobs:
     permissions:
       contents: read
       packages: write
-    runs-on: s390x
+    runs-on: ubuntu-24.04-s390x
     steps:
       - name: Login to Kata Containers ghcr.io
         uses: docker/login-action@74a5d142397b4f367a81961eba4e8cd7edddf772 # v3.4.0

.github/workflows/release.yaml

@@ -181,6 +181,23 @@ jobs:
           GH_TOKEN: ${{ github.token }}
           ARCHITECTURE: ppc64le
 
+      - name: Set KATA_TOOLS_STATIC_TARBALL env var
+        run: |
+          tarball=$(pwd)/kata-tools-static.tar.zst
+          echo "KATA_TOOLS_STATIC_TARBALL=${tarball}" >> "$GITHUB_ENV"
+
+      - name: Download amd64 tools artifacts
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        with:
+          name: kata-tools-static-tarball-amd64
+
+      - name: Upload amd64 static tarball tools to GitHub
+        run: |
+          ./tools/packaging/release/release.sh upload-kata-tools-static-tarball
+        env:
+          GH_TOKEN: ${{ github.token }}
+          ARCHITECTURE: amd64
+
   upload-versions-yaml:
     name: upload-versions-yaml
     needs: release

.github/workflows/run-containerd-guest-pull-stability-tests.yaml

@@ -1,167 +0,0 @@
-name: CI | Run containerd guest pull stability tests
-on:
-  schedule:
-    - cron: "0 */1 * * *" #run every hour
-
-permissions: {}
-
-# This job relies on k8s pre-installed using kubeadm
-jobs:
-  run-containerd-guest-pull-stability-tests:
-    name: run-containerd-guest-pull-stability-tests-${{ matrix.environment.test-type }}-${{ matrix.environment.containerd }}
-    strategy:
-      fail-fast: false
-      matrix:
-        environment: [
-          { test-type: multi-snapshotter, containerd: v2.2 },
-          { test-type: force-guest-pull, containerd: v1.7 },
-          { test-type: force-guest-pull, containerd: v2.0 },
-          { test-type: force-guest-pull, containerd: v2.1 },
-          { test-type: force-guest-pull, containerd: v2.2 },
-        ]
-    env:
-      # I don't want those to be inside double quotes, so I'm deliberately ignoring the double quotes here.
-      IMAGES_LIST: quay.io/mongodb/mongodb-community-server@sha256:8b73733842da21b6bbb6df4d7b2449229bb3135d2ec8c6880314d88205772a11 ghcr.io/edgelesssys/redis@sha256:ecb0a964c259a166a1eb62f0eb19621d42bd1cce0bc9bb0c71c828911d4ba93d
-    runs-on: containerd-${{ matrix.environment.test-type }}-${{ matrix.environment.containerd }}
-    steps:
-      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-        with:
-          persist-credentials: false
-
-      - name: Rotate the journal
-        run: sudo journalctl --rotate --vacuum-time 1s
-
-      - name: Pull the kata-deploy image to be used
-        run: sudo ctr -n k8s.io image pull quay.io/kata-containers/kata-deploy-ci:kata-containers-latest
-
-      - name: Deploy Kata Containers
-        run: bash tests/integration/kubernetes/gha-run.sh deploy-kata
-        env:
-          KATA_HYPERVISOR: qemu-coco-dev
-          KUBERNETES: vanilla
-          SNAPSHOTTER: ${{ matrix.environment.test-type == 'multi-snapshotter' && 'nydus' || '' }}
-          USE_EXPERIMENTAL_SETUP_SNAPSHOTTER: ${{ matrix.environment.test-type == 'multi-snapshotter' }}
-          EXPERIMENTAL_FORCE_GUEST_PULL: ${{ matrix.environment.test-type == 'force-guest-pull' && 'qemu-coco-dev' || '' }}
-
-      # This is needed as we may hit the createContainerTimeout
-      - name: Adjust Kata Containers' create_container_timeout
-        run: |
-          sudo sed -i -e 's/^\(create_container_timeout\).*=.*$/\1 = 600/g' /opt/kata/share/defaults/kata-containers/configuration-qemu-coco-dev.toml
-          grep "create_container_timeout.*=" /opt/kata/share/defaults/kata-containers/configuration-qemu-coco-dev.toml
-
-      # This is needed in order to have enough tmpfs space inside the guest to pull the image
-      - name: Adjust Kata Containers' default_memory
-        run: |
-          sudo sed -i -e 's/^\(default_memory\).*=.*$/\1 = 4096/g' /opt/kata/share/defaults/kata-containers/configuration-qemu-coco-dev.toml
-          grep "default_memory.*=" /opt/kata/share/defaults/kata-containers/configuration-qemu-coco-dev.toml
-
-      - name: Run a few containers using overlayfs
-        run: |
-          # I don't want those to be inside double quotes, so I'm deliberately ignoring the double quotes here
-          # shellcheck disable=SC2086
-          for img in ${IMAGES_LIST}; do
-            echo "overlayfs | Using on image: ${img}"
-            pod="$(echo ${img} | tr ':.@/' '-' | awk '{print substr($0,1,56)}')"
-            kubectl run "${pod}" \
-              -it --rm \
-              --restart=Never \
-              --image="${img}" \
-              --image-pull-policy=Always \
-              --pod-running-timeout=10m \
-              -- uname -r
-          done
-          
-      - name: Run a the same few containers using a different snapshotter
-        run: |
-          # I don't want those to be inside double quotes, so I'm deliberately ignoring the double quotes here
-          # shellcheck disable=SC2086
-          for img in ${IMAGES_LIST}; do
-            echo "nydus | Using on image: ${img}"
-            pod="kata-$(echo ${img} | tr ':.@/' '-' | awk '{print substr($0,1,56)}')"
-            kubectl run "${pod}" \
-              -it --rm \
-              --restart=Never \
-              --image="${img}" \
-              --image-pull-policy=Always \
-              --pod-running-timeout=10m \
-              --overrides='{
-                "spec": {
-                  "runtimeClassName": "kata-qemu-coco-dev"
-                }
-              }' \
-              -- uname -r
-          done
-
-      - name: Uninstall Kata Containers
-        run: bash tests/integration/kubernetes/gha-run.sh cleanup
-        env:
-          KATA_HYPERVISOR: qemu-coco-dev
-          KUBERNETES: vanilla
-          SNAPSHOTTER: nydus
-          USE_EXPERIMENTAL_SETUP_SNAPSHOTTER: true
-
-      - name: Run a few containers using overlayfs
-        run: |
-          # I don't want those to be inside double quotes, so I'm deliberately ignoring the double quotes here
-          # shellcheck disable=SC2086
-          for img in ${IMAGES_LIST}; do
-            echo "overlayfs | Using on image: ${img}"
-            pod="$(echo ${img} | tr ':.@/' '-' | awk '{print substr($0,1,56)}')"
-            kubectl run "${pod}" \
-              -it --rm \
-              --restart=Never \
-              --image=${img} \
-              --image-pull-policy=Always \
-              --pod-running-timeout=10m \
-              -- uname -r
-          done
-          
-      - name: Deploy Kata Containers
-        run: bash tests/integration/kubernetes/gha-run.sh deploy-kata
-        env:
-          KATA_HYPERVISOR: qemu-coco-dev
-          KUBERNETES: vanilla
-          SNAPSHOTTER: nydus
-          USE_EXPERIMENTAL_SETUP_SNAPSHOTTER: true
-
-      # This is needed as we may hit the createContainerTimeout
-      - name: Adjust Kata Containers' create_container_timeout
-        run: |
-          sudo sed -i -e 's/^\(create_container_timeout\).*=.*$/\1 = 600/g' /opt/kata/share/defaults/kata-containers/configuration-qemu-coco-dev.toml
-          grep "create_container_timeout.*=" /opt/kata/share/defaults/kata-containers/configuration-qemu-coco-dev.toml
-
-      # This is needed in order to have enough tmpfs space inside the guest to pull the image
-      - name: Adjust Kata Containers' default_memory
-        run: |
-          sudo sed -i -e 's/^\(default_memory\).*=.*$/\1 = 4096/g' /opt/kata/share/defaults/kata-containers/configuration-qemu-coco-dev.toml
-          grep "default_memory.*=" /opt/kata/share/defaults/kata-containers/configuration-qemu-coco-dev.toml
-
-      - name: Run a the same few containers using a different snapshotter
-        run: |
-          # I don't want those to be inside double quotes, so I'm deliberately ignoring the double quotes here
-          # shellcheck disable=SC2086
-          for img in ${IMAGES_LIST}; do
-            echo "nydus | Using on image: ${img}"
-            pod="kata-$(echo ${img} | tr ':.@/' '-' | awk '{print substr($0,1,56)}')"
-            kubectl run "${pod}" \
-              -it --rm \
-              --restart=Never \
-              --image="${img}" \
-              --image-pull-policy=Always \
-              --pod-running-timeout=10m \
-              --overrides='{
-                "spec": {
-                  "runtimeClassName": "kata-qemu-coco-dev"
-                }
-              }' \
-              -- uname -r
-          done
-
-      - name: Uninstall Kata Containers
-        run: bash tests/integration/kubernetes/gha-run.sh cleanup || true
-        if: always()
-        env:
-          KATA_HYPERVISOR: qemu-coco-dev
-          KUBERNETES: vanilla
-          SNAPSHOTTER: nydus
-          USE_EXPERIMENTAL_SETUP_SNAPSHOTTER: true

.github/workflows/run-k8s-tests-on-aks.yaml

@@ -93,14 +93,14 @@ jobs:
         env:
           TARGET_BRANCH: ${{ inputs.target-branch }}
 
-      - name: get-kata-tarball
+      - name: get-kata-tools-tarball
         uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
         with:
-          name: kata-static-tarball-amd64${{ inputs.tarball-suffix }}
-          path: kata-artifacts
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
 
-      - name: Install kata
-        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-artifacts
+      - name: Install kata-tools
+        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-tools-artifacts
 
       - name: Download Azure CLI
         uses: azure/setup-kubectl@776406bce94f63e41d621b960d78ee25c8b76ede # v4.0.1
@@ -142,6 +142,10 @@ jobs:
         timeout-minutes: 60
         run: bash tests/integration/kubernetes/gha-run.sh run-tests
 
+      - name: Report tests
+        if: always()
+        run: bash tests/integration/kubernetes/gha-run.sh report-tests
+
       - name: Refresh OIDC token in case access token expired
         if: always()
         uses: azure/login@a457da9ea143d694b1b9c7c869ebb04ebe844ef5 # v2.3.0

.github/workflows/run-k8s-tests-on-arm64.yaml

@@ -32,6 +32,7 @@ jobs:
       matrix:
         vmm:
           - qemu
+          - qemu-runtime-rs
         k8s:
           - kubeadm
     runs-on: arm64-k8s
@@ -68,6 +69,10 @@ jobs:
         timeout-minutes: 30
         run: bash tests/integration/kubernetes/gha-run.sh run-tests
 
+      - name: Report tests
+        if: always()
+        run: bash tests/integration/kubernetes/gha-run.sh report-tests
+
       - name: Collect artifacts ${{ matrix.vmm }}
         if: always()
         run: bash tests/integration/kubernetes/gha-run.sh collect-artifacts

.github/workflows/run-k8s-tests-on-nvidia-gpu.yaml

@@ -2,6 +2,9 @@ name: CI | Run NVIDIA GPU kubernetes tests on amd64
 on:
   workflow_call:
     inputs:
+      tarball-suffix:
+        required: true
+        type: string
       registry:
         required: true
         type: string
@@ -45,6 +48,7 @@ jobs:
       GH_PR_NUMBER: ${{ inputs.pr-number }}
       KATA_HYPERVISOR: ${{ matrix.environment.vmm }}
       KUBERNETES: kubeadm
+      KBS: ${{ matrix.environment.name == 'nvidia-gpu-snp' && 'true' || 'false' }}
       K8S_TEST_HOST_TYPE: baremetal
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
@@ -59,6 +63,15 @@ jobs:
         env:
           TARGET_BRANCH: ${{ inputs.target-branch }}
 
+      - name: get-kata-tools-tarball
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        with:
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
+
+      - name: Install kata-tools
+        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-tools-artifacts
+
       - name: Uninstall previous `kbs-client`
         if: matrix.environment.name != 'nvidia-gpu'
         timeout-minutes: 10
@@ -89,6 +102,11 @@ jobs:
         run: bash tests/integration/kubernetes/gha-run.sh run-nv-tests
         env:
           NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+
+      - name: Report tests
+        if: always()
+        run: bash tests/integration/kubernetes/gha-run.sh report-tests
+
       - name: Collect artifacts ${{ matrix.environment.vmm }}
         if: always()
         run: bash tests/integration/kubernetes/gha-run.sh collect-artifacts
@@ -108,5 +126,6 @@ jobs:
 
       - name: Delete CoCo KBS
         if: always() && matrix.environment.name != 'nvidia-gpu'
+        timeout-minutes: 10
         run: |
           bash tests/integration/kubernetes/gha-run.sh delete-coco-kbs

.github/workflows/run-k8s-tests-on-ppc64le.yaml

@@ -75,3 +75,7 @@ jobs:
       - name: Run tests
         timeout-minutes: 30
         run: bash tests/integration/kubernetes/gha-run.sh run-tests
+
+      - name: Report tests
+        if: always()
+        run: bash tests/integration/kubernetes/gha-run.sh report-tests

.github/workflows/run-k8s-tests-on-zvsi.yaml

@@ -131,12 +131,18 @@ jobs:
         timeout-minutes: 60
         run: bash tests/integration/kubernetes/gha-run.sh run-tests
 
+      - name: Report tests
+        if: always()
+        run: bash tests/integration/kubernetes/gha-run.sh report-tests
+
       - name: Delete kata-deploy
         if: always()
+        timeout-minutes: 10
         run: bash tests/integration/kubernetes/gha-run.sh cleanup-zvsi
 
       - name: Delete CoCo KBS
         if: always()
+        timeout-minutes: 10
         run: |
           if [ "${KBS}" == "true" ]; then
             bash tests/integration/kubernetes/gha-run.sh delete-coco-kbs

.github/workflows/run-kata-coco-stability-tests.yaml

@@ -84,14 +84,14 @@ jobs:
         env:
           TARGET_BRANCH: ${{ inputs.target-branch }}
 
-      - name: get-kata-tarball
+      - name: get-kata-tools-tarball
         uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
         with:
-          name: kata-static-tarball-amd64${{ inputs.tarball-suffix }}
-          path: kata-artifacts
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
 
-      - name: Install kata
-        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-artifacts
+      - name: Install kata-tools
+        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-tools-artifacts
 
       - name: Log into the Azure account
         uses: azure/login@a457da9ea143d694b1b9c7c869ebb04ebe844ef5 # v2.3.0
@@ -140,6 +140,10 @@ jobs:
         timeout-minutes: 300
         run: bash tests/stability/gha-stability-run.sh run-tests
 
+      - name: Report tests
+        if: always()
+        run: bash tests/integration/kubernetes/gha-run.sh report-tests
+
       - name: Refresh OIDC token in case access token expired
         if: always()
         uses: azure/login@a457da9ea143d694b1b9c7c869ebb04ebe844ef5 # v2.3.0

.github/workflows/run-kata-coco-tests.yaml

@@ -79,6 +79,15 @@ jobs:
         env:
           TARGET_BRANCH: ${{ inputs.target-branch }}
 
+      - name: get-kata-tools-tarball
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        with:
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
+
+      - name: Install kata-tools
+        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-tools-artifacts
+
       - name: Deploy Kata
         timeout-minutes: 20
         run: bash tests/integration/kubernetes/gha-run.sh deploy-kata
@@ -111,10 +120,12 @@ jobs:
 
       - name: Delete kata-deploy
         if: always()
+        timeout-minutes: 15
         run: bash tests/integration/kubernetes/gha-run.sh cleanup
 
       - name: Delete CoCo KBS
         if: always()
+        timeout-minutes: 10
         run: |
           [[ "${KATA_HYPERVISOR}" == "qemu-tdx" ]] && echo "ITA_KEY=${GH_ITA_KEY}" >> "${GITHUB_ENV}"
           bash tests/integration/kubernetes/gha-run.sh delete-coco-kbs
@@ -178,14 +189,14 @@ jobs:
         env:
           TARGET_BRANCH: ${{ inputs.target-branch }}
 
-      - name: get-kata-tarball
+      - name: get-kata-tools-tarball
         uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
         with:
-          name: kata-static-tarball-amd64${{ inputs.tarball-suffix }}
-          path: kata-artifacts
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
 
-      - name: Install kata
-        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-artifacts
+      - name: Install kata-tools
+        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-tools-artifacts
 
       - name: Log into the Azure account
         uses: azure/login@a457da9ea143d694b1b9c7c869ebb04ebe844ef5 # v2.3.0
@@ -301,6 +312,15 @@ jobs:
         env:
           TARGET_BRANCH: ${{ inputs.target-branch }}
 
+      - name: get-kata-tools-tarball
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        with:
+          name: kata-tools-static-tarball-amd64${{ inputs.tarball-suffix }}
+          path: kata-tools-artifacts
+
+      - name: Install kata-tools
+        run: bash tests/integration/kubernetes/gha-run.sh install-kata-tools kata-tools-artifacts
+
       - name: Remove unnecessary directories to free up space
         run: |
           sudo rm -rf /usr/local/.ghcup
@@ -309,7 +329,6 @@ jobs:
           sudo rm -rf /usr/share/dotnet
           sudo rm -rf /opt/ghc
           sudo rm -rf /usr/local/share/boost
-          sudo rm -rf "$AGENT_TOOLSDIRECTORY"
           sudo rm -rf /usr/lib/jvm
           sudo rm -rf /usr/share/swift
           sudo rm -rf /usr/local/share/powershell

.github/workflows/run-kata-deploy-tests-on-aks.yaml

@@ -102,6 +102,10 @@ jobs:
       - name: Run tests
         run: bash tests/functional/kata-deploy/gha-run.sh run-tests
 
+      - name: Report tests
+        if: always()
+        run: bash tests/integration/kubernetes/gha-run.sh report-tests
+
       - name: Refresh OIDC token in case access token expired
         if: always()
         uses: azure/login@a457da9ea143d694b1b9c7c869ebb04ebe844ef5 # v2.3.0

.github/workflows/run-kata-deploy-tests.yaml

@@ -66,7 +66,6 @@ jobs:
           sudo rm -rf /usr/share/dotnet
           sudo rm -rf /opt/ghc
           sudo rm -rf /usr/local/share/boost
-          sudo rm -rf "$AGENT_TOOLSDIRECTORY"
           sudo rm -rf /usr/lib/jvm
           sudo rm -rf /usr/share/swift
           sudo rm -rf /usr/local/share/powershell
@@ -85,3 +84,7 @@ jobs:
 
       - name: Run tests
         run: bash tests/functional/kata-deploy/gha-run.sh run-tests
+
+      - name: Report tests
+        if: always()
+        run: bash tests/functional/kata-deploy/gha-run.sh report-tests

.github/workflows/run-runk-tests.yaml

@@ -1,54 +0,0 @@
-name: CI | Run runk tests
-on:
-  workflow_call:
-    inputs:
-      tarball-suffix:
-        required: false
-        type: string
-      commit-hash:
-        required: false
-        type: string
-      target-branch:
-        required: false
-        type: string
-        default: ""
-
-permissions: {}
-
-jobs:
-  run-runk:
-    name: run-runk
-    # Skip runk tests as we have no maintainers. TODO: Decide when to remove altogether
-    if: false
-    runs-on: ubuntu-22.04
-    env:
-      CONTAINERD_VERSION: lts
-    steps:
-      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-        with:
-          ref: ${{ inputs.commit-hash }}
-          fetch-depth: 0
-          persist-credentials: false
-
-      - name: Rebase atop of the latest target branch
-        run: |
-          ./tests/git-helper.sh "rebase-atop-of-the-latest-target-branch"
-        env:
-          TARGET_BRANCH: ${{ inputs.target-branch }}
-
-      - name: Install dependencies
-        run: bash tests/integration/runk/gha-run.sh install-dependencies
-        env:
-          GH_TOKEN: ${{ github.token }}
-
-      - name: get-kata-tarball
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
-        with:
-          name: kata-static-tarball-amd64${{ inputs.tarball-suffix }}
-          path: kata-artifacts
-
-      - name: Install kata
-        run: bash tests/integration/runk/gha-run.sh install-kata kata-artifacts
-
-      - name: Run runk tests
-        run: bash tests/integration/runk/gha-run.sh run

.github/workflows/stale.yaml

@@ -6,14 +6,21 @@ on:
 
 permissions: {}
 
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
 jobs:
   stale:
     name: stale
     runs-on: ubuntu-22.04
+    permissions:
+      actions: write # Needed to manage caches for state persistence across runs
+      pull-requests: write # Needed to add/remove labels, post comments, or close PRs
     steps:
       - uses: actions/stale@5bef64f19d7facfb25b37b414482c7164d639639 # v9.1.0
         with:
-          stale-pr-message: 'This PR has been opened without with no activity for 180 days. Comment on the issue otherwise it will be closed in 7 days'
+          stale-pr-message: 'This PR has been opened without activity for 180 days. Please comment on the issue or it will be closed in 7 days.'
           days-before-pr-stale: 180
           days-before-pr-close: 7
           days-before-issue-stale: -1

.github/workflows/static-checks-self-hosted.yaml

@@ -29,7 +29,7 @@ jobs:
       matrix:
         instance:
           - "ubuntu-24.04-arm"
-          - "s390x"
+          - "ubuntu-24.04-s390x"
           - "ubuntu-24.04-ppc64le"
     uses: ./.github/workflows/build-checks.yaml
     with:

.github/workflows/zizmor.yaml

@@ -21,7 +21,7 @@ jobs:
           persist-credentials: false
 
       - name: Run zizmor
-        uses: zizmorcore/zizmor-action@e673c3917a1aef3c65c972347ed84ccd013ecda4 # v0.2.0
+        uses: zizmorcore/zizmor-action@135698455da5c3b3e55f73f4419e481ab68cdd95 # v0.4.1
         with:
           advanced-security: false
           annotations: true

.gitignore

@@ -18,3 +18,5 @@ src/tools/log-parser/kata-log-parser
 tools/packaging/static-build/agent/install_libseccomp.sh
 .envrc
 .direnv
+**/.DS_Store
+site/

Cargo.toml

@@ -2,39 +2,53 @@
 authors = ["The Kata Containers community <kata-dev@lists.katacontainers.io>"]
 edition = "2018"
 license = "Apache-2.0"
-rust-version = "1.85.1"
+rust-version = "1.88"
 
 [workspace]
 members = [
-    # Dragonball
-    "src/dragonball",
-    "src/dragonball/dbs_acpi",
-    "src/dragonball/dbs_address_space",
-    "src/dragonball/dbs_allocator",
-    "src/dragonball/dbs_arch",
-    "src/dragonball/dbs_boot",
-    "src/dragonball/dbs_device",
-    "src/dragonball/dbs_interrupt",
-    "src/dragonball/dbs_legacy_devices",
-    "src/dragonball/dbs_pci",
-    "src/dragonball/dbs_tdx",
-    "src/dragonball/dbs_upcall",
-    "src/dragonball/dbs_utils",
-    "src/dragonball/dbs_virtio_devices",
+  # Dragonball
+  "src/dragonball",
+  "src/dragonball/dbs_acpi",
+  "src/dragonball/dbs_address_space",
+  "src/dragonball/dbs_allocator",
+  "src/dragonball/dbs_arch",
+  "src/dragonball/dbs_boot",
+  "src/dragonball/dbs_device",
+  "src/dragonball/dbs_interrupt",
+  "src/dragonball/dbs_legacy_devices",
+  "src/dragonball/dbs_pci",
+  "src/dragonball/dbs_tdx",
+  "src/dragonball/dbs_upcall",
+  "src/dragonball/dbs_utils",
+  "src/dragonball/dbs_virtio_devices",
+
+  # runtime-rs
+  "src/runtime-rs",
+  "src/runtime-rs/crates/agent",
+  "src/runtime-rs/crates/hypervisor",
+  "src/runtime-rs/crates/persist",
+  "src/runtime-rs/crates/resource",
+  "src/runtime-rs/crates/runtimes",
+  "src/runtime-rs/crates/service",
+  "src/runtime-rs/crates/shim",
+  "src/runtime-rs/crates/shim-ctl",
+  "src/runtime-rs/tests/utils",
 ]
 resolver = "2"
 
 # TODO: Add all excluded crates to root workspace
 exclude = [
-    "src/agent",
-    "src/tools",
-    "src/libs",
-    "src/runtime-rs",
+  "src/agent",
+  "src/tools",
+  "src/libs",
 
-    # We are cloning and building rust packages under
-    # "tools/packaging/kata-deploy/local-build/build" folder, which may mislead
-    # those packages to think they are part of the kata root workspace
-    "tools/packaging/kata-deploy/local-build/build",
+  # kata-deploy binary is standalone and has its own Cargo.toml for now
+  "tools/packaging/kata-deploy/binary",
+
+  # We are cloning and building rust packages under
+  # "tools/packaging/kata-deploy/local-build/build" folder, which may mislead
+  # those packages to think they are part of the kata root workspace
+  "tools/packaging/kata-deploy/local-build/build",
 ]
 
 [workspace.dependencies]
@@ -54,6 +68,7 @@ vm-superio = "0.5.0"
 vmm-sys-util = "0.11.0"
 
 # Local dependencies from Dragonball Sandbox crates
+dragonball = { path = "src/dragonball" }
 dbs-acpi = { path = "src/dragonball/dbs_acpi" }
 dbs-address-space = { path = "src/dragonball/dbs_address_space" }
 dbs-allocator = { path = "src/dragonball/dbs_allocator" }
@@ -68,5 +83,58 @@ dbs-upcall = { path = "src/dragonball/dbs_upcall" }
 dbs-utils = { path = "src/dragonball/dbs_utils" }
 dbs-virtio-devices = { path = "src/dragonball/dbs_virtio_devices" }
 
+# Local dependencies from runtime-rs
+agent = { path = "src/runtime-rs/crates/agent" }
+hypervisor = { path = "src/runtime-rs/crates/hypervisor" }
+persist = { path = "src/runtime-rs/crates/persist" }
+resource = { path = "src/runtime-rs/crates/resource" }
+runtimes = { path = "src/runtime-rs/crates/runtimes" }
+service = { path = "src/runtime-rs/crates/service" }
+tests_utils = { path = "src/runtime-rs/tests/utils" }
+ch-config = { path = "src/runtime-rs/crates/hypervisor/ch-config" }
+common = { path = "src/runtime-rs/crates/runtimes/common" }
+linux_container = { path = "src/runtime-rs/crates/runtimes/linux_container" }
+virt_container = { path = "src/runtime-rs/crates/runtimes/virt_container" }
+wasm_container = { path = "src/runtime-rs/crates/runtimes/wasm_container" }
+
 # Local dependencies from `src/lib`
+kata-sys-util = { path = "src/libs/kata-sys-util" }
+kata-types = { path = "src/libs/kata-types", features = ["safe-path"] }
+logging = { path = "src/libs/logging" }
+protocols = { path = "src/libs/protocols", features = ["async"] }
+runtime-spec = { path = "src/libs/runtime-spec" }
+safe-path = { path = "src/libs/safe-path" }
+shim-interface = { path = "src/libs/shim-interface" }
 test-utils = { path = "src/libs/test-utils" }
+
+# Outside dependencies
+actix-rt = "2.7.0"
+anyhow = "1.0"
+async-trait = "0.1.48"
+containerd-shim = { version = "0.10.0", features = ["async"] }
+containerd-shim-protos = { version = "0.10.0", features = ["async"] }
+go-flag = "0.1.0"
+hyper = "0.14.20"
+hyperlocal = "0.8.0"
+lazy_static = "1.4"
+libc = "0.2"
+log = "0.4.14"
+netns-rs = "0.1.0"
+# Note: nix needs to stay sync'd with libs versions
+nix = "0.26.4"
+oci-spec = { version = "0.8.1", features = ["runtime"] }
+protobuf = "3.7.2"
+rand = "0.8.4"
+serde = { version = "1.0.145", features = ["derive"] }
+serde_json = "1.0.91"
+sha2 = "0.10.9"
+slog = "2.5.2"
+slog-scope = "4.4.0"
+strum = { version = "0.24.0", features = ["derive"] }
+tempfile = "3.19.1"
+thiserror = "1.0"
+tokio = "1.46.1"
+tracing = "0.1.41"
+tracing-opentelemetry = "0.18.0"
+ttrpc = "0.8.4"
+url = "2.5.4"

Makefile

@@ -18,7 +18,6 @@ TOOLS =
 TOOLS += agent-ctl
 TOOLS += kata-ctl
 TOOLS += log-parser
-TOOLS += runk
 TOOLS += trace-forwarder
 
 STANDARD_TARGETS = build check clean install static-checks-build test vendor
@@ -50,10 +49,14 @@ docs-url-alive-check:
 build-and-publish-kata-debug:
 	bash tools/packaging/kata-debug/kata-debug-build-and-upload-payload.sh ${KATA_DEBUG_REGISTRY} ${KATA_DEBUG_TAG} 
 
+docs-serve:
+	docker run --rm -p 8000:8000 -v ./docs:/docs:ro -v ${PWD}/zensical.toml:/zensical.toml:ro zensical/zensical serve --config-file /zensical.toml -a 0.0.0.0:8000 
+
 .PHONY: \
 	all \
 	kata-tarball \
 	install-tarball \
 	default \
 	static-checks \
-	docs-url-alive-check
+	docs-url-alive-check \
+	docs-serve

README.md

@@ -139,7 +139,6 @@ The table below lists the remaining parts of the project:
 | [`agent-ctl`](src/tools/agent-ctl) | utility | Tool that provides low-level access for testing the agent. |
 | [`kata-ctl`](src/tools/kata-ctl) | utility | Tool that provides advanced commands and debug facilities. |
 | [`trace-forwarder`](src/tools/trace-forwarder) | utility | Agent tracing helper. |
-| [`runk`](src/tools/runk) | utility | Standard OCI container runtime based on the agent. |
 | [`ci`](.github/workflows) | CI | Continuous Integration configuration files and scripts. |
 | [`ocp-ci`](ci/openshift-ci/README.md) | CI | Continuous Integration configuration for the OpenShift pipelines. |
 | [`katacontainers.io`](https://github.com/kata-containers/www.katacontainers.io) | Source for the [`katacontainers.io`](https://www.katacontainers.io) site. |

VERSION

@@ -1 +1 @@
-3.23.0
+3.26.0

ci/install_libseccomp.sh

@@ -11,6 +11,10 @@ script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 
 source "${script_dir}/../tests/common.bash"
 
+# Path to the ORAS cache helper for downloading tarballs (sourced when needed)
+# Use ORAS_CACHE_HELPER env var (set by build.sh in Docker) or fallback to repo path
+oras_cache_helper="${ORAS_CACHE_HELPER:-${script_dir}/../tools/packaging/scripts/download-with-oras-cache.sh}"
+
 # The following variables if set on the environment will change the behavior
 # of gperf and libseccomp configure scripts, that may lead this script to
 # fail. So let's ensure they are unset here.
@@ -44,6 +48,9 @@ fi
 gperf_tarball="gperf-${gperf_version}.tar.gz"
 gperf_tarball_url="${gperf_url}/${gperf_tarball}"
 
+# Use ORAS cache for gperf downloads (gperf upstream can be unreliable)
+USE_ORAS_CACHE="${USE_ORAS_CACHE:-yes}"
+
 # We need to build the libseccomp library from sources to create a static
 # library for the musl libc.
 # However, ppc64le, riscv64 and s390x have no musl targets in Rust. Hence, we do
@@ -68,7 +75,23 @@ trap finish EXIT
 build_and_install_gperf() {
 	echo "Build and install gperf version ${gperf_version}"
 	mkdir -p "${gperf_install_dir}"
-	curl -sLO "${gperf_tarball_url}"
+
+	# Use ORAS cache if available and enabled
+	if [[ "${USE_ORAS_CACHE}" == "yes" ]] && [[ -f "${oras_cache_helper}" ]]; then
+		echo "Using ORAS cache for gperf download"
+		source "${oras_cache_helper}"
+		local cached_tarball
+		cached_tarball=$(download_component gperf "$(pwd)")
+		if [[ -f "${cached_tarball}" ]]; then
+			gperf_tarball="${cached_tarball}"
+		else
+			echo "ORAS cache download failed, falling back to direct download"
+			curl -sLO "${gperf_tarball_url}"
+		fi
+	else
+		curl -sLO "${gperf_tarball_url}"
+	fi
+
 	tar -xf "${gperf_tarball}"
 	pushd "gperf-${gperf_version}"
 	# Unset $CC for configure, we will always use native for gperf

ci/openshift-ci/cleanup.sh

@@ -46,16 +46,12 @@ fi
 [[ ${SELINUX_PERMISSIVE} == "yes" ]] && oc delete -f "${deployments_dir}/machineconfig_selinux.yaml.in"
 
 # Delete kata-containers
-pushd "${katacontainers_repo_dir}/tools/packaging/kata-deploy" || { echo "Failed to push to ${katacontainers_repo_dir}/tools/packaging/kata-deploy"; exit 125; }
-oc delete -f kata-deploy/base/kata-deploy.yaml
+helm uninstall kata-deploy --wait --namespace kube-system
 oc -n kube-system wait --timeout=10m --for=delete -l name=kata-deploy pod
-oc apply -f kata-cleanup/base/kata-cleanup.yaml
 echo "Wait for all related pods to be gone"
 ( repeats=1; for _ in $(seq 1 600); do
   oc get pods -l name="kubelet-kata-cleanup" --no-headers=true -n kube-system 2>&1 | grep "No resources found" -q && ((repeats++)) || repeats=1
   [[ "${repeats}" -gt 5 ]] && echo kata-cleanup finished && break
   sleep 1
 done) || { echo "There are still some kata-cleanup related pods after 600 iterations"; oc get all -n kube-system; exit 1; }
-oc delete -f kata-cleanup/base/kata-cleanup.yaml
-oc delete -f kata-rbac/base/kata-rbac.yaml
 oc delete -f runtimeclasses/kata-runtimeClasses.yaml

ci/openshift-ci/cluster/install_kata.sh

@@ -51,13 +51,13 @@ apply_kata_deploy() {
 
 	oc label --overwrite ns kube-system pod-security.kubernetes.io/enforce=privileged pod-security.kubernetes.io/warn=baseline pod-security.kubernetes.io/audit=baseline
 	local version chart
-	version=$(curl -sSL https://api.github.com/repos/kata-containers/kata-containers/releases/latest | jq .tag_name | tr -d '"')
+	version='0.0.0-dev'
 	chart="oci://ghcr.io/kata-containers/kata-deploy-charts/kata-deploy"
 
 	# Ensure any potential leftover is cleaned up ... and this secret usually is not in case of previous failures
 	oc delete secret sh.helm.release.v1.kata-deploy.v1 -n kube-system || true
 
-	echo "Installing kata using helm ${chart} ${version}"
+	echo "Installing kata using helm ${chart} ${version} (sha printed in helm output)"
 	helm install kata-deploy --wait --namespace kube-system --set "image.reference=${KATA_DEPLOY_IMAGE%%:*},image.tag=${KATA_DEPLOY_IMAGE##*:}" "${chart}" --version "${version}"
 }
 

ci/openshift-ci/peer-pods-azure.sh

@@ -157,6 +157,16 @@ if [[ -z "${CAA_IMAGE}" ]]; then
 fi
 
 # Get latest PP image
+#
+# You can list the CI images by:
+#     az sig image-version list-community --location "eastus" --public-gallery-name "cocopodvm-d0e4f35f-5530-4b9c-8596-112487cdea85" --gallery-image-definition "podvm_image0" --output table
+# or the release images by:
+#     az sig image-version list-community --location "eastus" --public-gallery-name "cococommunity-42d8482d-92cd-415b-b332-7648bd978eff" --gallery-image-definition "peerpod-podvm-fedora" --output table
+# or the release debug images by:
+#     az sig image-version list-community --location "eastus" --public-gallery-name "cococommunity-42d8482d-92cd-415b-b332-7648bd978eff" --gallery-image-definition "peerpod-podvm-fedora-debug" --output table
+#
+# Note there are other flavours of the released images, you can list them by:
+#     az sig image-definition list-community --location "eastus" --public-gallery-name "cococommunity-42d8482d-92cd-415b-b332-7648bd978eff" --output table
 if [[ -z "${PP_IMAGE_ID}" ]]; then
 	SUCCESS_TIME=$(curl -s \
 	  -H "Accept: application/vnd.github+json" \

docs/Developer-Guide.md

@@ -125,7 +125,7 @@ If you want to enable SELinux in Permissive mode, add `enforcing=0` to the kerne
 Enable full debug as follows:
 
 ```bash
-$ sudo sed -i -e 's/^# *\(enable_debug\).*=.*$/\1 = true/g' /etc/kata-containers/configuration.toml
+$ sudo sed -i -E 's/^(\s*enable_debug\s*=\s*)false/\1true/' /etc/kata-containers/configuration.toml
 $ sudo sed -i -e 's/^kernel_params = "\(.*\)"/kernel_params = "\1 agent.log=debug initcall_debug"/g' /etc/kata-containers/configuration.toml
 ```
 

docs/README.md

@@ -83,3 +83,7 @@ Documents that help to understand and contribute to Kata Containers.
 If you have a suggestion for how we can improve the
 [website](https://katacontainers.io), please raise an issue (or a PR) on
 [the repository that holds the source for the website](https://github.com/OpenStackweb/kata-netlify-refresh).
+
+### Toolchain Guidance
+
+* [Toolchain Guidance](./Toochain-Guidance.md)

docs/Toochain-Guidance.md

@@ -0,0 +1,39 @@
+# Toolchains
+
+As a community we want to strike a balance between having up-to-date toolchains, to receive the
+latest security fixes and to be able to benefit from new features and packages, whilst not being
+too bleeding edge and disrupting downstream and other consumers. As a result we have the following
+guidelines (note, not hard rules) for our go and rust toolchains that we are attempting to try out:
+
+## Go toolchain
+
+Go is released [every six months](https://go.dev/wiki/Go-Release-Cycle) with support for the
+[last two major release versions](https://go.dev/doc/devel/release#policy). We always want to
+ensure that we are on a supported version so we receive security fixes. To try and make
+things easier for some of our users, we aim to be using the older of the two supported major
+versions, unless there is a compelling reason to adopt the newer version.
+
+In practice this means that we bump our major version of the go toolchain every six months to
+version (1.x-1) in response to a new version (1.x) coming out, which makes our current version
+(1.x-2) no longer supported. We will bump the minor version whenever required to satisfy
+dependency updates, or security fixes.
+
+Our go toolchain version is recorded in [`versions.yaml`](../versions.yaml) under
+`.languages.golang.version` and should match with the version in our `go.mod` files.
+
+## Rust toolchain
+
+Rust has a [six week](https://doc.rust-lang.org/book/appendix-05-editions.html#:~:text=The%20Rust%20language%20and%20compiler,these%20tiny%20changes%20add%20up.)
+release cycle and they only support the latest stable release, so if we wanted to remain on a
+supported release we would only ever build with the latest stable and bump every 6 weeks.
+However feedback from our community has indicated that this is a challenge as downstream consumers
+often want to get rust from their distro, or downstream fork and these struggle to keep up with
+the six week release schedule. As a result the community has agreed to try out a policy of
+"stable-2", where we aim to build with a rust version that is two versions behind the latest stable
+version.
+
+In practice this should mean that we bump our rust toolchain every six weeks, to version
+1.x-2 when 1.x is released as stable and we should be picking up the latest point release
+of that version, if there were any.
+
+The rust-toolchain that we are using is recorded in [`rust-toolchain.toml`](../rust-toolchain.toml).

docs/Unit-Test-Advice.md

@@ -198,7 +198,7 @@ fn join_params_with_dash(str: &str, num: i32) -> Result<String> {
         return Err("number must be positive");
     }
 
-    let result = format!("{}-{}", str, num);
+    let result = format!("{str}-{num}");
 
     Ok(result)
 }
@@ -253,13 +253,13 @@ mod tests {
         // Run the tests
         for (i, d) in tests.iter().enumerate() {
             // Create a string containing details of the test
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             // Call the function under test
             let result = join_params_with_dash(d.str, d.num);
 
             // Update the test details string with the results of the call
-            let msg = format!("{}, result: {:?}", msg, result);
+            let msg = format!("{msg}, result: {result:?}");
 
             // Perform the checks
             if d.result.is_ok() {
@@ -267,8 +267,8 @@ mod tests {
                 continue;
             }
 
-            let expected_error = format!("{}", d.result.as_ref().unwrap_err());
-            let actual_error = format!("{}", result.unwrap_err());
+            let expected_error = format!("{d.result.as_ref().unwrap_err()}");
+            let actual_error = format!("{result.unwrap_err()}");
             assert!(actual_error == expected_error, msg);
         }
     }

docs/assets/favicon.svg

@@ -0,0 +1,9 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 32 32">
+  <!-- Dark background matching the site -->
+  <rect width="32" height="32" rx="4" fill="#1a1a2e"/>
+  
+  <!-- Kata logo scaled and centered -->
+  <g transform="translate(-27, -2) scale(0.75)">
+    <path d="M70.925 25.22L58.572 37.523 46.27 25.22l2.192-2.192 10.11 10.11 10.11-10.11zm-6.575-.2l-3.188-3.188 3.188-3.188 3.188 3.188zm-4.93-2.54l3.736 3.736-3.736 3.736zm-1.694 7.422l-8.07-8.07 8.07-8.07zm1.694-16.14l3.686 3.686-3.686 3.686zm-13.15 4.682L58.572 6.143l12.353 12.303-2.192 2.192-10.16-10.11-10.11 10.11zm26.997 0L58.572 3.752 43.878 18.446l3.387 3.387-3.387 3.387 14.694 14.694L73.266 25.22l-3.337-3.387z" fill="#f15b3e"/>
+  </g>
+</svg>

docs/design/architecture/storage.md

@@ -51,6 +51,7 @@ containers started after the VM has been launched.
 Users can check to see if the container uses the `devicemapper` block
 device as its rootfs by calling `mount(8)` within the container. If
 the `devicemapper` block device is used, the root filesystem (`/`)
-will be mounted from `/dev/vda`. Users can disable direct mounting of
-the underlying block device through the runtime
-[configuration](README.md#configuration).
+will be mounted from `/dev/vda`. Users can enable direct mounting of
+the underlying block device by setting the runtime
+[configuration](README.md#configuration) flag `disable_block_device_use` to
+`false`.

docs/how-to/how-to-pull-images-in-guest-with-kata.md

@@ -256,7 +256,7 @@ spec:
             values:
             - NODE_NAME
   volumes:
-    - name: trusted-storage
+    - name: trusted-image-storage
       persistentVolumeClaim:
         claimName: trusted-pvc
   containers:

docs/how-to/how-to-set-sandbox-config-kata.md

@@ -50,7 +50,7 @@ There are several kinds of Kata configurations and they are listed below.
 | `io.katacontainers.config.hypervisor.default_max_vcpus` | uint32| the maximum number of vCPUs allocated for the VM by the hypervisor |
 | `io.katacontainers.config.hypervisor.default_memory` | uint32| the memory assigned for a VM by the hypervisor in `MiB` |
 | `io.katacontainers.config.hypervisor.default_vcpus` | float32| the default vCPUs assigned for a VM by the hypervisor |
-| `io.katacontainers.config.hypervisor.disable_block_device_use` | `boolean` | disallow a block device from being used |
+| `io.katacontainers.config.hypervisor.disable_block_device_use` | `boolean` | disable hotplugging host block devices to guest VMs for container rootfs |
 | `io.katacontainers.config.hypervisor.disable_image_nvdimm` | `boolean` | specify if a `nvdimm` device should be used as rootfs for the guest (QEMU) |
 | `io.katacontainers.config.hypervisor.disable_vhost_net` | `boolean` | specify if `vhost-net` is not available on the host |
 | `io.katacontainers.config.hypervisor.enable_hugepages` | `boolean` | if the memory should be `pre-allocated` from huge pages |
@@ -97,6 +97,8 @@ There are several kinds of Kata configurations and they are listed below.
 | `io.katacontainers.config.hypervisor.use_legacy_serial` | `boolean` | uses legacy serial device for guest's console (QEMU) |
 | `io.katacontainers.config.hypervisor.default_gpus` | uint32 | the minimum number of GPUs required for the VM. Only used by remote hypervisor to help with instance selection |
 | `io.katacontainers.config.hypervisor.default_gpu_model` | string | the GPU model required for the VM. Only used by remote hypervisor to help with instance selection |
+| `io.katacontainers.config.hypervisor.block_device_num_queues` | `usize` | The number of queues to use for block devices (runtime-rs only) |
+| `io.katacontainers.config.hypervisor.block_device_queue_size` | uint32 | The size of the of the queue to use for block devices (runtime-rs only) |
 
 ## Container Options
 | Key | Value Type | Comments |

docs/install/minikube-installation-guide.md

@@ -103,48 +103,8 @@ $ minikube ssh "grep -c -E 'vmx|svm' /proc/cpuinfo"
 
 ## Installing Kata Containers
 
-You can now install the Kata Containers runtime components. You will need a local copy of some Kata
-Containers components to help with this, and then use `kubectl` on the host (that Minikube has already
-configured for you) to deploy them:
-
-```sh
-$ git clone https://github.com/kata-containers/kata-containers.git
-$ cd kata-containers/tools/packaging/kata-deploy
-$ kubectl apply -f kata-rbac/base/kata-rbac.yaml
-$ kubectl apply -f kata-deploy/base/kata-deploy.yaml
-```
-
-This installs the Kata Containers components into `/opt/kata` inside the Minikube node. It can take
-a few minutes for the operation to complete. You can check the installation has worked by checking
-the status of the `kata-deploy` pod, which will be executing
-[this script](../../tools/packaging/kata-deploy/scripts/kata-deploy.sh),
-and will be executing a `sleep infinity` once it has successfully completed its work.
-You can accomplish this by running the following:
-
-```sh
-$ podname=$(kubectl -n kube-system get pods -o=name | grep -F kata-deploy | sed 's?pod/??')
-$ kubectl -n kube-system exec ${podname} -- ps -ef | grep -F infinity
-```
-
-> *NOTE:* This check only works for single node clusters, which is the default for Minikube.
-> For multi-node clusters, the check would need to be adapted to check `kata-deploy` had
-> completed on all nodes.
-
-## Enabling Kata Containers
-
-Now you have installed the Kata Containers components in the Minikube node. Next, you need to configure
-Kubernetes `RuntimeClass` to know when to use Kata Containers to run a pod.
-
-### Register the runtime
-
-Now register the `kata qemu` runtime with that class. This should result in no errors:
-
-```sh
-$ cd kata-containers/tools/packaging/kata-deploy/runtimeclasses
-$ kubectl apply -f kata-runtimeClasses.yaml
-```
-
-The Kata Containers installation process should be complete and enabled in the Minikube cluster.
+You can now install the Kata Containers runtime components
+[following the official instructions](../../tools/packaging/kata-deploy/helm-chart).
 
 ## Testing Kata Containers
 

docs/presentations/unit-testing/kata-containers-unit-testing.md

@@ -48,7 +48,7 @@ $ make test
 - Run a test in the current package in verbose mode:
 
   ```bash
-  # Example 
+  # Example
   $ test="config::tests::test_get_log_level"
 
   $ cargo test "$test" -vv -- --exact --nocapture
@@ -223,7 +223,7 @@ What's wrong with this function?
 
 ```rust
 fn foo(config: &Config, path_prefix: String, container_id: String, pid: String) -> Result<()> {
-    let mut full_path = format!("{}/{}", path_prefix, container_id);
+    let mut full_path = format!("{path_prefix}/{container_id}");
 
     let _ = remove_recursively(&mut full_path);
 

docs/use-cases/GPU-passthrough-and-Kata.md

@@ -3,4 +3,4 @@
 Kata Containers supports passing certain GPUs from the host into the container. Select the GPU vendor for detailed information:
 
 - [Intel Discrete GPUs](Intel-Discrete-GPU-passthrough-and-Kata.md)/[Intel Integrated GPUs](Intel-GPU-passthrough-and-Kata.md)
-- [NVIDIA](NVIDIA-GPU-passthrough-and-Kata.md)
+- [NVIDIA GPUs](NVIDIA-GPU-passthrough-and-Kata.md) and [Enabling NVIDIA GPU workloads using GPU passthrough with Kata Containers](NVIDIA-GPU-passthrough-and-Kata-QEMU.md)

docs/use-cases/NVIDIA-GPU-passthrough-and-Kata-QEMU.md

@@ -0,0 +1,569 @@
+# Enabling NVIDIA GPU workloads using GPU passthrough with Kata Containers
+
+This page provides:
+1. A description of the components involved when running GPU workloads with
+   Kata Containers using the NVIDIA TEE and non-TEE GPU runtime classes.
+1. An explanation of the orchestration flow on a Kubernetes node for this
+   scenario.
+1. A deployment guide enabling to utilize these runtime classes.
+
+The goal is to educate readers familiar with Kubernetes and Kata Containers
+on NVIDIA's reference implementation which is reflected in Kata CI's build
+and test framework. With this, we aim to enable readers to leverage this
+stack, or to use the principles behind this stack in order to run GPU
+workloads on their variant of the Kata Containers stack.
+
+We assume the reader is familiar with Kubernetes, Kata Containers, and
+Confidential Containers.
+
+> **Note:**
+>
+> The current supported mode for enabling GPU workloads in the TEE scenario
+> is single GPU passthrough (one GPU per pod) on AMD64 platforms (AMD SEV-SNP
+> being the only supported TEE scenario so far with support for Intel TDX being
+> on the way).
+
+## Component Overview
+
+Before providing deployment guidance, we describe the components involved to
+support running GPU workloads. We start from a top to bottom perspective
+from the NVIDIA GPU operator via the Kata runtime to the components within
+the NVIDIA GPU Utility Virtual Machine (UVM) root filesystem.
+
+### NVIDIA GPU Operator
+
+A central component is the
+[NVIDIA GPU operator](https://github.com/NVIDIA/gpu-operator) which can be
+deployed onto your cluster as a helm chart. Installing the GPU operator
+delivers various operands on your nodes in the form of Kubernetes DaemonSets.
+These operands are vital to support the flow of orchestrating pod manifests
+using NVIDIA GPU runtime classes with GPU passthrough on your nodes. Without
+getting into the details, the most important operands and their
+responsibilities are:
+
+- **nvidia-vfio-manager:** Binding discovered NVIDIA GPUs to the `vfio-pci`
+  driver for VFIO passthrough.
+- **nvidia-cc-manager:** Transitioning GPUs into confidential computing (CC)
+  and non-CC mode (see the
+  [NVIDIA/k8s-cc-manager](https://github.com/NVIDIA/k8s-cc-manager)
+  repository).
+- **nvidia-kata-manager:** Creating host-side CDI specifications for GPU
+  passthrough, resulting in the file `/var/run/cdi/nvidia.yaml`, containing
+  `kind: nvidia.com/pgpu` (see the
+  [NVIDIA/k8s-kata-manager](https://github.com/NVIDIA/k8s-kata-manager)
+  repository).
+- **nvidia-sandbox-device-plugin** (see the
+  [NVIDIA/sandbox-device-plugin](https://github.com/NVIDIA/sandbox-device-plugin)
+  repository):
+  - Allocating GPUs during pod deployment.
+  - Discovering NVIDIA GPUs, their capabilities, and advertising these to
+    the Kubernetes control plane (allocatable resources as type
+    `nvidia.com/pgpu` resources will appear for the node and GPU Device IDs
+    will be registered with Kubelet). These GPUs can thus be allocated as
+    container resources in your pod manifests. See below GPU operator
+    deployment instructions for the use of the key `pgpu`, controlled via a
+    variable.
+
+To summarize, the GPU operator manages the GPUs on each node, allowing for
+simple orchestration of pod manifests using Kata Containers. Once the cluster
+with GPU operator and Kata bits is up and running, the end user can schedule
+Kata NVIDIA GPU workloads, using resource limits and the
+`kata-qemu-nvidia-gpu` or `kata-qemu-nvidia-gpu-snp` runtime classes, for
+example:
+
+```yaml
+apiVersion: v1
+kind: Pod
+...
+spec:
+  ...
+  runtimeClassName: kata-qemu-nvidia-gpu-snp
+  ...
+    resources:
+      limits:
+        "nvidia.com/pgpu": 1
+...
+```
+
+When this happens, the Kubelet calls into the sandbox device plugin to
+allocate a GPU. The sandbox device plugin returns `DeviceSpec` entries to the
+Kubelet for the allocated GPU. The Kubelet uses internal device IDs for
+tracking of allocated GPUs and includes the device specifications in the CRI
+request when scheduling the pod through containerd. Containerd processes the
+device specifications and includes the device configuration in the OCI
+runtime spec used to invoke the Kata runtime during the create container
+request.
+
+### Kata runtime
+
+The Kata runtime for the NVIDIA GPU handlers is configured to cold-plug VFIO
+devices (`cold_plug_vfio` is set to `root-port` while
+`hot_plug_vfio` is set to `no-port`). Cold-plug is by design the only
+supported mode for NVIDIA GPU passthrough of the NVIDIA reference stack.
+
+With cold-plug, the Kata runtime attaches the GPU at VM launch time, when
+creating the pod sandbox. This happens *before* the create container request,
+i.e., before the Kata runtime receives the OCI spec including device
+configurations from containerd. Thus, a mechanism to acquire the device
+information is required. This is done by the runtime calling the
+`coldPlugDevices()` function during sandbox creation. In this function,
+the runtime queries Kubelet's Pod Resources API to discover allocated GPU
+device IDs (e.g., `nvidia.com/pgpu = [vfio0]`). The runtime formats these as
+CDI device identifiers and injects them into the OCI spec using
+`config.InjectCDIDevices()`. The runtime then consults the host CDI
+specifications and determines the device path the GPU is backed by
+(e.g., `/dev/vfio/devices/vfio0`). Finally, the runtime resolves the device's
+PCI BDF (e.g., `0000:21:00`) and cold-plugs the GPU by launching QEMU with
+relevant parameters for device passthrough (e.g.,
+`-device vfio-pci,host=0000:21:00.0,x-pci-vendor-id=0x10de,x-pci-device-id=0x2321,bus=rp0,iommufd=iommufdvfio-faf829f2ea7aec330`).
+
+The runtime also creates *inner runtime* CDI annotations
+which map host VFIO devices to guest GPU devices. These are annotations
+intended for the kata-agent, here referred to as the inner runtime (inside the
+UVM), to properly handle GPU passthrough into containers. These annotations
+serve as metadata providing the kata-agent with the information needed to
+attach the passthrough devices to the correct container.
+The annotations are key-value pairs consisting of `cdi.k8s.io/vfio<num>` keys
+(derived from the host VFIO device path, e.g., `/dev/vfio/devices/vfio1`) and
+`nvidia.com/gpu=<index>` values (referencing the corresponding device in the
+guest CDI spec). These annotations are injected by the runtime during container
+creation via the `annotateContainerWithVFIOMetadata` function (see
+`container.go`).
+
+We continue describing the orchestration flow inside the UVM in the next
+section.
+
+### Kata NVIDIA GPU UVM
+
+#### UVM composition
+
+To better understand the orchestration flow inside the NVIDIA GPU UVM, we
+first look at the components its root filesystem contains. Should you decide
+to use your own root filesystem to enable NVIDIA GPU scenarios, this should
+give you a good idea on what ingredients you need.
+
+From a file system perspective, the UVM is composed of two files: a standard
+Kata kernel image and the NVIDIA GPU rootfs in initrd or disk image format.
+These two files are being utilized for the QEMU launch command when the UVM
+is created.
+
+The two most important pieces in Kata Container's build recipes for the
+NVIDIA GPU root filesystem are the `nvidia_chroot.sh` and `nvidia_rootfs.sh`
+files. The build follows a two-stage process. In the first stage, a
+full-fledged Ubuntu-based root filesystem is composed within a chroot
+environment. In this stage, NVIDIA kernel modules are built and signed
+against the current Kata kernel and relevant NVIDIA packages are installed.
+In the second stage, a chiseled build is performed: Only relevant contents
+from the first stage are copied and compressed into a new distro-less root
+filesystem folder. Kata's build infrastructure then turns this root
+filesystem into the NVIDIA initrd and image files.
+
+The resulting root filesystem contains the following software components:
+
+- NVRC - the
+  [NVIDIA Runtime Container init system](https://github.com/NVIDIA/nvrc/tree/main)
+- NVIDIA drivers (kernel modules)
+- NVIDIA user space driver libraries
+- NVIDIA user space tools
+- kata-agent
+- confidential computing guest components: the attestation agent,
+  confidential data hub and api-server-rest binaries
+- CRI-O pause container (for the guest image-pull method)
+- BusyBox utilities (provides a base set of libraries and binaries, and a
+  linker)
+- some supporting files, such as file containing a list of supported GPU
+  device IDs which NVRC reads
+
+#### UVM orchestration flow
+
+When the Kata runtime asks QEMU to launch the VM, the UVM's Linux kernel
+boots and mounts the root filesystem. After this, NVRC starts as the initial
+process.
+
+NVRC scans for NVIDIA GPUs on the PCI bus, loads the
+NVIDIA kernel modules, waits for driver initialization, creates the device nodes,
+and initializes the GPU hardware (using the `nvidia-smi` binary). NVRC also
+creates the guest-side CDI specification file (using the
+`nvidia-ctk cdi generate` command). This file specifies devices of
+`kind: nvidia.com/gpu`, i.e., GPUs appearing to be physical GPUs on regular
+bare metal systems. The guest CDI specification also contains `containerEdits`
+for each device, specifying device nodes (e.g., `/dev/nvidia0`,
+`/dev/nvidiactl`), library mounts, and environment variables to be mounted
+into the container which receives the passthrough GPU.
+
+Then, NVRC forks the Kata agent while continuing to run as the
+init system. This allows NVRC to handle ongoing GPU management tasks
+while kata-agent focuses on container lifecycle management. See the
+[NVRC sources](https://github.com/NVIDIA/nvrc/blob/main/src/main.rs) for an
+overview on the steps carried out by NVRC.
+
+When the Kata runtime sends the create container request, the Kata agent
+parses the inner runtime CDI annotation. For example, for the inner runtime
+annotation `"cdi.k8s.io/vfio1": "nvidia.com/gpu=0"`, the agent looks up device
+`0` in the guest CDI specification with `kind: nvidia.com/gpu`.
+
+The Kata agent also reads the guest CDI specification's `containerEdits`
+section and injects relevant contents into the OCI spec of the respective
+container. The kata agent then creates and starts a `rustjail` container
+based on the final OCI spec. The container now has relevant device nodes,
+binaries and low-level libraries available, and can start a user application
+linked against the CUDA runtime API (e.g., `libcudart.so` and other
+libraries). When used, the CUDA runtime API in turn calls the CUDA driver
+API and kernel drivers, interacting with the pass-through GPU device.
+
+An additional step is exercised in our CI samples: when using images from an
+authenticated registry, the guest-pull mechanism triggers attestation using
+trustee's Key Broker Service (KBS) for secure release of the NGC API
+authentication key used to access the NVCR container registry. As part of
+this, the attestation agent exercises composite attestation and transitions
+the GPU into `Ready` state (without this, the GPU has to explicitly be
+transitioned into `Ready` state by passing the `nvrc.smi.srs=1` kernel
+parameter via the shim config, causing NVRC to transition the GPU into the
+`Ready` state).
+
+## Deployment Guidance
+
+This guidance assumes you use bare-metal machines with proper support for
+Kata's non-TEE and TEE GPU workload deployment scenarios for your Kubernetes
+nodes. We provide guidance based on the upstream Kata CI procedures for the
+NVIDIA GPU CI validation jobs. Note that, this setup:
+
+- uses the guest image pull method to pull container image layers
+- uses the genpolicy tool to attach Kata agent security policies to the pod
+  manifest
+- has dedicated (composite) attestation tests, a CUDA vectorAdd test, and a
+  NIM/RA test sample with secure API key release
+
+A similar deployment guide and scenario description can be found in NVIDIA resources
+under
+[Early Access: NVIDIA GPU Operator with Confidential Containers based on Kata](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/confidential-containers.html).
+
+### Requirements
+
+The requirements for the TEE scenario are:
+
+- Ubuntu 25.10 as host OS
+- CPU with AMD SEV-SNP support with proper BIOS/UEFI version and settings
+- CC-capable Hopper/Blackwell GPU with proper VBIOS version.
+
+BIOS and VBIOS configuration is out of scope for this guide. Other resources,
+such as the documentation found on the
+[NVIDIA Trusted Computing Solutions](https://docs.nvidia.com/nvtrust/index.html)
+page and the above linked NVIDIA documentation, provide guidance on
+selecting proper hardware and on properly configuring its firmware and OS.
+
+### Installation
+
+#### Containerd and Kubernetes
+
+First, set up your Kubernetes cluster. For instance, in Kata CI, our NVIDIA
+jobs use a single-node vanilla Kubernetes cluster with a 2.x containerd
+version and Kata's current supported Kubernetes version. We set this cluster
+up using the `deploy_k8s` function from `tests/integration/kubernetes/gha-run.sh`
+as follows:
+
+```bash
+$ export KUBERNETES="vanilla"
+$ export CONTAINER_ENGINE="containerd"
+$ export CONTAINER_ENGINE_VERSION="v2.1"
+$ source tests/gha-run-k8s-common.sh
+$ deploy_k8s
+```
+
+> **Note:**
+>
+> We recommend to configure your Kubelet with a higher
+> `runtimeRequestTimeout` timeout value than the two minute default timeout.
+> Using the guest-pull mechanism, pulling large images may take a significant
+> amount of time and may delay container start, possibly leading your Kubelet
+> to de-allocate your pod before it transitions from the *container created*
+> to the *container running* state.
+
+> **Note:**
+>
+> The NVIDIA GPU runtime classes use VFIO cold-plug which, as
+> described above, requires the Kata runtime to query Kubelet's Pod Resources
+> API to discover allocated GPU devices during sandbox creation. For
+> Kubernetes versions **older than 1.34**, you must explicitly enable the
+> `KubeletPodResourcesGet` feature gate in your Kubelet configuration. For
+> Kubernetes 1.34 and later, this feature is enabled by default.
+
+#### GPU Operator
+
+Assuming you have the helm tools installed, deploy the latest version of the
+GPU Operator as a helm chart (minimum version: `v25.10.0`):
+
+```bash
+$ helm repo add nvidia https://helm.ngc.nvidia.com/nvidia && helm repo update
+$ helm install --wait --generate-name \
+    -n gpu-operator --create-namespace \
+    nvidia/gpu-operator \
+    --set sandboxWorkloads.enabled=true \
+    --set sandboxWorkloads.defaultWorkload=vm-passthrough \
+    --set kataManager.enabled=true \
+    --set kataManager.config.runtimeClasses=null \
+    --set kataManager.repository=nvcr.io/nvidia/cloud-native \
+    --set kataManager.image=k8s-kata-manager \
+    --set kataManager.version=v0.2.4 \
+    --set ccManager.enabled=true \
+    --set ccManager.defaultMode=on \
+    --set ccManager.repository=nvcr.io/nvidia/cloud-native \
+    --set ccManager.image=k8s-cc-manager \
+    --set ccManager.version=v0.2.0 \
+    --set sandboxDevicePlugin.repository=nvcr.io/nvidia/cloud-native \
+    --set sandboxDevicePlugin.image=nvidia-sandbox-device-plugin \
+    --set sandboxDevicePlugin.version=v0.0.1 \
+    --set 'sandboxDevicePlugin.env[0].name=P_GPU_ALIAS' \
+    --set 'sandboxDevicePlugin.env[0].value=pgpu' \
+    --set nfd.enabled=true \
+    --set nfd.nodefeaturerules=true
+```
+
+> **Note:**
+>
+> For heterogeneous clusters with different GPU types, you can omit
+> the `P_GPU_ALIAS` environment variable lines. This will cause the sandbox
+> device plugin to create GPU model-specific resource types (e.g.,
+> `nvidia.com/GH100_H100L_94GB`) instead of the generic `nvidia.com/pgpu`,
+> which in turn can be used by pods through respective resource limits.
+> For simplicity, this guide uses the generic alias.
+
+> **Note:**
+>
+> Using `--set sandboxWorkloads.defaultWorkload=vm-passthrough` causes all
+> your nodes to be labeled for GPU VM passthrough. Remove this parameter if
+> you intend to only use selected nodes for this scenario, and label these
+> nodes by hand, using:
+> `kubectl label node <node-name> nvidia.com/gpu.workload.config=vm-passthrough`.
+
+#### Kata Containers
+
+Install the latest Kata Containers helm chart, similar to
+[existing documentation](https://github.com/kata-containers/kata-containers/blob/main/tools/packaging/kata-deploy/helm-chart/README.md)
+(minimum version: `3.24.0`).
+
+```bash
+$ export VERSION=$(curl -sSL https://api.github.com/repos/kata-containers/kata-containers/releases/latest | jq .tag_name | tr -d '"')
+$ export CHART="oci://ghcr.io/kata-containers/kata-deploy-charts/kata-deploy"
+
+$ helm install kata-deploy \
+    --namespace kata-system \
+    --create-namespace \
+    -f "https://raw.githubusercontent.com/kata-containers/kata-containers/refs/tags/${VERSION}/tools/packaging/kata-deploy/helm-chart/kata-deploy/try-kata-nvidia-gpu.values.yaml" \
+    --set nfd.enabled=false \
+    --set shims.qemu-nvidia-gpu-tdx.enabled=false \
+    --wait --timeout 10m --atomic \
+    "${CHART}" --version "${VERSION}"
+```
+
+#### Trustee's KBS for remote attestation
+
+For our Kata CI runners we use Trustee's KBS for composite attestation for
+secure key release, for instance, for test scenarios which use authenticated
+container images. In such scenarios, the credentials to access the
+authenticated container registry are only released to the confidential guest
+after successful attestation. Please see the section below for more
+information about this.
+
+```bash
+$ export NVIDIA_VERIFIER_MODE="remote"
+$ export KBS_INGRESS="nodeport"
+$ bash tests/integration/kubernetes/gha-run.sh deploy-coco-kbs
+$ bash tests/integration/kubernetes/gha-run.sh install-kbs-client
+```
+
+Please note, that Trustee can also be deployed via any other upstream
+mechanism as documented by the
+[confidential-containers repository](https://github.com/confidential-containers/trustee).
+For our architecture it is important to set up KBS in the remote verifier
+mode which requires entering a licensing agreement with NVIDIA, see the
+[notes in confidential-containers repository](https://github.com/confidential-containers/trustee/blob/main/deps/verifier/src/nvidia/README.md).
+
+### Cluster validation and preparation
+
+If you did not use the `sandboxWorkloads.defaultWorkload=vm-passthrough`
+parameter during GPU operator deployment, label your nodes for GPU VM
+passthrough, for the example of using all nodes for GPU passthrough, run:
+
+```bash
+$ kubectl label nodes --all nvidia.com/gpu.workload.config=vm-passthrough --overwrite
+```
+
+Check if the `nvidia-cc-manager` pod is running if you intend to run GPU TEE
+scenarios. If not, you need to manually label the node as CC capable. Current
+GPU Operator node feature rules do not yet recognize all CC capable GPU PCI
+IDs. Run the following command:
+
+```bash
+$ kubectl label nodes --all nvidia.com/cc.capable=true
+```
+
+After this, assure the `nvidia-cc-manager` pod is running. With the suggested
+parameters for GPU Operator deployment, the `nvidia-cc-manager` will
+automatically transition the GPU into CC mode.
+
+After deployment, you can transition your node(s) to the desired CC state,
+using either the `on` or `off` value, depending on your scenario. For the
+non-CC scenario, transition to the `off` state via:
+`kubectl label nodes --all nvidia.com/cc.mode=off` and wait until all pods
+are back running. When an actual change is exercised, various GPU operator
+operands will be restarted.
+
+Ensure all pods are running:
+
+```bash
+$ kubectl get pods -A
+```
+
+On your node(s), ensure for correct driver binding. Your GPU device should be
+bound to the VFIO driver, i.e., showing `Kernel driver in use: vfio-pci`
+when running:
+
+```bash
+$ lspci -nnk -d 10de:
+```
+
+### Run the CUDA vectorAdd sample
+
+Create the following file:
+
+```yaml
+apiVersion: v1
+kind: Pod
+metadata:
+  name: cuda-vectoradd-kata
+  namespace: default
+  annotations:
+    io.katacontainers.config.hypervisor.kernel_params: "nvrc.smi.srs=1"
+spec:
+  runtimeClassName: ${GPU_RUNTIME_CLASS_NAME}
+  restartPolicy: Never
+  containers:
+  - name: cuda-vectoradd
+    image: "nvcr.io/nvidia/k8s/cuda-sample:vectoradd-cuda12.5.0-ubuntu22.04"
+    resources:
+      limits:
+        nvidia.com/pgpu: "1"
+        memory: 16Gi
+```
+
+Depending on your scenario and on the CC state, export your desired runtime
+class name define the environment variable:
+
+```bash
+$ export GPU_RUNTIME_CLASS_NAME="kata-qemu-nvidia-gpu-snp"
+```
+
+Then, deploy the sample Kubernetes pod manifest and observe the pod logs:
+
+```bash
+$ envsubst < ./cuda-vectoradd-kata.yaml.in | kubectl apply -f -
+$ kubectl wait --for=condition=Ready pod/cuda-vectoradd-kata --timeout=60s
+$ kubectl logs -n default cuda-vectoradd-kata
+```
+
+Expect the following output:
+
+```
+[Vector addition of 50000 elements]
+Copy input data from the host memory to the CUDA device
+CUDA kernel launch with 196 blocks of 256 threads
+Copy output data from the CUDA device to the host memory
+Test PASSED
+Done
+```
+
+To stop the pod, run: `kubectl delete pod cuda-vectoradd-kata`.
+
+### Next steps
+
+#### Transition between CC and non-CC mode
+
+Use the previously described node labeling approach to transition between
+the CC and non-CC mode. In case of the non-CC mode, you can use the
+`kata-qemu-nvidia-gpu` value for the `GPU_RUNTIME_CLASS_NAME` runtime class
+variable in the above CUDA vectorAdd sample. The `kata-qemu-nvidia-gpu-snp`
+runtime class will **NOT** work in this mode - and vice versa.
+
+#### Run Kata CI tests locally
+
+Upstream Kata CI runs the CUDA vectorAdd test, a composite attestation test,
+and a basic NIM/RAG deployment. Running CI tests for the TEE GPU scenario
+requires KBS to be deployed (except for the CUDA vectorAdd test). The best
+place to get started running these tests locally is to look into our
+[NVIDIA CI workflow manifest](https://github.com/kata-containers/kata-containers/blob/main/.github/workflows/run-k8s-tests-on-nvidia-gpu.yaml)
+and into the underling
+[run_kubernetes_nv_tests.sh](https://github.com/kata-containers/kata-containers/blob/main/tests/integration/kubernetes/run_kubernetes_nv_tests.sh)
+script. For example, to run the CUDA vectorAdd scenario against the TEE GPU
+runtime class use the following commands:
+
+```bash
+# create the kata runtime class the test framework uses
+$ export KATA_HYPERVISOR=qemu-nvidia-gpu-snp
+$ kubectl delete runtimeclass kata --ignore-not-found
+$ kubectl get runtimeclass "kata-${KATA_HYPERVISOR}" -o json | \
+    jq '.metadata.name = "kata" | del(.metadata.uid, .metadata.resourceVersion, .metadata.creationTimestamp)' | \
+    kubectl apply -f -
+$ cd tests/integration/kubernetes
+$ K8S_TEST_NV="k8s-nvidia-cuda.bats" ./gha-run.sh run-nv-tests
+```
+
+> **Note:**
+>
+> The other scenarios require an NGC API key to run, i.e., to export the
+> `NGC_API_KEY` variable with a valid NGC API key.
+
+#### Deploy pods using attestation
+
+Attestation is a fundamental piece of the confidential containers solution.
+In our upstream CI we use attestation at the example of leveraging the
+authenticated container image pull mechanism where container images reside
+in the authenticated NVCR registry (`k8s-nvidia-nim.bats`), and for
+requesting secrets from KBS (`k8s-confidential-attestation.bats`). KBS will
+release the image pull secret to a confidential guest. To get the
+authentication credentials from inside the guest, KBS must already be
+deployed and configured. In our CI samples, we configure KBS with the guest
+image pull secret, a resource policy, and launch the pod with certain kernel
+command line parameters:
+`"agent.image_registry_auth=kbs:///default/credentials/nvcr agent.aa_kbc_params=cc_kbc::${CC_KBS_ADDR}"`.
+
+The `agent.aa_kbc_params` option is a general configuration for attestation.
+For your use case, you need to set the IP address and port under which KBS
+is reachable through the `CC_KBS_ADDR` variable (see our CI sample). This
+tells the guest how to reach KBS. Something like this must be set whenever
+attestation is used, but on its own this parameter does not trigger
+attestation. The `agent.image_registry_auth` option tells the guest to ask
+for a resource from KBS and use it as the authentication configuration. When
+this is set, the guest will request this resource at boot (and trigger
+attestation) regardless of which image is being pulled.
+
+To deploy your own pods using authenticated container images, or secure key
+release for attestation, follow steps similar to our mentioned CI samples.
+
+#### Deploy pods with Kata agent security policies
+
+With GPU passthrough being supported by the
+[genpolicy tool](https://github.com/kata-containers/kata-containers/tree/main/src/tools/genpolicy),
+you can use the tool to create a Kata agent security policy. Our CI deploys
+all sample pod manifests with a Kata agent security policy.
+
+#### Deploy pods using your own containers and manifests
+
+You can author pod manifests leveraging your own containers, for instance,
+containers built using the CUDA container toolkit. We recommend to start
+with a CUDA base container.
+
+The GPU is transitioned into the `Ready` state via attestation, for instance,
+when pulling authenticated images. If your deployment scenario does not use
+attestation, please refer back to the CUDA vectorAdd pod manifest. In this
+manifest, we ensure that NVRC sets the GPU to `Ready` state by adding the
+following annotation in the manifest:
+`io.katacontainers.config.hypervisor.kernel_params: "nvrc.smi.srs=1"`
+
+> **Notes:**
+>
+> - musl-based container images (e.g., using Alpine), or distro-less
+>   containers are not supported.
+> - for the TEE scenario, only single-GPU passthrough per pod is supported,
+>   so your pod resource limit must be: `nvidia.com/pgpu: "1"` (on a system
+>   with multiple GPUs, you can thus pass through one GPU per pod).

docs/use-cases/NVIDIA-GPU-passthrough-and-Kata.md

@@ -1,10 +1,25 @@
 # Using NVIDIA GPU device with Kata Containers
 
+This page gives an overview on the different modes in which GPUs can be passed
+to a Kata Containers container, provides host system requirements, explains how
+Kata Containers guest components can be built to support the NVIDIA GPU
+scenario, and gives practical usage examples using `ctr`.
+
+Please see the guide
+[Enabling NVIDIA GPU workloads using GPU passthrough with Kata Containers](NVIDIA-GPU-passthrough-and-Kata-QEMU.md)
+for a documentation of an end-to-end reference implementation of a Kata
+Containers stack for GPU passthrough using QEMU, the go-based Kata Runtime,
+and an NVIDIA-specific root filesystem. This reference implementation is built
+and validated in Kata's CI, and it can be used to test GPU workloads with Kata
+components and Kubernetes out of the box.
+
+## Comparison between Passthrough and vGPU Modes
+
 An NVIDIA GPU device can be passed to a Kata Containers container using GPU
-passthrough (NVIDIA GPU pass-through mode) as well as GPU mediated passthrough
+passthrough (NVIDIA GPU passthrough mode) as well as GPU mediated passthrough
 (NVIDIA `vGPU` mode).
 
-NVIDIA GPU pass-through mode, an entire physical GPU is directly assigned to one
+NVIDIA GPU passthrough mode, an entire physical GPU is directly assigned to one
 VM, bypassing the NVIDIA Virtual GPU Manager. In this mode of operation, the GPU
 is accessed exclusively by the NVIDIA driver running in the VM to which it is
 assigned. The GPU is not shared among VMs.
@@ -20,18 +35,20 @@ with [MIG-slices](https://docs.nvidia.com/datacenter/tesla/mig-user-guide/).
 
 | Technology | Description | Behavior | Detail |
 | --- | --- | --- | --- |
-| NVIDIA GPU pass-through mode | GPU passthrough | Physical GPU assigned to a single VM | Direct GPU assignment to VM without limitation |
+| NVIDIA GPU passthrough mode | GPU passthrough | Physical GPU assigned to a single VM | Direct GPU assignment to VM without limitation |
 | NVIDIA vGPU time-sliced | GPU time-sliced | Physical GPU time-sliced for multiple VMs | Mediated passthrough |
 | NVIDIA vGPU MIG-backed | GPU with MIG-slices | Physical GPU MIG-sliced for multiple VMs | Mediated passthrough |
 
-## Hardware Requirements
+## Host Requirements
 
-NVIDIA GPUs Recommended for Virtualization:
+### Hardware
+
+NVIDIA GPUs recommended for virtualization:
 
 - NVIDIA Tesla (T4, M10, P6, V100 or newer)
 - NVIDIA Quadro RTX 6000/8000
 
-## Host BIOS Requirements
+### Firmware
 
 Some hardware requires a larger PCI BARs window, for example, NVIDIA Tesla P100,
 K40m
@@ -55,9 +72,7 @@ Some hardware vendors use a different name in BIOS, such as:
 If one is using a GPU based on the Ampere architecture and later additionally
 SR-IOV needs to be enabled for the `vGPU` use-case.
 
-The following steps outline the workflow for using an NVIDIA GPU with Kata.
-
-## Host Kernel Requirements
+### Kernel
 
 The following configurations need to be enabled on your host kernel:
 
@@ -70,7 +85,13 @@ The following configurations need to be enabled on your host kernel:
 Your host kernel needs to be booted with `intel_iommu=on` on the kernel command
 line.
 
-## Install and configure Kata Containers
+## Build the Kata Components
+
+This section explains how to build an environment with Kata Containers bits
+supporting the GPU scenario. We first deploy and configure the regular Kata
+components, then describe how to build the guest kernel and root filesystem.
+
+### Install and configure Kata Containers
 
 To use non-large BARs devices (for example, NVIDIA Tesla T4), you need Kata
 version 1.3.0 or above. Follow the [Kata Containers setup
@@ -101,7 +122,7 @@ hotplug_vfio_on_root_bus = true
 pcie_root_port = 1
 ```
 
-## Build Kata Containers kernel with GPU support
+### Build guest kernel with GPU support
 
 The default guest kernel installed with Kata Containers does not provide GPU
 support. To use an NVIDIA GPU with Kata Containers, you need to build a kernel
@@ -160,11 +181,11 @@ code, using `Dragonball VMM` for NVIDIA GPU `hot-plug/hot-unplug` requires apply
 addition to the above kernel configuration items. Follow these steps to build for NVIDIA GPU `hot-[un]plug`
 for `Dragonball`:
 
-```sh 
-# Prepare .config to support both upcall and nvidia gpu 
+```sh
+# Prepare .config to support both upcall and nvidia gpu
 $ ./build-kernel.sh -v 5.10.25 -e -t dragonball -g nvidia -f setup
 
-# Build guest kernel to support both upcall and nvidia gpu 
+# Build guest kernel to support both upcall and nvidia gpu
 $ ./build-kernel.sh -v 5.10.25 -e -t dragonball -g nvidia build
 
 # Install guest kernel to support both upcall and nvidia gpu
@@ -196,303 +217,7 @@ Before using the new guest kernel, please update the `kernel` parameters in
 kernel = "/usr/share/kata-containers/vmlinuz-nvidia-gpu.container"
 ```
 
-## NVIDIA GPU pass-through mode with Kata Containers
-
-Use the following steps to pass an NVIDIA GPU device in pass-through mode with Kata:
-
-1. Find the Bus-Device-Function (BDF) for the GPU device on the host:
-
-   ```sh
-   $ sudo lspci -nn -D | grep -i nvidia
-   0000:d0:00.0 3D controller [0302]: NVIDIA Corporation Device [10de:20b9] (rev a1)
-   ```
-
-   > PCI address `0000:d0:00.0` is assigned to the hardware GPU device.
-   > `10de:20b9` is the device ID of the hardware GPU device.
-
-2. Find the IOMMU group for the GPU device:
-
-   ```sh
-   $ BDF="0000:d0:00.0"
-   $ readlink -e /sys/bus/pci/devices/$BDF/iommu_group
-   ```
-
-   The previous output shows that the GPU belongs to IOMMU group 192. The next
-   step is to bind the GPU to the VFIO-PCI driver.
-
-   ```sh
-   $ BDF="0000:d0:00.0"
-   $ DEV="/sys/bus/pci/devices/$BDF"
-   $ echo "vfio-pci" > $DEV/driver_override
-   $ echo $BDF > $DEV/driver/unbind
-   $ echo $BDF > /sys/bus/pci/drivers_probe
-   # To return the device to the standard driver, we simply clear the
-   # driver_override and reprobe the device, ex:
-   $ echo > $DEV/preferred_driver
-   $ echo $BDF > $DEV/driver/unbind
-   $ echo $BDF > /sys/bus/pci/drivers_probe
-   ```
-
-3. Check the IOMMU group number under `/dev/vfio`:
-
-   ```sh
-   $ ls -l /dev/vfio
-   total 0
-   crw------- 1 zvonkok zvonkok 243,   0 Mar 18 03:06 192
-   crw-rw-rw- 1 root    root     10, 196 Mar 18 02:27 vfio
-   ```
-
-4. Start a Kata container with the GPU device:
-
-   ```sh
-   # You may need to `modprobe vhost-vsock` if you get
-   # host system doesn't support vsock: stat /dev/vhost-vsock
-   $ sudo ctr --debug run --runtime "io.containerd.kata.v2"  --device /dev/vfio/192  --rm -t  "docker.io/library/archlinux:latest" arch uname -r
-   ```
-
-5. Run `lspci` within the container to verify the GPU device is seen in the list
-   of the PCI devices. Note the vendor-device id of the GPU (`10de:20b9`) in the `lspci` output.
-
-   ```sh
-   $ sudo ctr --debug run --runtime "io.containerd.kata.v2"  --device /dev/vfio/192  --rm -t  "docker.io/library/archlinux:latest" arch sh -c "lspci -nn | grep '10de:20b9'"
-   ```
-
-6. Additionally, you can check the PCI BARs space of ​​the NVIDIA GPU device in the container:
-
-   ```sh
-   $ sudo ctr --debug run --runtime "io.containerd.kata.v2"  --device /dev/vfio/192  --rm -t  "docker.io/library/archlinux:latest" arch sh -c "lspci -s 02:00.0 -vv | grep Region"
-   ```
-
-   > **Note**: If you see a message similar to the above, the BAR space of the NVIDIA
-   > GPU has been successfully allocated.
-
-## NVIDIA vGPU mode with Kata Containers
-
-NVIDIA vGPU is a licensed product on all supported GPU boards. A software license
-is required to enable all vGPU features within the guest VM. NVIDIA vGPU manager
-needs to be installed on the host to configure GPUs in vGPU mode. See [NVIDIA Virtual GPU Software Documentation v14.0 through 14.1](https://docs.nvidia.com/grid/14.0/) for more details.
-
-### NVIDIA vGPU time-sliced
-
-In the time-sliced mode, the GPU is not partitioned and the workload uses the
-whole GPU and shares access to the GPU engines. Processes are scheduled in
-series. The best effort scheduler is the default one and can be exchanged by
-other scheduling policies see the documentation above how to do that.
-
-Beware if you had `MIG` enabled before to disable `MIG` on the GPU if you want
-to use `time-sliced` `vGPU`.
-
-```sh
-$ sudo nvidia-smi -mig 0
-```
-
-Enable the virtual functions for the physical GPU in the `sysfs` file system.
-
-```sh
-$ sudo /usr/lib/nvidia/sriov-manage -e 0000:41:00.0
-```
-
-Get the `BDF` of the available virtual function on the GPU, and choose one for the
-following steps.
-
-```sh
-$ cd /sys/bus/pci/devices/0000:41:00.0/
-$ ls -l |  grep virtfn
-```
-
-#### List all available vGPU instances
-
-The following shell snippet will walk the `sysfs` and only print instances
-that are available, that can be created.
-
-```sh
-# The 00.0 is often the PF of the device the VFs will have the funciont in the
-# BDF incremented by some values so e.g. the very first VF is 0000:41:00.4
-
-cd /sys/bus/pci/devices/0000:41:00.0/
-
-for vf in $(ls -d virtfn*)
-do
-        BDF=$(basename $(readlink -f $vf))
-        for md in $(ls -d $vf/mdev_supported_types/*)
-        do
-                AVAIL=$(cat $md/available_instances)
-                NAME=$(cat $md/name)
-                DIR=$(basename $md)
-
-                if [ $AVAIL -gt 0 ]; then
-                        echo "| BDF          | INSTANCES | NAME           | DIR        |"
-                        echo "+--------------+-----------+----------------+------------+"
-                        printf "| %12s |%10d |%15s | %10s |\n\n" "$BDF" "$AVAIL" "$NAME" "$DIR"
-                fi
-
-        done
-done
-```
-
-If there are available instances you get something like this (for the first VF),
-beware that the output is highly dependent on the GPU you have, if there is no
-output check again if `MIG` is really disabled.
-
-```sh
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.4 |         1 |  GRID A100D-4C | nvidia-692 |
-
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.4 |         1 |  GRID A100D-8C | nvidia-693 |
-
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.4 |         1 | GRID A100D-10C | nvidia-694 |
-
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.4 |         1 | GRID A100D-16C | nvidia-695 |
-
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.4 |         1 | GRID A100D-20C | nvidia-696 |
-
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.4 |         1 | GRID A100D-40C | nvidia-697 |
-
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.4 |         1 | GRID A100D-80C | nvidia-698 |
-
-```
-
-Change to the `mdev_supported_types` directory for the virtual function on which
-you want to create the `vGPU`. Taking the first output as an example:
-
-```sh
-$ cd virtfn0/mdev_supported_types/nvidia-692
-$ UUIDGEN=$(uuidgen)
-$ sudo bash -c "echo $UUIDGEN > create"
-```
-
-Confirm that the `vGPU` was created. You should see the `UUID` pointing to a
-subdirectory of the `sysfs` space.
-
-```sh
-$ ls -l /sys/bus/mdev/devices/
-```
-
-Get the `IOMMU` group number and verify there is a `VFIO` device created to use
-with Kata.
-
-```sh
-$ ls -l /sys/bus/mdev/devices/*/
-$ ls -l /dev/vfio
-```
-
-Use the `VFIO` device created in the same way as in the pass-through use-case.
-Beware that the guest needs the NVIDIA guest drivers, so one would need to build
-a new guest `OS` image.
-
-### NVIDIA vGPU MIG-backed
-
-We're not going into detail what `MIG` is but briefly it is a technology to
-partition the hardware into independent instances with guaranteed quality of
-service. For more details see [NVIDIA Multi-Instance GPU User Guide](https://docs.nvidia.com/datacenter/tesla/mig-user-guide/).
-
-First enable `MIG` mode for a GPU, depending on the platform you're running
-a reboot would be necessary. Some platforms support GPU reset.
-
-```sh
-$ sudo nvidia-smi -mig 1
-```
-
-If the platform supports a GPU reset one can run, otherwise you will get a
-warning to reboot the server.
-
-```sh
-$ sudo nvidia-smi --gpu-reset
-```
-
-The driver per default provides a number of profiles that users can opt-in when
-configuring the MIG feature.
-
-```sh
-$ sudo nvidia-smi mig -lgip
-+-----------------------------------------------------------------------------+
-| GPU instance profiles:                                                      |
-| GPU   Name             ID    Instances   Memory     P2P    SM    DEC   ENC  |
-|                              Free/Total   GiB              CE    JPEG  OFA  |
-|=============================================================================|
-|   0  MIG 1g.10gb       19     7/7        9.50       No     14     0     0   |
-|                                                             1     0     0   |
-+-----------------------------------------------------------------------------+
-|   0  MIG 1g.10gb+me    20     1/1        9.50       No     14     1     0   |
-|                                                             1     1     1   |
-+-----------------------------------------------------------------------------+
-|   0  MIG 2g.20gb       14     3/3        19.50      No     28     1     0   |
-|                                                             2     0     0   |
-+-----------------------------------------------------------------------------+
-                              ...
-```
-
-Create the GPU instances that correspond to the `vGPU` types of the `MIG-backed`
-`vGPUs` that you will create [NVIDIA A100 PCIe 80GB Virtual GPU Types](https://docs.nvidia.com/grid/13.0/grid-vgpu-user-guide/index.html#vgpu-types-nvidia-a100-pcie-80gb).
-
-```sh
-# MIG 1g.10gb --> vGPU A100D-1-10C
-$ sudo nvidia-smi mig -cgi 19
-```
-
-List the GPU instances and get the GPU instance id to create the compute
-instance.
-
-```sh
-$ sudo nvidia-smi mig -lgi # list the created GPU instances
-$ sudo nvidia-smi mig -cci -gi 9 # each GPU instance can have several compute
-                                 # instances. Instance -> Workload
-```
-
-Verify that the compute instances were created within the GPU instance
-
-```sh
-$ nvidia-smi
-                              ... snip ...
-+-----------------------------------------------------------------------------+
-| MIG devices:                                                                |
-+------------------+----------------------+-----------+-----------------------+
-| GPU  GI  CI  MIG |         Memory-Usage |        Vol|         Shared        |
-|      ID  ID  Dev |           BAR1-Usage | SM     Unc| CE  ENC  DEC  OFA  JPG|
-|                  |                      |        ECC|                       |
-|==================+======================+===========+=======================|
-|  0    9   0   0  |      0MiB /  9728MiB | 14      0 |  1   0    0    0    0 |
-|                  |      0MiB /  4095MiB |           |                       |
-+------------------+----------------------+-----------+-----------------------+
-                              ... snip ...
-```
-
-We can use the [snippet](#list-all-available-vgpu-instances) from before to list
-the available `vGPU` instances, this time `MIG-backed`.
-
-```sh
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.4 |         1 |GRID A100D-1-10C | nvidia-699 |
-
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:00.5 |         1 |GRID A100D-1-10C | nvidia-699 |
-
-| BDF          | INSTANCES | NAME           | DIR        |
-+--------------+-----------+----------------+------------+
-| 0000:41:01.6 |         1 |GRID A100D-1-10C | nvidia-699 |
-                       ... snip ...
-```
-
-Repeat the steps after the [snippet](#list-all-available-vgpu-instances) listing
-to create the corresponding `mdev` device and use the guest `OS` created in the
-previous section with `time-sliced` `vGPUs`.
-
-## Install NVIDIA Driver + Toolkit in Kata Containers Guest OS
+### Build Guest OS with NVIDIA Driver and Toolkit
 
 Consult the [Developer-Guide](https://github.com/kata-containers/kata-containers/blob/main/docs/Developer-Guide.md#create-a-rootfs-image) on how to create a
 rootfs base image for a distribution of your choice. This is going to be used as
@@ -583,9 +308,12 @@ Enable the `guest_hook_path` in Kata's `configuration.toml`
 guest_hook_path = "/usr/share/oci/hooks"
 ```
 
+As the last step one can remove the additional packages and files that were added
+to the `$ROOTFS_DIR` to keep it as small as possible.
+
 One has built a NVIDIA rootfs, kernel and now we can run any GPU container
 without installing the drivers into the container. Check NVIDIA device status
-with `nvidia-smi`
+with `nvidia-smi`:
 
 ```sh
 $  sudo ctr --debug run --runtime "io.containerd.kata.v2"  --device /dev/vfio/192  --rm -t "docker.io/nvidia/cuda:11.6.0-base-ubuntu20.04" cuda nvidia-smi
@@ -611,8 +339,309 @@ Fri Mar 18 10:36:59 2022
 +-----------------------------------------------------------------------------+
 ```
 
-As the last step one can remove the additional packages and files that were added
-to the `$ROOTFS_DIR` to keep it as small as possible.
+## Usage Examples with Kata Containers
+
+The following sections give usage examples for this based on the different modes.
+
+### NVIDIA GPU passthrough mode
+
+Use the following steps to pass an NVIDIA GPU device in passthrough mode with Kata:
+
+1. Find the Bus-Device-Function (BDF) for the GPU device on the host:
+
+   ```sh
+   $ sudo lspci -nn -D | grep -i nvidia
+   0000:d0:00.0 3D controller [0302]: NVIDIA Corporation Device [10de:20b9] (rev a1)
+   ```
+
+   > PCI address `0000:d0:00.0` is assigned to the hardware GPU device.
+   > `10de:20b9` is the device ID of the hardware GPU device.
+
+2. Find the IOMMU group for the GPU device:
+
+   ```sh
+   $ BDF="0000:d0:00.0"
+   $ readlink -e /sys/bus/pci/devices/$BDF/iommu_group
+   ```
+
+   The previous output shows that the GPU belongs to IOMMU group 192. The next
+   step is to bind the GPU to the VFIO-PCI driver.
+
+   ```sh
+   $ BDF="0000:d0:00.0"
+   $ DEV="/sys/bus/pci/devices/$BDF"
+   $ echo "vfio-pci" > $DEV/driver_override
+   $ echo $BDF > $DEV/driver/unbind
+   $ echo $BDF > /sys/bus/pci/drivers_probe
+   # To return the device to the standard driver, we simply clear the
+   # driver_override and reprobe the device, ex:
+   $ echo > $DEV/preferred_driver
+   $ echo $BDF > $DEV/driver/unbind
+   $ echo $BDF > /sys/bus/pci/drivers_probe
+   ```
+
+3. Check the IOMMU group number under `/dev/vfio`:
+
+   ```sh
+   $ ls -l /dev/vfio
+   total 0
+   crw------- 1 zvonkok zvonkok 243,   0 Mar 18 03:06 192
+   crw-rw-rw- 1 root    root     10, 196 Mar 18 02:27 vfio
+   ```
+
+4. Start a Kata container with the GPU device:
+
+   ```sh
+   # You may need to `modprobe vhost-vsock` if you get
+   # host system doesn't support vsock: stat /dev/vhost-vsock
+   $ sudo ctr --debug run --runtime "io.containerd.kata.v2"  --device /dev/vfio/192  --rm -t  "docker.io/library/archlinux:latest" arch uname -r
+   ```
+
+5. Run `lspci` within the container to verify the GPU device is seen in the list
+   of the PCI devices. Note the vendor-device id of the GPU (`10de:20b9`) in the `lspci` output.
+
+   ```sh
+   $ sudo ctr --debug run --runtime "io.containerd.kata.v2"  --device /dev/vfio/192  --rm -t  "docker.io/library/archlinux:latest" arch sh -c "lspci -nn | grep '10de:20b9'"
+   ```
+
+6. Additionally, you can check the PCI BARs space of ​​the NVIDIA GPU device in the container:
+
+   ```sh
+   $ sudo ctr --debug run --runtime "io.containerd.kata.v2"  --device /dev/vfio/192  --rm -t  "docker.io/library/archlinux:latest" arch sh -c "lspci -s 02:00.0 -vv | grep Region"
+   ```
+
+   > **Note**: If you see a message similar to the above, the BAR space of the NVIDIA
+   > GPU has been successfully allocated.
+
+### NVIDIA vGPU mode
+
+NVIDIA vGPU is a licensed product on all supported GPU boards. A software license
+is required to enable all vGPU features within the guest VM. NVIDIA vGPU manager
+needs to be installed on the host to configure GPUs in vGPU mode. See
+[NVIDIA Virtual GPU Software Documentation v14.0 through 14.1](https://docs.nvidia.com/grid/14.0/)
+for more details.
+
+#### NVIDIA vGPU time-sliced
+
+In the time-sliced mode, the GPU is not partitioned and the workload uses the
+whole GPU and shares access to the GPU engines. Processes are scheduled in
+series. The best effort scheduler is the default one and can be exchanged by
+other scheduling policies see the documentation above how to do that.
+
+Beware if you had `MIG` enabled before to disable `MIG` on the GPU if you want
+to use `time-sliced` `vGPU`.
+
+```sh
+$ sudo nvidia-smi -mig 0
+```
+
+Enable the virtual functions for the physical GPU in the `sysfs` file system.
+
+```sh
+$ sudo /usr/lib/nvidia/sriov-manage -e 0000:41:00.0
+```
+
+Get the `BDF` of the available virtual function on the GPU, and choose one for the
+following steps.
+
+```sh
+$ cd /sys/bus/pci/devices/0000:41:00.0/
+$ ls -l |  grep virtfn
+```
+
+##### List all available vGPU instances
+
+The following shell snippet will walk the `sysfs` and only print instances
+that are available, that can be created.
+
+```sh
+# The 00.0 is often the PF of the device. The VFs will have the function in the
+# BDF incremented by some values so e.g. the very first VF is 0000:41:00.4
+
+cd /sys/bus/pci/devices/0000:41:00.0/
+
+for vf in $(ls -d virtfn*)
+do
+        BDF=$(basename $(readlink -f $vf))
+        for md in $(ls -d $vf/mdev_supported_types/*)
+        do
+                AVAIL=$(cat $md/available_instances)
+                NAME=$(cat $md/name)
+                DIR=$(basename $md)
+
+                if [ $AVAIL -gt 0 ]; then
+                        echo "| BDF          | INSTANCES | NAME           | DIR        |"
+                        echo "+--------------+-----------+----------------+------------+"
+                        printf "| %12s |%10d |%15s | %10s |\n\n" "$BDF" "$AVAIL" "$NAME" "$DIR"
+                fi
+
+        done
+done
+```
+
+If there are available instances you get something like this (for the first VF),
+beware that the output is highly dependent on the GPU you have, if there is no
+output check again if `MIG` is really disabled.
+
+```sh
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.4 |         1 |  GRID A100D-4C | nvidia-692 |
+
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.4 |         1 |  GRID A100D-8C | nvidia-693 |
+
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.4 |         1 | GRID A100D-10C | nvidia-694 |
+
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.4 |         1 | GRID A100D-16C | nvidia-695 |
+
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.4 |         1 | GRID A100D-20C | nvidia-696 |
+
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.4 |         1 | GRID A100D-40C | nvidia-697 |
+
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.4 |         1 | GRID A100D-80C | nvidia-698 |
+
+```
+
+Change to the `mdev_supported_types` directory for the virtual function on which
+you want to create the `vGPU`. Taking the first output as an example:
+
+```sh
+$ cd virtfn0/mdev_supported_types/nvidia-692
+$ UUIDGEN=$(uuidgen)
+$ sudo bash -c "echo $UUIDGEN > create"
+```
+
+Confirm that the `vGPU` was created. You should see the `UUID` pointing to a
+subdirectory of the `sysfs` space.
+
+```sh
+$ ls -l /sys/bus/mdev/devices/
+```
+
+Get the `IOMMU` group number and verify there is a `VFIO` device created to use
+with Kata.
+
+```sh
+$ ls -l /sys/bus/mdev/devices/*/
+$ ls -l /dev/vfio
+```
+
+Use the `VFIO` device created in the same way as in the passthrough use-case.
+Beware that the guest needs the NVIDIA guest drivers, so one would need to build
+a new guest `OS` image.
+
+#### NVIDIA vGPU MIG-backed
+
+We're not going into detail what `MIG` is but briefly it is a technology to
+partition the hardware into independent instances with guaranteed quality of
+service. For more details see
+[NVIDIA Multi-Instance GPU User Guide](https://docs.nvidia.com/datacenter/tesla/mig-user-guide/).
+
+First enable `MIG` mode for a GPU, depending on the platform you're running
+a reboot would be necessary. Some platforms support GPU reset.
+
+```sh
+$ sudo nvidia-smi -mig 1
+```
+
+If the platform supports a GPU reset one can run, otherwise you will get a
+warning to reboot the server.
+
+```sh
+$ sudo nvidia-smi --gpu-reset
+```
+
+The driver per default provides a number of profiles that users can opt-in when
+configuring the MIG feature.
+
+```sh
+$ sudo nvidia-smi mig -lgip
++-----------------------------------------------------------------------------+
+| GPU instance profiles:                                                      |
+| GPU   Name             ID    Instances   Memory     P2P    SM    DEC   ENC  |
+|                              Free/Total   GiB              CE    JPEG  OFA  |
+|=============================================================================|
+|   0  MIG 1g.10gb       19     7/7        9.50       No     14     0     0   |
+|                                                             1     0     0   |
++-----------------------------------------------------------------------------+
+|   0  MIG 1g.10gb+me    20     1/1        9.50       No     14     1     0   |
+|                                                             1     1     1   |
++-----------------------------------------------------------------------------+
+|   0  MIG 2g.20gb       14     3/3        19.50      No     28     1     0   |
+|                                                             2     0     0   |
++-----------------------------------------------------------------------------+
+                              ...
+```
+
+Create the GPU instances that correspond to the `vGPU` types of the `MIG-backed`
+`vGPUs` that you will create
+[NVIDIA A100 PCIe 80GB Virtual GPU Types](https://docs.nvidia.com/grid/13.0/grid-vgpu-user-guide/index.html#vgpu-types-nvidia-a100-pcie-80gb).
+
+```sh
+# MIG 1g.10gb --> vGPU A100D-1-10C
+$ sudo nvidia-smi mig -cgi 19
+```
+
+List the GPU instances and get the GPU instance id to create the compute
+instance.
+
+```sh
+$ sudo nvidia-smi mig -lgi # list the created GPU instances
+$ sudo nvidia-smi mig -cci -gi 9 # each GPU instance can have several compute
+                                 # instances. Instance -> Workload
+```
+
+Verify that the compute instances were created within the GPU instance
+
+```sh
+$ nvidia-smi
+                              ... snip ...
++-----------------------------------------------------------------------------+
+| MIG devices:                                                                |
++------------------+----------------------+-----------+-----------------------+
+| GPU  GI  CI  MIG |         Memory-Usage |        Vol|         Shared        |
+|      ID  ID  Dev |           BAR1-Usage | SM     Unc| CE  ENC  DEC  OFA  JPG|
+|                  |                      |        ECC|                       |
+|==================+======================+===========+=======================|
+|  0    9   0   0  |      0MiB /  9728MiB | 14      0 |  1   0    0    0    0 |
+|                  |      0MiB /  4095MiB |           |                       |
++------------------+----------------------+-----------+-----------------------+
+                              ... snip ...
+```
+
+We can use the [snippet](#list-all-available-vgpu-instances) from before to list
+the available `vGPU` instances, this time `MIG-backed`.
+
+```sh
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.4 |         1 |GRID A100D-1-10C | nvidia-699 |
+
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:00.5 |         1 |GRID A100D-1-10C | nvidia-699 |
+
+| BDF          | INSTANCES | NAME           | DIR        |
++--------------+-----------+----------------+------------+
+| 0000:41:01.6 |         1 |GRID A100D-1-10C | nvidia-699 |
+                       ... snip ...
+```
+
+Repeat the steps after the [snippet](#list-all-available-vgpu-instances) listing
+to create the corresponding `mdev` device and use the guest `OS` created in the
+previous section with `time-sliced` `vGPUs`.
 
 ## References
 

docs/use-cases/using-Intel-QAT-and-kata.md

@@ -1,24 +1,20 @@
 # Table of Contents
 
+**Note:**: This guide used to contain an end-to-end flow to build a
+custom Kata containers root filesystem with QAT out-of-tree SR-IOV virtual
+function driver and run QAT enabled containers. The former is no longer necessary
+so the instructions are dropped. If the use-case is still of interest, please file
+an issue in either of the QAT Kubernetes specific repos linked below.
+
 # Introduction
 
 Intel® QuickAssist Technology (QAT) provides hardware acceleration
-for security (cryptography) and compression. These instructions cover the
-steps for the latest [Ubuntu LTS release](https://ubuntu.com/download/desktop)
-which already include the QAT host driver. These instructions can be adapted to
-any Linux distribution. These instructions guide the user on how to download
-the kernel sources, compile kernel driver modules against those sources, and
-load them onto the host as well as preparing a specially built Kata Containers
-kernel and custom Kata Containers rootfs.
+for security (cryptography) and compression. Kata Containers can enable
+these acceleration functions for containers using QAT SR-IOV with the
+support from [Intel QAT Device Plugin for Kubernetes](https://github.com/intel/intel-device-plugins-for-kubernetes)
+or [Intel QAT DRA Resource Driver for Kubernetes](https://github.com/intel/intel-resource-drivers-for-kubernetes).
 
-* Download kernel sources
-* Compile Kata kernel
-* Compile kernel driver modules against those sources
-* Download rootfs
-* Add driver modules to rootfs
-* Build rootfs image
-
-## Helpful Links before starting
+## More Information
 
 [Intel® QuickAssist Technology at `01.org`](https://www.intel.com/content/www/us/en/developer/topic-technology/open/quick-assist-technology/overview.html)
 
@@ -26,554 +22,6 @@ kernel and custom Kata Containers rootfs.
 
 [Intel Device Plugin for Kubernetes](https://github.com/intel/intel-device-plugins-for-kubernetes)
 
+[Intel DRA Resource Driver for Kubernetes](https://github.com/intel/intel-resource-drivers-for-kubernetes)
+
 [Intel® QuickAssist Technology for Crypto Poll Mode Driver](https://dpdk-docs.readthedocs.io/en/latest/cryptodevs/qat.html)
-
-## Steps to enable Intel® QAT in Kata Containers
-
-There are some steps to complete only once, some steps to complete with every
-reboot, and some steps to complete when the host kernel changes.
-
-## Script variables
-
-The following list of variables must be set before running through the
-scripts. These variables refer to locations to store modules and configuration
-files on the host and links to the drivers to use. Modify these as
-needed to point to updated drivers or different install locations.
-
-### Set environment variables (Every Reboot)
-
-Make sure to check [`01.org`](https://www.intel.com/content/www/us/en/developer/topic-technology/open/quick-assist-technology/overview.html) for
-the latest driver.
-
-```bash
-$ export QAT_DRIVER_VER=qat1.7.l.4.14.0-00031.tar.gz
-$ export QAT_DRIVER_URL=https://downloadmirror.intel.com/30178/eng/${QAT_DRIVER_VER}
-$ export QAT_CONF_LOCATION=~/QAT_conf
-$ export QAT_DOCKERFILE=https://raw.githubusercontent.com/intel/intel-device-plugins-for-kubernetes/main/demo/openssl-qat-engine/Dockerfile
-$ export QAT_SRC=~/src/QAT
-$ export GOPATH=~/src/go
-$ export KATA_KERNEL_LOCATION=~/kata
-$ export KATA_ROOTFS_LOCATION=~/kata
-```
-
-## Prepare the Ubuntu Host
-
-The host could be a bare metal instance or a virtual machine. If using a
-virtual machine, make sure that KVM nesting is enabled. The following
-instructions reference an Intel® C62X chipset. Some of the instructions must be
-modified if using a different Intel® QAT device. The Intel® QAT chipset can be
-identified by executing the following.
-
-### Identify which PCI Bus the Intel® QAT card is on
-
-```bash
-$ for i in 0434 0435 37c8 1f18 1f19; do lspci -d 8086:$i; done
-```
-
-### Install necessary packages for Ubuntu
-
-These packages are necessary to compile the Kata kernel, Intel® QAT driver, and to
-prepare the rootfs for Kata. [Docker](https://docs.docker.com/engine/install/ubuntu/)
-also needs to be installed to be able to build the rootfs. To test that
-everything works a Kubernetes pod is started requesting Intel® QAT resources. For the
-pass through of the virtual functions the kernel boot parameter needs to have
-`INTEL_IOMMU=on`.
-
-```bash
-$ sudo apt update
-$ sudo apt install -y golang-go build-essential python pkg-config zlib1g-dev libudev-dev bison libelf-dev flex libtool automake autotools-dev autoconf bc libpixman-1-dev coreutils libssl-dev
-$ sudo sed -i 's/GRUB_CMDLINE_LINUX_DEFAULT=""/GRUB_CMDLINE_LINUX_DEFAULT="intel_iommu=on"/' /etc/default/grub
-$ sudo update-grub
-$ sudo reboot
-```
-
-### Download Intel® QAT drivers
-
-This will download the [Intel® QAT drivers](https://www.intel.com/content/www/us/en/developer/topic-technology/open/quick-assist-technology/overview.html).
-Make sure to check the website for the latest version.
-
-```bash
-$ mkdir -p $QAT_SRC
-$ cd $QAT_SRC
-$ curl -L $QAT_DRIVER_URL | tar zx
-```
-
-### Copy Intel® QAT configuration files and enable virtual functions
-
-Modify the instructions below as necessary if using a different Intel® QAT hardware
-platform. You can learn more about customizing configuration files at the
-[Intel® QAT Engine repository](https://github.com/intel/QAT_Engine/#copy-the-correct-intel-quickassist-technology-driver-config-files)
-This section starts from a base config file and changes the `SSL` section to
-`SHIM` to support the OpenSSL engine. There are more tweaks that you can make
-depending on the use case and how many Intel® QAT engines should be run. You
-can find more information about how to customize in the
-[Intel® QuickAssist Technology Software for Linux* - Programmer's Guide.](https://www.intel.com/content/www/us/en/content-details/709196/intel-quickassist-technology-api-programmer-s-guide.html)
-
-> **Note: This section assumes that a Intel® QAT `c6xx` platform is used.**
-
-```bash
-$ mkdir -p $QAT_CONF_LOCATION
-$ cp $QAT_SRC/quickassist/utilities/adf_ctl/conf_files/c6xxvf_dev0.conf.vm $QAT_CONF_LOCATION/c6xxvf_dev0.conf
-$ sed -i 's/\[SSL\]/\[SHIM\]/g' $QAT_CONF_LOCATION/c6xxvf_dev0.conf
-```
-
-### Expose and Bind Intel® QAT virtual functions to VFIO-PCI (Every reboot)
-
-To enable virtual functions, the host OS should have IOMMU groups enabled. In
-the UEFI Firmware Intel® Virtualization Technology for Directed I/O
-(Intel® VT-d) must be enabled. Also, the kernel boot parameter should be
-`intel_iommu=on` or `intel_iommu=ifgx_off`. This should have been set from
-the instructions above. Check the output of `/proc/cmdline` to confirm. The
-following commands assume you installed an Intel® QAT card, IOMMU is on, and
-VT-d is enabled. The vendor and device ID add to the `VFIO-PCI` driver so that
-each exposed virtual function can be bound to the `VFIO-PCI` driver. Once
-complete, each virtual function passes into a Kata Containers container using
-the PCIe device passthrough feature. For Kubernetes, the
-[Intel device plugin](https://github.com/intel/intel-device-plugins-for-kubernetes)
-for Kubernetes handles the binding of the driver, but the VF’s still must be
-enabled.
-
-```bash
-$ sudo modprobe vfio-pci
-$ QAT_PCI_BUS_PF_NUMBERS=$((lspci -d :435 && lspci -d :37c8 && lspci -d :19e2 && lspci -d :6f54) | cut -d ' ' -f 1)
-$ QAT_PCI_BUS_PF_1=$(echo $QAT_PCI_BUS_PF_NUMBERS | cut -d ' ' -f 1)
-$ echo 16 | sudo tee /sys/bus/pci/devices/0000:$QAT_PCI_BUS_PF_1/sriov_numvfs
-$ QAT_PCI_ID_VF=$(cat /sys/bus/pci/devices/0000:${QAT_PCI_BUS_PF_1}/virtfn0/uevent | grep PCI_ID)
-$ QAT_VENDOR_AND_ID_VF=$(echo ${QAT_PCI_ID_VF/PCI_ID=} | sed 's/:/ /')
-$ echo $QAT_VENDOR_AND_ID_VF | sudo tee --append /sys/bus/pci/drivers/vfio-pci/new_id
-```
-
-Loop through all the virtual functions and bind to the VFIO driver
-
-```bash
-$ for f in /sys/bus/pci/devices/0000:$QAT_PCI_BUS_PF_1/virtfn*
-  do QAT_PCI_BUS_VF=$(basename $(readlink $f))
-   echo $QAT_PCI_BUS_VF | sudo tee --append /sys/bus/pci/drivers/c6xxvf/unbind
-   echo $QAT_PCI_BUS_VF | sudo tee --append /sys/bus/pci/drivers/vfio-pci/bind
-  done
-```
-
-### Check Intel® QAT virtual functions are enabled
-
-If the following command returns empty, then the virtual functions are not
-properly enabled. This command checks the enumerated device IDs for just the
-virtual functions. Using the Intel® QAT as an example, the physical device ID
-is `37c8` and virtual function device ID is `37c9`. The following command checks
-if VF's are enabled for any of the currently known Intel® QAT device ID's. The
-following `ls` command should show the 16 VF's bound to `VFIO-PCI`.
-
-```bash
-$ for i in 0442 0443 37c9 19e3; do lspci -d 8086:$i; done
-```
-
-Another way to check is to see what PCI devices that `VFIO-PCI` is mapped to.
-It should match the device ID's of the VF's.
-
-```bash
-$ ls -la /sys/bus/pci/drivers/vfio-pci
-```
-
-## Prepare Kata Containers
-
-### Download Kata kernel Source
-
-This example automatically uses the latest Kata kernel supported by Kata. It
-follows the instructions from the
-[packaging kernel repository](../../tools/packaging/kernel)
-and uses the latest Kata kernel
-[config](../../tools/packaging/kernel/configs).
-There are some patches that must be installed as well, which the
-`build-kernel.sh` script should automatically apply. If you are using a
-different kernel version, then you might need to manually apply them. Since
-the Kata Containers kernel has a minimal set of kernel flags set, you must
-create a Intel® QAT kernel fragment with the necessary `CONFIG_CRYPTO_*` options set.
-Update the config to set some of the `CRYPTO` flags to enabled. This might
-change with different kernel versions. The following instructions were tested
-with kernel `v5.4.0-64-generic`.
-
-```bash
-$ mkdir -p $GOPATH
-$ cd $GOPATH
-$ go get -v github.com/kata-containers/kata-containers
-$ cat << EOF > $GOPATH/src/github.com/kata-containers/kata-containers/tools/packaging/kernel/configs/fragments/common/qat.conf
-CONFIG_PCIEAER=y
-CONFIG_UIO=y
-CONFIG_CRYPTO_HW=y
-CONFIG_CRYPTO_DEV_QAT_C62XVF=m
-CONFIG_CRYPTO_CBC=y
-CONFIG_MODULES=y
-CONFIG_MODULE_SIG=y
-CONFIG_CRYPTO_AUTHENC=y
-CONFIG_CRYPTO_DH=y
-EOF
-$ $GOPATH/src/github.com/kata-containers/kata-containers/tools/packaging/kernel/build-kernel.sh setup
-```
-
-### Build Kata kernel
-
-```bash
-$ cd $GOPATH
-$ export LINUX_VER=$(ls -d kata-linux-*)
-$ sed -i 's/EXTRAVERSION =/EXTRAVERSION = .qat.container/' $LINUX_VER/Makefile
-$ $GOPATH/src/github.com/kata-containers/kata-containers/tools/packaging/kernel/build-kernel.sh build
-```
-
-### Copy Kata kernel
-
-```bash
-$ export KATA_KERNEL_NAME=vmlinux-${LINUX_VER}_qat
-$ mkdir -p $KATA_KERNEL_LOCATION
-$ cp ${GOPATH}/${LINUX_VER}/vmlinux ${KATA_KERNEL_LOCATION}/${KATA_KERNEL_NAME}
-```
-
-### Prepare Kata root filesystem
-
-These instructions build upon the OS builder instructions located in the
-[Developer Guide](../Developer-Guide.md). At this point it is recommended that
-[Docker](https://docs.docker.com/engine/install/ubuntu/) is installed first, and
-then [Kata-deploy](../../tools/packaging/kata-deploy)
-is use to install Kata. This will make sure that the correct `agent` version
-is installed into the rootfs in the steps below.
-
-The following instructions use Ubuntu as the root filesystem with systemd as
-the init and will add in the `kmod` binary, which is not a standard binary in
-a Kata rootfs image. The `kmod` binary is necessary to load the Intel® QAT
-kernel modules when the virtual machine rootfs boots.
-
-```bash
-$ export OSBUILDER=$GOPATH/src/github.com/kata-containers/kata-containers/tools/osbuilder
-$ export ROOTFS_DIR=${OSBUILDER}/rootfs-builder/rootfs
-$ export EXTRA_PKGS='kmod'
-```
-
-Make sure that the `kata-agent` version matches the installed `kata-runtime`
-version. Also make sure the `kata-runtime` install location is in your `PATH`
-variable. The following `AGENT_VERSION` can be set manually to match
-the `kata-runtime` version if the following commands don't work.
-
-```bash
-$ export PATH=$PATH:/opt/kata/bin
-$ cd $GOPATH
-$ export AGENT_VERSION=$(kata-runtime version | head -n 1 | grep -o "[0-9.]\+")
-$ cd ${OSBUILDER}/rootfs-builder
-$ sudo rm -rf ${ROOTFS_DIR}
-$ script -fec 'sudo -E GOPATH=$GOPATH USE_DOCKER=true SECCOMP=no ./rootfs.sh ubuntu'
-```
-
-### Compile Intel® QAT drivers for Kata Containers kernel and add to Kata Containers rootfs
-
-After the Kata Containers kernel builds with the proper configuration flags,
-you must build the Intel® QAT drivers against that Kata Containers kernel
-version in a similar way they were previously built for the host OS. You must
-set the `KERNEL_SOURCE_ROOT` variable to the Kata Containers kernel source
-directory and build the Intel® QAT drivers again. The  `make` command will
-install the Intel® QAT modules into the Kata rootfs.
-
-```bash
-$ cd $GOPATH
-$ export LINUX_VER=$(ls -d kata*)
-$ export KERNEL_MAJOR_VERSION=$(awk '/^VERSION =/{print $NF}' $GOPATH/$LINUX_VER/Makefile)
-$ export KERNEL_PATHLEVEL=$(awk '/^PATCHLEVEL =/{print $NF}' $GOPATH/$LINUX_VER/Makefile)
-$ export KERNEL_SUBLEVEL=$(awk '/^SUBLEVEL =/{print $NF}' $GOPATH/$LINUX_VER/Makefile)
-$ export KERNEL_EXTRAVERSION=$(awk '/^EXTRAVERSION =/{print $NF}' $GOPATH/$LINUX_VER/Makefile)
-$ export KERNEL_ROOTFS_DIR=${KERNEL_MAJOR_VERSION}.${KERNEL_PATHLEVEL}.${KERNEL_SUBLEVEL}${KERNEL_EXTRAVERSION}
-$ cd $QAT_SRC
-$ KERNEL_SOURCE_ROOT=$GOPATH/$LINUX_VER ./configure --enable-icp-sriov=guest
-$ sudo -E make all -j $(nproc)
-$ sudo -E make INSTALL_MOD_PATH=$ROOTFS_DIR qat-driver-install -j $(nproc)
-```
-
-The `usdm_drv` module also needs to be copied into the rootfs modules path and
-`depmod` should be run.
-
-```bash
-$ sudo cp $QAT_SRC/build/usdm_drv.ko $ROOTFS_DIR/lib/modules/${KERNEL_ROOTFS_DIR}/updates/drivers
-$ sudo depmod -a -b ${ROOTFS_DIR} ${KERNEL_ROOTFS_DIR}
-$ cd ${OSBUILDER}/image-builder
-$ script -fec 'sudo -E USE_DOCKER=true ./image_builder.sh ${ROOTFS_DIR}'
-```
-
-> **Note: Ignore any errors on modules.builtin and modules.order when running
-> `depmod`.**
-
-### Copy Kata rootfs
-
-```bash
-$ mkdir -p $KATA_ROOTFS_LOCATION
-$ cp ${OSBUILDER}/image-builder/kata-containers.img $KATA_ROOTFS_LOCATION
-```
-
-## Verify Intel® QAT works in a container
-
-The following instructions uses a OpenSSL Dockerfile that builds the
-Intel® QAT engine to allow OpenSSL to offload crypto functions. It is a
-convenient way to test that VFIO device passthrough for the Intel® QAT VF’s are
-working properly with the Kata Containers VM.
-
-### Build OpenSSL Intel® QAT engine container
-
-Use the OpenSSL Intel® QAT [Dockerfile](https://github.com/intel/intel-device-plugins-for-kubernetes/tree/main/demo/openssl-qat-engine)
-to build a container image with an optimized OpenSSL engine for
-Intel® QAT. Using `docker build` with the Kata Containers runtime can sometimes
-have issues. Therefore, make sure that `runc` is the default Docker container
-runtime.
-
-```bash
-$ cd $QAT_SRC
-$ curl -O $QAT_DOCKERFILE
-$ sudo docker build -t openssl-qat-engine .
-```
-
-> **Note: The Intel® QAT driver version in this container might not match the
-> Intel® QAT driver compiled and loaded on the host when compiling.**
-
-### Test Intel® QAT with the ctr tool
-
-The `ctr` tool can be used to interact with the containerd daemon. It may be
-more convenient to use this tool to verify the kernel and image instead of
-setting up a Kubernetes cluster. The correct Kata runtimes need to be added
-to the containerd `config.toml`. Below is a sample snippet that can be added
-to allow QEMU and Cloud Hypervisor (CLH) to work with `ctr`.
-
-```
-[plugins.cri.containerd.runtimes.kata-qemu]
-  runtime_type = "io.containerd.kata-qemu.v2"
-  privileged_without_host_devices = true
-  pod_annotations = ["io.katacontainers.*"]
-  [plugins.cri.containerd.runtimes.kata-qemu.options]
-    ConfigPath = "/opt/kata/share/defaults/kata-containers/configuration-qemu.toml"
-[plugins.cri.containerd.runtimes.kata-clh]
-  runtime_type = "io.containerd.kata-clh.v2"
-  privileged_without_host_devices = true
-  pod_annotations = ["io.katacontainers.*"]
-  [plugins.cri.containerd.runtimes.kata-clh.options]
-    ConfigPath = "/opt/kata/share/defaults/kata-containers/configuration-clh.toml"
-```
-
-In addition, containerd expects the binary to be in `/usr/local/bin` so add
-this small script so that it redirects to be able to use either QEMU or
-Cloud Hypervisor with Kata.
-
-```bash
-$ echo '#!/usr/bin/env bash' | sudo tee /usr/local/bin/containerd-shim-kata-qemu-v2
-$ echo 'KATA_CONF_FILE=/opt/kata/share/defaults/kata-containers/configuration-qemu.toml /opt/kata/bin/containerd-shim-kata-v2 $@' | sudo tee -a /usr/local/bin/containerd-shim-kata-qemu-v2
-$ sudo chmod +x /usr/local/bin/containerd-shim-kata-qemu-v2
-$ echo '#!/usr/bin/env bash' | sudo tee /usr/local/bin/containerd-shim-kata-clh-v2
-$ echo 'KATA_CONF_FILE=/opt/kata/share/defaults/kata-containers/configuration-clh.toml /opt/kata/bin/containerd-shim-kata-v2 $@' | sudo tee -a /usr/local/bin/containerd-shim-kata-clh-v2
-$ sudo chmod +x /usr/local/bin/containerd-shim-kata-clh-v2
-```
-
-After the OpenSSL image is built and imported into containerd, a Intel® QAT
-virtual function exposed in the step above can be added to the `ctr` command.
-Make sure to change the `/dev/vfio` number to one that actually exists on the
-host system. When using the `ctr` tool, the`configuration.toml` for Kata needs
-to point to the custom Kata kernel and rootfs built above and the Intel® QAT
-modules in the Kata rootfs need to load at boot. The following steps assume that
-`kata-deploy` was used to install Kata and QEMU is being tested. If using a
-different hypervisor, different install method for Kata, or a different
-Intel® QAT chipset then the command will need to be modified.
-
-> **Note: The following was tested with
-[containerd v1.4.6](https://github.com/containerd/containerd/releases/tag/v1.4.6).**
-
-```bash
-$ config_file="/opt/kata/share/defaults/kata-containers/configuration-qemu.toml"
-$ sudo sed -i "/kernel =/c kernel = "\"${KATA_ROOTFS_LOCATION}/${KATA_KERNEL_NAME}\""" $config_file
-$ sudo sed -i "/image =/c image = "\"${KATA_KERNEL_LOCATION}/kata-containers.img\""" $config_file
-$ sudo sed -i -e 's/^kernel_params = "\(.*\)"/kernel_params = "\1 modules-load=usdm_drv,qat_c62xvf"/g' $config_file
-$ sudo docker save -o openssl-qat-engine.tar openssl-qat-engine:latest
-$ sudo ctr images import openssl-qat-engine.tar
-$ sudo ctr run --runtime io.containerd.run.kata-qemu.v2 --privileged -t --rm --device=/dev/vfio/180 --mount type=bind,src=/dev,dst=/dev,options=rbind:rw --mount type=bind,src=${QAT_CONF_LOCATION}/c6xxvf_dev0.conf,dst=/etc/c6xxvf_dev0.conf,options=rbind:rw  docker.io/library/openssl-qat-engine:latest bash
-```
-
-Below are some commands to run in the container image to verify Intel® QAT is
-working
-
-```sh
-root@67561dc2757a/ # cat /proc/modules
-qat_c62xvf 16384 - - Live 0xffffffffc00d9000 (OE)
-usdm_drv 86016 - - Live 0xffffffffc00e8000 (OE)
-intel_qat 249856 - - Live 0xffffffffc009b000 (OE)
-
-root@67561dc2757a/ # adf_ctl restart
-Restarting all devices.
-Processing /etc/c6xxvf_dev0.conf
-
-root@67561dc2757a/ # adf_ctl status
-Checking status of all devices.
-There is 1 QAT acceleration device(s) in the system:
- qat_dev0 - type: c6xxvf,  inst_id: 0,  node_id: 0,  bsf: 0000:01:01.0,  #accel: 1 #engines: 1 state: up
-
-root@67561dc2757a/ # openssl engine -c -t qat-hw
-(qat-hw) Reference implementation of QAT crypto engine v0.6.1
- [RSA, DSA, DH, AES-128-CBC-HMAC-SHA1, AES-128-CBC-HMAC-SHA256, AES-256-CBC-HMAC-SHA1, AES-256-CBC-HMAC-SHA256, TLS1-PRF, HKDF, X25519, X448]
-     [ available ]
-```
-
-### Test Intel® QAT in Kubernetes
-
-Start a Kubernetes cluster with containerd as the CRI. The host should
-already be setup with 16 virtual functions of the Intel® QAT card bound to
-`VFIO-PCI`. Verify this by looking in `/dev/vfio` for a listing of devices.
-You might need to disable Docker before initializing Kubernetes. Be aware
-that the OpenSSL container image built above will need to be exported from
-Docker and imported into containerd.
-
-If Kata is installed through [`kata-deploy`](../../tools/packaging/kata-deploy/helm-chart/README.md)
-there will be multiple `configuration.toml` files associated with different
-hypervisors. Rather than add in the custom Kata kernel, Kata rootfs, and
-kernel modules to each `configuration.toml` as the default, instead use
-[annotations](../how-to/how-to-load-kernel-modules-with-kata.md)
-in the Kubernetes YAML file to tell Kata which kernel and rootfs to use. The
-easy way to do this is to use `kata-deploy` which will install the Kata binaries
-to `/opt` and properly configure the `/etc/containerd/config.toml` with annotation
-support. However, the `configuration.toml` needs to enable support for
-annotations as well. The following configures both QEMU and Cloud Hypervisor
-`configuration.toml` files that are currently available with Kata Container
-versions 2.0 and higher.
-
-```bash
-$ sudo sed -i 's/enable_annotations\s=\s\[\]/enable_annotations = [".*"]/' /opt/kata/share/defaults/kata-containers/configuration-qemu.toml
-$ sudo sed -i 's/enable_annotations\s=\s\[\]/enable_annotations = [".*"]/' /opt/kata/share/defaults/kata-containers/configuration-clh.toml
-```
-
-Export the OpenSSL image from Docker and import into containerd.
-
-```bash
-$ sudo docker save -o openssl-qat-engine.tar openssl-qat-engine:latest
-$ sudo ctr -n=k8s.io images import openssl-qat-engine.tar
-```
-
-The [Intel® QAT Plugin](https://github.com/intel/intel-device-plugins-for-kubernetes/blob/main/cmd/qat_plugin/README.md)
-needs to be started so that the virtual functions can be discovered and
-used by Kubernetes.
-
-The following YAML file can be used to start a Kata container with Intel® QAT
-support. If Kata is installed with `kata-deploy`, then the containerd
-`configuration.toml` should have all of the Kata runtime classes already
-populated and annotations supported. To use a Intel® QAT virtual function, the
-Intel® QAT plugin needs to be started after the VF's are bound to `VFIO-PCI` as
-described [above](#expose-and-bind-intel-qat-virtual-functions-to-vfio-pci-every-reboot).
-Edit the following to point to the correct Kata kernel and rootfs location
-built with Intel® QAT support.
-
-```bash
-$ cat << EOF > kata-openssl-qat.yaml
-apiVersion: v1
-kind: Pod
-metadata:
-  name: kata-openssl-qat
-  labels:
-    app: kata-openssl-qat
-  annotations:
-    io.katacontainers.config.hypervisor.kernel: "$KATA_KERNEL_LOCATION/$KATA_KERNEL_NAME"
-    io.katacontainers.config.hypervisor.image: "$KATA_ROOTFS_LOCATION/kata-containers.img"
-    io.katacontainers.config.hypervisor.kernel_params: "modules-load=usdm_drv,qat_c62xvf"
-spec:
-  runtimeClassName: kata-qemu
-  containers:
-  - name: kata-openssl-qat
-    image: docker.io/library/openssl-qat-engine:latest
-    imagePullPolicy: IfNotPresent
-    resources:
-      limits:
-        qat.intel.com/generic: 1
-        cpu: 1
-    securityContext:
-      capabilities:
-        add: ["IPC_LOCK", "SYS_ADMIN"]
-    volumeMounts:
-      - mountPath: /etc/c6xxvf_dev0.conf
-        name: etc-mount
-      - mountPath: /dev
-        name: dev-mount
-  volumes:
-    - name: dev-mount
-      hostPath:
-        path: /dev
-    - name: etc-mount
-      hostPath:
-        path: $QAT_CONF_LOCATION/c6xxvf_dev0.conf
-EOF
-```
-
-Use `kubectl` to start the pod. Verify that Intel® QAT card acceleration is
-working with the Intel® QAT engine.
-```bash
-$ kubectl apply -f kata-openssl-qat.yaml
-```
-
-```sh
-$ kubectl exec -it kata-openssl-qat -- adf_ctl restart
-Restarting all devices.
-Processing /etc/c6xxvf_dev0.conf
-
-$ kubectl exec -it kata-openssl-qat -- adf_ctl status
-Checking status of all devices.
-There is 1 QAT acceleration device(s) in the system:
- qat_dev0 - type: c6xxvf,  inst_id: 0,  node_id: 0,  bsf: 0000:01:01.0,  #accel: 1 #engines: 1 state: up
-
-$ kubectl exec -it kata-openssl-qat -- openssl engine -c -t qat-hw
-(qat-hw) Reference implementation of QAT crypto engine v0.6.1
- [RSA, DSA, DH, AES-128-CBC-HMAC-SHA1, AES-128-CBC-HMAC-SHA256, AES-256-CBC-HMAC-SHA1, AES-256-CBC-HMAC-SHA256, TLS1-PRF, HKDF, X25519, X448]
-     [ available ]
-```
-
-### Troubleshooting
-
-* Check that `/dev/vfio` has VF’s enabled.
-
-```sh
-$ ls /dev/vfio
-57  58  59  60  61  62  63  64  65  66  67  68  69  70  71  72  vfio
-```
-
-* Check that the modules load when inside the Kata Container.
-
-```sh
-bash-5.0# grep -E "qat|usdm_drv" /proc/modules
-qat_c62xvf 16384 - - Live 0x0000000000000000 (O)
-usdm_drv 86016 - - Live 0x0000000000000000 (O)
-intel_qat 184320 - - Live 0x0000000000000000 (O)
-```
-
-* Verify that at least the first `c6xxvf_dev0.conf` file mounts inside the
-container image in `/etc`. You will need one configuration file for each VF
-passed into the container.
-
-```sh
-bash-5.0# ls /etc
-c6xxvf_dev0.conf   c6xxvf_dev11.conf  c6xxvf_dev14.conf  c6xxvf_dev3.conf  c6xxvf_dev6.conf  c6xxvf_dev9.conf  resolv.conf
-c6xxvf_dev1.conf   c6xxvf_dev12.conf  c6xxvf_dev15.conf  c6xxvf_dev4.conf  c6xxvf_dev7.conf  hostname
-c6xxvf_dev10.conf  c6xxvf_dev13.conf  c6xxvf_dev2.conf   c6xxvf_dev5.conf c6xxvf_dev8.conf  hosts
-```
-
-* Check `dmesg` inside the container to see if there are any issues with the
-Intel® QAT driver.
-
-* If there are issues building the OpenSSL Intel® QAT container image, then
-check to make sure that runc is the default runtime for building container.
-
-```sh
-$ cat /etc/systemd/system/docker.service.d/50-runtime.conf
-[Service]
-Environment="DOCKER_DEFAULT_RUNTIME=--default-runtime runc"
-```
-
-## Optional Scripts
-
-### Verify Intel® QAT card counters are incremented
-
-To check the built in firmware counters, the Intel® QAT driver has to be compiled
-and installed to the host and can't rely on the built in host driver. The
-counters will increase when the accelerator is actively being used. To verify
-Intel® QAT is actively accelerating the containerized application, use the
-following instructions to check if any of the counters increment. Make
-sure to change the PCI Device ID to match whats in the system.
-
-```bash
-$ for i in 0434 0435 37c8 1f18 1f19; do lspci -d 8086:$i; done
-$ sudo watch cat /sys/kernel/debug/qat_c6xx_0000\:b1\:00.0/fw_counters
-$ sudo watch cat /sys/kernel/debug/qat_c6xx_0000\:b3\:00.0/fw_counters
-$ sudo watch cat /sys/kernel/debug/qat_c6xx_0000\:b5\:00.0/fw_counters
-```

rust-toolchain.toml

@@ -1,3 +1,3 @@
 [toolchain]
 # Keep in sync with versions.yaml
-channel = "1.85.1"
+channel = "1.89"

src/agent/Cargo.lock

@@ -780,9 +780,9 @@ dependencies = [
 
 [[package]]
 name = "container-device-interface"
-version = "0.1.0"
+version = "0.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "653849f0c250f73d9afab4b2a9a6b07adaee1f34c44ffa6f2d2c3f9392002c1a"
+checksum = "2605001b0e8214dae8af146a43ccaa965d960403e330f174c21327154530df8b"
 dependencies = [
  "anyhow",
  "clap",
@@ -1207,9 +1207,9 @@ dependencies = [
 
 [[package]]
 name = "fancy-regex"
-version = "0.14.0"
+version = "0.16.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6e24cb5a94bcae1e5408b0effca5cd7172ea3c5755049c5f3af4cd283a165298"
+checksum = "998b056554fbe42e03ae0e152895cd1a7e1002aec800fdc6635d20270260c46f"
 dependencies = [
  "bit-set",
  "regex-automata 0.4.9",
@@ -2007,9 +2007,9 @@ dependencies = [
 
 [[package]]
 name = "jsonschema"
-version = "0.30.0"
+version = "0.33.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f1b46a0365a611fbf1d2143104dcf910aada96fafd295bab16c60b802bf6fa1d"
+checksum = "d46662859bc5f60a145b75f4632fbadc84e829e45df6c5de74cfc8e05acb96b5"
 dependencies = [
  "ahash 0.8.12",
  "base64 0.22.1",
@@ -3405,9 +3405,9 @@ dependencies = [
 
 [[package]]
 name = "referencing"
-version = "0.30.0"
+version = "0.33.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c8eff4fa778b5c2a57e85c5f2fe3a709c52f0e60d23146e2151cbef5893f420e"
+checksum = "9e9c261f7ce75418b3beadfb3f0eb1299fe8eb9640deba45ffa2cb783098697d"
 dependencies = [
  "ahash 0.8.12",
  "fluent-uri 0.3.2",
@@ -4305,6 +4305,7 @@ checksum = "8f50febec83f5ee1df3015341d8bd429f2d1cc62bcba7ea2076759d315084683"
 name = "test-utils"
 version = "0.1.0"
 dependencies = [
+ "libc",
  "nix 0.26.4",
 ]
 

src/agent/Cargo.toml

@@ -186,7 +186,7 @@ base64 = "0.22"
 sha2 = "0.10.8"
 async-compression = { version = "0.4.22", features = ["tokio", "gzip"] }
 
-container-device-interface = "0.1.0"
+container-device-interface = "0.1.1"
 
 [target.'cfg(target_arch = "s390x")'.dependencies]
 pv_core = { git = "https://github.com/ibm-s390-linux/s390-tools", rev = "4942504a9a2977d49989a5e5b7c1c8e07dc0fa41", package = "s390_pv_core" }

src/agent/rustjail/src/capabilities.rs

@@ -21,7 +21,7 @@ fn to_capshashset(cfd_log: RawFd, capabilities: &Option<HashSet<LinuxCapability>
     let binding: HashSet<LinuxCapability> = HashSet::new();
     let caps = capabilities.as_ref().unwrap_or(&binding);
     for cap in caps.iter() {
-        match Capability::from_str(&format!("CAP_{}", cap)) {
+        match Capability::from_str(&format!("CAP_{cap}")) {
             Err(_) => {
                 log_child!(cfd_log, "{} is not a cap", &cap.to_string());
                 continue;

src/agent/rustjail/src/cgroups/fs/mod.rs

@@ -1097,7 +1097,7 @@ impl Manager {
                             devices_group_info
                         );
                         Self::setup_allowed_all_mode(pod_cg).with_context(|| {
-                            format!("Setup allowed all devices mode for {}", pod_cpath)
+                            format!("Setup allowed all devices mode for {pod_cpath}")
                         })?;
                         devices_group_info.allowed_all = true;
                     }
@@ -1109,11 +1109,11 @@ impl Manager {
 
                     if !is_allowded_all {
                         Self::setup_devcg_whitelist(pod_cg).with_context(|| {
-                            format!("Setup device cgroup whitelist for {}", pod_cpath)
+                            format!("Setup device cgroup whitelist for {pod_cpath}")
                         })?;
                     } else {
                         Self::setup_allowed_all_mode(pod_cg)
-                            .with_context(|| format!("Setup allowed all mode for {}", pod_cpath))?;
+                            .with_context(|| format!("Setup allowed all mode for {pod_cpath}"))?;
                         devices_group_info.allowed_all = true;
                     }
 
@@ -1132,7 +1132,7 @@ impl Manager {
         if let Some(devices_group_info) = devices_group_info.as_ref() {
             if !devices_group_info.allowed_all {
                 Self::setup_devcg_whitelist(&cg)
-                    .with_context(|| format!("Setup device cgroup whitelist for {}", cpath))?;
+                    .with_context(|| format!("Setup device cgroup whitelist for {cpath}"))?;
             }
         }
 

src/agent/rustjail/src/cgroups/systemd/cgroups_path.rs

@@ -57,8 +57,8 @@ fn parse_parent(slice: String) -> Result<String> {
         if subslice.is_empty() {
             return Err(anyhow!("invalid slice name: {}", slice));
         }
-        slice_path = format!("{}/{}{}{}", slice_path, prefix, subslice, SLICE_SUFFIX);
-        prefix = format!("{}{}-", prefix, subslice);
+        slice_path = format!("{slice_path}/{prefix}{subslice}{SLICE_SUFFIX}");
+        prefix = format!("{prefix}{subslice}-");
     }
     slice_path.remove(0);
     Ok(slice_path)
@@ -68,9 +68,9 @@ fn get_unit_name(prefix: String, name: String) -> String {
     if name.ends_with(SLICE_SUFFIX) {
         name
     } else if prefix.is_empty() {
-        format!("{}{}", name, SCOPE_SUFFIX)
+        format!("{name}{SCOPE_SUFFIX}")
     } else {
-        format!("{}-{}{}", prefix, name, SCOPE_SUFFIX)
+        format!("{prefix}-{name}{SCOPE_SUFFIX}")
     }
 }
 

src/agent/rustjail/src/container.rs

@@ -346,7 +346,7 @@ pub fn init_child() {
         Ok(_) => log_child!(cfd_log, "temporary parent process exit successfully"),
         Err(e) => {
             log_child!(cfd_log, "temporary parent process exit:child exit: {:?}", e);
-            let _ = write_sync(cwfd, SYNC_FAILED, format!("{:?}", e).as_str());
+            let _ = write_sync(cwfd, SYNC_FAILED, format!("{e:?}").as_str());
         }
     }
 }
@@ -544,13 +544,13 @@ fn do_init_child(cwfd: RawFd) -> Result<()> {
         sched::setns(fd, s).or_else(|e| {
             if s == CloneFlags::CLONE_NEWUSER {
                 if e != Errno::EINVAL {
-                    let _ = write_sync(cwfd, SYNC_FAILED, format!("{:?}", e).as_str());
+                    let _ = write_sync(cwfd, SYNC_FAILED, format!("{e:?}").as_str());
                     return Err(e);
                 }
 
                 Ok(())
             } else {
-                let _ = write_sync(cwfd, SYNC_FAILED, format!("{:?}", e).as_str());
+                let _ = write_sync(cwfd, SYNC_FAILED, format!("{e:?}").as_str());
                 Err(e)
             }
         })?;
@@ -685,7 +685,7 @@ fn do_init_child(cwfd: RawFd) -> Result<()> {
             let _ = write_sync(
                 cwfd,
                 SYNC_FAILED,
-                format!("setgroups failed: {:?}", e).as_str(),
+                format!("setgroups failed: {e:?}").as_str(),
             );
         })?;
     }
@@ -808,7 +808,7 @@ fn do_init_child(cwfd: RawFd) -> Result<()> {
 
     if init {
         let fd = fcntl::open(
-            format!("/proc/self/fd/{}", fifofd).as_str(),
+            format!("/proc/self/fd/{fifofd}").as_str(),
             OFlag::O_RDONLY | OFlag::O_CLOEXEC,
             Mode::from_bits_truncate(0),
         )?;
@@ -1171,14 +1171,14 @@ impl BaseContainer for LinuxContainer {
             .stderr(child_stderr)
             .env(INIT, format!("{}", p.init))
             .env(NO_PIVOT, format!("{}", self.config.no_pivot_root))
-            .env(CRFD_FD, format!("{}", crfd))
-            .env(CWFD_FD, format!("{}", cwfd))
-            .env(CLOG_FD, format!("{}", cfd_log))
+            .env(CRFD_FD, format!("{crfd}"))
+            .env(CWFD_FD, format!("{cwfd}"))
+            .env(CLOG_FD, format!("{cfd_log}"))
             .env(CONSOLE_SOCKET_FD, console_name)
             .env(PIDNS_ENABLED, format!("{}", pidns.enabled));
 
         if p.init {
-            child = child.env(FIFO_FD, format!("{}", fifofd));
+            child = child.env(FIFO_FD, format!("{fifofd}"));
         }
 
         if pidns.fd.is_some() {
@@ -1588,9 +1588,11 @@ async fn join_namespaces(
         cm.apply(p.pid)?;
     }
 
-    if p.init && res.is_some() {
-        info!(logger, "set properties to cgroups!");
-        cm.set(res.unwrap(), false)?;
+    if p.init {
+        if let Some(resource) = res {
+            info!(logger, "set properties to cgroups!");
+            cm.set(resource, false)?;
+        }
     }
 
     info!(logger, "notify child to continue");
@@ -1687,7 +1689,7 @@ impl LinuxContainer {
                 return anyhow!(e).context(format!("container {} already exists", id.as_str()));
             }
 
-            anyhow!(e).context(format!("fail to create container directory {}", root))
+            anyhow!(e).context(format!("fail to create container directory {root}"))
         })?;
 
         unistd::chown(
@@ -1695,7 +1697,7 @@ impl LinuxContainer {
             Some(unistd::getuid()),
             Some(unistd::getgid()),
         )
-        .context(format!("Cannot change owner of container {} root", id))?;
+        .context(format!("Cannot change owner of container {id} root"))?;
 
         let spec = config.spec.as_ref().unwrap();
         let linux_cgroups_path = spec

src/agent/rustjail/src/mount.rs

@@ -528,7 +528,7 @@ pub fn pivot_rootfs<P: ?Sized + NixPath + std::fmt::Debug>(path: &P) -> Result<(
 
     // Change to the new root so that the pivot_root actually acts on it.
     unistd::fchdir(newroot)?;
-    pivot_root(".", ".").context(format!("failed to pivot_root on {:?}", path))?;
+    pivot_root(".", ".").context(format!("failed to pivot_root on {path:?}"))?;
 
     // Currently our "." is oldroot (according to the current kernel code).
     // However, purely for safety, we will fchdir(oldroot) since there isn't
@@ -752,15 +752,6 @@ fn parse_mount(m: &Mount) -> (MsFlags, MsFlags, String) {
     (flags, pgflags, data.join(","))
 }
 
-// This function constructs a canonicalized path by combining the `rootfs` and `unsafe_path` elements.
-// The resulting path is guaranteed to be ("below" / "in a directory under") the `rootfs` directory.
-//
-// Parameters:
-//
-// - `rootfs` is the absolute path to the root of the containers root filesystem directory.
-// - `unsafe_path` is path inside a container. It is unsafe since it may try to "escape" from the containers
-//    rootfs by using one or more "../" path elements or is its a symlink to path.
-
 fn mount_from(
     cfd_log: RawFd,
     m: &Mount,
@@ -929,7 +920,7 @@ fn create_devices(devices: &[LinuxDevice], bind: bool) -> Result<()> {
     for dev in DEFAULT_DEVICES.iter() {
         let dev_path = dev.path().display().to_string();
         let path = Path::new(&dev_path[1..]);
-        op(dev, path).context(format!("Creating container device {:?}", dev))?;
+        op(dev, path).context(format!("Creating container device {dev:?}"))?;
     }
     for dev in devices {
         let dev_path = &dev.path();
@@ -941,9 +932,9 @@ fn create_devices(devices: &[LinuxDevice], bind: bool) -> Result<()> {
             anyhow!(msg)
         })?;
         if let Some(dir) = path.parent() {
-            fs::create_dir_all(dir).context(format!("Creating container device {:?}", dev))?;
+            fs::create_dir_all(dir).context(format!("Creating container device {dev:?}"))?;
         }
-        op(dev, path).context(format!("Creating container device {:?}", dev))?;
+        op(dev, path).context(format!("Creating container device {dev:?}"))?;
     }
     stat::umask(old);
     Ok(())

src/agent/rustjail/src/selinux.rs

@@ -18,10 +18,10 @@ pub fn is_enabled() -> Result<bool> {
 
 pub fn add_mount_label(data: &mut String, label: &str) {
     if data.is_empty() {
-        let context = format!("context=\"{}\"", label);
+        let context = format!("context=\"{label}\"");
         data.push_str(&context);
     } else {
-        let context = format!(",context=\"{}\"", label);
+        let context = format!(",context=\"{label}\"");
         data.push_str(&context);
     }
 }

src/agent/src/ap.rs

@@ -68,7 +68,7 @@ mod tests {
     #[test]
     fn test_from_str() {
         let device = Address::from_str("a.1").unwrap();
-        assert_eq!(format!("{}", device), "0a.0001");
+        assert_eq!(format!("{device}"), "0a.0001");
 
         assert!(Address::from_str("").is_err());
         assert!(Address::from_str(".").is_err());

src/agent/src/ccw.rs

@@ -102,10 +102,10 @@ mod tests {
     fn test_new_device() {
         // Valid devices
         let device = Device::new(0, 0).unwrap();
-        assert_eq!(format!("{}", device), "0.0.0000");
+        assert_eq!(format!("{device}"), "0.0.0000");
 
         let device = Device::new(3, 0xffff).unwrap();
-        assert_eq!(format!("{}", device), "0.3.ffff");
+        assert_eq!(format!("{device}"), "0.3.ffff");
 
         // Invalid device
         let device = Device::new(4, 0);
@@ -116,13 +116,13 @@ mod tests {
     fn test_device_from_str() {
         // Valid devices
         let device = Device::from_str("0.0.0").unwrap();
-        assert_eq!(format!("{}", device), "0.0.0000");
+        assert_eq!(format!("{device}"), "0.0.0000");
 
         let device = Device::from_str("0.0.0000").unwrap();
-        assert_eq!(format!("{}", device), "0.0.0000");
+        assert_eq!(format!("{device}"), "0.0.0000");
 
         let device = Device::from_str("0.3.ffff").unwrap();
-        assert_eq!(format!("{}", device), "0.3.ffff");
+        assert_eq!(format!("{device}"), "0.3.ffff");
 
         // Invalid devices
         let device = Device::from_str("0.0");

src/agent/src/confidential_data_hub/mod.rs

@@ -110,7 +110,7 @@ impl CDHClient {
 
     pub async fn get_resource(&self, resource_path: &str) -> Result<Vec<u8>> {
         let req = GetResourceRequest {
-            ResourcePath: format!("kbs://{}", resource_path),
+            ResourcePath: format!("kbs://{resource_path}"),
             ..Default::default()
         };
         let res = self

src/agent/src/config.rs

@@ -260,7 +260,7 @@ impl Default for AgentConfig {
             debug_console_vport: 0,
             log_vport: 0,
             container_pipe_size: DEFAULT_CONTAINER_PIPE_SIZE,
-            server_addr: format!("{}:{}", VSOCK_ADDR, DEFAULT_AGENT_VSOCK_PORT),
+            server_addr: format!("{VSOCK_ADDR}:{DEFAULT_AGENT_VSOCK_PORT}"),
             passfd_listener_port: 0,
             cgroup_no_v1: String::from(""),
             unified_cgroup_hierarchy: false,
@@ -269,7 +269,7 @@ impl Default for AgentConfig {
             no_proxy: String::from(""),
             guest_components_rest_api: GuestComponentsFeatures::default(),
             guest_components_procs: GuestComponentsProcs::default(),
-            secure_storage_integrity: false,
+            secure_storage_integrity: true,
             #[cfg(feature = "agent-policy")]
             policy_file: String::from(""),
             mem_agent: None,
@@ -417,7 +417,7 @@ impl AgentConfig {
             // generate our config from it.
             // The agent will fail to start if the configuration file is not present,
             // or if it can't be parsed properly.
-            if param.starts_with(format!("{}=", CONFIG_FILE).as_str()) {
+            if param.starts_with(format!("{CONFIG_FILE}=").as_str()) {
                 let config_file = get_string_value(param)?;
                 return AgentConfig::from_config_file(&config_file)
                     .context("AgentConfig from kernel cmdline");
@@ -651,7 +651,7 @@ impl AgentConfig {
     #[instrument]
     pub fn from_config_file(file: &str) -> Result<AgentConfig> {
         let config = fs::read_to_string(file)
-            .with_context(|| format!("Failed to read config file {}", file))?;
+            .with_context(|| format!("Failed to read config file {file}"))?;
         AgentConfig::from_str(&config)
     }
 
@@ -668,7 +668,7 @@ impl AgentConfig {
         }
 
         if let Ok(value) = env::var(TRACING_ENV_VAR) {
-            let name_value = format!("{}={}", TRACING_ENV_VAR, value);
+            let name_value = format!("{TRACING_ENV_VAR}={value}");
 
             self.tracing = get_bool_value(&name_value).unwrap_or(false);
         }
@@ -911,7 +911,7 @@ mod tests {
                     no_proxy: "",
                     guest_components_rest_api: GuestComponentsFeatures::default(),
                     guest_components_procs: GuestComponentsProcs::default(),
-                    secure_storage_integrity: false,
+                    secure_storage_integrity: true,
                     #[cfg(feature = "agent-policy")]
                     policy_file: "",
                     mem_agent: None,
@@ -1364,7 +1364,7 @@ mod tests {
             },
             TestData {
                 contents: "",
-                secure_storage_integrity: false,
+                secure_storage_integrity: true,
                 ..Default::default()
             },
             TestData {
@@ -1442,7 +1442,7 @@ mod tests {
         // Now, test various combinations of file contents and environment
         // variables.
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let file_path = dir.path().join("cmdline");
 
@@ -1470,40 +1470,36 @@ mod tests {
             let config =
                 AgentConfig::from_cmdline(filename, vec![]).expect("Failed to parse command line");
 
-            assert_eq!(d.debug_console, config.debug_console, "{}", msg);
-            assert_eq!(d.dev_mode, config.dev_mode, "{}", msg);
-            assert_eq!(d.cgroup_no_v1, config.cgroup_no_v1, "{}", msg);
+            assert_eq!(d.debug_console, config.debug_console, "{msg}");
+            assert_eq!(d.dev_mode, config.dev_mode, "{msg}");
+            assert_eq!(d.cgroup_no_v1, config.cgroup_no_v1, "{msg}");
             assert_eq!(
                 d.unified_cgroup_hierarchy, config.unified_cgroup_hierarchy,
-                "{}",
-                msg
+                "{msg}"
             );
-            assert_eq!(d.log_level, config.log_level, "{}", msg);
-            assert_eq!(d.hotplug_timeout, config.hotplug_timeout, "{}", msg);
-            assert_eq!(d.container_pipe_size, config.container_pipe_size, "{}", msg);
-            assert_eq!(d.server_addr, config.server_addr, "{}", msg);
-            assert_eq!(d.tracing, config.tracing, "{}", msg);
-            assert_eq!(d.https_proxy, config.https_proxy, "{}", msg);
-            assert_eq!(d.no_proxy, config.no_proxy, "{}", msg);
+            assert_eq!(d.log_level, config.log_level, "{msg}");
+            assert_eq!(d.hotplug_timeout, config.hotplug_timeout, "{msg}");
+            assert_eq!(d.container_pipe_size, config.container_pipe_size, "{msg}");
+            assert_eq!(d.server_addr, config.server_addr, "{msg}");
+            assert_eq!(d.tracing, config.tracing, "{msg}");
+            assert_eq!(d.https_proxy, config.https_proxy, "{msg}");
+            assert_eq!(d.no_proxy, config.no_proxy, "{msg}");
             assert_eq!(
                 d.guest_components_rest_api, config.guest_components_rest_api,
-                "{}",
-                msg
+                "{msg}"
             );
             assert_eq!(
                 d.guest_components_procs, config.guest_components_procs,
-                "{}",
-                msg
+                "{msg}"
             );
             assert_eq!(
                 d.secure_storage_integrity, config.secure_storage_integrity,
-                "{}",
-                msg
+                "{msg}"
             );
             #[cfg(feature = "agent-policy")]
-            assert_eq!(d.policy_file, config.policy_file, "{}", msg);
+            assert_eq!(d.policy_file, config.policy_file, "{msg}");
 
-            assert_eq!(d.mem_agent, config.mem_agent, "{}", msg);
+            assert_eq!(d.mem_agent, config.mem_agent, "{msg}");
 
             for v in vars_to_unset {
                 env::remove_var(v);
@@ -1568,7 +1564,7 @@ mod tests {
     #[case("panic", Ok(slog::Level::Critical))]
     fn test_logrus_to_slog_level(#[case] input: &str, #[case] expected: Result<slog::Level>) {
         let result = logrus_to_slog_level(input);
-        let msg = format!("expected: {:?}, result: {:?}", expected, result);
+        let msg = format!("expected: {expected:?}, result: {result:?}");
         assert_result!(expected, result, msg);
     }
 
@@ -1593,7 +1589,7 @@ mod tests {
     #[case("agent.log=panic", Ok(slog::Level::Critical))]
     fn test_get_log_level(#[case] input: &str, #[case] expected: Result<slog::Level>) {
         let result = get_log_level(input);
-        let msg = format!("expected: {:?}, result: {:?}", expected, result);
+        let msg = format!("expected: {expected:?}, result: {result:?}");
         assert_result!(expected, result, msg);
     }
 
@@ -1636,7 +1632,7 @@ Caused by:
     #[case("agent.cdi_timeout=320", Ok(time::Duration::from_secs(320)))]
     fn test_timeout(#[case] param: &str, #[case] expected: Result<time::Duration>) {
         let result = get_timeout(param);
-        let msg = format!("expected: {:?}, result: {:?}", expected, result);
+        let msg = format!("expected: {expected:?}, result: {result:?}");
         assert_result!(expected, result, msg);
     }
 
@@ -1676,7 +1672,7 @@ Caused by:
     )))]
     fn test_get_container_pipe_size(#[case] param: &str, #[case] expected: Result<i32>) {
         let result = get_container_pipe_size(param);
-        let msg = format!("expected: {:?}, result: {:?}", expected, result);
+        let msg = format!("expected: {expected:?}, result: {result:?}");
         assert_result!(expected, result, msg);
     }
 
@@ -1697,7 +1693,7 @@ Caused by:
     #[case("x= = ", Ok(" = ".into()))]
     fn test_get_string_value(#[case] param: &str, #[case] expected: Result<String>) {
         let result = get_string_value(param);
-        let msg = format!("expected: {:?}, result: {:?}", expected, result);
+        let msg = format!("expected: {expected:?}, result: {result:?}");
         assert_result!(expected, result, msg);
     }
 
@@ -1716,7 +1712,7 @@ Caused by:
         #[case] expected: Result<GuestComponentsFeatures>,
     ) {
         let result = get_guest_components_features_value(input);
-        let msg = format!("expected: {:?}, result: {:?}", expected, result);
+        let msg = format!("expected: {expected:?}, result: {result:?}");
         assert_result!(expected, result, msg);
     }
 
@@ -1739,7 +1735,7 @@ Caused by:
         #[case] expected: Result<GuestComponentsProcs>,
     ) {
         let result = get_guest_components_procs_value(param);
-        let msg = format!("expected: {:?}, result: {:?}", expected, result);
+        let msg = format!("expected: {expected:?}, result: {result:?}");
         assert_result!(expected, result, msg);
     }
 

src/agent/src/device/block_device_handler.rs

@@ -16,7 +16,10 @@ use crate::pci;
 use crate::sandbox::Sandbox;
 use crate::uevent::{wait_for_uevent, Uevent, UeventMatcher};
 use anyhow::{anyhow, Context, Result};
-use kata_types::device::{DRIVER_BLK_CCW_TYPE, DRIVER_BLK_MMIO_TYPE, DRIVER_BLK_PCI_TYPE};
+
+#[cfg(target_arch = "s390x")]
+use kata_types::device::DRIVER_BLK_CCW_TYPE;
+use kata_types::device::{DRIVER_BLK_MMIO_TYPE, DRIVER_BLK_PCI_TYPE};
 use protocols::agent::Device;
 use regex::Regex;
 use std::path::Path;
@@ -28,6 +31,7 @@ use tracing::instrument;
 #[derive(Debug)]
 pub struct VirtioBlkPciDeviceHandler {}
 
+#[cfg(target_arch = "s390x")]
 #[derive(Debug)]
 pub struct VirtioBlkCcwDeviceHandler {}
 
@@ -52,6 +56,7 @@ impl DeviceHandler for VirtioBlkPciDeviceHandler {
     }
 }
 
+#[cfg(target_arch = "s390x")]
 #[async_trait::async_trait]
 impl DeviceHandler for VirtioBlkCcwDeviceHandler {
     #[instrument]
@@ -164,7 +169,7 @@ pub struct VirtioBlkPciMatcher {
 impl VirtioBlkPciMatcher {
     pub fn new(relpath: &str) -> VirtioBlkPciMatcher {
         let root_bus = create_pci_root_bus_path();
-        let re = format!(r"^{}{}/virtio[0-9]+/block/", root_bus, relpath);
+        let re = format!(r"^{root_bus}{relpath}/virtio[0-9]+/block/");
 
         VirtioBlkPciMatcher {
             rex: Regex::new(&re).expect("BUG: failed to compile VirtioBlkPciMatcher regex"),
@@ -186,7 +191,7 @@ pub struct VirtioBlkMmioMatcher {
 impl VirtioBlkMmioMatcher {
     pub fn new(devname: &str) -> VirtioBlkMmioMatcher {
         VirtioBlkMmioMatcher {
-            suffix: format!(r"/block/{}", devname),
+            suffix: format!(r"/block/{devname}"),
         }
     }
 }
@@ -206,10 +211,8 @@ pub struct VirtioBlkCCWMatcher {
 #[cfg(target_arch = "s390x")]
 impl VirtioBlkCCWMatcher {
     pub fn new(root_bus_path: &str, device: &ccw::Device) -> Self {
-        let re = format!(
-            r"^{}/0\.[0-3]\.[0-9a-f]{{1,4}}/{}/virtio[0-9]+/block/",
-            root_bus_path, device
-        );
+        let re =
+            format!(r"^{root_bus_path}/0\.[0-3]\.[0-9a-f]{{1,4}}/{device}/virtio[0-9]+/block/");
         VirtioBlkCCWMatcher {
             rex: Regex::new(&re).expect("BUG: failed to compile VirtioBlkCCWMatcher regex"),
         }
@@ -238,12 +241,12 @@ mod tests {
         uev_a.action = crate::linux_abi::U_EVENT_ACTION_ADD.to_string();
         uev_a.subsystem = BLOCK.to_string();
         uev_a.devname = devname.to_string();
-        uev_a.devpath = format!("{}{}/virtio4/block/{}", root_bus, relpath_a, devname);
+        uev_a.devpath = format!("{root_bus}{relpath_a}/virtio4/block/{devname}");
         let matcher_a = VirtioBlkPciMatcher::new(relpath_a);
 
         let mut uev_b = uev_a.clone();
         let relpath_b = "/0000:00:0a.0/0000:00:0b.0";
-        uev_b.devpath = format!("{}{}/virtio0/block/{}", root_bus, relpath_b, devname);
+        uev_b.devpath = format!("{root_bus}{relpath_b}/virtio0/block/{devname}");
         let matcher_b = VirtioBlkPciMatcher::new(relpath_b);
 
         assert!(matcher_a.is_match(&uev_a));
@@ -264,10 +267,7 @@ mod tests {
         uev.action = crate::linux_abi::U_EVENT_ACTION_ADD.to_string();
         uev.subsystem = subsystem.to_string();
         uev.devname = devname.to_string();
-        uev.devpath = format!(
-            "{}/0.0.0001/{}/virtio1/{}/{}",
-            root_bus, relpath, subsystem, devname
-        );
+        uev.devpath = format!("{root_bus}/0.0.0001/{relpath}/virtio1/{subsystem}/{devname}");
 
         // Valid path
         let device = ccw::Device::from_str(relpath).unwrap();
@@ -275,40 +275,25 @@ mod tests {
         assert!(matcher.is_match(&uev));
 
         // Invalid paths
-        uev.devpath = format!(
-            "{}/0.0.0001/0.0.0003/virtio1/{}/{}",
-            root_bus, subsystem, devname
-        );
+        uev.devpath = format!("{root_bus}/0.0.0001/0.0.0003/virtio1/{subsystem}/{devname}");
         assert!(!matcher.is_match(&uev));
 
-        uev.devpath = format!("0.0.0001/{}/virtio1/{}/{}", relpath, subsystem, devname);
+        uev.devpath = format!("0.0.0001/{relpath}/virtio1/{subsystem}/{devname}");
         assert!(!matcher.is_match(&uev));
 
-        uev.devpath = format!(
-            "{}/0.0.0001/{}/virtio/{}/{}",
-            root_bus, relpath, subsystem, devname
-        );
+        uev.devpath = format!("{root_bus}/0.0.0001/{relpath}/virtio/{subsystem}/{devname}");
         assert!(!matcher.is_match(&uev));
 
-        uev.devpath = format!("{}/0.0.0001/{}/virtio1", root_bus, relpath);
+        uev.devpath = format!("{root_bus}/0.0.0001/{relpath}/virtio1");
         assert!(!matcher.is_match(&uev));
 
-        uev.devpath = format!(
-            "{}/1.0.0001/{}/virtio1/{}/{}",
-            root_bus, relpath, subsystem, devname
-        );
+        uev.devpath = format!("{root_bus}/1.0.0001/{relpath}/virtio1/{subsystem}/{devname}");
         assert!(!matcher.is_match(&uev));
 
-        uev.devpath = format!(
-            "{}/0.4.0001/{}/virtio1/{}/{}",
-            root_bus, relpath, subsystem, devname
-        );
+        uev.devpath = format!("{root_bus}/0.4.0001/{relpath}/virtio1/{subsystem}/{devname}");
         assert!(!matcher.is_match(&uev));
 
-        uev.devpath = format!(
-            "{}/0.0.10000/{}/virtio1/{}/{}",
-            root_bus, relpath, subsystem, devname
-        );
+        uev.devpath = format!("{root_bus}/0.0.10000/{relpath}/virtio1/{subsystem}/{devname}");
         assert!(!matcher.is_match(&uev));
     }
 
@@ -321,17 +306,13 @@ mod tests {
         uev_a.action = crate::linux_abi::U_EVENT_ACTION_ADD.to_string();
         uev_a.subsystem = BLOCK.to_string();
         uev_a.devname = devname_a.to_string();
-        uev_a.devpath = format!(
-            "/sys/devices/virtio-mmio-cmdline/virtio-mmio.0/virtio0/block/{}",
-            devname_a
-        );
+        uev_a.devpath =
+            format!("/sys/devices/virtio-mmio-cmdline/virtio-mmio.0/virtio0/block/{devname_a}");
         let matcher_a = VirtioBlkMmioMatcher::new(devname_a);
 
         let mut uev_b = uev_a.clone();
-        uev_b.devpath = format!(
-            "/sys/devices/virtio-mmio-cmdline/virtio-mmio.4/virtio4/block/{}",
-            devname_b
-        );
+        uev_b.devpath =
+            format!("/sys/devices/virtio-mmio-cmdline/virtio-mmio.4/virtio4/block/{devname_b}");
         let matcher_b = VirtioBlkMmioMatcher::new(devname_b);
 
         assert!(matcher_a.is_match(&uev_a));

src/agent/src/device/mod.rs

@@ -425,7 +425,7 @@ pub fn update_env_pci(
         let mut guest_addrs = Vec::<String>::new();
         for host_addr_str in val.split(',') {
             let host_addr = pci::Address::from_str(host_addr_str)
-                .with_context(|| format!("Can't parse {} environment variable", name))?;
+                .with_context(|| format!("Can't parse {name} environment variable"))?;
             let host_guest = pcimap
                 .get(cid)
                 .ok_or_else(|| anyhow!("No PCI mapping found for container {}", cid))?;
@@ -433,11 +433,11 @@ pub fn update_env_pci(
                 .get(&host_addr)
                 .ok_or_else(|| anyhow!("Unable to translate host PCI address {}", host_addr))?;
 
-            guest_addrs.push(format!("{}", guest_addr));
-            addr_map.insert(host_addr_str.to_string(), format!("{}", guest_addr));
+            guest_addrs.push(format!("{guest_addr}"));
+            addr_map.insert(host_addr_str.to_string(), format!("{guest_addr}"));
         }
 
-        pci_dev_map.insert(format!("{}_INFO", name), addr_map);
+        pci_dev_map.insert(format!("{name}_INFO"), addr_map);
 
         envvar.replace_range(eqpos + 1.., guest_addrs.join(",").as_str());
     }
@@ -526,7 +526,7 @@ fn update_spec_devices(
             "Missing devices in OCI spec: {:?}",
             updates
                 .keys()
-                .map(|d| format!("{:?}", d))
+                .map(|d| format!("{d:?}"))
                 .collect::<Vec<_>>()
                 .join(" ")
         ));
@@ -572,7 +572,7 @@ pub fn pcipath_to_sysfs(root_bus_sysfs: &str, pcipath: &pci::Path) -> Result<Str
     for i in 0..pcipath.len() {
         let bdf = format!("{}:{}", bus, pcipath[i]);
 
-        relpath = format!("{}/{}", relpath, bdf);
+        relpath = format!("{relpath}/{bdf}");
 
         if i == pcipath.len() - 1 {
             // Final device need not be a bridge
@@ -580,7 +580,7 @@ pub fn pcipath_to_sysfs(root_bus_sysfs: &str, pcipath: &pci::Path) -> Result<Str
         }
 
         // Find out the bus exposed by bridge
-        let bridgebuspath = format!("{}{}/pci_bus", root_bus_sysfs, relpath);
+        let bridgebuspath = format!("{root_bus_sysfs}{relpath}/pci_bus");
         let mut files: Vec<_> = fs::read_dir(&bridgebuspath)?.collect();
 
         match files.pop() {
@@ -1120,7 +1120,7 @@ mod tests {
 
         // Create mock sysfs files to indicate that 0000:00:02.0 is a bridge to bus 01
         let bridge2bus = "0000:01";
-        let bus2path = format!("{}/pci_bus/{}", bridge2path, bridge2bus);
+        let bus2path = format!("{bridge2path}/pci_bus/{bridge2bus}");
 
         fs::create_dir_all(bus2path).unwrap();
 
@@ -1134,9 +1134,9 @@ mod tests {
         assert!(relpath.is_err());
 
         // Create mock sysfs files for a bridge at 0000:01:03.0 to bus 02
-        let bridge3path = format!("{}/0000:01:03.0", bridge2path);
+        let bridge3path = format!("{bridge2path}/0000:01:03.0");
         let bridge3bus = "0000:02";
-        let bus3path = format!("{}/pci_bus/{}", bridge3path, bridge3bus);
+        let bus3path = format!("{bridge3path}/pci_bus/{bridge3bus}");
 
         fs::create_dir_all(bus3path).unwrap();
 
@@ -1169,7 +1169,7 @@ mod tests {
         let devname = "vda";
         let root_bus = create_pci_root_bus_path();
         let relpath = "/0000:00:0a.0/0000:03:0b.0";
-        let devpath = format!("{}{}/virtio4/block/{}", root_bus, relpath, devname);
+        let devpath = format!("{root_bus}{relpath}/virtio4/block/{devname}");
 
         let mut uev = crate::uevent::Uevent::default();
         uev.action = crate::linux_abi::U_EVENT_ACTION_ADD.to_string();
@@ -1272,7 +1272,7 @@ mod tests {
             cdi_timeout,
         )
         .await;
-        println!("modfied spec {:?}", spec);
+        println!("modfied spec {spec:?}");
         assert!(res.is_ok(), "{}", res.err().unwrap());
 
         let linux = spec.linux().as_ref().unwrap();

src/agent/src/device/network_device_handler.rs

@@ -69,7 +69,7 @@ impl NetPciMatcher {
         let root_bus = create_pci_root_bus_path();
 
         NetPciMatcher {
-            devpath: format!("{}{}", root_bus, relpath),
+            devpath: format!("{root_bus}{relpath}"),
         }
     }
 }
@@ -106,10 +106,7 @@ struct NetCcwMatcher {
 #[cfg(target_arch = "s390x")]
 impl NetCcwMatcher {
     pub fn new(root_bus_path: &str, device: &ccw::Device) -> Self {
-        let re = format!(
-            r"{}/0\.[0-3]\.[0-9a-f]{{1,4}}/{}/virtio[0-9]+/net/",
-            root_bus_path, device
-        );
+        let re = format!(r"{root_bus_path}/0\.[0-3]\.[0-9a-f]{{1,4}}/{device}/virtio[0-9]+/net/");
         NetCcwMatcher {
             re: Regex::new(&re).expect("BUG: failed to compile NetCCWMatcher regex"),
         }
@@ -139,7 +136,7 @@ mod tests {
 
         let mut uev_a = crate::uevent::Uevent::default();
         uev_a.action = crate::linux_abi::U_EVENT_ACTION_ADD.to_string();
-        uev_a.devpath = format!("{}{}", root_bus, relpath_a);
+        uev_a.devpath = format!("{root_bus}{relpath_a}");
         uev_a.subsystem = String::from("net");
         uev_a.interface = String::from("eth0");
         let matcher_a = NetPciMatcher::new(relpath_a);
@@ -147,7 +144,7 @@ mod tests {
 
         let relpath_b = "/0000:00:02.0/0000:01:02.0";
         let mut uev_b = uev_a.clone();
-        uev_b.devpath = format!("{}{}", root_bus, relpath_b);
+        uev_b.devpath = format!("{root_bus}{relpath_b}");
         let matcher_b = NetPciMatcher::new(relpath_b);
 
         assert!(matcher_a.is_match(&uev_a));
@@ -158,7 +155,7 @@ mod tests {
         let relpath_c = "/0000:00:02.0/0000:01:03.0";
         let net_substr = "/net/eth0";
         let mut uev_c = uev_a.clone();
-        uev_c.devpath = format!("{}{}{}", root_bus, relpath_c, net_substr);
+        uev_c.devpath = format!("{root_bus}{relpath_c}{net_substr}");
         let matcher_c = NetPciMatcher::new(relpath_c);
 
         assert!(matcher_c.is_match(&uev_c));

src/agent/src/device/nvdimm_device_handler.rs

@@ -67,7 +67,7 @@ pub struct PmemBlockMatcher {
 
 impl PmemBlockMatcher {
     pub fn new(devname: &str) -> PmemBlockMatcher {
-        let suffix = format!(r"/block/{}", devname);
+        let suffix = format!(r"/block/{devname}");
 
         PmemBlockMatcher { suffix }
     }

src/agent/src/device/scsi_device_handler.rs

@@ -58,7 +58,7 @@ pub struct ScsiBlockMatcher {
 
 impl ScsiBlockMatcher {
     pub fn new(scsi_addr: &str) -> ScsiBlockMatcher {
-        let search = format!(r"/0:0:{}/block/", scsi_addr);
+        let search = format!(r"/0:0:{scsi_addr}/block/");
 
         ScsiBlockMatcher { search }
     }
@@ -118,18 +118,14 @@ mod tests {
         uev_a.action = crate::linux_abi::U_EVENT_ACTION_ADD.to_string();
         uev_a.subsystem = BLOCK.to_string();
         uev_a.devname = devname.to_string();
-        uev_a.devpath = format!(
-            "{}/0000:00:00.0/virtio0/host0/target0:0:0/0:0:{}/block/sda",
-            root_bus, addr_a
-        );
+        uev_a.devpath =
+            format!("{root_bus}/0000:00:00.0/virtio0/host0/target0:0:0/0:0:{addr_a}/block/sda");
         let matcher_a = ScsiBlockMatcher::new(addr_a);
 
         let mut uev_b = uev_a.clone();
         let addr_b = "2:0";
-        uev_b.devpath = format!(
-            "{}/0000:00:00.0/virtio0/host0/target0:0:2/0:0:{}/block/sdb",
-            root_bus, addr_b
-        );
+        uev_b.devpath =
+            format!("{root_bus}/0000:00:00.0/virtio0/host0/target0:0:2/0:0:{addr_b}/block/sdb");
         let matcher_b = ScsiBlockMatcher::new(addr_b);
 
         assert!(matcher_a.is_match(&uev_a));

src/agent/src/device/vfio_device_handler.rs

@@ -170,7 +170,7 @@ pub struct VfioMatcher {
 impl VfioMatcher {
     pub fn new(grp: IommuGroup) -> VfioMatcher {
         VfioMatcher {
-            syspath: format!("/devices/virtual/vfio/{}", grp),
+            syspath: format!("/devices/virtual/vfio/{grp}"),
         }
     }
 }
@@ -215,7 +215,7 @@ impl PciMatcher {
     pub fn new(relpath: &str) -> Result<PciMatcher> {
         let root_bus = create_pci_root_bus_path();
         Ok(PciMatcher {
-            devpath: format!("{}{}", root_bus, relpath),
+            devpath: format!("{root_bus}{relpath}"),
         })
     }
 }
@@ -425,12 +425,12 @@ mod tests {
 
         let mut uev_a = crate::uevent::Uevent::default();
         uev_a.action = crate::linux_abi::U_EVENT_ACTION_ADD.to_string();
-        uev_a.devname = format!("vfio/{}", grpa);
-        uev_a.devpath = format!("/devices/virtual/vfio/{}", grpa);
+        uev_a.devname = format!("vfio/{grpa}");
+        uev_a.devpath = format!("/devices/virtual/vfio/{grpa}");
         let matcher_a = VfioMatcher::new(grpa);
 
         let mut uev_b = uev_a.clone();
-        uev_b.devpath = format!("/devices/virtual/vfio/{}", grpb);
+        uev_b.devpath = format!("/devices/virtual/vfio/{grpb}");
         let matcher_b = VfioMatcher::new(grpb);
 
         assert!(matcher_a.is_match(&uev_a));
@@ -531,12 +531,12 @@ mod tests {
     async fn test_vfio_ap_matcher() {
         let subsystem = "ap";
         let card = "0a";
-        let relpath = format!("{}.0001", card);
+        let relpath = format!("{card}.0001");
 
         let mut uev = Uevent::default();
         uev.action = U_EVENT_ACTION_ADD.to_string();
         uev.subsystem = subsystem.to_string();
-        uev.devpath = format!("{}/card{}/{}", AP_ROOT_BUS_PATH, card, relpath);
+        uev.devpath = format!("{AP_ROOT_BUS_PATH}/card{card}/{relpath}");
 
         let ap_address = ap::Address::from_str(&relpath).unwrap();
         let matcher = ApMatcher::new(ap_address);
@@ -548,7 +548,7 @@ mod tests {
         assert!(!matcher.is_match(&uev_remove));
 
         let mut uev_other_device = uev.clone();
-        uev_other_device.devpath = format!("{}/card{}/{}.0002", AP_ROOT_BUS_PATH, card, card);
+        uev_other_device.devpath = format!("{AP_ROOT_BUS_PATH}/card{card}/{card}.0002");
         assert!(!matcher.is_match(&uev_other_device));
     }
 }

src/agent/src/main.rs

@@ -301,12 +301,12 @@ async fn real_main(init_mode: bool) -> std::result::Result<(), Box<dyn std::erro
         tracer::end_tracing();
     }
 
-    eprintln!("{} shutdown complete", NAME);
+    eprintln!("{NAME} shutdown complete");
 
     let mut wait_errors: Vec<tokio::task::JoinError> = vec![];
     for result in results {
         if let Err(e) = result {
-            eprintln!("wait task error: {:#?}", e);
+            eprintln!("wait task error: {e:#?}");
             wait_errors.push(e);
         }
     }
@@ -746,7 +746,7 @@ mod tests {
                 skip_if_root!();
             }
 
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
             let (rfd, wfd) = unistd::pipe2(OFlag::O_CLOEXEC).unwrap();
             defer!({
                 // XXX: Never try to close rfd, because it will be closed by PipeStream in
@@ -759,7 +759,7 @@ mod tests {
             shutdown_tx.send(true).unwrap();
             let result = create_logger_task(rfd, d.vsock_port, shutdown_rx).await;
 
-            let msg = format!("{}, result: {:?}", msg, result);
+            let msg = format!("{msg}, result: {result:?}");
             assert_result!(d.result, result, msg);
         }
     }

src/agent/src/metrics.rs

@@ -239,7 +239,7 @@ fn update_guest_metrics() {
         Ok(kernel_stats) => {
             set_gauge_vec_cpu_time(&GUEST_CPU_TIME, "total", &kernel_stats.total);
             for (i, cpu_time) in kernel_stats.cpu_time.iter().enumerate() {
-                set_gauge_vec_cpu_time(&GUEST_CPU_TIME, format!("{}", i).as_str(), cpu_time);
+                set_gauge_vec_cpu_time(&GUEST_CPU_TIME, format!("{i}").as_str(), cpu_time);
             }
         }
     }

src/agent/src/mount.rs

@@ -88,7 +88,7 @@ pub fn baremount(
 
     let destination_str = destination.to_string_lossy();
     if let Ok(m) = get_linux_mount_info(destination_str.deref()) {
-        if m.fs_type == fs_type {
+        if m.fs_type == fs_type && !flags.contains(MsFlags::MS_REMOUNT) {
             slog_info!(logger, "{source:?} is already mounted at {destination:?}");
             return Ok(());
         }
@@ -177,7 +177,7 @@ pub fn get_mount_fs_type_from_file(mount_file: &str, mount_point: &str) -> Resul
     let content = fs::read_to_string(mount_file)
         .map_err(|e| anyhow!("read mount file {}: {}", mount_file, e))?;
 
-    let re = Regex::new(format!("device .+ mounted on {} with fstype (.+)", mount_point).as_str())?;
+    let re = Regex::new(format!("device .+ mounted on {mount_point} with fstype (.+)").as_str())?;
 
     // Read the file line by line using the lines() iterator from std::io::BufRead.
     for line in content.lines() {
@@ -355,8 +355,7 @@ mod tests {
             let flags = MsFlags::MS_RDONLY;
             let options = "mode=755";
             println!(
-                "testing if already mounted baremount({:?} {:?} {:?})",
-                source, destination, fs_type
+                "testing if already mounted baremount({source:?} {destination:?} {fs_type:?})"
             );
             assert!(baremount(source, destination, fs_type, flags, options, &logger).is_ok());
         }
@@ -461,7 +460,7 @@ mod tests {
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             skip_loop_by_user!(msg, d.test_user);
 
@@ -505,7 +504,7 @@ mod tests {
 
             let result = baremount(src, dest, d.fs_type, d.flags, d.options, &logger);
 
-            let msg = format!("{}: result: {:?}", msg, result);
+            let msg = format!("{msg}: result: {result:?}");
 
             if d.error_contains.is_empty() {
                 assert!(result.is_ok(), "{}", msg);
@@ -517,7 +516,7 @@ mod tests {
             }
 
             let err = result.unwrap_err();
-            let error_msg = format!("{}", err);
+            let error_msg = format!("{err}");
 
             assert!(
                 error_msg.contains(d.error_contains),
@@ -619,11 +618,11 @@ mod tests {
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let result = remove_mounts(&d.mounts);
 
-            let msg = format!("{}: result: {:?}", msg, result);
+            let msg = format!("{msg}: result: {result:?}");
 
             if d.error_contains.is_empty() {
                 assert!(result.is_ok(), "{}", msg);
@@ -699,15 +698,15 @@ mod tests {
             .iter()
             .enumerate()
         {
-            let msg = format!("missing mount file test[{}] with mountpoint: {}", i, mp);
+            let msg = format!("missing mount file test[{i}] with mountpoint: {mp}");
 
             let result = get_mount_fs_type_from_file("", mp);
             let err = result.unwrap_err();
 
-            let msg = format!("{}: error: {}", msg, err);
+            let msg = format!("{msg}: error: {err}");
 
             assert!(
-                format!("{}", err).contains("No such file or directory"),
+                format!("{err}").contains("No such file or directory"),
                 "{}",
                 msg
             );
@@ -715,7 +714,7 @@ mod tests {
 
         // Now, test various combinations of file contents
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let file_path = dir.path().join("mount_stats");
 
@@ -732,7 +731,7 @@ mod tests {
             let result = get_mount_fs_type_from_file(filename, d.mount_point);
 
             // add more details if an assertion fails
-            let msg = format!("{}: result: {:?}", msg, result);
+            let msg = format!("{msg}: result: {result:?}");
 
             if d.error_contains.is_empty() {
                 let fs_type = result.unwrap();
@@ -875,7 +874,7 @@ mod tests {
         );
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let file_path = dir.path().join("cgroups");
             let filename = file_path
@@ -889,7 +888,7 @@ mod tests {
                 .unwrap_or_else(|_| panic!("{}: failed to write file contents", msg));
 
             let result = get_cgroup_mounts(&logger, filename, false);
-            let msg = format!("{}: result: {:?}", msg, result);
+            let msg = format!("{msg}: result: {result:?}");
 
             if !d.error_contains.is_empty() {
                 assert!(result.is_err(), "{}", msg);
@@ -974,7 +973,7 @@ mod tests {
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
             skip_loop_by_user!(msg, d.test_user);
 
             let drain = slog::Discard;
@@ -1011,7 +1010,7 @@ mod tests {
                 nix::mount::umount(&dest).unwrap();
             }
 
-            let msg = format!("{}: result: {:?}", msg, result);
+            let msg = format!("{msg}: result: {result:?}");
             if d.error_contains.is_empty() {
                 assert!(result.is_ok(), "{}", msg);
             } else {
@@ -1062,14 +1061,14 @@ mod tests {
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let result = parse_mount_options(&d.options_vec).unwrap();
 
-            let msg = format!("{}: result: {:?}", msg, result);
+            let msg = format!("{msg}: result: {result:?}");
 
             let expected_result = (d.result.0, d.result.1.to_owned());
-            assert_eq!(expected_result, result, "{}", msg);
+            assert_eq!(expected_result, result, "{msg}");
         }
     }
 }

src/agent/src/netlink.rs

@@ -41,9 +41,9 @@ pub enum LinkFilter<'a> {
 impl fmt::Display for LinkFilter<'_> {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
         match self {
-            LinkFilter::Name(name) => write!(f, "Name: {}", name),
-            LinkFilter::Index(idx) => write!(f, "Index: {}", idx),
-            LinkFilter::Address(addr) => write!(f, "Address: {}", addr),
+            LinkFilter::Name(name) => write!(f, "Name: {name}"),
+            LinkFilter::Index(idx) => write!(f, "Index: {idx}"),
+            LinkFilter::Address(addr) => write!(f, "Address: {addr}"),
         }
     }
 }
@@ -272,7 +272,7 @@ impl Handle {
             use LinkAttribute as Nla;
 
             let mac_addr = parse_mac_address(addr)
-                .with_context(|| format!("Failed to parse MAC address: {}", addr))?;
+                .with_context(|| format!("Failed to parse MAC address: {addr}"))?;
 
             // Hardware filter might not be supported by netlink,
             // we may have to dump link list and then find the target link.
@@ -401,11 +401,10 @@ impl Handle {
                 }
 
                 if let RouteAttribute::Oif(index) = attribute {
-                    route.device = self
-                        .find_link(LinkFilter::Index(*index))
-                        .await
-                        .context(format!("error looking up device {index}"))?
-                        .name();
+                    route.device = match self.find_link(LinkFilter::Index(*index)).await {
+                        Ok(link) => link.name(),
+                        Err(_) => String::new(),
+                    };
                 }
             }
 
@@ -925,7 +924,7 @@ mod tests {
     /// Helper function to check if the result is a netlink EACCES error
     fn is_netlink_permission_error<T>(result: &Result<T>) -> bool {
         if let Err(e) = result {
-            let error_string = format!("{:?}", e);
+            let error_string = format!("{e:?}");
             if error_string.contains("code: Some(-13)") {
                 println!("INFO: skipping test - netlink operations are restricted in this environment (EACCES)");
                 return true;
@@ -1001,14 +1000,10 @@ mod tests {
             .unwrap()
             .list_routes()
             .await
-            .context(format!("available devices: {:?}", devices))
+            .context(format!("available devices: {devices:?}"))
             .expect("Failed to list routes");
 
         assert_ne!(all.len(), 0);
-
-        for r in &all {
-            assert_ne!(r.device.len(), 0);
-        }
     }
 
     #[tokio::test]
@@ -1191,7 +1186,7 @@ mod tests {
         );
         assert_eq!(
             stdout.trim(),
-            format!("{} lladdr {} PERMANENT", TEST_ARP_IP, mac)
+            format!("{TEST_ARP_IP} lladdr {mac} PERMANENT")
         );
 
         clean_env_for_test_add_one_arp_neighbor(TEST_DUMMY_INTERFACE, TEST_ARP_IP);

src/agent/src/pci.rs

@@ -191,13 +191,13 @@ mod tests {
     fn test_slotfn() {
         // Valid slots
         let sf = SlotFn::new(0x00, 0x0).unwrap();
-        assert_eq!(format!("{}", sf), "00.0");
+        assert_eq!(format!("{sf}"), "00.0");
 
         let sf = SlotFn::from_str("00.0").unwrap();
-        assert_eq!(format!("{}", sf), "00.0");
+        assert_eq!(format!("{sf}"), "00.0");
 
         let sf = SlotFn::from_str("00").unwrap();
-        assert_eq!(format!("{}", sf), "00.0");
+        assert_eq!(format!("{sf}"), "00.0");
 
         let sf = SlotFn::new(31, 7).unwrap();
         let sf2 = SlotFn::from_str("1f.7").unwrap();
@@ -256,12 +256,12 @@ mod tests {
         let sf1f_7 = SlotFn::new(0x1f, 7).unwrap();
 
         let addr = Address::new(0, 0, sf0_0);
-        assert_eq!(format!("{}", addr), "0000:00:00.0");
+        assert_eq!(format!("{addr}"), "0000:00:00.0");
         let addr2 = Address::from_str("0000:00:00.0").unwrap();
         assert_eq!(addr, addr2);
 
         let addr = Address::new(0xffff, 0xff, sf1f_7);
-        assert_eq!(format!("{}", addr), "ffff:ff:1f.7");
+        assert_eq!(format!("{addr}"), "ffff:ff:1f.7");
         let addr2 = Address::from_str("ffff:ff:1f.7").unwrap();
         assert_eq!(addr, addr2);
 
@@ -299,7 +299,7 @@ mod tests {
 
         // Valid paths
         let pcipath = Path::new(vec![sf3_0]).unwrap();
-        assert_eq!(format!("{}", pcipath), "03.0");
+        assert_eq!(format!("{pcipath}"), "03.0");
         let pcipath2 = Path::from_str("03.0").unwrap();
         assert_eq!(pcipath, pcipath2);
         let pcipath2 = Path::from_str("03").unwrap();
@@ -308,7 +308,7 @@ mod tests {
         assert_eq!(pcipath[0], sf3_0);
 
         let pcipath = Path::new(vec![sf3_0, sf4_0]).unwrap();
-        assert_eq!(format!("{}", pcipath), "03.0/04.0");
+        assert_eq!(format!("{pcipath}"), "03.0/04.0");
         let pcipath2 = Path::from_str("03.0/04.0").unwrap();
         assert_eq!(pcipath, pcipath2);
         let pcipath2 = Path::from_str("03/04").unwrap();
@@ -318,7 +318,7 @@ mod tests {
         assert_eq!(pcipath[1], sf4_0);
 
         let pcipath = Path::new(vec![sf3_0, sf4_0, sf5_0]).unwrap();
-        assert_eq!(format!("{}", pcipath), "03.0/04.0/05.0");
+        assert_eq!(format!("{pcipath}"), "03.0/04.0/05.0");
         let pcipath2 = Path::from_str("03.0/04.0/05.0").unwrap();
         assert_eq!(pcipath, pcipath2);
         let pcipath2 = Path::from_str("03/04/05").unwrap();
@@ -329,7 +329,7 @@ mod tests {
         assert_eq!(pcipath[2], sf5_0);
 
         let pcipath = Path::new(vec![sfa_5, sfb_6, sfc_7]).unwrap();
-        assert_eq!(format!("{}", pcipath), "0a.5/0b.6/0c.7");
+        assert_eq!(format!("{pcipath}"), "0a.5/0b.6/0c.7");
         let pcipath2 = Path::from_str("0a.5/0b.6/0c.7").unwrap();
         assert_eq!(pcipath, pcipath2);
         assert_eq!(pcipath.len(), 3);

src/agent/src/rpc.rs

@@ -72,7 +72,7 @@ use crate::network::setup_guest_dns;
 use crate::passfd_io;
 use crate::pci;
 use crate::random;
-use crate::sandbox::Sandbox;
+use crate::sandbox::{Sandbox, SandboxError};
 use crate::storage::{add_storages, update_ephemeral_mounts, STORAGE_HANDLERS};
 use crate::util;
 use crate::version::{AGENT_VERSION, API_VERSION};
@@ -138,7 +138,17 @@ fn sl() -> slog::Logger {
 
 // Convenience function to wrap an error and response to ttrpc client
 pub fn ttrpc_error(code: ttrpc::Code, err: impl Debug) -> ttrpc::Error {
-    get_rpc_status(code, format!("{:?}", err))
+    get_rpc_status(code, format!("{err:?}"))
+}
+
+/// Convert SandboxError to ttrpc error with appropriate code.
+/// Process not found errors map to NOT_FOUND, others to INVALID_ARGUMENT.
+fn sandbox_err_to_ttrpc(err: SandboxError) -> ttrpc::Error {
+    let code = match &err {
+        SandboxError::InitProcessNotFound | SandboxError::InvalidExecId => ttrpc::Code::NOT_FOUND,
+        SandboxError::InvalidContainerId => ttrpc::Code::INVALID_ARGUMENT,
+    };
+    ttrpc_error(code, err)
 }
 
 #[cfg(not(feature = "agent-policy"))]
@@ -460,7 +470,9 @@ impl AgentService {
         let mut sig: libc::c_int = req.signal as libc::c_int;
         {
             let mut sandbox = self.sandbox.lock().await;
-            let p = sandbox.find_container_process(cid.as_str(), eid.as_str())?;
+            let p = sandbox
+                .find_container_process(cid.as_str(), eid.as_str())
+                .map_err(sandbox_err_to_ttrpc)?;
             // For container initProcess, if it hasn't installed handler for "SIGTERM" signal,
             // it will ignore the "SIGTERM" signal sent to it, thus send it "SIGKILL" signal
             // instead of "SIGTERM" to terminate it.
@@ -568,7 +580,9 @@ impl AgentService {
         let (exit_send, mut exit_recv) = tokio::sync::mpsc::channel(100);
         let exit_rx = {
             let mut sandbox = self.sandbox.lock().await;
-            let p = sandbox.find_container_process(cid.as_str(), eid.as_str())?;
+            let p = sandbox
+                .find_container_process(cid.as_str(), eid.as_str())
+                .map_err(sandbox_err_to_ttrpc)?;
 
             p.exit_watchers.push(exit_send);
             pid = p.pid;
@@ -665,7 +679,9 @@ impl AgentService {
         let term_exit_notifier;
         let reader = {
             let mut sandbox = self.sandbox.lock().await;
-            let p = sandbox.find_container_process(cid.as_str(), eid.as_str())?;
+            let p = sandbox
+                .find_container_process(cid.as_str(), eid.as_str())
+                .map_err(sandbox_err_to_ttrpc)?;
 
             term_exit_notifier = p.term_exit_notifier.clone();
 
@@ -947,12 +963,7 @@ impl agent_ttrpc::AgentService for AgentService {
 
         let p = sandbox
             .find_container_process(cid.as_str(), eid.as_str())
-            .map_err(|e| {
-                ttrpc_error(
-                    ttrpc::Code::INVALID_ARGUMENT,
-                    format!("invalid argument: {:?}", e),
-                )
-            })?;
+            .map_err(sandbox_err_to_ttrpc)?;
 
         p.close_stdin().await;
 
@@ -970,12 +981,7 @@ impl agent_ttrpc::AgentService for AgentService {
         let mut sandbox = self.sandbox.lock().await;
         let p = sandbox
             .find_container_process(req.container_id(), req.exec_id())
-            .map_err(|e| {
-                ttrpc_error(
-                    ttrpc::Code::UNAVAILABLE,
-                    format!("invalid argument: {:?}", e),
-                )
-            })?;
+            .map_err(sandbox_err_to_ttrpc)?;
 
         let fd = p
             .term_master
@@ -990,7 +996,7 @@ impl agent_ttrpc::AgentService for AgentService {
         let err = unsafe { libc::ioctl(fd, TIOCSWINSZ, &win) };
         Errno::result(err)
             .map(drop)
-            .map_ttrpc_err(|e| format!("ioctl error: {:?}", e))?;
+            .map_ttrpc_err(|e| format!("ioctl error: {e:?}"))?;
 
         Ok(Empty::new())
     }
@@ -1014,20 +1020,20 @@ impl agent_ttrpc::AgentService for AgentService {
             #[cfg(not(target_arch = "s390x"))]
             {
                 let pcipath = pci::Path::from_str(&interface.devicePath).map_ttrpc_err(|e| {
-                    format!("Unexpected pci-path for network interface: {:?}", e)
+                    format!("Unexpected pci-path for network interface: {e:?}")
                 })?;
                 wait_for_pci_net_interface(&self.sandbox, &pcipath)
                     .await
-                    .map_ttrpc_err(|e| format!("interface not available: {:?}", e))?;
+                    .map_ttrpc_err(|e| format!("interface not available: {e:?}"))?;
             }
             #[cfg(target_arch = "s390x")]
             {
                 let ccw_dev = ccw::Device::from_str(&interface.devicePath).map_ttrpc_err(|e| {
-                    format!("Unexpected CCW path for network interface: {:?}", e)
+                    format!("Unexpected CCW path for network interface: {e:?}")
                 })?;
                 wait_for_ccw_net_interface(&self.sandbox, &ccw_dev)
                     .await
-                    .map_ttrpc_err(|e| format!("interface not available: {:?}", e))?;
+                    .map_ttrpc_err(|e| format!("interface not available: {e:?}"))?;
             }
         }
 
@@ -1037,7 +1043,7 @@ impl agent_ttrpc::AgentService for AgentService {
             .rtnl
             .update_interface(&interface)
             .await
-            .map_ttrpc_err(|e| format!("update interface: {:?}", e))?;
+            .map_ttrpc_err(|e| format!("update interface: {e:?}"))?;
 
         Ok(interface)
     }
@@ -1062,13 +1068,13 @@ impl agent_ttrpc::AgentService for AgentService {
             .rtnl
             .update_routes(new_routes)
             .await
-            .map_ttrpc_err(|e| format!("Failed to update routes: {:?}", e))?;
+            .map_ttrpc_err(|e| format!("Failed to update routes: {e:?}"))?;
 
         let list = sandbox
             .rtnl
             .list_routes()
             .await
-            .map_ttrpc_err(|e| format!("Failed to list routes after update: {:?}", e))?;
+            .map_ttrpc_err(|e| format!("Failed to list routes after update: {e:?}"))?;
 
         Ok(protocols::agent::Routes {
             Routes: list,
@@ -1086,7 +1092,7 @@ impl agent_ttrpc::AgentService for AgentService {
 
         update_ephemeral_mounts(sl(), &req.storages, &self.sandbox)
             .await
-            .map_ttrpc_err(|e| format!("Failed to update mounts: {:?}", e))?;
+            .map_ttrpc_err(|e| format!("Failed to update mounts: {e:?}"))?;
         Ok(Empty::new())
     }
 
@@ -1237,7 +1243,7 @@ impl agent_ttrpc::AgentService for AgentService {
             .rtnl
             .list_interfaces()
             .await
-            .map_ttrpc_err(|e| format!("Failed to list interfaces: {:?}", e))?;
+            .map_ttrpc_err(|e| format!("Failed to list interfaces: {e:?}"))?;
 
         Ok(protocols::agent::Interfaces {
             Interfaces: list,
@@ -1260,7 +1266,7 @@ impl agent_ttrpc::AgentService for AgentService {
             .rtnl
             .list_routes()
             .await
-            .map_ttrpc_err(|e| format!("list routes: {:?}", e))?;
+            .map_ttrpc_err(|e| format!("list routes: {e:?}"))?;
 
         Ok(protocols::agent::Routes {
             Routes: list,
@@ -1377,7 +1383,7 @@ impl agent_ttrpc::AgentService for AgentService {
             .rtnl
             .add_arp_neighbors(neighs)
             .await
-            .map_ttrpc_err(|e| format!("Failed to add ARP neighbours: {:?}", e))?;
+            .map_ttrpc_err(|e| format!("Failed to add ARP neighbours: {e:?}"))?;
 
         Ok(Empty::new())
     }
@@ -1597,7 +1603,7 @@ impl agent_ttrpc::AgentService for AgentService {
             ma.memcg_set_config_async(mem_agent_memcgconfig_to_memcg_optionconfig(&config))
                 .await
                 .map_err(|e| {
-                    let estr = format!("ma.memcg_set_config_async fail: {}", e);
+                    let estr = format!("ma.memcg_set_config_async fail: {e}");
                     error!(sl(), "{}", estr);
                     ttrpc::Error::RpcStatus(ttrpc::get_status(ttrpc::Code::INTERNAL, estr))
                 })?;
@@ -1621,7 +1627,7 @@ impl agent_ttrpc::AgentService for AgentService {
             ma.compact_set_config_async(mem_agent_compactconfig_to_compact_optionconfig(&config))
                 .await
                 .map_err(|e| {
-                    let estr = format!("ma.compact_set_config_async fail: {}", e);
+                    let estr = format!("ma.compact_set_config_async fail: {e}");
                     error!(sl(), "{}", estr);
                     ttrpc::Error::RpcStatus(ttrpc::get_status(ttrpc::Code::INTERNAL, estr))
                 })?;
@@ -2233,10 +2239,8 @@ fn load_kernel_module(module: &protocols::agent::KernelModule) -> Result<()> {
         Some(code) => {
             let std_out = String::from_utf8_lossy(&output.stdout);
             let std_err = String::from_utf8_lossy(&output.stderr);
-            let msg = format!(
-                "load_kernel_module return code: {} stdout:{} stderr:{}",
-                code, std_out, std_err
-            );
+            let msg =
+                format!("load_kernel_module return code: {code} stdout:{std_out} stderr:{std_err}");
             Err(anyhow!(msg))
         }
         None => Err(anyhow!("Process terminated by signal")),
@@ -2485,9 +2489,9 @@ mod tests {
         // Skip test if loading kernel modules is not permitted
         // or kernel module is not found
         if let Err(e) = &result {
-            let error_string = format!("{:?}", e);
+            let error_string = format!("{e:?}");
             // Let's print out the error message first
-            println!("DEBUG: error: {}", error_string);
+            println!("DEBUG: error: {error_string}");
             if error_string.contains("Operation not permitted")
                 || error_string.contains("EPERM")
                 || error_string.contains("Permission denied")
@@ -2629,12 +2633,12 @@ mod tests {
             },
             TestData {
                 create_container: false,
-                result: Err(anyhow!(crate::sandbox::ERR_INVALID_CONTAINER_ID)),
+                result: Err(anyhow!(crate::sandbox::SandboxError::InvalidContainerId)),
                 ..Default::default()
             },
             TestData {
                 container_id: "8181",
-                result: Err(anyhow!(crate::sandbox::ERR_INVALID_CONTAINER_ID)),
+                result: Err(anyhow!(crate::sandbox::SandboxError::InvalidContainerId)),
                 ..Default::default()
             },
             TestData {
@@ -2648,7 +2652,7 @@ mod tests {
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let logger = slog::Logger::root(slog::Discard, o!());
             let mut sandbox = Sandbox::new(&logger).unwrap();
@@ -2719,7 +2723,7 @@ mod tests {
             // the fd will be closed on Process's dropping.
             // unistd::close(wfd).unwrap();
 
-            let msg = format!("{}, result: {:?}", msg, result);
+            let msg = format!("{msg}, result: {result:?}");
             assert_result!(d.result, result, msg);
         }
     }
@@ -2823,7 +2827,7 @@ mod tests {
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let logger = slog::Logger::root(slog::Discard, o!());
             let mut sandbox = Sandbox::new(&logger).unwrap();
@@ -2843,15 +2847,14 @@ mod tests {
 
             let result = update_container_namespaces(&sandbox, &mut oci, d.use_sandbox_pidns);
 
-            let msg = format!("{}, result: {:?}", msg, result);
+            let msg = format!("{msg}, result: {result:?}");
 
             assert_result!(d.result, result, msg);
             if let Some(linux) = oci.linux() {
                 assert_eq!(
                     d.expected_namespaces,
                     linux.namespaces().clone().unwrap(),
-                    "{}",
-                    msg
+                    "{msg}"
                 );
             }
         }
@@ -2944,7 +2947,7 @@ mod tests {
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let dir = tempdir().expect("failed to make tempdir");
             let block_size_path = dir.path().join("block_size_bytes");
@@ -2964,7 +2967,7 @@ mod tests {
                 hotplug_probe_path.to_str().unwrap(),
             );
 
-            let msg = format!("{}, result: {:?}", msg, result);
+            let msg = format!("{msg}, result: {result:?}");
 
             assert_result!(d.result, result, msg);
         }
@@ -3074,7 +3077,7 @@ OtherField:other
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let dir = tempdir().expect("failed to make tempdir");
             let proc_status_file_path = dir.path().join("status");
@@ -3085,9 +3088,9 @@ OtherField:other
 
             let result = is_signal_handled(proc_status_file_path.to_str().unwrap(), d.signum);
 
-            let msg = format!("{}, result: {:?}", msg, result);
+            let msg = format!("{msg}, result: {result:?}");
 
-            assert_eq!(d.result, result, "{}", msg);
+            assert_eq!(d.result, result, "{msg}");
         }
     }
 
@@ -3386,9 +3389,9 @@ COMMIT
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
             let result = is_sealed_secret_path(d.source_path);
-            assert_eq!(d.result, result, "{}", msg);
+            assert_eq!(d.result, result, "{msg}");
         }
     }
 }

src/agent/src/sandbox.rs

@@ -32,6 +32,7 @@ use rustjail::container::BaseContainer;
 use rustjail::container::LinuxContainer;
 use rustjail::process::Process;
 use slog::Logger;
+use thiserror::Error;
 use tokio::sync::mpsc::{channel, Receiver, Sender};
 use tokio::sync::oneshot;
 use tokio::sync::Mutex;
@@ -47,7 +48,16 @@ use crate::storage::StorageDeviceGeneric;
 use crate::uevent::{Uevent, UeventMatcher};
 use crate::watcher::BindWatcher;
 
-pub const ERR_INVALID_CONTAINER_ID: &str = "Invalid container id";
+/// Errors that can occur when looking up processes in the sandbox.
+#[derive(Debug, Error)]
+pub enum SandboxError {
+    #[error("Invalid container id")]
+    InvalidContainerId,
+    #[error("Process not found: init process missing")]
+    InitProcessNotFound,
+    #[error("Process not found: invalid exec id")]
+    InvalidExecId,
+}
 
 type UeventWatcher = (Box<dyn UeventMatcher>, oneshot::Sender<Uevent>);
 
@@ -248,7 +258,7 @@ impl Sandbox {
             }
 
             let mut pid_ns = Namespace::new(&self.logger).get_pid();
-            pid_ns.path = format!("/proc/{}/ns/pid", init_pid);
+            pid_ns.path = format!("/proc/{init_pid}/ns/pid");
 
             self.sandbox_pidns = Some(pid_ns);
         }
@@ -282,10 +292,14 @@ impl Sandbox {
         None
     }
 
-    pub fn find_container_process(&mut self, cid: &str, eid: &str) -> Result<&mut Process> {
+    pub fn find_container_process(
+        &mut self,
+        cid: &str,
+        eid: &str,
+    ) -> Result<&mut Process, SandboxError> {
         let ctr = self
             .get_container(cid)
-            .ok_or_else(|| anyhow!(ERR_INVALID_CONTAINER_ID))?;
+            .ok_or(SandboxError::InvalidContainerId)?;
 
         if eid.is_empty() {
             let init_pid = ctr.init_process_pid;
@@ -293,10 +307,11 @@ impl Sandbox {
                 .processes
                 .values_mut()
                 .find(|p| p.pid == init_pid)
-                .ok_or_else(|| anyhow!("cannot find init process!"));
+                .ok_or(SandboxError::InitProcessNotFound);
         }
 
-        ctr.get_process(eid).map_err(|_| anyhow!("Invalid exec id"))
+        ctr.get_process(eid)
+            .map_err(|_| SandboxError::InvalidExecId)
     }
 
     #[instrument]
@@ -711,8 +726,7 @@ mod tests {
         let ref_count = new_storage.ref_count().await;
         assert_eq!(
             ref_count, 1,
-            "Invalid refcount, got {} expected 1.",
-            ref_count
+            "Invalid refcount, got {ref_count} expected 1."
         );
 
         // Use the existing sandbox storage
@@ -723,8 +737,7 @@ mod tests {
         let ref_count = new_storage.ref_count().await;
         assert_eq!(
             ref_count, 2,
-            "Invalid refcount, got {} expected 2.",
-            ref_count
+            "Invalid refcount, got {ref_count} expected 2."
         );
     }
 
@@ -811,11 +824,7 @@ mod tests {
         // Reference counter should decrement to 1.
         let storage = &s.storages[storage_path];
         let refcount = storage.ref_count().await;
-        assert_eq!(
-            refcount, 1,
-            "Invalid refcount, got {} expected 1.",
-            refcount
-        );
+        assert_eq!(refcount, 1, "Invalid refcount, got {refcount} expected 1.");
 
         assert!(
             s.remove_sandbox_storage(storage_path).await.unwrap(),
@@ -959,7 +968,7 @@ mod tests {
 
         assert!(s.sandbox_pidns.is_some());
 
-        let ns_path = format!("/proc/{}/ns/pid", test_pid);
+        let ns_path = format!("/proc/{test_pid}/ns/pid");
         assert_eq!(s.sandbox_pidns.unwrap().path, ns_path);
     }
 
@@ -1271,7 +1280,7 @@ mod tests {
         let tmpdir_path = tmpdir.path().to_str().unwrap();
 
         for (i, d) in tests.iter().enumerate() {
-            let current_test_dir_path = format!("{}/test_{}", tmpdir_path, i);
+            let current_test_dir_path = format!("{tmpdir_path}/test_{i}");
             fs::create_dir(&current_test_dir_path).unwrap();
 
             // create numbered directories and fill using root name
@@ -1280,7 +1289,7 @@ mod tests {
                     "{}/{}{}",
                     current_test_dir_path, d.directory_autogen_name, j
                 );
-                let subfile_path = format!("{}/{}", subdir_path, SYSFS_ONLINE_FILE);
+                let subfile_path = format!("{subdir_path}/{SYSFS_ONLINE_FILE}");
                 fs::create_dir(&subdir_path).unwrap();
                 let mut subfile = File::create(subfile_path).unwrap();
                 subfile.write_all(b"0").unwrap();
@@ -1307,18 +1316,15 @@ mod tests {
                 result.is_ok()
             );
 
-            assert_eq!(result.is_ok(), d.result.is_ok(), "{}", msg);
+            assert_eq!(result.is_ok(), d.result.is_ok(), "{msg}");
 
             if d.result.is_ok() {
                 let test_result_val = *d.result.as_ref().ok().unwrap();
                 let result_val = result.ok().unwrap();
 
-                msg = format!(
-                    "test[{}]: {:?}, expected {}, actual {}",
-                    i, d, test_result_val, result_val
-                );
+                msg = format!("test[{i}]: {d:?}, expected {test_result_val}, actual {result_val}");
 
-                assert_eq!(test_result_val, result_val, "{}", msg);
+                assert_eq!(test_result_val, result_val, "{msg}");
             }
         }
     }

src/agent/src/signal.rs

@@ -44,7 +44,7 @@ async fn handle_sigchild(logger: Logger, sandbox: Arc<Mutex<Sandbox>>) -> Result
 
         if let Some(pid) = wait_status.pid() {
             let raw_pid = pid.as_raw();
-            let child_pid = format!("{}", raw_pid);
+            let child_pid = format!("{raw_pid}");
 
             let logger = logger.new(o!("child-pid" => child_pid));
 

src/agent/src/storage/block_handler.rs

@@ -11,9 +11,10 @@ use std::str::FromStr;
 use std::sync::Arc;
 
 use anyhow::{anyhow, Context, Result};
+#[cfg(target_arch = "s390x")]
+use kata_types::device::DRIVER_BLK_CCW_TYPE;
 use kata_types::device::{
-    DRIVER_BLK_CCW_TYPE, DRIVER_BLK_MMIO_TYPE, DRIVER_BLK_PCI_TYPE, DRIVER_NVDIMM_TYPE,
-    DRIVER_SCSI_TYPE,
+    DRIVER_BLK_MMIO_TYPE, DRIVER_BLK_PCI_TYPE, DRIVER_NVDIMM_TYPE, DRIVER_SCSI_TYPE,
 };
 use kata_types::mount::StorageDevice;
 use protocols::agent::Storage;
@@ -93,9 +94,11 @@ impl StorageHandler for VirtioBlkPciHandler {
     }
 }
 
+#[cfg(target_arch = "s390x")]
 #[derive(Debug)]
 pub struct VirtioBlkCcwHandler {}
 
+#[cfg(target_arch = "s390x")]
 #[async_trait::async_trait]
 impl StorageHandler for VirtioBlkCcwHandler {
     #[instrument]

src/agent/src/storage/mod.rs

@@ -467,7 +467,7 @@ mod tests {
         ];
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             skip_loop_by_user!(msg, d.test_user);
 
@@ -515,7 +515,7 @@ mod tests {
                 nix::mount::umount(&mount_point).unwrap();
             }
 
-            let msg = format!("{}: result: {:?}", msg, result);
+            let msg = format!("{msg}: result: {result:?}");
             if d.error_contains.is_empty() {
                 assert!(result.is_ok(), "{}", msg);
             } else {
@@ -576,7 +576,7 @@ mod tests {
         let tempdir = tempdir().expect("failed to create tmpdir");
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let mount_dir = tempdir.path().join(d.mount_path);
             fs::create_dir(&mount_dir)
@@ -663,7 +663,7 @@ mod tests {
         let tempdir = tempdir().expect("failed to create tmpdir");
 
         for (i, d) in tests.iter().enumerate() {
-            let msg = format!("test[{}]: {:?}", i, d);
+            let msg = format!("test[{i}]: {d:?}");
 
             let mount_dir = tempdir.path().join(d.path);
             fs::create_dir(&mount_dir)
@@ -674,12 +674,12 @@ mod tests {
 
             // create testing directories and files
             for n in 1..COUNT {
-                let nest_dir = mount_dir.join(format!("nested{}", n));
+                let nest_dir = mount_dir.join(format!("nested{n}"));
                 fs::create_dir(&nest_dir)
                     .unwrap_or_else(|_| panic!("{}: failed to create nest directory", msg));
 
                 for f in 1..COUNT {
-                    let filename = nest_dir.join(format!("file{}", f));
+                    let filename = nest_dir.join(format!("file{f}"));
                     File::create(&filename)
                         .unwrap_or_else(|_| panic!("{}: failed to create file", msg));
                     file_mode = filename.as_path().metadata().unwrap().permissions().mode();
@@ -707,9 +707,9 @@ mod tests {
             );
 
             for n in 1..COUNT {
-                let nest_dir = mount_dir.join(format!("nested{}", n));
+                let nest_dir = mount_dir.join(format!("nested{n}"));
                 for f in 1..COUNT {
-                    let filename = nest_dir.join(format!("file{}", f));
+                    let filename = nest_dir.join(format!("file{f}"));
                     let file = Path::new(&filename);
 
                     assert_eq!(file.metadata().unwrap().gid(), d.gid);

src/agent/src/util.rs

@@ -147,7 +147,7 @@ mod tests {
 
             let v = vec_locked
                 .as_deref_mut()
-                .map_err(|e| std::io::Error::new(std::io::ErrorKind::Other, e.to_string()))?;
+                .map_err(|e| std::io::Error::other(e.to_string()))?;
 
             std::io::Write::flush(v)
         }

src/agent/src/watcher.rs

@@ -556,9 +556,9 @@ mod tests {
     use test_utils::skip_if_not_root;
 
     async fn create_test_storage(dir: &Path, id: &str) -> Result<(protos::Storage, PathBuf)> {
-        let src_path = dir.join(format!("src{}", id));
+        let src_path = dir.join(format!("src{id}"));
         let src_filename = src_path.to_str().expect("failed to create src filename");
-        let dest_path = dir.join(format!("dest{}", id));
+        let dest_path = dir.join(format!("dest{id}"));
         let dest_filename = dest_path.to_str().expect("failed to create dest filename");
 
         std::fs::create_dir_all(src_filename).expect("failed to create path");
@@ -682,7 +682,7 @@ mod tests {
 
         // setup storage0: too many files
         for i in 1..21 {
-            fs::write(src0_path.join(format!("{}.txt", i)), "original").unwrap();
+            fs::write(src0_path.join(format!("{i}.txt")), "original").unwrap();
         }
 
         // setup storage1: two small files
@@ -700,7 +700,7 @@ mod tests {
 
         // setup storage3: many files, but still watchable
         for i in 1..MAX_ENTRIES_PER_STORAGE {
-            fs::write(src3_path.join(format!("{}.txt", i)), "original").unwrap();
+            fs::write(src3_path.join(format!("{i}.txt")), "original").unwrap();
         }
 
         let logger = slog::Logger::root(slog::Discard, o!());
@@ -919,7 +919,7 @@ mod tests {
 
         // Up to 15 files should be okay (can watch 15 files + 1 directory)
         for i in 1..MAX_ENTRIES_PER_STORAGE {
-            fs::write(source_dir.path().join(format!("{}.txt", i)), "original").unwrap();
+            fs::write(source_dir.path().join(format!("{i}.txt")), "original").unwrap();
         }
 
         assert_eq!(
@@ -1387,7 +1387,7 @@ mod tests {
         assert!(!dest_dir.path().exists());
 
         for i in 1..21 {
-            fs::write(source_dir.path().join(format!("{}.txt", i)), "fluff").unwrap();
+            fs::write(source_dir.path().join(format!("{i}.txt")), "fluff").unwrap();
         }
 
         // verify non-watched storage is cleaned up correctly

src/agent/vsock-exporter/src/lib.rs

@@ -70,7 +70,7 @@ impl ExportError for Error {
 }
 
 fn make_io_error(desc: String) -> std::io::Error {
-    std::io::Error::new(ErrorKind::Other, desc)
+    std::io::Error::other(desc)
 }
 
 // Send a trace span to the forwarder running on the host.

src/dragonball/Makefile

@@ -44,7 +44,7 @@ test:
 ifdef SUPPORT_VIRTUALIZATION
 	@set -e; \
 	for dir in $(PROJECT_DIRS); do \
-		bash -c "pushd $${dir} && RUST_BACKTRACE=1 cargo test --all-features --target $(TRIPLE) -- --nocapture --test-threads=1 && popd"; \
+		bash -c "pushd $${dir} && RUST_BACKTRACE=1 cargo test --all-features --target $(TRIPLE) --no-fail-fast -- --nocapture --test-threads=1 --skip bindgen && popd"; \
 	done
 else
 	@echo "INFO: skip testing dragonball, it need virtualization support."

src/dragonball/dbs_address_space/src/memory/hybrid.rs

@@ -207,7 +207,7 @@ impl<B: Bitmap> GuestMemoryRegion for GuestRegionHybrid<B> {
         &self,
         offset: MemoryRegionAddress,
         count: usize,
-    ) -> guest_memory::Result<VolatileSlice<BS<B>>> {
+    ) -> guest_memory::Result<VolatileSlice<'_, BS<'_, B>>> {
         match self {
             GuestRegionHybrid::Mmap(region) => region.get_slice(offset, count),
             GuestRegionHybrid::Raw(region) => region.get_slice(offset, count),
@@ -246,8 +246,8 @@ impl<B: Bitmap> GuestMemoryHybrid<B> {
     /// # Arguments
     ///
     /// * `regions` - The vector of regions.
-    ///               The regions shouldn't overlap and they should be sorted
-    ///               by the starting address.
+    ///   The regions shouldn't overlap and they should be sorted
+    ///   by the starting address.
     pub fn from_regions(mut regions: Vec<GuestRegionHybrid<B>>) -> Result<Self, Error> {
         Self::from_arc_regions(regions.drain(..).map(Arc::new).collect())
     }
@@ -262,8 +262,8 @@ impl<B: Bitmap> GuestMemoryHybrid<B> {
     /// # Arguments
     ///
     /// * `regions` - The vector of `Arc` regions.
-    ///               The regions shouldn't overlap and they should be sorted
-    ///               by the starting address.
+    ///   The regions shouldn't overlap and they should be sorted
+    ///   by the starting address.
     pub fn from_arc_regions(regions: Vec<Arc<GuestRegionHybrid<B>>>) -> Result<Self, Error> {
         if regions.is_empty() {
             return Err(Error::NoMemoryRegion);
@@ -359,7 +359,7 @@ impl<B: Bitmap + 'static> GuestMemory for GuestMemoryHybrid<B> {
         index.map(|x| self.regions[x].as_ref())
     }
 
-    fn iter(&self) -> Iter<B> {
+    fn iter(&self) -> Iter<'_, B> {
         Iter(self.regions.iter())
     }
 }

src/dragonball/dbs_address_space/src/memory/raw_region.rs

@@ -202,7 +202,7 @@ impl<B: Bitmap> GuestMemoryRegion for GuestRegionRaw<B> {
         &self,
         offset: MemoryRegionAddress,
         count: usize,
-    ) -> guest_memory::Result<VolatileSlice<BS<B>>> {
+    ) -> guest_memory::Result<VolatileSlice<'_, BS<'_, B>>> {
         let offset = offset.raw_value() as usize;
         let end = compute_offset(offset, count)?;
         if end > self.size {

src/dragonball/dbs_arch/Cargo.toml

@@ -21,6 +21,8 @@ libc = ">=0.2.39"
 
 [dev-dependencies]
 vm-memory = { workspace = true, features = ["backend-mmap"] }
+test-utils = { workspace = true }
+nix = { workspace = true }
 
 [package.metadata.docs.rs]
 all-features = true

src/dragonball/dbs_arch/src/aarch64/gic/mod.rs

@@ -205,12 +205,12 @@ pub fn create_gic(vm: &VmFd, vcpu_count: u64) -> Result<Box<dyn GICDevice>> {
 
 #[cfg(test)]
 mod tests {
-
     use super::*;
     use kvm_ioctls::Kvm;
 
     #[test]
     fn test_create_gic() {
+        test_utils::skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         let vm = kvm.create_vm().unwrap();
         assert!(create_gic(&vm, 1).is_ok());

src/dragonball/dbs_arch/src/aarch64/pmu.rs

@@ -150,6 +150,7 @@ mod tests {
 
     #[test]
     fn test_create_pmu() {
+        test_utils::skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         let vm = kvm.create_vm().unwrap();
         let vcpu = vm.create_vcpu(0).unwrap();

src/dragonball/dbs_arch/src/aarch64/regs.rs

@@ -166,9 +166,11 @@ pub fn read_mpidr(vcpu: &VcpuFd) -> Result<u64> {
 mod tests {
     use super::*;
     use kvm_ioctls::Kvm;
+    use test_utils::skip_if_kvm_unaccessable;
 
     #[test]
     fn test_setup_regs() {
+        skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         let vm = kvm.create_vm().unwrap();
         let vcpu = vm.create_vcpu(0).unwrap();
@@ -185,6 +187,7 @@ mod tests {
 
     #[test]
     fn test_read_mpidr() {
+        skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         let vm = kvm.create_vm().unwrap();
         let vcpu = vm.create_vcpu(0).unwrap();

src/dragonball/dbs_arch/src/x86_64/interrupts.rs

@@ -78,6 +78,7 @@ pub fn set_lint(vcpu: &VcpuFd) -> Result<()> {
 mod tests {
     use super::*;
     use kvm_ioctls::Kvm;
+    use test_utils::skip_if_kvm_unaccessable;
 
     const KVM_APIC_REG_SIZE: usize = 0x400;
 
@@ -100,6 +101,7 @@ mod tests {
 
     #[test]
     fn test_setlint() {
+        skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         assert!(kvm.check_extension(kvm_ioctls::Cap::Irqchip));
         let vm = kvm.create_vm().unwrap();
@@ -126,6 +128,7 @@ mod tests {
 
     #[test]
     fn test_setlint_fails() {
+        skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         let vm = kvm.create_vm().unwrap();
         let vcpu = vm.create_vcpu(0).unwrap();

src/dragonball/dbs_arch/src/x86_64/regs.rs

@@ -271,6 +271,7 @@ mod tests {
     use super::*;
     use crate::x86_64::gdt::gdt_entry;
     use kvm_ioctls::Kvm;
+    use test_utils::skip_if_kvm_unaccessable;
     use vm_memory::{Bytes, GuestAddress, GuestMemoryMmap};
 
     const BOOT_GDT_OFFSET: u64 = 0x500;
@@ -334,6 +335,7 @@ mod tests {
 
     #[test]
     fn test_setup_fpu() {
+        skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         let vm = kvm.create_vm().unwrap();
         let vcpu = vm.create_vcpu(0).unwrap();
@@ -356,6 +358,7 @@ mod tests {
     #[test]
     #[allow(clippy::cast_ptr_alignment)]
     fn test_setup_msrs() {
+        skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         let vm = kvm.create_vm().unwrap();
         let vcpu = vm.create_vcpu(0).unwrap();
@@ -384,6 +387,7 @@ mod tests {
 
     #[test]
     fn test_setup_regs() {
+        skip_if_kvm_unaccessable!();
         let kvm = Kvm::new().unwrap();
         let vm = kvm.create_vm().unwrap();
         let vcpu = vm.create_vcpu(0).unwrap();

src/dragonball/dbs_boot/Cargo.toml

@@ -24,3 +24,5 @@ vm-fdt = {workspace= true}
 vm-memory = { workspace = true, features = ["backend-mmap"] }
 device_tree = ">=1.1.0"
 dbs-device = { workspace = true }
+test-utils = { workspace = true }
+nix = { workspace = true }

src/dragonball/dbs_boot/src/aarch64/fdt.rs

@@ -399,6 +399,7 @@ mod tests {
     use device_tree::DeviceTree;
     use kvm_bindings::{kvm_vcpu_init, KVM_ARM_VCPU_PMU_V3, KVM_ARM_VCPU_PSCI_0_2};
     use kvm_ioctls::{Kvm, VcpuFd, VmFd};
+    use test_utils::skip_if_kvm_unaccessable;
     use vm_memory::GuestMemoryMmap;
 
     use super::super::tests::MMIODeviceInfo;
@@ -460,6 +461,7 @@ mod tests {
 
     #[test]
     fn test_create_fdt_with_devices() {
+        skip_if_kvm_unaccessable!();
         let regions = arch_memory_regions(FDT_MAX_SIZE + 0x1000);
         let mem = GuestMemoryMmap::<()>::from_ranges(&regions).expect("Cannot initialize memory");
         let dev_info: HashMap<(DeviceType, String), MMIODeviceInfo> = [
@@ -498,6 +500,7 @@ mod tests {
 
     #[test]
     fn test_create_fdt() {
+        skip_if_kvm_unaccessable!();
         let regions = arch_memory_regions(FDT_MAX_SIZE + 0x1000);
         let mem = GuestMemoryMmap::<()>::from_ranges(&regions).expect("Cannot initialize memory");
         let kvm = Kvm::new().unwrap();
@@ -532,6 +535,7 @@ mod tests {
 
     #[test]
     fn test_create_fdt_with_initrd() {
+        skip_if_kvm_unaccessable!();
         let regions = arch_memory_regions(FDT_MAX_SIZE + 0x1000);
         let mem = GuestMemoryMmap::<()>::from_ranges(&regions).expect("Cannot initialize memory");
         let kvm = Kvm::new().unwrap();
@@ -570,6 +574,7 @@ mod tests {
 
     #[test]
     fn test_create_fdt_with_pmu() {
+        skip_if_kvm_unaccessable!();
         let regions = arch_memory_regions(FDT_MAX_SIZE + 0x1000);
         let mem = GuestMemoryMmap::<()>::from_ranges(&regions).expect("Cannot initialize memory");
         let kvm = Kvm::new().unwrap();