build: Release v22.1 (bug fix release)

Signed-off-by: Rob Bradford <robert.bradford@intel.com>

.github/dependabot.yml

@@ -4,7 +4,7 @@ updates:
   directory: "/"
   schedule:
     interval: daily
-  open-pull-requests-limit: 1
+  open-pull-requests-limit: 3
   allow:
   - dependency-type: direct
   - dependency-type: indirect
@@ -12,7 +12,7 @@ updates:
   directory: "/fuzz"
   schedule:
     interval: daily
-  open-pull-requests-limit: 1
+  open-pull-requests-limit: 3
   allow:
   - dependency-type: direct
   - dependency-type: indirect

.github/workflows/build.yaml

@@ -13,7 +13,7 @@ jobs:
           - stable
           - beta
           - nightly
-          - "1.64"
+          - 1.54
         target:
           - x86_64-unknown-linux-gnu
           - x86_64-unknown-linux-musl
@@ -29,33 +29,32 @@ jobs:
       - name: Install Rust toolchain (${{ matrix.rust }})
         uses: actions-rs/toolchain@v1
         with:
-          toolchain: ${{ matrix.rust }}
-          target: ${{ matrix.target }}
-          override: true
+            toolchain: ${{ matrix.rust }}
+            target: ${{ matrix.target }}
+            override: true
 
-      - name: Build (default features)
-        run: cargo rustc --locked --bin cloud-hypervisor -- -D warnings -D clippy::undocumented_unsafe_blocks
+      - name: Debug Check (default features)
+        run: |
+          git rev-list origin/main..$GITHUB_SHA | xargs -t -I % sh -c 'git checkout %; cargo check --all --target=${{ matrix.target }}'
+          git checkout $GITHUB_SHA
+
+      - name: Build (default + tdx)
+        run: cargo rustc --bin cloud-hypervisor --features "tdx" -- -D warnings
+
+      - name: Build (acpi,kvm)
+        run: cargo rustc --bin cloud-hypervisor --no-default-features --features "acpi,kvm"  -- -D warnings
 
       - name: Build (kvm)
-        run: cargo rustc --locked --bin cloud-hypervisor --no-default-features --features "kvm"  -- -D warnings -D clippy::undocumented_unsafe_blocks
+        run: cargo rustc --bin cloud-hypervisor --no-default-features --features "kvm"  -- -D warnings
 
-      - name: Build (default features + tdx)
-        run: cargo rustc --locked --bin cloud-hypervisor --features "tdx" -- -D warnings -D clippy::undocumented_unsafe_blocks
-
-      - name: Build (default features + dbus_api)
-        run: cargo rustc --locked --bin cloud-hypervisor --features "dbus_api" -- -D warnings -D clippy::undocumented_unsafe_blocks
-
-      - name: Build (default features + guest_debug)
-        run: cargo rustc --locked --bin cloud-hypervisor --features "guest_debug" -- -D warnings -D clippy::undocumented_unsafe_blocks
+      - name: Build (acpi,mshv)
+        run: cargo rustc --bin cloud-hypervisor --no-default-features --features "acpi,mshv"  -- -D warnings
 
       - name: Build (mshv)
-        run: cargo rustc --locked --bin cloud-hypervisor --no-default-features --features "mshv"  -- -D warnings -D clippy::undocumented_unsafe_blocks
-
-      - name: Build (mshv + kvm)
-        run: cargo rustc --locked --bin cloud-hypervisor --no-default-features --features "mshv,kvm"  -- -D warnings -D clippy::undocumented_unsafe_blocks
-
+        run: cargo rustc --bin cloud-hypervisor --no-default-features --features "mshv"  -- -D warnings
+        
       - name: Release Build (default features)
-        run: cargo build --locked --all --release --target=${{ matrix.target }}
+        run: cargo build --all --release --target=${{ matrix.target }}
 
       - name: Check build did not modify any files
         run: test -z "$(git status --porcelain)"

.github/workflows/docker-image.yaml

@@ -7,10 +7,6 @@ on:
   pull_request:
     paths: resources/Dockerfile
 
-env:
-  REGISTRY: ghcr.io
-  IMAGE_NAME: ${{ github.repository }}
-
 jobs:
   main:
     runs-on: ubuntu-latest
@@ -24,19 +20,19 @@ jobs:
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v1
 
-      - name: Login to ghcr
-        uses: docker/login-action@v2
+      - name: Login to DockerHub
+        if: ${{ github.event_name == 'push' }}
+        uses: docker/login-action@v1 
         with:
-          registry: ${{ env.REGISTRY }}
-          username: ${{ github.actor }}
-          password: ${{ secrets.GITHUB_TOKEN }}
-
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
 
       - name: Docker meta
         id: meta
-        uses: docker/metadata-action@v4
+        uses: docker/metadata-action@v3
         with:
-          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+          # list of Docker images to use as base name for tags
+          images: cloudhypervisor/dev
           # generate Docker tags based on the following events/attributes
           tags: |
             type=raw,value={{date 'YYYYMMDD'}}-0

.github/workflows/hadolint.yaml

@@ -1,26 +0,0 @@
-name: Lint Dockerfile
-
-on:
-  push:
-    paths:
-    - resources/Dockerfile
-  pull_request:
-    paths:
-    - resources/Dockerfile
-
-jobs:
-  hadolint:
-    name: Run Hadolint Dockerfile Linter
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v3
-
-      - name: Lint Dockerfile
-        uses: hadolint/hadolint-action@master
-        with:
-          dockerfile: ./resources/Dockerfile
-          format: tty
-          no-fail: false
-          verbose: true
-          failure-threshold: info

.github/workflows/quality-aarch64.yaml

@@ -0,0 +1,61 @@
+name: Cloud Hypervisor Quality Checks
+on: [pull_request, create]
+
+jobs:
+  build:
+    if: github.event_name == 'pull_request'
+    name: Quality (clippy, rustfmt)
+    runs-on: ubuntu-latest
+    continue-on-error: ${{ matrix.experimental }}
+    strategy:
+      fail-fast: false
+      matrix:
+        rust:
+          - stable
+        target:
+          - aarch64-unknown-linux-gnu
+        experimental: [false]
+        include:
+          - rust: beta
+            target: aarch64-unknown-linux-gnu
+            experimental: true
+    steps:
+      - name: Code checkout
+        uses: actions/checkout@v2
+      - name: Install Rust toolchain (${{ matrix.rust }})
+        uses: actions-rs/toolchain@v1
+        with:
+            toolchain: ${{ matrix.rust }}
+            target: ${{ matrix.target }}
+            override: true
+            components: rustfmt, clippy
+      - name: Formatting (rustfmt)
+        run: cargo fmt -- --check
+
+      - name: Clippy (kvm)
+        uses: actions-rs/cargo@v1
+        with:
+          use-cross: true
+          command: clippy
+          args: --target=${{ matrix.target }} --all --no-default-features --features "kvm" -- -D warnings
+
+      - name: Clippy (kvm,acpi)
+        uses: actions-rs/cargo@v1
+        with:
+          use-cross: true
+          command: clippy
+          args: --target=${{ matrix.target }} --all --no-default-features --features "kvm,acpi" -- -D warnings
+
+      - name: Clippy (all features,kvm)
+        uses: actions-rs/cargo@v1
+        with:
+          use-cross: true
+          command: clippy
+          args: --target=${{ matrix.target }} --all --no-default-features --features "common,kvm" -- -D warnings
+
+      - name: Clippy (default)
+        uses: actions-rs/cargo@v1
+        with:
+          use-cross: true
+          command: clippy
+          args: --target=${{ matrix.target }} --all -- -D warnings

.github/workflows/quality.yaml

@@ -13,31 +13,15 @@ jobs:
         rust:
           - stable
         target:
-          - aarch64-unknown-linux-gnu
-          - aarch64-unknown-linux-musl
           - x86_64-unknown-linux-gnu
-          - x86_64-unknown-linux-musl
-
         experimental: [false]
         include:
-          - rust: beta
-            target: aarch64-unknown-linux-gnu
-            experimental: true
-          - rust: beta
-            target: aarch64-unknown-linux-musl
-            experimental: true
           - rust: beta
             target: x86_64-unknown-linux-gnu
             experimental: true
-          - rust: beta
-            target: x86_64-unknown-linux-musl
-            experimental: true
     steps:
       - name: Code checkout
-        uses: actions/checkout@v3
-        with:
-          fetch-depth: 0
-
+        uses: actions/checkout@v2
       - name: Install Rust toolchain (${{ matrix.rust }})
         uses: actions-rs/toolchain@v1
         with:
@@ -46,68 +30,32 @@ jobs:
           override: true
           components: rustfmt, clippy
 
-      - name: Debug Check (default features)
-        if: ${{ matrix.target == 'x86_64-unknown-linux-gnu' }}
-        run: |
-          set -e
-          commits=$(git rev-list origin/${{ github.base_ref }}..${{ github.sha }})
-          for commit in $commits; do git checkout $commit; cargo check --tests --examples --all --target=${{ matrix.target }}; done
-          git checkout ${{ github.sha }}
-
       - name: Formatting (rustfmt)
         run: cargo fmt -- --check
 
+      - name: Clippy (all features,kvm)
+        run: cargo clippy --all --all-targets --no-default-features --tests --features "common,kvm" -- -D warnings
+
+      - name: Clippy (all features,mshv)
+        run: cargo clippy --all --all-targets --no-default-features --tests --features "common,mshv" -- -D warnings
+
+      - name: Clippy (acpi,kvm)
+        run: cargo clippy --all --all-targets --no-default-features --tests --features "acpi,kvm" -- -D warnings
+
+      - name: Clippy (acpi,kvm,tdx)
+        run: cargo clippy --all --all-targets --no-default-features --tests --features "acpi,kvm,tdx" -- -D warnings
+
       - name: Clippy (kvm)
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: ${{ matrix.target != 'x86_64-unknown-linux-gnu' }}
-          command: clippy
-          args: --target=${{ matrix.target }} --locked --all --all-targets --no-default-features --tests --examples --features "kvm" -- -D warnings -D clippy::undocumented_unsafe_blocks
+        run: cargo clippy --all --all-targets --no-default-features --tests --features "kvm" -- -D warnings
 
-      - name: Clippy (default features)
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: ${{ matrix.target != 'x86_64-unknown-linux-gnu' }}
-          command: clippy
-          args: --target=${{ matrix.target }} --locked --all --all-targets --tests --examples -- -D warnings -D clippy::undocumented_unsafe_blocks
-
-      - name: Clippy (default features + guest_debug)
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: ${{ matrix.target != 'x86_64-unknown-linux-gnu' }}
-          command: clippy
-          args: --target=${{ matrix.target }} --locked --all --all-targets --tests --examples --features "guest_debug" -- -D warnings -D clippy::undocumented_unsafe_blocks
-
-      - name: Clippy (default features + tracing)
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: ${{ matrix.target != 'x86_64-unknown-linux-gnu' }}
-          command: clippy
-          args: --target=${{ matrix.target }} --locked --all --all-targets --tests --examples --features "tracing" -- -D warnings -D clippy::undocumented_unsafe_blocks
+      - name: Clippy (acpi,mshv)
+        run: cargo clippy --all --all-targets --no-default-features --tests --features "acpi,mshv" -- -D warnings
 
       - name: Clippy (mshv)
-        if: ${{ matrix.target == 'x86_64-unknown-linux-gnu' }}
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: ${{ matrix.target != 'x86_64-unknown-linux-gnu' }}
-          command: clippy
-          args: --target=${{ matrix.target }} --locked --all --all-targets --no-default-features --tests --examples --features "mshv" -- -D warnings -D clippy::undocumented_unsafe_blocks
+        run: cargo clippy --all --all-targets --no-default-features --tests --features "mshv" -- -D warnings
 
-      - name: Clippy (mshv + kvm)
-        if: ${{ matrix.target == 'x86_64-unknown-linux-gnu' }}
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: ${{ matrix.target != 'x86_64-unknown-linux-gnu' }}
-          command: clippy
-          args: --target=${{ matrix.target }} --locked --all --all-targets --no-default-features --tests --examples --features "mshv,kvm" -- -D warnings -D clippy::undocumented_unsafe_blocks
-
-      - name: Clippy (kvm + tdx)
-        if: ${{ matrix.target == 'x86_64-unknown-linux-gnu' }}
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: ${{ matrix.target != 'x86_64-unknown-linux-gnu' }}
-          command: clippy
-          args: --target=${{ matrix.target }} --locked --all --all-targets --no-default-features --tests --examples --features "tdx,kvm" -- -D warnings -D clippy::undocumented_unsafe_blocks
+      - name: Clippy (integration tests)
+        run: cargo clippy --all --all-targets --tests -- -D warnings
 
       - name: Check build did not modify any files
         run: test -z "$(git status --porcelain)"

.github/workflows/release.yaml

@@ -1,48 +1,54 @@
 name: Cloud Hypervisor Release
-on: [pull_request, create]
+on: [create]
 
 jobs:
   release:
-    if: (github.event_name == 'create' && github.event.ref_type == 'tag') || github.event_name == 'pull_request'
+    if: github.event_name == 'create' && github.event.ref_type == 'tag'
     name: Release
     runs-on: ubuntu-latest
     steps:
       - name: Code checkout
         uses: actions/checkout@v2
       - name: Install musl-gcc
-        run: sudo apt install -y musl-tools
+        run: sudo apt install -y musl-tools      
       - name: Create release directory
         run: rsync -rv --exclude=.git . ../cloud-hypervisor-${{ github.event.ref }}
       - name: Install Rust toolchain (x86_64-unknown-linux-gnu)
         uses: actions-rs/toolchain@v1
         with:
-          toolchain: "1.67.1"
-          target: x86_64-unknown-linux-gnu
+            toolchain: stable
+            target: x86_64-unknown-linux-gnu
       - name: Install Rust toolchain (x86_64-unknown-linux-musl)
         uses: actions-rs/toolchain@v1
         with:
-          toolchain: "1.67.1"
-          target: x86_64-unknown-linux-musl
+            toolchain: stable
+            target: x86_64-unknown-linux-musl
       - name: Build
-        uses: actions-rs/cargo@v1
-        with:
-          toolchain: "1.67.1"
-          command: build
-          args: --all --release --features mshv --target=x86_64-unknown-linux-gnu
+        run: cargo build --all --release --target=x86_64-unknown-linux-gnu
       - name: Static Build
-        uses: actions-rs/cargo@v1
-        with:
-          toolchain: "1.67.1"
-          command: build
-          args: --all --release --features mshv --target=x86_64-unknown-linux-musl
+        run: cargo build --all --release --target=x86_64-unknown-linux-musl
+      - name: Strip cloud-hypervisor binaries
+        run: strip target/*/release/cloud-hypervisor
       - name: Install Rust toolchain (aarch64-unknown-linux-musl)
         uses: actions-rs/toolchain@v1
         with:
-          toolchain: "1.67.1"
-          target: aarch64-unknown-linux-musl
-          override: true
+            toolchain: stable
+            target: aarch64-unknown-linux-musl
+            override: true
+      - name: Static Build (AArch64)
+        uses: actions-rs/cargo@v1
+        with:
+            use-cross: true
+            command: build
+            args: --all --release --target=aarch64-unknown-linux-musl
+      - name: Vendor
+        working-directory: ../cloud-hypervisor-${{ github.event.ref }}
+        run: |
+          mkdir ../vendor-cargo-home
+          export CARGO_HOME=$(realpath ../vendor-cargo-home)
+          mkdir .cargo
+          cargo vendor > .cargo/config.toml
       - name: Create Release
-        if: github.event_name == 'create' && github.event.ref_type == 'tag'
         id: create_release
         uses: actions/create-release@v1
         env:
@@ -52,8 +58,20 @@ jobs:
           release_name: ${{ github.ref }}
           draft: true
           prerelease: true
+      - name: Create vendored source archive
+        working-directory: ../
+        run:  tar cJf cloud-hypervisor-${{ github.event.ref }}.tar.xz cloud-hypervisor-${{ github.event.ref }}
+      - name: Upload cloud-hypervisor vendored source archive
+        id: upload-release-cloud-hypervisor-vendored-sources
+        uses: actions/upload-release-asset@v1
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        with:
+          upload_url: ${{ steps.create_release.outputs.upload_url }}
+          asset_path: ../cloud-hypervisor-${{ github.event.ref }}.tar.xz
+          asset_name: cloud-hypervisor-${{ github.event.ref }}.tar.xz
+          asset_content_type: application/x-xz
       - name: Upload cloud-hypervisor
-        if: github.event_name == 'create' && github.event.ref_type == 'tag'
         id: upload-release-cloud-hypervisor
         uses: actions/upload-release-asset@v1
         env:
@@ -64,7 +82,6 @@ jobs:
           asset_name: cloud-hypervisor
           asset_content_type: application/octet-stream
       - name: Upload static cloud-hypervisor
-        if: github.event_name == 'create' && github.event.ref_type == 'tag'
         id: upload-release-static-cloud-hypervisor
         uses: actions/upload-release-asset@v1
         env:
@@ -75,7 +92,6 @@ jobs:
           asset_name: cloud-hypervisor-static
           asset_content_type: application/octet-stream
       - name: Upload ch-remote
-        if: github.event_name == 'create' && github.event.ref_type == 'tag'
         id: upload-release-ch-remote
         uses: actions/upload-release-asset@v1
         env:
@@ -86,7 +102,6 @@ jobs:
           asset_name: ch-remote
           asset_content_type: application/octet-stream
       - name: Upload static-ch-remote
-        if: github.event_name == 'create' && github.event.ref_type == 'tag'
         id: upload-release-static-ch-remote
         uses: actions/upload-release-asset@v1
         env:
@@ -96,18 +111,7 @@ jobs:
           asset_path: target/x86_64-unknown-linux-musl/release/ch-remote
           asset_name: ch-remote-static
           asset_content_type: application/octet-stream
-      - name: Clean build tree ahead of cross build
-        uses: actions-rs/cargo@v1
-        with:
-          command: clean
-      - name: Static Build (AArch64)
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: true
-          command: build
-          args: --all --release --target=aarch64-unknown-linux-musl
       - name: Upload static AArch64 cloud-hypervisor
-        if: github.event_name == 'create' && github.event.ref_type == 'tag'
         id: upload-release-static-aarch64-cloud-hypervisor
         uses: actions/upload-release-asset@v1
         env:
@@ -118,7 +122,6 @@ jobs:
           asset_name: cloud-hypervisor-static-aarch64
           asset_content_type: application/octet-stream
       - name: Upload static AArch64 ch-remote
-        if: github.event_name == 'create' && github.event.ref_type == 'tag'
         id: upload-release-static-aarch64-ch-remote
         uses: actions/upload-release-asset@v1
         env:
@@ -128,24 +131,3 @@ jobs:
           asset_path: target/aarch64-unknown-linux-musl/release/ch-remote
           asset_name: ch-remote-static-aarch64
           asset_content_type: application/octet-stream
-      - name: Vendor
-        working-directory: ../cloud-hypervisor-${{ github.event.ref }}
-        run: |
-          mkdir ../vendor-cargo-home
-          export CARGO_HOME=$(realpath ../vendor-cargo-home)
-          mkdir .cargo
-          cargo vendor > .cargo/config.toml
-      - name: Create vendored source archive
-        working-directory: ../
-        run: tar cJf cloud-hypervisor-${{ github.event.ref }}.tar.xz cloud-hypervisor-${{ github.event.ref }}
-      - name: Upload cloud-hypervisor vendored source archive
-        if: github.event_name == 'create' && github.event.ref_type == 'tag'
-        id: upload-release-cloud-hypervisor-vendored-sources
-        uses: actions/upload-release-asset@v1
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        with:
-          upload_url: ${{ steps.create_release.outputs.upload_url }}
-          asset_path: ../cloud-hypervisor-${{ github.event.ref }}.tar.xz
-          asset_name: cloud-hypervisor-${{ github.event.ref }}.tar.xz
-          asset_content_type: application/x-xz

.gitignore

@@ -4,5 +4,3 @@
 **/*.rs.bk
 **/Cargo.lock
 **/rusty-tags.vi
-/rpm/SOURCES
-/.vscode

.rustfmt.toml

@@ -1 +1 @@
-edition = "2021"
+edition = "2018"

CODEOWNERS

@@ -1,2 +0,0 @@
-# Add the list of code owners here (using their GitHub username)
-* @cloud-hypervisor/cloud-hypervisor-reviewers

CONTRIBUTING.md

@@ -2,14 +2,12 @@
 
 Cloud Hypervisor is an open source project licensed under the [Apache v2
 License](https://opensource.org/licenses/Apache-2.0) and the [BSD 3
-Clause](https://opensource.org/licenses/BSD-3-Clause) license. Individual files
-contain details of their licensing and changes to that file are under the same
-license unless the contribution changes the license of the file. When importing
-code from a third party project (e.g. Firecracker or CrosVM) please respect the
-license of those projects.
-
-New code should be under the [Apache v2
-License](https://opensource.org/licenses/Apache-2.0).
+Clause](https://opensource.org/licenses/BSD-3-Clause) license. Contributions
+can be made under either license or both. Individual files contain details of
+their licensing and changes to that file are under the same license unless the
+contribution changes the license of the file. When importing code from a third
+party project (e.g. Firecracker or CrosVM) please respect the license of those
+projects.
 
 ## Coding Style
 
@@ -17,23 +15,6 @@ We follow the [Rust Style](https://github.com/rust-dev-tools/fmt-rfcs/blob/maste
 convention and enforce it through the Continuous Integration (CI) process calling into `rustfmt`
 for each submitted Pull Request (PR).
 
-## Basic Checks
-
-Please consider creating the following hook as `.git/hooks/pre-commit` in order
-to ensure basic correctness of your code. You can extend this further if you
-have specific features that you regularly develop against.
-
-```sh
-#!/bin/sh
-
-cargo fmt -- --check || exit 1
-cargo check --locked --all --all-targets --tests || exit 1
-cargo clippy --locked --all --all-targets --tests -- -D warnings || exit 1
-```
-
-You will need to `chmod +x .git/hooks/pre-commit` to have it run on every
-commit you make.
-
 ## Certificate of Origin
 
 In order to get a clear contribution chain of trust we use the [signed-off-by language](https://01.org/community/signed-process)
@@ -82,9 +63,11 @@ you want to merge your changes to `cloud-hypervisor`:
 2. Within your fork, create a branch for your contribution.
 3. [Create a pull request](https://help.github.com/articles/creating-a-pull-request-from-a-fork/)
    against the main branch of the Cloud Hypervisor repository.
-4. To update your pull request amend existing commits whenever applicable and
+4. Add reviewers to your pull request and then work with your reviewers to address
+   any comments and obtain minimum of 2 [maintainers](MAINTAINERS.md) approvals.
+   To update your pull request amend existing commits whenever applicable and
    then push the new changes to your pull request branch.
-5. Once the pull request is approved it can be integrated.
+5. Once the pull request is approved, one of the maintainers will merge it.
 
 ## Issue tracking
 

Cargo.toml

@@ -1,8 +1,8 @@
 [package]
 name = "cloud-hypervisor"
-version = "34.0.0"
+version = "22.1.0"
 authors = ["The Cloud Hypervisor Authors"]
-edition = "2021"
+edition = "2018"
 default-run = "cloud-hypervisor"
 build = "build.rs"
 license = "LICENSE-APACHE & LICENSE-BSD-3-Clause"
@@ -10,77 +10,65 @@ description = "Open source Virtual Machine Monitor (VMM) that runs on top of KVM
 homepage = "https://github.com/cloud-hypervisor/cloud-hypervisor"
 # Minimum buildable version:
 # Keep in sync with version in .github/workflows/build.yaml
-# Policy on MSRV (see #4318):
-# Can only be bumped by:
-# a.) A dependency requires it,
-# b.) If we want to use a new feature and that MSRV is at least 6 months old,
-# c.) There is a security issue that is addressed by the toolchain update.
-rust-version = "1.64"
+rust-version = "1.54"
 
 [profile.release]
 lto = true
-codegen-units = 1
-opt-level = "s"
-strip = true
-
-[profile.profiling]
-inherits = "release"
-strip = false
-debug = true
 
 [dependencies]
-anyhow = "1.0.71"
+anyhow = "1.0.55"
 api_client = { path = "api_client" }
-argh = "0.1.9"
-dhat = { version = "0.3.2", optional = true }
-epoll = "4.3.3"
+clap = { version = "3.1.5", features = ["wrap_help","cargo"] }
+epoll = "4.3.1"
 event_monitor = { path = "event_monitor" }
 hypervisor = { path = "hypervisor" }
-libc = "0.2.139"
-log = { version = "0.4.17", features = ["std"] }
+libc = "0.2.119"
+log = { version = "0.4.14", features = ["std"] }
 option_parser = { path = "option_parser" }
-seccompiler = "0.3.0"
-serde_json = "1.0.96"
-signal-hook = "0.3.17"
-thiserror = "1.0.40"
-tpm = { path = "tpm"}
-tracer = { path = "tracer" }
+seccompiler = "0.2.0"
+serde_json = "1.0.79"
+signal-hook = "0.3.13"
+thiserror = "1.0.30"
 vmm = { path = "vmm" }
-vmm-sys-util = "0.11.0"
-vm-memory = "0.11.0"
-zbus = { version = "3.11.1", optional = true }
+vmm-sys-util = "0.9.0"
+vm-memory = "0.7.0"
+
+[build-dependencies]
+clap = { version = "3.1.5", features = ["wrap_help"] }
 
 # List of patched crates
 [patch.crates-io]
-kvm-bindings = { git = "https://github.com/cloud-hypervisor/kvm-bindings", branch = "ch-v0.6.0-tdx" }
+kvm-bindings = { git = "https://github.com/cloud-hypervisor/kvm-bindings", branch = "ch-v0.5.0-tdx" }
 kvm-ioctls = { git = "https://github.com/rust-vmm/kvm-ioctls", branch = "main" }
 versionize_derive = { git = "https://github.com/cloud-hypervisor/versionize_derive", branch = "ch" }
-vhost = { git = "https://github.com/rust-vmm/vhost", branch = "main" }
+virtio-queue = { git = "https://github.com/rust-vmm/vm-virtio", branch = "main" }
 
 [dev-dependencies]
-dirs = "5.0.0"
+dirs = "4.0.0"
+lazy_static= "1.4.0"
 net_util = { path = "net_util" }
-once_cell = "1.18.0"
-serde_json = "1.0.96"
+serde_json = "1.0.79"
 test_infra = { path = "test_infra" }
 wait-timeout = "0.2.0"
 
 [features]
-default = ["kvm", "io_uring"]
-dbus_api = ["zbus", "vmm/dbus_api"]
-dhat-heap = ["dhat"] # For heap profiling
-guest_debug = ["vmm/guest_debug"]
-io_uring = ["vmm/io_uring"]
+default = ["common", "kvm"]
+# Common features for all hypervisors
+common = ["acpi", "cmos", "fwdebug"]
+acpi = ["vmm/acpi"]
+cmos = ["vmm/cmos"]
+fwdebug = ["vmm/fwdebug"]
+gdb = ["vmm/gdb"]
 kvm = ["vmm/kvm"]
 mshv = ["vmm/mshv"]
 tdx = ["vmm/tdx"]
-tracing = ["vmm/tracing", "tracer/tracing"]
 
 [workspace]
 members = [
+    "acpi_tables",
     "api_client",
     "arch",
-    "block",
+    "block_util",
     "devices",
     "event_monitor",
     "hypervisor",
@@ -89,10 +77,10 @@ members = [
     "option_parser",
     "pci",
     "performance-metrics",
+    "qcow",
     "rate_limiter",
-    "serial_buffer",
-    "test_infra",
-    "tracer",
+    "vfio_user",
+    "vhdx",
     "vhost_user_block",
     "vhost_user_net",
     "virtio-devices",

Jenkinsfile

@@ -1,509 +1,396 @@
 def runWorkers = true
-pipeline {
-    agent none
-    options {
-        timeout(time: 4, unit: 'HOURS')
-    }
-    stages {
-        stage('Early checks') {
-            agent { node { label 'built-in' } }
-            stages {
-                stage('Checkout') {
-                    steps {
-                        checkout scm
-                    }
-                }
-                stage('Check if worker build can be skipped') {
-                    when {
-                        expression {
-                            return skipWorkerBuild()
-                        }
-                    }
-                    steps {
-                        script {
-                            runWorkers = false
-                            echo 'No changes requring a build'
-                        }
-                    }
-                }
-                stage('Check for RFC/WIP builds') {
-                    when {
-                        changeRequest comparator: 'REGEXP', title: '.*(rfc|RFC|wip|WIP).*'
-                        beforeAgent true
-                    }
-                    steps {
-                        error('Failing as this is marked as a WIP or RFC PR.')
-                    }
-                }
-                stage('Cancel older builds') {
-                    when { not { branch 'main' } }
-                    steps {
-                        cancelPreviousBuilds()
-                    }
-                }
-            }
-        }
-        stage('Build') {
-            parallel {
-                stage('Worker build') {
-                    agent { node { label 'jammy' } }
-                    when {
-                        beforeAgent true
-                        expression {
-                            return runWorkers
-                        }
-                    }
-                    stages {
-                        stage('Checkout') {
-                            steps {
-                                checkout scm
-                            }
-                        }
-                        stage('Prepare environment') {
-                            steps {
-                                sh 'scripts/prepare_vdpa.sh'
-                            }
-                        }
-                        stage('Run OpenAPI tests') {
-                            steps {
-                                sh 'scripts/run_openapi_tests.sh'
-                            }
-                        }
-                        stage('Run unit tests') {
-                            steps {
-                                sh 'scripts/dev_cli.sh tests --unit'
-                            }
-                        }
-                        stage('Run integration tests') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration'
-                            }
-                        }
-                        stage('Run live-migration integration tests') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration-live-migration'
-                            }
-                        }
-                        stage('Run unit tests for musl') {
-                            steps {
-                                sh 'scripts/dev_cli.sh tests --unit --libc musl'
-                            }
-                        }
-                        stage('Run integration tests for musl') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration --libc musl'
-                            }
-                        }
-                        stage('Run live-migration integration tests for musl') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration-live-migration --libc musl'
-                            }
-                        }
-                    }
-                }
-                stage('Worker build - AMD') {
-                    agent { node { label 'jammy-amd' } }
-                    when {
-                        beforeAgent true
-                        expression {
-                            return runWorkers
-                        }
-                    }
-                    stages {
-                        stage('Checkout') {
-                            steps {
-                                checkout scm
-                            }
-                        }
-                        stage('Prepare environment') {
-                            steps {
-                                sh 'scripts/prepare_vdpa.sh'
-                            }
-                        }
-                        stage('Run integration tests') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration'
-                            }
-                        }
-                        stage('Run live-migration integration tests') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration-live-migration --integration -- -- --skip live_migration::live_migration_parallel::test_live_upgrade_watchdog --skip live_migration::live_migration_parallel::test_live_upgrade_watchdog_local'
-                            }
-                        }
-                        stage('Run integration tests for musl') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration --libc musl'
-                            }
-                        }
-                        stage('Run live-migration integration tests for musl') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration-live-migration --libc musl -- -- --skip live_migration::live_migration_parallel::test_live_upgrade_watchdog --skip live_migration::live_migration_parallel::test_live_upgrade_watchdog_local'
-                            }
-                        }
-                    }
-                }
-                stage('AArch64 worker build') {
-                    agent { node { label 'bionic-arm64' } }
-                    when {
-                        beforeAgent true
-                        expression {
-                            return runWorkers
-                        }
-                    }
-                    environment {
-                            AZURE_CONNECTION_STRING = credentials('46b4e7d6-315f-4cc1-8333-b58780863b9b')
-                    }
-                    stages {
-                        stage('Checkout') {
-                            steps {
-                                checkout scm
-                            }
-                        }
-                        stage('Run unit tests') {
-                            steps {
-                                sh 'scripts/dev_cli.sh tests --unit --libc musl'
-                            }
-                        }
-                        stage('Run integration tests') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'sudo modprobe openvswitch'
-                                sh 'scripts/dev_cli.sh tests --integration --libc musl'
-                            }
-                        }
-                        stage('Install azure-cli') {
-                            steps {
-                                installAzureCli('bionic', 'arm64')
-                            }
-                        }
-                        stage('Download Windows image') {
-                            steps {
-                                sh '''#!/bin/bash -x
-                                    IMG_BASENAME=windows-11-iot-enterprise-aarch64.raw
-                                    IMG_PATH=$HOME/workloads/$IMG_BASENAME
-                                    IMG_GZ_PATH=$HOME/workloads/$IMG_BASENAME.gz
-                                    IMG_GZ_BLOB_NAME=windows-11-iot-enterprise-aarch64-9-min.raw.gz
-                                    cp "scripts/$IMG_BASENAME.sha1" "$HOME/workloads/"
-                                    pushd "$HOME/workloads"
-                                    if sha1sum "$IMG_BASENAME.sha1" --check; then
-                                        exit
-                                    fi
-                                    popd
-                                    mkdir -p "$HOME/workloads"
-                                    az storage blob download \
-                                        --container-name private-images \
-                                        --file "$IMG_GZ_PATH" \
-                                        --name "$IMG_GZ_BLOB_NAME" \
-                                        --connection-string "$AZURE_CONNECTION_STRING"
-                                    gzip -d $IMG_GZ_PATH
-                                '''
-                            }
-                        }
-                        stage('Run Windows guest integration tests') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'scripts/dev_cli.sh tests --integration-windows --libc musl'
-                            }
-                        }
-                    }
-                    post {
-                        always {
-                            sh "sudo chown -R jenkins.jenkins ${WORKSPACE}"
-                            deleteDir()
-                        }
-                    }
-                }
-                stage('Worker build - Windows guest') {
-                    agent { node { label 'jammy' } }
-                    when {
-                        beforeAgent true
-                        expression {
-                            return runWorkers
-                        }
-                    }
-                    environment {
-                            AZURE_CONNECTION_STRING = credentials('46b4e7d6-315f-4cc1-8333-b58780863b9b')
-                    }
-                    stages {
-                        stage('Checkout') {
-                            steps {
-                                checkout scm
-                            }
-                        }
-                        stage('Install azure-cli') {
-                            steps {
-                                installAzureCli('jammy', 'amd64')
-                            }
-                        }
-                        stage('Download assets') {
-                            steps {
-                                sh "mkdir ${env.HOME}/workloads"
-                                sh 'az storage blob download --container-name private-images --file "$HOME/workloads/windows-server-2022-amd64-2.raw" --name windows-server-2022-amd64-2.raw --connection-string "$AZURE_CONNECTION_STRING"'
-                            }
-                        }
-                        stage('Run Windows guest integration tests') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'scripts/dev_cli.sh tests --integration-windows'
-                            }
-                        }
-                        stage('Run Windows guest integration tests for musl') {
-                            options {
-                                timeout(time: 1, unit: 'HOURS')
-                            }
-                            steps {
-                                sh 'scripts/dev_cli.sh tests --integration-windows --libc musl'
-                            }
-                        }
-                    }
-                }
-                // stage('Worker build - Metrics') {
-                //     agent { node { label 'jammy-metrics' } }
-                //     when {
-                //         branch 'main'
-                //         beforeAgent true
-                //         expression {
-                //             return runWorkers
-                //         }
-                //     }
-                //     environment {
-                //         METRICS_PUBLISH_KEY = credentials('52e0945f-ce7a-43d1-87af-67d1d87cc40f')
-                //     }
-                //     stages {
-                //         stage('Checkout') {
-                //             steps {
-                //                 checkout scm
-                //             }
-                //         }
-                //         stage('Run metrics tests') {
-                //             options {
-                //                 timeout(time: 1, unit: 'HOURS')
-                //             }
-                //             steps {
-                //                 sh 'scripts/dev_cli.sh tests --metrics -- -- --report-file /root/workloads/metrics.json'
-                //             }
-                //         }
-                //         stage('Upload metrics report') {
-                //             steps {
-                //                 sh 'curl -X PUT https://cloud-hypervisor-metrics.azurewebsites.net/api/publishmetrics -H "x-functions-key: $METRICS_PUBLISH_KEY" -T ~/workloads/metrics.json'
-                //             }
-                //         }
-                //     }
-                // }
-                // stage('Worker build - Rate Limiter') {
-                //     agent { node { label 'focal-metrics' } }
-                //     when {
-                //         branch 'main'
-                //         beforeAgent true
-                //         expression {
-                //             return runWorkers
-                //         }
-                //     }
-                //     stages {
-                //         stage('Checkout') {
-                //             steps {
-                //                 checkout scm
-                //             }
-                //         }
-                //         stage('Run rate-limiter integration tests') {
-                //             options {
-                //                 timeout(time: 10, unit: 'MINUTES')
-                //             }
-                //             steps {
-                //                 sh 'scripts/dev_cli.sh tests --integration-rate-limiter'
-                //             }
-                //         }
-                //     }
-                // }
-                // stage('Worker build - SGX') {
-                //     agent { node { label 'jammy-sgx' } }
-                //     when {
-                //         beforeAgent true
-                //         allOf {
-                //             branch 'main'
-                //             expression {
-                //                 return runWorkers
-                //             }
-                //         }
-                //     }
-                //     stages {
-                //         stage('Checkout') {
-                //             steps {
-                //                 checkout scm
-                //             }
-                //         }
-                //         stage('Run SGX integration tests') {
-                //             options {
-                //                 timeout(time: 1, unit: 'HOURS')
-                //             }
-                //             steps {
-                //                 sh 'scripts/dev_cli.sh tests --integration-sgx'
-                //             }
-                //         }
-                //         stage('Run SGX integration tests for musl') {
-                //             options {
-                //                 timeout(time: 1, unit: 'HOURS')
-                //             }
-                //             steps {
-                //                 sh 'scripts/dev_cli.sh tests --integration-sgx --libc musl'
-                //             }
-                //         }
-                //     }
-                //     post {
-                //         always {
-                //             sh "sudo chown -R jenkins.jenkins ${WORKSPACE}"
-                //             deleteDir()
-                //         }
-                //     }
-                // }
-                // stage('Worker build - VFIO') {
-                //     agent { node { label 'jammy-vfio' } }
-                //     when {
-                //         beforeAgent true
-                //         allOf {
-                //             branch 'main'
-                //             expression {
-                //                 return runWorkers
-                //             }
-                //         }
-                //     }
-                //     stages {
-                //         stage('Checkout') {
-                //             steps {
-                //                 checkout scm
-                //             }
-                //         }
-                //         stage('Run VFIO integration tests') {
-                //             options {
-                //                 timeout(time: 1, unit: 'HOURS')
-                //             }
-                //             steps {
-                //                 sh 'scripts/dev_cli.sh tests --integration-vfio'
-                //             }
-                //         }
-                //         stage('Run VFIO integration tests for musl') {
-                //             options {
-                //                 timeout(time: 1, unit: 'HOURS')
-                //             }
-                //             steps {
-                //                 sh 'scripts/dev_cli.sh tests --integration-vfio --libc musl'
-                //             }
-                //         }
-                //     }
-                //     post {
-                //         always {
-                //             sh "sudo chown -R jenkins.jenkins ${WORKSPACE}"
-                //             deleteDir()
-                //         }
-                //     }
-                // }
-            }
-        }
-    }
-    post {
-        regression {
-            script {
-                if (env.BRANCH_NAME == 'main') {
-                    slackSend(color: '#ff0000', message: '"main" branch build is now failing', channel: '#jenkins-ci')
-                }
-            }
-        }
-        fixed {
-            script {
-                if (env.BRANCH_NAME == 'main') {
-                    slackSend(color: '#00ff00', message: '"main" branch build is now fixed', channel: '#jenkins-ci')
-                }
-            }
-        }
-    }
+pipeline{
+	agent none
+	stages {
+		stage ('Early checks') {
+			agent { node { label 'built-in' } }
+			stages {
+				stage ('Checkout') {
+					steps {
+						checkout scm
+					}
+				}
+				stage ('Check for documentation only changes') {
+					when {
+						expression {
+							return docsFileOnly()
+						}
+					}
+					steps {
+						script {
+							runWorkers = false
+							echo "Documentation only changes, no need to run the CI"
+						}
+					}
+				}
+				stage ('Check for RFC/WIP builds') {
+					when {
+  						changeRequest comparator: 'REGEXP', title: '.*(rfc|RFC|wip|WIP).*'
+  						beforeAgent true
+					}
+					steps {
+						error("Failing as this is marked as a WIP or RFC PR.")
+					}
+				}
+				stage ('Cancel older builds') {
+					when { not { branch 'main' } }
+					steps {
+						cancelPreviousBuilds()
+					}
+				}
+			}
+		}
+		stage ('Build') {
+			parallel {
+				stage ('Worker build') {
+					agent { node { label 'hirsute' } }
+					when {
+						beforeAgent true
+						expression {
+							return runWorkers
+						}
+					}
+					stages {
+						stage ('Checkout') {
+							steps {
+								checkout scm
+							}
+						}
+
+						stage ('Run OpenAPI tests') {
+							steps {
+								sh "scripts/run_openapi_tests.sh"
+							}
+						}
+						stage ('Run unit tests') {
+							steps {
+								sh "scripts/dev_cli.sh tests --unit"
+							}
+						}
+						stage ('Run integration tests') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "sudo modprobe openvswitch"
+								sh "scripts/dev_cli.sh tests --integration"
+							}
+						}
+					}
+				}
+				stage ('AArch64 worker build') {
+					agent { node { label 'bionic-arm64' } }
+					when {
+						beforeAgent true
+						expression {
+							return runWorkers
+						}
+					}
+					stages {
+						stage ('Checkout') {
+							steps {
+								checkout scm
+							}
+						}
+						stage ('Run unit tests') {
+							steps {
+								sh "scripts/dev_cli.sh tests --unit --libc musl"
+							}
+						}
+						stage ('Run integration tests') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "sudo modprobe openvswitch"
+								sh "scripts/dev_cli.sh tests --integration --libc musl"
+							}
+						}
+					}
+					post {
+						always {
+							sh "sudo chown -R jenkins.jenkins ${WORKSPACE}"
+							deleteDir()
+						}
+					}
+				}
+				stage ('Worker build (musl)') {
+					agent { node { label 'hirsute' } }
+					when {
+						beforeAgent true
+						expression {
+							return runWorkers
+						}
+					}
+					stages {
+						stage ('Checkout') {
+							steps {
+								checkout scm
+							}
+						}
+
+						stage ('Run unit tests for musl') {
+							steps {
+								sh "scripts/dev_cli.sh tests --unit --libc musl"
+							}
+						}
+						stage ('Run integration tests for musl') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "sudo modprobe openvswitch"
+								sh "scripts/dev_cli.sh tests --integration --libc musl"
+							}
+						}
+					}
+				}
+				stage ('Worker build SGX') {
+					agent { node { label 'bionic-sgx' } }
+					when {
+						beforeAgent true
+						allOf {
+							branch 'main'
+							expression {
+								return runWorkers
+							}
+						}
+					}
+					stages {
+						stage ('Checkout') {
+							steps {
+								checkout scm
+							}
+						}
+						stage ('Run SGX integration tests') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "scripts/dev_cli.sh tests --integration-sgx"
+							}
+						}
+						stage ('Run SGX integration tests for musl') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "scripts/dev_cli.sh tests --integration-sgx --libc musl"
+							}
+						}
+					}
+					post {
+						always {
+							sh "sudo chown -R jenkins.jenkins ${WORKSPACE}"
+							deleteDir()
+						}
+					}
+				}
+				stage ('Worker build VFIO') {
+					agent { node { label 'bionic-vfio' } }
+					when {
+						beforeAgent true
+						allOf {
+							branch 'main'
+							expression {
+								return runWorkers
+							}
+						}
+					}
+					stages {
+						stage ('Checkout') {
+							steps {
+								checkout scm
+							}
+						}
+						stage ('Run VFIO integration tests') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "scripts/dev_cli.sh tests --integration-vfio"
+							}
+						}
+						stage ('Run VFIO integration tests for musl') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "scripts/dev_cli.sh tests --integration-vfio --libc musl"
+							}
+						}
+					}
+					post {
+						always {
+							sh "sudo chown -R jenkins.jenkins ${WORKSPACE}"
+							deleteDir()
+						}
+					}
+				}
+				stage ('Worker build - Windows guest') {
+					agent { node { label 'hirsute' } }
+					when {
+						beforeAgent true
+						expression {
+							return runWorkers
+						}
+					}
+					environment {
+        					AZURE_CONNECTION_STRING = credentials('46b4e7d6-315f-4cc1-8333-b58780863b9b')
+					}
+					stages {
+						stage ('Checkout') {
+							steps {
+								checkout scm
+							}
+						}
+						stage ('Install azure-cli') {
+							steps {
+								installAzureCli()
+							}
+						}
+						stage ('Download assets') {
+							steps {
+								sh "mkdir ${env.HOME}/workloads"
+								sh 'az storage blob download --container-name private-images --file "$HOME/workloads/windows-server-2019.raw" --name windows-server-2019.raw --connection-string "$AZURE_CONNECTION_STRING"'
+							}
+						}
+						stage ('Run Windows guest integration tests') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "scripts/dev_cli.sh tests --integration-windows"
+							}
+						}
+						stage ('Run Windows guest integration tests for musl') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "scripts/dev_cli.sh tests --integration-windows --libc musl"
+							}
+						}
+					}
+				}
+				stage ('Worker build - Live Migration') {
+					agent { node { label 'hirsute-small' } }
+					when {
+						beforeAgent true
+						expression {
+							return runWorkers
+						}
+					}
+					stages {
+						stage ('Checkout') {
+							steps {
+								checkout scm
+							}
+						}
+						stage ('Run live-migration integration tests') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "sudo modprobe openvswitch"
+								sh "scripts/dev_cli.sh tests --integration-live-migration"
+							}
+						}
+						stage ('Run live-migration integration tests for musl') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh "sudo modprobe openvswitch"
+								sh "scripts/dev_cli.sh tests --integration-live-migration --libc musl"
+							}
+						}
+					}
+				}
+				stage ('Worker build - Metrics') {
+					agent { node { label 'focal-metrics' } }
+					when {
+						branch 'main'
+						beforeAgent true
+						expression {
+							return runWorkers
+						}
+					}
+					environment {
+						METRICS_PUBLISH_KEY = credentials('52e0945f-ce7a-43d1-87af-67d1d87cc40f')
+					}
+					stages {
+						stage ('Checkout') {
+							steps {
+								checkout scm
+							}
+						}
+						stage ('Run metrics tests') {
+							options {
+								timeout(time: 1, unit: 'HOURS')
+							}
+							steps {
+								sh 'scripts/dev_cli.sh tests --metrics -- -- --report-file /root/workloads/metrics.json'
+							}
+						}
+						stage ('Upload metrics report') {
+							steps {
+								sh 'curl -X PUT https://cloud-hypervisor-metrics.azurewebsites.net/api/publishmetrics -H "x-functions-key: $METRICS_PUBLISH_KEY" -T ~/workloads/metrics.json'
+							}
+						}
+					}
+				}
+			}
+		}
+	}
+	post {
+		regression {
+			script {
+				if (env.BRANCH_NAME == 'main') {
+					slackSend (color: '#ff0000', message: '"main" branch build is now failing')
+				}
+			}
+		}
+		fixed {
+			script {
+				if (env.BRANCH_NAME == 'main') {
+					slackSend (color: '#00ff00', message: '"main" branch build is now fixed')
+				}
+			}
+		}
+	}
 }
 
 def cancelPreviousBuilds() {
-    // Check for other instances of this particular build, cancel any that are older than the current one
-    def jobName = env.JOB_NAME
-    def currentBuildNumber = env.BUILD_NUMBER.toInteger()
-    def currentJob = Jenkins.instance.getItemByFullName(jobName)
+	// Check for other instances of this particular build, cancel any that are older than the current one
+	def jobName = env.JOB_NAME
+		def currentBuildNumber = env.BUILD_NUMBER.toInteger()
+		def currentJob = Jenkins.instance.getItemByFullName(jobName)
 
-    // Loop through all instances of this particular job/branch
-    for (def build : currentJob.builds) {
-        if (build.isBuilding() && (build.number.toInteger() < currentBuildNumber)) {
-            echo "Older build still queued. Sending kill signal to build number: ${build.number}"
-            build.doStop()
-        }
-    }
+		// Loop through all instances of this particular job/branch
+		for (def build : currentJob.builds) {
+			if (build.isBuilding() && (build.number.toInteger() < currentBuildNumber)) {
+				echo "Older build still queued. Sending kill signal to build number: ${build.number}"
+				build.doStop()
+			}
+		}
 }
 
-def installAzureCli(distro, arch) {
-    sh 'sudo apt install -y ca-certificates curl apt-transport-https lsb-release gnupg'
-    sh 'curl -sL https://packages.microsoft.com/keys/microsoft.asc | gpg --dearmor | sudo tee /etc/apt/trusted.gpg.d/microsoft.gpg > /dev/null'
-    sh "echo \"deb [arch=${arch}] https://packages.microsoft.com/repos/azure-cli/ ${distro} main\" | sudo tee /etc/apt/sources.list.d/azure-cli.list"
-    sh 'sudo apt update'
-    sh 'sudo apt install -y azure-cli'
+def installAzureCli() {
+	sh "sudo apt install -y ca-certificates curl apt-transport-https lsb-release gnupg"
+	sh "curl -sL https://packages.microsoft.com/keys/microsoft.asc | gpg --dearmor | sudo tee /etc/apt/trusted.gpg.d/microsoft.gpg > /dev/null"
+	sh "echo \"deb [arch=amd64] https://packages.microsoft.com/repos/azure-cli/ hirsute main\" | sudo tee /etc/apt/sources.list.d/azure-cli.list"
+	sh "sudo apt update"
+	sh "sudo apt install -y azure-cli"
 }
 
-def boolean skipWorkerBuild() {
+def boolean docsFileOnly() {
     if (env.CHANGE_TARGET == null) {
-        return false
+        return false;
     }
 
-    if (sh(
+    return sh(
         returnStatus: true,
         script: "git diff --name-only origin/${env.CHANGE_TARGET}... | grep -v '\\.md'"
-    ) != 0) {
-        return true
-    }
-
-    if (sh(
-        returnStatus: true,
-        script: "git diff --name-only origin/${env.CHANGE_TARGET}... | grep -v -E 'fuzz/'"
-    ) != 0) {
-        return true
-    }
-
-    if (sh(
-        returnStatus: true,
-        script: "git diff --name-only origin/${env.CHANGE_TARGET}... | grep -v -E '.github/'"
-    ) != 0) {
-        return true
-    }
-
-    return false
+    ) != 0
 }

MAINTAINERS.md

@@ -2,7 +2,6 @@
 
 - Sebastien Boeuf - @sboeuf
 - Robert Bradford - @rbradford
-- Bo Chen - @likebreath
 - Samuel Ortiz - @sameo
 - Wei Liu - @liuw
 - Michael Zhao - @michael2012z

README.md

@@ -4,43 +4,42 @@
     - [Architectures](#architectures)
     - [Guest OS](#guest-os)
 - [2. Getting Started](#2-getting-started)
-  - [Host OS](#host-os)
-  - [Use Pre-built Binaries](#use-pre-built-binaries)
-  - [Packages](#packages)
-  - [Building from Source](#building-from-source)
-  - [Booting Linux](#booting-linux)
-    - [Firmware Booting](#firmware-booting)
-    - [Custom Kernel and Disk Image](#custom-kernel-and-disk-image)
-      - [Building your Kernel](#building-your-kernel)
+  - [Preparation](#preparation)
+  - [Install prerequisites](#install-prerequisites)
+  - [Clone and build](#clone-and-build)
+    - [Containerized builds and tests](#containerized-builds-and-tests)
+  - [Run](#run)
+    - [Cloud image](#cloud-image)
+    - [Custom kernel and disk image](#custom-kernel-and-disk-image)
+      - [Building your kernel](#building-your-kernel)
       - [Disk image](#disk-image)
       - [Booting the guest VM](#booting-the-guest-vm)
 - [3. Status](#3-status)
   - [Hot Plug](#hot-plug)
   - [Device Model](#device-model)
-  - [Roadmap](#roadmap)
-- [4. Relationship with _Rust VMM_ Project](#4-relationship-with-rust-vmm-project)
-  - [Differences with Firecracker and crosvm](#differences-with-firecracker-and-crosvm)
+  - [TODO](#todo)
+- [4. `rust-vmm` project dependency](#4-rust-vmm-project-dependency)
+  - [Firecracker and crosvm](#firecracker-and-crosvm)
 - [5. Community](#5-community)
   - [Contribute](#contribute)
-  - [Slack](#slack)
-  - [Mailing list](#mailing-list)
+  - [Join us](#join-us)
   - [Security issues](#security-issues)
 
 # 1. What is Cloud Hypervisor?
 
 Cloud Hypervisor is an open source Virtual Machine Monitor (VMM) that runs on
-top of the [KVM](https://www.kernel.org/doc/Documentation/virtual/kvm/api.txt)
-hypervisor and the Microsoft Hypervisor (MSHV).
+top of [KVM](https://www.kernel.org/doc/Documentation/virtual/kvm/api.txt)
+hypervisor and Microsoft Hypervisor (MSHV).
 
-The project focuses on running modern, _Cloud Workloads_, on specific, common,
-hardware architectures. In this case _Cloud Workloads_ refers to those that are
-run by customers inside a Cloud Service Provider. This means modern operating
-systems with most I/O handled by
-paravirtualised devices (e.g. _virtio_), no requirement for legacy devices, and
+The project focuses on exclusively running modern, cloud workloads, on top of
+a limited set of hardware architectures and platforms. Cloud workloads refers
+to those that are usually run by customers inside a cloud provider. For our
+purposes this means modern operating systems with most I/O handled by
+paravirtualised devices (i.e. virtio), no requirement for legacy devices, and
 64-bit CPUs.
 
 Cloud Hypervisor is implemented in [Rust](https://www.rust-lang.org/) and is
-based on the [Rust VMM](https://github.com/rust-vmm) crates.
+based on the [rust-vmm](https://github.com/rust-vmm) crates.
 
 ## Objectives
 
@@ -60,7 +59,7 @@ based on the [Rust VMM](https://github.com/rust-vmm) crates.
 ### Architectures
 
 Cloud Hypervisor supports the `x86-64` and `AArch64` architectures. There are
-minor differences in functionality between the two architectures
+some small differences in functionality between the two architectures
 (see [#1125](https://github.com/cloud-hypervisor/cloud-hypervisor/issues/1125)).
 
 ### Guest OS
@@ -69,189 +68,189 @@ Cloud Hypervisor supports `64-bit Linux` and Windows 10/Windows Server 2019.
 
 # 2. Getting Started
 
-The following sections describe how to build and run Cloud Hypervisor.
+Below sections describe how to build and run Cloud Hypervisor on the `x86_64`
+platform. For getting started on the `AArch64` platform, please refer to the
+[Arm64 documentation](docs/arm64.md).
 
-## Prerequisites for AArch64
+## Preparation
 
-- AArch64 servers (recommended) or development boards equipped with the GICv3
-  interrupt controller.
-
-## Host OS
-
-For required KVM functionality and adequate performance the recommended host
-kernel version is 5.13. The majority of the CI currently tests with kernel
-version 5.15.
-
-## Use Pre-built Binaries
-
-The recommended approach to getting started with Cloud Hypervisor is by using a
-pre-built binary. Binaries are available for the [latest
-release](https://github.com/cloud-hypervisor/cloud-hypervisor/releases/latest).
-Use `cloud-hypervisor-static` for `x86-64` or `cloud-hypervisor-static-aarch64`
-for `AArch64` platform.
-
-## Packages
-
-For convenience, packages are also available targeting some popular Linux
-distributions. This is thanks to the [Open Build
-Service](https://build.opensuse.org). The [OBS
-README](https://github.com/cloud-hypervisor/obs-packaging) explains how to
-enable the repository in a supported Linux distribution and install Cloud Hypervisor
-and accompanying packages. Please report any packaging issues in the
-[obs-packaging](https://github.com/cloud-hypervisor/obs-packaging) repository.
-
-## Building from Source
-
-Please see the [instructions for building from source](docs/building.md) if you
-do not wish to use the pre-built binaries.
-
-## Booting Linux
-
-Cloud Hypervisor supports direct kernel boot (the x86-64 kernel requires the kernel
-built with PVH support) or booting via a firmware (either [Rust Hypervisor
-Firmware](https://github.com/cloud-hypervisor/rust-hypervisor-firmware) or an
-edk2 UEFI firmware called `CLOUDHV` / `CLOUDHV_EFI`.)
-
-Binary builds of the firmware files are available for the latest release of
-[Rust Hypervisor
-Firmware](https://github.com/cloud-hypervisor/rust-hypervisor-firmware/releases/latest)
-and [our edk2
-repository](https://github.com/cloud-hypervisor/edk2/releases/latest)
-
-The choice of firmware depends on your guest OS choice; some experimentation
-may be required.
-
-### Firmware Booting
-
-Cloud Hypervisor supports booting disk images containing all needed components
-to run cloud workloads, a.k.a. cloud images.
-
-The following sample commands will download an Ubuntu Cloud image, converting
-it into a format that Cloud Hypervisor can use and a firmware to boot the image
-with.
+We create a folder to build and run `cloud-hypervisor` at `$HOME/cloud-hypervisor`
 
 ```shell
-$ wget https://cloud-images.ubuntu.com/focal/current/focal-server-cloudimg-amd64.img
-$ qemu-img convert -p -f qcow2 -O raw focal-server-cloudimg-amd64.img focal-server-cloudimg-amd64.raw
-$ wget https://github.com/cloud-hypervisor/rust-hypervisor-firmware/releases/download/0.4.2/hypervisor-fw
+$ export CLOUDH=$HOME/cloud-hypervisor
+$ mkdir $CLOUDH
 ```
 
-The Ubuntu cloud images do not ship with a default password so it necessary to
-use a `cloud-init` disk image to customise the image on the first boot. A basic
-`cloud-init` image is generated by this [script](scripts/create-cloud-init.sh).
-This seeds the image with a default username/password of `cloud/cloud123`. It
-is only necessary to add this disk image on the first boot. Script also assigns
-default IP address using `test_data/cloud-init/ubuntu/local/network-config` details
-with `--net "mac=12:34:56:78:90:ab,tap="` option. Then the matching mac address
-interface will be enabled as per `network-config` details.
+## Install prerequisites
+
+You need to install some prerequisite packages in order to build and test Cloud
+Hypervisor. Here, all the steps are based on Ubuntu, for other Linux
+distributions please replace the package manager and package name.
 
 ```shell
-$ sudo setcap cap_net_admin+ep ./cloud-hypervisor
-$ ./create-cloud-init.sh
-$ ./cloud-hypervisor \
+# Install git
+$ sudo apt install git
+# Install rust tool chain
+$ curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
+# Install build-essential
+$ sudo apt install build-essential
+# If you want to build statically linked binary please add musl target
+$ rustup target add x86_64-unknown-linux-musl
+```
+
+## Clone and build
+
+First you need to clone and build the cloud-hypervisor repo:
+
+```shell
+$ pushd $CLOUDH
+$ git clone https://github.com/cloud-hypervisor/cloud-hypervisor.git
+$ cd cloud-hypervisor
+$ cargo build --release
+
+# We need to give the cloud-hypervisor binary the NET_ADMIN capabilities for it to set TAP interfaces up on the host.
+$ sudo setcap cap_net_admin+ep ./target/release/cloud-hypervisor
+
+# If you want to build statically linked binary
+$ cargo build --release --target=x86_64-unknown-linux-musl --all
+$ popd
+```
+
+This will build a `cloud-hypervisor` binary under
+`$CLOUDH/cloud-hypervisor/target/release/cloud-hypervisor`.
+
+### Containerized builds and tests
+
+If you want to build and test Cloud Hypervisor without having to install all the
+required dependencies (The rust toolchain, cargo tools, etc), you can also use
+Cloud Hypervisor's development script: `dev_cli.sh`. Please note that upon its
+first invocation, this script will pull a fairly large container image.
+
+For example, to build the Cloud Hypervisor release binary:
+
+```shell
+$ pushd $CLOUDH
+$ cd cloud-hypervisor
+$ ./scripts/dev_cli.sh build --release
+```
+
+With `dev_cli.sh`, one can also run the Cloud Hypervisor CI locally. This can be
+very convenient for debugging CI errors without having to fully rely on the
+Cloud Hypervisor CI infrastructure.
+
+For example, to run the Cloud Hypervisor unit tests:
+
+```shell
+$ ./scripts/dev_cli.sh tests --unit
+```
+
+Run the `./scripts/dev_cli.sh --help` command to view all the supported
+development script commands and their related options.
+
+## Run
+
+You can run a guest VM by either using an existing cloud image or booting into
+your own kernel and disk image.
+
+### Cloud image
+
+Cloud Hypervisor supports booting disk images containing all needed
+components to run cloud workloads, a.k.a. cloud images. To do that we rely on
+the [Rust Hypervisor
+Firmware](https://github.com/cloud-hypervisor/rust-hypervisor-firmware) project
+to provide an ELF formatted KVM firmware for `cloud-hypervisor` to directly
+boot into.
+
+We need to get the latest `rust-hypervisor-firmware` release and also a working
+cloud image. Here we will use a Ubuntu image:
+
+```shell
+$ pushd $CLOUDH
+$ wget https://cloud-images.ubuntu.com/focal/current/focal-server-cloudimg-amd64.img
+$ qemu-img convert -p -f qcow2 -O raw focal-server-cloudimg-amd64.img focal-server-cloudimg-amd64.raw
+$ wget https://github.com/cloud-hypervisor/rust-hypervisor-firmware/releases/download/0.3.2/hypervisor-fw
+$ popd
+```
+
+```shell
+$ pushd $CLOUDH
+$ sudo setcap cap_net_admin+ep ./cloud-hypervisor/target/release/cloud-hypervisor
+$ ./cloud-hypervisor/target/release/cloud-hypervisor \
 	--kernel ./hypervisor-fw \
-	--disk path=focal-server-cloudimg-amd64.raw --disk path=/tmp/ubuntu-cloudinit.img \
+	--disk path=focal-server-cloudimg-amd64.raw \
 	--cpus boot=4 \
 	--memory size=1024M \
 	--net "tap=,mac=,ip=,mask="
+$ popd
 ```
 
-If access to the firmware messages or interaction with the boot loader (e.g.
-GRUB) is required then it necessary to switch to the serial console instead of
-`virtio-console`.
+Multiple arguments can be given to the `--disk` parameter.
 
-```shell
-$ ./cloud-hypervisor \
-	--kernel ./hypervisor-fw \
-	--disk path=focal-server-cloudimg-amd64.raw --disk path=/tmp/ubuntu-cloudinit.img \
-	--cpus boot=4 \
-	--memory size=1024M \
-	--net "tap=,mac=,ip=,mask=" \
-	--serial tty \
-	--console off
-```
+### Custom kernel and disk image
 
-### Custom Kernel and Disk Image
+#### Building your kernel
 
-#### Building your Kernel
-
-Cloud Hypervisor also supports direct kernel boot. For x86-64, a `vmlinux` ELF kernel (compiled with PVH support) is needed. In order to support development there is a custom branch; however provided the required options are enabled any recent kernel will suffice.
+Cloud Hypervisor also supports direct kernel boot into a `vmlinux` ELF kernel.
+In order to support virtio-watchdog we have our own development branch. You are
+of course able to use your own kernel but these instructions will continue with
+the version that we develop and test against.
 
 To build the kernel:
 
 ```shell
+
 # Clone the Cloud Hypervisor Linux branch
-$ git clone --depth 1 https://github.com/cloud-hypervisor/linux.git -b ch-6.2 linux-cloud-hypervisor
+$ pushd $CLOUDH
+$ git clone --depth 1 https://github.com/cloud-hypervisor/linux.git -b ch-5.15.12 linux-cloud-hypervisor
 $ pushd linux-cloud-hypervisor
-# Use the x86-64 cloud-hypervisor kernel config to build your kernel for x86-64
-$ wget https://raw.githubusercontent.com/cloud-hypervisor/cloud-hypervisor/main/resources/linux-config-x86_64
-# Use the AArch64 cloud-hypervisor kernel config to build your kernel for AArch64
-$ wget https://raw.githubusercontent.com/cloud-hypervisor/cloud-hypervisor/main/resources/linux-config-aarch64
-$ cp linux-config-x86_64 .config  # x86-64
-$ cp linux-config-aarch64 .config # AArch64
-# Do native build of the x86-64 kernel
+
+# Use the cloud-hypervisor kernel config to build your kernel
+$ cp $CLOUDH/cloud-hypervisor/resources/linux-config-x86_64 .config
 $ KCFLAGS="-Wa,-mx86-used-note=no" make bzImage -j `nproc`
-# Do native build of the AArch64 kernel
-$ make -j `nproc`
 $ popd
 ```
 
-For x86-64, the `vmlinux` kernel image will then be located at
+The `vmlinux` kernel image will then be located at
 `linux-cloud-hypervisor/arch/x86/boot/compressed/vmlinux.bin`.
-For AArch64, the `Image` kernel image will then be located at
-`linux-cloud-hypervisor/arch/arm64/boot/Image`.
 
 #### Disk image
 
-For the disk image the same Ubuntu image as before can be used. This contains
-an `ext4` root filesystem.
+For the disk image, we will use a Ubuntu cloud image that contains a root
+partition:
 
 ```shell
-$ wget https://cloud-images.ubuntu.com/focal/current/focal-server-cloudimg-amd64.img # x86-64
-$ wget https://cloud-images.ubuntu.com/focal/current/focal-server-cloudimg-arm64.img # AArch64
-$ qemu-img convert -p -f qcow2 -O raw focal-server-cloudimg-amd64.img focal-server-cloudimg-amd64.raw # x86-64
-$ qemu-img convert -p -f qcow2 -O raw focal-server-cloudimg-arm64.img focal-server-cloudimg-arm64.raw # AArch64
+$ pushd $CLOUDH
+$ wget https://cloud-images.ubuntu.com/focal/current/focal-server-cloudimg-amd64.img
+$ qemu-img convert -p -f qcow2 -O raw focal-server-cloudimg-amd64.img focal-server-cloudimg-amd64.raw
+$ popd
 ```
 
 #### Booting the guest VM
 
-These sample commands boot the disk image using the custom kernel whilst also
-supplying the desired kernel command line.
-
-- x86-64
+Now we can directly boot into our custom kernel and make it use the Ubuntu root
+partition. If we want to have 4 vCPUs and 1024 MBytes of memory:
 
 ```shell
-$ sudo setcap cap_net_admin+ep ./cloud-hypervisor
-$ ./create-cloud-init.sh
-$ ./cloud-hypervisor \
+$ pushd $CLOUDH
+$ sudo setcap cap_net_admin+ep ./cloud-hypervisor/target/release/cloud-hypervisor
+$ ./cloud-hypervisor/target/release/cloud-hypervisor \
 	--kernel ./linux-cloud-hypervisor/arch/x86/boot/compressed/vmlinux.bin \
-	--disk path=focal-server-cloudimg-amd64.raw --disk path=/tmp/ubuntu-cloudinit.img \
+	--disk path=focal-server-cloudimg-amd64.raw \
 	--cmdline "console=hvc0 root=/dev/vda1 rw" \
 	--cpus boot=4 \
 	--memory size=1024M \
 	--net "tap=,mac=,ip=,mask="
 ```
 
-- AArch64
+The above example use the `virtio-console` device as the guest console, and this
+device may not be enabled soon enough by the guest kernel to get early kernel
+debug messages.
+
+When in need for earlier debug messages, using the legacy serial device based
+console is preferred:
 
-```shell
-$ sudo setcap cap_net_admin+ep ./cloud-hypervisor
-$ ./create-cloud-init.sh
-$ ./cloud-hypervisor \
-	--kernel ./linux-cloud-hypervisor/arch/arm64/boot/Image \
-	--disk path=focal-server-cloudimg-arm64.raw --disk path=/tmp/ubuntu-cloudinit.img \
-	--cmdline "console=hvc0 root=/dev/vda1 rw" \
-	--cpus boot=4 \
-	--memory size=1024M \
-	--net "tap=,mac=,ip=,mask="
 ```
-
-If earlier kernel messages are required the serial console should be used instead of `virtio-console`.
-
-- x86-64
-
-```shell
-$ ./cloud-hypervisor \
+$ ./cloud-hypervisor/target/release/cloud-hypervisor \
 	--kernel ./linux-cloud-hypervisor/arch/x86/boot/compressed/vmlinux.bin \
 	--console off \
 	--serial tty \
@@ -262,52 +261,35 @@ $ ./cloud-hypervisor \
 	--net "tap=,mac=,ip=,mask="
 ```
 
-- AArch64
-
-```shell
-$ ./cloud-hypervisor \
-	--kernel ./linux-cloud-hypervisor/arch/arm64/boot/Image \
-	--console off \
-	--serial tty \
-	--disk path=focal-server-cloudimg-arm64.raw \
-	--cmdline "console=ttyAMA0 root=/dev/vda1 rw" \
-	--cpus boot=4 \
-	--memory size=1024M \
-	--net "tap=,mac=,ip=,mask="
-```
-
 # 3. Status
 
-Cloud Hypervisor is under active development. The following stability
-guarantees are currently made:
+Cloud Hypervisor is under active development. The following stability guarantees
+are currently made:
 
 * The API (including command line options) will not be removed or changed in a
-  breaking way without a minimum of 2 major releases notice. Where possible
-  warnings will be given about the use of deprecated functionality and the
-  deprecations will be documented in the release notes.
-
+  breaking way without a minimum of 2 releases notice. Where possible warnings
+  will be given about the use of deprecated functionality and the deprecations
+  will be documented in the release notes.
 * Point releases will be made between individual releases where there are
-  substantial bug fixes or security issues that need to be fixed. These point
-  releases will only include bug fixes.
+  substantial bug fixes or security issues that need to be fixed.
 
 Currently the following items are **not** guaranteed across updates:
 
 * Snapshot/restore is not supported across different versions
 * Live migration is not supported across different versions
 * The following features are considered experimental and may change
-  substantially between releases: TDX, vfio-user, vDPA.
+  substantially between releases: TDX, SGX.
 
-Further details can be found in the [release documentation](docs/releases.md).
 
-As of 2023-01-03, the following cloud images are supported:
+As of 2021-04-29, the following cloud images are supported:
 
-- [Ubuntu Focal](https://cloud-images.ubuntu.com/focal/current/) (focal-server-cloudimg-{amd64,arm64}.img)
-- [Ubuntu Jammy](https://cloud-images.ubuntu.com/jammy/current/) (jammy-server-cloudimg-{amd64,arm64}.img )
-- [Fedora 36](https://fedora.mirrorservice.org/fedora/linux/releases/36/Cloud/) ([Fedora-Cloud-Base-36-1.5.x86_64.raw.xz](https://fedora.mirrorservice.org/fedora/linux/releases/36/Cloud/x86_64/images/) / [Fedora-Cloud-Base-36-1.5.aarch64.raw.xz](https://fedora.mirrorservice.org/fedora/linux/releases/36/Cloud/aarch64/images/))
+- [Ubuntu Bionic](https://cloud-images.ubuntu.com/bionic/current/) (cloudimg)
+- [Ubuntu Focal](https://cloud-images.ubuntu.com/focal/current/) (cloudimg)
+- [Ubuntu Groovy](https://cloud-images.ubuntu.com/groovy/current/) (cloudimg)
+- [Ubuntu Hirsute](https://cloud-images.ubuntu.com/hirsute/current/) (cloudimg)
 
 Direct kernel boot to userspace should work with a rootfs from most
-distributions although you may need to enable exotic filesystem types in the
-reference kernel configuration (e.g. XFS or btrfs.)
+distributions.
 
 ## Hot Plug
 
@@ -320,12 +302,14 @@ Cloud Hypervisor supports hotplug of CPUs, passthrough devices (VFIO),
 Details of the device model can be found in this
 [documentation](docs/device_model.md).
 
-## Roadmap
+## TODO
 
-The project roadmap is tracked through a [GitHub
-project](https://github.com/orgs/cloud-hypervisor/projects/6).
+We are not tracking the Cloud Hypervisor TODO list from a specific git tracked
+file but through
+[github issues](https://github.com/cloud-hypervisor/cloud-hypervisor/issues/new)
+instead.
 
-# 4. Relationship with _Rust VMM_ Project
+# 4. `rust-vmm` project dependency
 
 In order to satisfy the design goal of having a high-performance,
 security-focused hypervisor the decision was made to use the
@@ -334,26 +318,39 @@ focus on memory and thread safety makes it an ideal candidate for implementing
 VMMs.
 
 Instead of implementing the VMM components from scratch, Cloud Hypervisor is
-importing the [Rust VMM](https://github.com/rust-vmm) crates, and sharing code
+importing the [rust-vmm](https://github.com/rust-vmm) crates, and sharing code
 and architecture together with other VMMs like e.g. Amazon's
 [Firecracker](https://firecracker-microvm.github.io/) and Google's
 [crosvm](https://chromium.googlesource.com/chromiumos/platform/crosvm/).
 
-Cloud Hypervisor embraces the _Rust VMM_ project's goals, which is to be able
-to share and re-use as many virtualization crates as possible.
+Cloud Hypervisor embraces the rust-vmm project goals, which is to be able to
+share and re-use as many virtualization crates as possible. As such, the Cloud
+Hypervisor relationship with the rust-vmm project is twofold:
 
-## Differences with Firecracker and crosvm
+1. It will use as much of the rust-vmm code as possible. Any new rust-vmm crate
+   that's relevant to the project goals will be integrated as soon as possible.
+2. As it is likely that the rust-vmm project will lack some of the features that
+   Cloud Hypervisor needs (e.g. ACPI, VFIO, vhost-user, etc), we will be using
+   the Cloud Hypervisor VMM to implement and test them, and contribute them back
+   to the rust-vmm project.
+
+## Firecracker and crosvm
 
 A large part of the Cloud Hypervisor code is based on either the Firecracker or
-the crosvm project's implementations. Both of these are VMMs written in Rust
-with a focus on safety and security, like Cloud Hypervisor.
+the crosvm projects implementations. Both of these are VMMs written in Rust with
+a focus on safety and security, like Cloud Hypervisor.
 
-The goal of the Cloud Hypervisor project differs from the aforementioned
-projects in that it aims to be a general purpose VMM for _Cloud Workloads_ and
-not limited to container/serverless or client workloads.
+However we want to emphasize that the Cloud Hypervisor project is neither a fork
+nor a reimplementation of any of those projects. The goals and use cases we're
+trying to meet are different. We're aiming at supporting cloud workloads, i.e.
+those modern, full Linux distribution images currently being run by Cloud
+Service Provider (CSP) tenants.
 
-The Cloud Hypervisor community thanks the communities of both the Firecracker
-and crosvm projects for their excellent work.
+Our primary target is not to support client or serverless use cases, and as such
+our code base already diverges from the crosvm and Firecracker ones. As we add
+more features to support our use cases, we believe that the divergence will
+increase while at the same time sharing as much of the fundamental
+virtualization code through the rust-vmm project crates as possible.
 
 # 5. Community
 
@@ -363,27 +360,20 @@ repository.
 
 ## Contribute
 
-The project strongly believes in building a global, diverse and collaborative
-community around the Cloud Hypervisor project. Anyone who is interested in
+We are working on building a global, diverse and collaborative community around
+the Cloud Hypervisor project. Anyone who is interested in
 [contributing](CONTRIBUTING.md) to the project is welcome to participate.
 
-Contributing to a open source project like Cloud Hypervisor covers a lot more
-than just sending code. Testing, documentation, pull request
+We believe that contributing to a open source project like Cloud Hypervisor
+covers a lot more than just sending code. Testing, documentation, pull request
 reviews, bug reports, feature requests, project improvement suggestions, etc,
 are all equal and welcome means of contribution. See the
 [CONTRIBUTING](CONTRIBUTING.md) document for more details.
 
-## Slack
+## Join us
 
-Get an [invite to our Slack channel](https://join.slack.com/t/cloud-hypervisor/shared_invite/enQtNjY3MTE3MDkwNDQ4LWQ1MTA1ZDVmODkwMWQ1MTRhYzk4ZGNlN2UwNTI3ZmFlODU0OTcwOWZjMTkwZDExYWE3YjFmNzgzY2FmNDAyMjI),
- [join us on Slack](https://cloud-hypervisor.slack.com/), and [participate in our community activities](https://cloud-hypervisor.slack.com/archives/C04R5DUQVBN).
-
-## Mailing list
-
-Please report bugs using the [GitHub issue
-tracker](https://github.com/cloud-hypervisor/cloud-hypervisor/issues) but for
-broader community discussions you may use our [mailing
-list](https://lists.cloudhypervisor.org/g/dev/).
+Get an [invite to our Slack channel](https://join.slack.com/t/cloud-hypervisor/shared_invite/enQtNjY3MTE3MDkwNDQ4LWQ1MTA1ZDVmODkwMWQ1MTRhYzk4ZGNlN2UwNTI3ZmFlODU0OTcwOWZjMTkwZDExYWE3YjFmNzgzY2FmNDAyMjI)
+and [join us on Slack](https://cloud-hypervisor.slack.com/).
 
 ## Security issues
 

acpi_tables/Cargo.toml

@@ -0,0 +1,8 @@
+[package]
+name = "acpi_tables"
+version = "0.1.0"
+authors = ["The Cloud Hypervisor Authors"]
+edition = "2018"
+
+[dependencies]
+vm-memory = "0.7.0"

acpi_tables/src/lib.rs

@@ -0,0 +1,12 @@
+// Copyright © 2019 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+//
+
+pub mod aml;
+pub mod rsdp;
+pub mod sdt;
+
+fn generate_checksum(data: &[u8]) -> u8 {
+    (255 - data.iter().fold(0u8, |acc, x| acc.wrapping_add(*x))).wrapping_add(1)
+}

acpi_tables/src/rsdp.rs

@@ -0,0 +1,68 @@
+// Copyright © 2019 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+//
+
+use vm_memory::ByteValued;
+
+#[repr(packed)]
+#[derive(Clone, Copy, Default)]
+pub struct Rsdp {
+    pub signature: [u8; 8],
+    pub checksum: u8,
+    pub oem_id: [u8; 6],
+    pub revision: u8,
+    _rsdt_addr: u32,
+    pub length: u32,
+    pub xsdt_addr: u64,
+    pub extended_checksum: u8,
+    _reserved: [u8; 3],
+}
+
+// SAFETY: Rsdp only contains a series of integers
+unsafe impl ByteValued for Rsdp {}
+
+impl Rsdp {
+    pub fn new(oem_id: [u8; 6], xsdt_addr: u64) -> Self {
+        let mut rsdp = Rsdp {
+            signature: *b"RSD PTR ",
+            checksum: 0,
+            oem_id,
+            revision: 2,
+            _rsdt_addr: 0,
+            length: std::mem::size_of::<Rsdp>() as u32,
+            xsdt_addr,
+            extended_checksum: 0,
+            _reserved: [0; 3],
+        };
+
+        rsdp.checksum = super::generate_checksum(&rsdp.as_slice()[0..19]);
+        rsdp.extended_checksum = super::generate_checksum(rsdp.as_slice());
+        rsdp
+    }
+
+    pub fn len() -> usize {
+        std::mem::size_of::<Rsdp>()
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::Rsdp;
+    use vm_memory::bytes::ByteValued;
+
+    #[test]
+    fn test_rsdp() {
+        let rsdp = Rsdp::new(*b"CHYPER", 0xdead_beef);
+        let sum = rsdp
+            .as_slice()
+            .iter()
+            .fold(0u8, |acc, x| acc.wrapping_add(*x));
+        assert_eq!(sum, 0);
+        let sum: u8 = rsdp
+            .as_slice()
+            .iter()
+            .fold(0u8, |acc, x| acc.wrapping_add(*x));
+        assert_eq!(sum, 0);
+    }
+}

acpi_tables/src/sdt.rs

@@ -0,0 +1,146 @@
+// Copyright © 2019 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#[repr(packed)]
+pub struct GenericAddress {
+    pub address_space_id: u8,
+    pub register_bit_width: u8,
+    pub register_bit_offset: u8,
+    pub access_size: u8,
+    pub address: u64,
+}
+
+impl GenericAddress {
+    pub fn io_port_address<T>(address: u16) -> Self {
+        GenericAddress {
+            address_space_id: 1,
+            register_bit_width: 8 * std::mem::size_of::<T>() as u8,
+            register_bit_offset: 0,
+            access_size: std::mem::size_of::<T>() as u8,
+            address: u64::from(address),
+        }
+    }
+    pub fn mmio_address<T>(address: u64) -> Self {
+        GenericAddress {
+            address_space_id: 0,
+            register_bit_width: 8 * std::mem::size_of::<T>() as u8,
+            register_bit_offset: 0,
+            access_size: std::mem::size_of::<T>() as u8,
+            address,
+        }
+    }
+}
+
+pub struct Sdt {
+    data: Vec<u8>,
+}
+
+#[allow(clippy::len_without_is_empty)]
+impl Sdt {
+    pub fn new(
+        signature: [u8; 4],
+        length: u32,
+        revision: u8,
+        oem_id: [u8; 6],
+        oem_table: [u8; 8],
+        oem_revision: u32,
+    ) -> Self {
+        assert!(length >= 36);
+
+        let mut data = Vec::with_capacity(length as usize);
+        data.extend_from_slice(&signature);
+        data.extend_from_slice(&length.to_le_bytes());
+        data.push(revision);
+        data.push(0); // checksum
+        data.extend_from_slice(&oem_id);
+        data.extend_from_slice(&oem_table);
+        data.extend_from_slice(&oem_revision.to_le_bytes());
+        data.extend_from_slice(b"CLDH");
+        data.extend_from_slice(&0u32.to_le_bytes());
+        assert_eq!(data.len(), 36);
+
+        data.resize(length as usize, 0);
+        let mut sdt = Sdt { data };
+
+        sdt.update_checksum();
+        sdt
+    }
+
+    pub fn update_checksum(&mut self) {
+        self.data[9] = 0;
+        let checksum = super::generate_checksum(self.data.as_slice());
+        self.data[9] = checksum
+    }
+
+    pub fn as_slice(&self) -> &[u8] {
+        self.data.as_slice()
+    }
+
+    pub fn append<T>(&mut self, value: T) {
+        let orig_length = self.data.len();
+        let new_length = orig_length + std::mem::size_of::<T>();
+        self.data.resize(new_length, 0);
+        self.write_u32(4, new_length as u32);
+        self.write(orig_length, value);
+    }
+
+    pub fn append_slice(&mut self, data: &[u8]) {
+        let orig_length = self.data.len();
+        let new_length = orig_length + data.len();
+        self.write_u32(4, new_length as u32);
+        self.data.extend_from_slice(data);
+        self.update_checksum();
+    }
+
+    /// Write a value at the given offset
+    pub fn write<T>(&mut self, offset: usize, value: T) {
+        assert!((offset + (std::mem::size_of::<T>() - 1)) < self.data.len());
+        unsafe {
+            *(((self.data.as_mut_ptr() as usize) + offset) as *mut T) = value;
+        }
+        self.update_checksum();
+    }
+
+    pub fn write_u8(&mut self, offset: usize, val: u8) {
+        self.write(offset, val);
+    }
+
+    pub fn write_u16(&mut self, offset: usize, val: u16) {
+        self.write(offset, val);
+    }
+
+    pub fn write_u32(&mut self, offset: usize, val: u32) {
+        self.write(offset, val);
+    }
+
+    pub fn write_u64(&mut self, offset: usize, val: u64) {
+        self.write(offset, val);
+    }
+
+    pub fn len(&self) -> usize {
+        self.data.len()
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::Sdt;
+
+    #[test]
+    fn test_sdt() {
+        let mut sdt = Sdt::new(*b"TEST", 40, 1, *b"CLOUDH", *b"TESTTEST", 1);
+        let sum: u8 = sdt
+            .as_slice()
+            .iter()
+            .fold(0u8, |acc, x| acc.wrapping_add(*x));
+        assert_eq!(sum, 0);
+        sdt.write_u32(36, 0x12345678);
+        let sum: u8 = sdt
+            .as_slice()
+            .iter()
+            .fold(0u8, |acc, x| acc.wrapping_add(*x));
+        assert_eq!(sum, 0);
+    }
+}

api_client/Cargo.toml

@@ -2,7 +2,7 @@
 name = "api_client"
 version = "0.1.0"
 authors = ["The Cloud Hypervisor Authors"]
-edition = "2021"
+edition = "2018"
 
 [dependencies]
-vmm-sys-util = "0.11.0"
+vmm-sys-util = "0.9.0"

api_client/src/lib.rs

@@ -22,16 +22,16 @@ impl fmt::Display for Error {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         use Error::*;
         match self {
-            Socket(e) => write!(f, "Error writing to or reading from HTTP socket: {e}"),
-            SocketSendFds(e) => write!(f, "Error writing to or reading from HTTP socket: {e}"),
-            StatusCodeParsing(e) => write!(f, "Error parsing HTTP status code: {e}"),
+            Socket(e) => write!(f, "Error writing to or reading from HTTP socket: {}", e),
+            SocketSendFds(e) => write!(f, "Error writing to or reading from HTTP socket: {}", e),
+            StatusCodeParsing(e) => write!(f, "Error parsing HTTP status code: {}", e),
             MissingProtocol => write!(f, "HTTP output is missing protocol statement"),
-            ContentLengthParsing(e) => write!(f, "Error parsing HTTP Content-Length field: {e}"),
+            ContentLengthParsing(e) => write!(f, "Error parsing HTTP Content-Length field: {}", e),
             ServerResponse(s, o) => {
                 if let Some(o) = o {
-                    write!(f, "Server responded with an error: {s:?}: {o}")
+                    write!(f, "Server responded with an error: {:?}: {}", s, o)
                 } else {
-                    write!(f, "Server responded with an error: {s:?}")
+                    write!(f, "Server responded with an error: {:?}", s)
                 }
             }
         }
@@ -79,7 +79,7 @@ impl StatusCode {
 }
 
 fn get_header<'a>(res: &'a str, header: &'a str) -> Option<&'a str> {
-    let header_str = format!("{header}: ");
+    let header_str = format!("{}: ", header);
     res.find(&header_str)
         .map(|o| &res[o + header_str.len()..o + res[o..].find('\r').unwrap()])
 }
@@ -101,10 +101,6 @@ fn parse_http_response(socket: &mut dyn Read) -> Result<Option<String>, Error> {
     loop {
         let mut bytes = vec![0; 256];
         let count = socket.read(&mut bytes).map_err(Error::Socket)?;
-        // If the return value is 0, the peer has performed an orderly shutdown.
-        if count == 0 {
-            break;
-        }
         res.push_str(std::str::from_utf8(&bytes[0..count]).unwrap());
 
         // End of headers
@@ -131,7 +127,7 @@ fn parse_http_response(socket: &mut dyn Read) -> Result<Option<String>, Error> {
             }
         }
     }
-    let body_string = content_length.and(body_offset.map(|o| String::from(&res[o..])));
+    let body_string = content_length.and(Some(String::from(&res[body_offset.unwrap()..])));
     let status_code = get_status_code(&res)?;
 
     if status_code.is_server_error() {
@@ -141,19 +137,18 @@ fn parse_http_response(socket: &mut dyn Read) -> Result<Option<String>, Error> {
     }
 }
 
-/// Make an API request using the fully qualified command name.
-/// For example, full_command could be "vm.create" or "vmm.ping".
-pub fn simple_api_full_command_with_fds_and_response<T: Read + Write + ScmSocket>(
+pub fn simple_api_command_with_fds<T: Read + Write + ScmSocket>(
     socket: &mut T,
     method: &str,
-    full_command: &str,
+    c: &str,
     request_body: Option<&str>,
     request_fds: Vec<RawFd>,
-) -> Result<Option<String>, Error> {
+) -> Result<(), Error> {
     socket
         .send_with_fds(
             &[format!(
-                "{method} /api/v1/{full_command} HTTP/1.1\r\nHost: localhost\r\nAccept: */*\r\n"
+                "{} /api/v1/vm.{} HTTP/1.1\r\nHost: localhost\r\nAccept: */*\r\n",
+                method, c
             )
             .as_bytes()],
             &request_fds,
@@ -176,69 +171,12 @@ pub fn simple_api_full_command_with_fds_and_response<T: Read + Write + ScmSocket
 
     socket.flush().map_err(Error::Socket)?;
 
-    parse_http_response(socket)
-}
-
-pub fn simple_api_full_command_with_fds<T: Read + Write + ScmSocket>(
-    socket: &mut T,
-    method: &str,
-    full_command: &str,
-    request_body: Option<&str>,
-    request_fds: Vec<RawFd>,
-) -> Result<(), Error> {
-    let response = simple_api_full_command_with_fds_and_response(
-        socket,
-        method,
-        full_command,
-        request_body,
-        request_fds,
-    )?;
-
-    if response.is_some() {
-        println!("{}", response.unwrap());
+    if let Some(body) = parse_http_response(socket)? {
+        println!("{}", body);
     }
-
     Ok(())
 }
 
-pub fn simple_api_full_command<T: Read + Write + ScmSocket>(
-    socket: &mut T,
-    method: &str,
-    full_command: &str,
-    request_body: Option<&str>,
-) -> Result<(), Error> {
-    simple_api_full_command_with_fds(socket, method, full_command, request_body, Vec::new())
-}
-
-pub fn simple_api_full_command_and_response<T: Read + Write + ScmSocket>(
-    socket: &mut T,
-    method: &str,
-    full_command: &str,
-    request_body: Option<&str>,
-) -> Result<Option<String>, Error> {
-    simple_api_full_command_with_fds_and_response(
-        socket,
-        method,
-        full_command,
-        request_body,
-        Vec::new(),
-    )
-}
-
-pub fn simple_api_command_with_fds<T: Read + Write + ScmSocket>(
-    socket: &mut T,
-    method: &str,
-    c: &str,
-    request_body: Option<&str>,
-    request_fds: Vec<RawFd>,
-) -> Result<(), Error> {
-    // Create the full VM command. For VMM commands, use
-    // simple_api_full_command().
-    let full_command = format!("vm.{c}");
-
-    simple_api_full_command_with_fds(socket, method, &full_command, request_body, request_fds)
-}
-
 pub fn simple_api_command<T: Read + Write + ScmSocket>(
     socket: &mut T,
     method: &str,

arch/Cargo.toml

@@ -2,28 +2,30 @@
 name = "arch"
 version = "0.1.0"
 authors = ["The Chromium OS Authors"]
-edition = "2021"
+edition = "2018"
 
 [features]
 default = []
+acpi = ["acpi_tables"]
 tdx = []
 
 [dependencies]
-anyhow = "1.0.71"
+acpi_tables = { path = "../acpi_tables", optional = true }
+anyhow = "1.0.55"
 byteorder = "1.4.3"
 hypervisor = { path = "../hypervisor" }
-libc = "0.2.139"
-linux-loader = { version = "0.9.0", features = ["elf", "bzimage", "pe"] }
-log = "0.4.17"
-serde = { version = "1.0.164", features = ["rc", "derive"] }
-thiserror = "1.0.40"
-uuid = "1.3.4"
-versionize = "0.1.10"
+libc = "0.2.119"
+linux-loader = { version = "0.4.0", features = ["elf", "bzimage", "pe"] }
+log = "0.4.14"
+serde = { version = "1.0.136", features = ["rc"] }
+serde_derive = "1.0.136"
+thiserror = "1.0.30"
+versionize = "0.1.6"
 versionize_derive = "0.1.4"
-vm-memory = { version = "0.11.0", features = ["backend-mmap", "backend-bitmap"] }
+vm-memory = { version = "0.7.0", features = ["backend-mmap", "backend-bitmap"] }
 vm-migration = { path = "../vm-migration" }
-vmm-sys-util = { version = "0.11.0", features = ["with-serde"] }
+vmm-sys-util = { version = "0.9.0", features = ["with-serde"] }
 
 [target.'cfg(target_arch = "aarch64")'.dependencies]
-fdt_parser = { version = "0.1.4", package = "fdt" }
+fdt_parser = { version = "0.1.3", package = 'fdt'}
 vm-fdt = { git = "https://github.com/rust-vmm/vm-fdt", branch = "main" }

arch/src/aarch64/fdt.rs

@@ -8,26 +8,24 @@
 
 use crate::{NumaNodes, PciSpaceInfo};
 use byteorder::{BigEndian, ByteOrder};
-use hypervisor::arch::aarch64::gic::Vgic;
 use std::cmp;
 use std::collections::HashMap;
 use std::ffi::CStr;
 use std::fmt::Debug;
 use std::result;
 use std::str;
-use std::sync::{Arc, Mutex};
 
 use super::super::DeviceType;
 use super::super::GuestMemoryMmap;
 use super::super::InitramfsConfig;
+use super::get_fdt_addr;
+use super::gic::GicDevice;
 use super::layout::{
     IRQ_BASE, MEM_32BIT_DEVICES_SIZE, MEM_32BIT_DEVICES_START, MEM_PCI_IO_SIZE, MEM_PCI_IO_START,
     PCI_HIGH_BASE, PCI_MMIO_CONFIG_SIZE_PER_SEGMENT,
 };
-use std::fs;
-use std::path::Path;
 use vm_fdt::{FdtWriter, FdtWriterResult};
-use vm_memory::{Address, Bytes, GuestMemory, GuestMemoryError, GuestMemoryRegion};
+use vm_memory::{Address, Bytes, GuestAddress, GuestMemory, GuestMemoryError, GuestMemoryRegion};
 
 // This is a value for uniquely identifying the FDT node declaring the interrupt controller.
 const GIC_PHANDLE: u32 = 1;
@@ -42,12 +40,8 @@ const VIRTIO_IOMMU_PHANDLE: u32 = 5;
 // NOTE: Keep FIRST_VCPU_PHANDLE the last PHANDLE defined.
 // This is a value for uniquely identifying the FDT node containing the first vCPU.
 // The last number of vCPU phandle depends on the number of vCPUs.
-const FIRST_VCPU_PHANDLE: u32 = 8;
+const FIRST_VCPU_PHANDLE: u32 = 6;
 
-// This is a value for uniquely identifying the FDT node containing the L2 cache info
-const L2_CACHE_PHANDLE: u32 = 6;
-// This is a value for uniquely identifying the FDT node containing the L3 cache info
-const L3_CACHE_PHANDLE: u32 = 7;
 // Read the documentation specified when appending the root node to the FDT.
 const ADDRESS_CELLS: u32 = 0x2;
 const SIZE_CELLS: u32 = 0x2;
@@ -57,6 +51,8 @@ const SIZE_CELLS: u32 = 0x2;
 // Look for "The 1st cell..."
 const GIC_FDT_IRQ_TYPE_SPI: u32 = 0;
 const GIC_FDT_IRQ_TYPE_PPI: u32 = 1;
+const GIC_FDT_IRQ_PPI_CPU_SHIFT: u32 = 8;
+const GIC_FDT_IRQ_PPI_CPU_MASK: u32 = 0xff << GIC_FDT_IRQ_PPI_CPU_SHIFT;
 
 // From https://elixir.bootlin.com/linux/v4.9.62/source/include/dt-bindings/interrupt-controller/irq.h#L17
 const IRQ_TYPE_EDGE_RISING: u32 = 1;
@@ -87,97 +83,6 @@ pub enum Error {
 }
 type Result<T> = result::Result<T, Error>;
 
-pub enum CacheLevel {
-    /// L1 data cache
-    L1D = 0,
-    /// L1 instruction cache
-    L1I = 1,
-    /// L2 cache
-    L2 = 2,
-    /// L3 cache
-    L3 = 3,
-}
-
-/// NOTE: CACHE SIZE file directory example,
-/// "/sys/devices/system/cpu/cpu0/cache/index0/size".
-pub fn get_cache_size(cache_level: CacheLevel) -> u32 {
-    let mut file_directory: String = "/sys/devices/system/cpu/cpu0/cache".to_string();
-    match cache_level {
-        CacheLevel::L1D => file_directory += "/index0/size",
-        CacheLevel::L1I => file_directory += "/index1/size",
-        CacheLevel::L2 => file_directory += "/index2/size",
-        CacheLevel::L3 => file_directory += "/index3/size",
-    }
-
-    let file_path = Path::new(&file_directory);
-    if !file_path.exists() {
-        error!("File: {} not exist.", file_directory);
-        0
-    } else {
-        info!("File: {} exist.", file_directory);
-
-        let src = fs::read_to_string(file_directory).expect("File not exists or file corrupted.");
-        // The content of the file is as simple as a size, like: "32K"
-        let src = src.trim();
-        let src_digits: u32 = src[0..src.len() - 1].parse().unwrap();
-        let src_unit = &src[src.len() - 1..];
-
-        src_digits
-            * match src_unit {
-                "K" => 1024,
-                "M" => 1024u32.pow(2),
-                "G" => 1024u32.pow(3),
-                _ => 1,
-            }
-    }
-}
-
-/// NOTE: CACHE COHERENCY LINE SIZE file directory example,
-/// "/sys/devices/system/cpu/cpu0/cache/index0/coherency_line_size".
-pub fn get_cache_coherency_line_size(cache_level: CacheLevel) -> u32 {
-    let mut file_directory: String = "/sys/devices/system/cpu/cpu0/cache".to_string();
-    match cache_level {
-        CacheLevel::L1D => file_directory += "/index0/coherency_line_size",
-        CacheLevel::L1I => file_directory += "/index1/coherency_line_size",
-        CacheLevel::L2 => file_directory += "/index2/coherency_line_size",
-        CacheLevel::L3 => file_directory += "/index3/coherency_line_size",
-    }
-
-    let file_path = Path::new(&file_directory);
-    if !file_path.exists() {
-        error!("File: {} not exist.", file_directory);
-        0
-    } else {
-        info!("File: {} exist.", file_directory);
-
-        let src = fs::read_to_string(file_directory).expect("File not exists or file corrupted.");
-        src.trim().parse::<u32>().unwrap()
-    }
-}
-
-/// NOTE: CACHE NUMBER OF SETS file directory example,
-/// "/sys/devices/system/cpu/cpu0/cache/index0/number_of_sets".
-pub fn get_cache_number_of_sets(cache_level: CacheLevel) -> u32 {
-    let mut file_directory: String = "/sys/devices/system/cpu/cpu0/cache".to_string();
-    match cache_level {
-        CacheLevel::L1D => file_directory += "/index0/number_of_sets",
-        CacheLevel::L1I => file_directory += "/index1/number_of_sets",
-        CacheLevel::L2 => file_directory += "/index2/number_of_sets",
-        CacheLevel::L3 => file_directory += "/index3/number_of_sets",
-    }
-
-    let file_path = Path::new(&file_directory);
-    if !file_path.exists() {
-        error!("File: {} not exist.", file_directory);
-        0
-    } else {
-        info!("File: {} exist.", file_directory);
-
-        let src = fs::read_to_string(file_directory).expect("File not exists or file corrupted.");
-        src.trim().parse::<u32>().unwrap()
-    }
-}
-
 /// Creates the flattened device tree for this aarch64 VM.
 #[allow(clippy::too_many_arguments)]
 pub fn create_fdt<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::BuildHasher>(
@@ -186,7 +91,7 @@ pub fn create_fdt<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::BuildHash
     vcpu_mpidr: Vec<u64>,
     vcpu_topology: Option<(u8, u8, u8)>,
     device_info: &HashMap<(DeviceType, String), T, S>,
-    gic_device: &Arc<Mutex<dyn Vgic>>,
+    gic_device: &dyn GicDevice,
     initrd: &Option<InitramfsConfig>,
     pci_space_info: &[PciSpaceInfo],
     numa_nodes: &NumaNodes,
@@ -216,7 +121,7 @@ pub fn create_fdt<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::BuildHash
     create_gic_node(&mut fdt, gic_device)?;
     create_timer_node(&mut fdt)?;
     if pmu_supported {
-        create_pmu_node(&mut fdt)?;
+        create_pmu_node(&mut fdt, vcpu_mpidr.len())?;
     }
     create_clock_node(&mut fdt)?;
     create_psci_node(&mut fdt)?;
@@ -236,8 +141,9 @@ pub fn create_fdt<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::BuildHash
 
 pub fn write_fdt_to_memory(fdt_final: Vec<u8>, guest_mem: &GuestMemoryMmap) -> Result<()> {
     // Write FDT to memory.
+    let fdt_address = GuestAddress(get_fdt_addr());
     guest_mem
-        .write_slice(fdt_final.as_slice(), super::layout::FDT_START)
+        .write_slice(fdt_final.as_slice(), fdt_address)
         .map_err(Error::WriteFdtToMemory)?;
     Ok(())
 }
@@ -256,56 +162,8 @@ fn create_cpu_nodes(
 
     let num_cpus = vcpu_mpidr.len();
 
-    // Add cache info.
-    // L1 Data Cache Info.
-    let mut l1_d_cache_size: u32 = 0;
-    let mut l1_d_cache_line_size: u32 = 0;
-    let mut l1_d_cache_sets: u32 = 0;
-
-    // L1 Instruction Cache Info.
-    let mut l1_i_cache_size: u32 = 0;
-    let mut l1_i_cache_line_size: u32 = 0;
-    let mut l1_i_cache_sets: u32 = 0;
-
-    // L2 Cache Info.
-    let mut l2_cache_size: u32 = 0;
-    let mut l2_cache_line_size: u32 = 0;
-    let mut l2_cache_sets: u32 = 0;
-
-    // L3 Cache Info.
-    let mut l3_cache_size: u32 = 0;
-    let mut l3_cache_line_size: u32 = 0;
-    let mut l3_cache_sets: u32 = 0;
-
-    let cache_path = Path::new("/sys/devices/system/cpu/cpu0/cache");
-    let cache_exist: bool = cache_path.exists();
-    if !cache_exist {
-        error!("cache sysfs system does not exist.");
-    } else {
-        info!("cache sysfs system exists.");
-        // L1 Data Cache Info.
-        l1_d_cache_size = get_cache_size(CacheLevel::L1D);
-        l1_d_cache_line_size = get_cache_coherency_line_size(CacheLevel::L1D);
-        l1_d_cache_sets = get_cache_number_of_sets(CacheLevel::L1D);
-
-        // L1 Instruction Cache Info.
-        l1_i_cache_size = get_cache_size(CacheLevel::L1I);
-        l1_i_cache_line_size = get_cache_coherency_line_size(CacheLevel::L1I);
-        l1_i_cache_sets = get_cache_number_of_sets(CacheLevel::L1I);
-
-        // L2 Cache Info.
-        l2_cache_size = get_cache_size(CacheLevel::L2);
-        l2_cache_line_size = get_cache_coherency_line_size(CacheLevel::L2);
-        l2_cache_sets = get_cache_number_of_sets(CacheLevel::L2);
-
-        // L3 Cache Info.
-        l3_cache_size = get_cache_size(CacheLevel::L3);
-        l3_cache_line_size = get_cache_coherency_line_size(CacheLevel::L3);
-        l3_cache_sets = get_cache_number_of_sets(CacheLevel::L3);
-    }
-
     for (cpu_id, mpidr) in vcpu_mpidr.iter().enumerate().take(num_cpus) {
-        let cpu_name = format!("cpu@{cpu_id:x}");
+        let cpu_name = format!("cpu@{:x}", cpu_id);
         let cpu_node = fdt.begin_node(&cpu_name)?;
         fdt.property_string("device_type", "cpu")?;
         fdt.property_string("compatible", "arm,arm-v8")?;
@@ -318,21 +176,6 @@ fn create_cpu_nodes(
         fdt.property_u32("reg", (mpidr & 0x7FFFFF) as u32)?;
         fdt.property_u32("phandle", cpu_id as u32 + FIRST_VCPU_PHANDLE)?;
 
-        if cache_exist && l1_d_cache_size != 0 && l1_i_cache_size != 0 {
-            // Add cache info.
-            fdt.property_u32("d-cache-size", l1_d_cache_size)?;
-            fdt.property_u32("d-cache-line-size", l1_d_cache_line_size)?;
-            fdt.property_u32("d-cache-sets", l1_d_cache_sets)?;
-
-            fdt.property_u32("i-cache-size", l1_i_cache_size)?;
-            fdt.property_u32("i-cache-line-size", l1_i_cache_line_size)?;
-            fdt.property_u32("i-cache-sets", l1_i_cache_sets)?;
-
-            if l2_cache_size != 0 {
-                fdt.property_u32("next-level-cache", L2_CACHE_PHANDLE)?;
-            }
-        }
-
         // Add `numa-node-id` property if there is any numa config.
         if numa_nodes.len() > 1 {
             for numa_node_idx in 0..numa_nodes.len() {
@@ -346,57 +189,23 @@ fn create_cpu_nodes(
         fdt.end_node(cpu_node)?;
     }
 
-    if cache_exist && l2_cache_size != 0 {
-        let l2_cache_name = "l2-cache0";
-        let l2_cache_node = fdt.begin_node(l2_cache_name)?;
-        fdt.property_u32("phandle", L2_CACHE_PHANDLE)?;
-        fdt.property_string("compatible", "cache")?;
-        fdt.property_u32("cache-size", l2_cache_size)?;
-        fdt.property_u32("cache-line-size", l2_cache_line_size)?;
-        fdt.property_u32("cache-sets", l2_cache_sets)?;
-        fdt.property_u32("cache-level", 2)?;
-
-        if l3_cache_size != 0 {
-            fdt.property_u32("next-level-cache", L3_CACHE_PHANDLE)?;
-        }
-
-        fdt.end_node(l2_cache_node)?;
-    }
-
-    if cache_exist && l3_cache_size != 0 {
-        let l3_cache_name = "l3-cache0";
-        let l3_cache_node = fdt.begin_node(l3_cache_name)?;
-        fdt.property_u32("phandle", L3_CACHE_PHANDLE)?;
-        fdt.property_string("compatible", "cache")?;
-        fdt.property_null("cache-unified")?;
-        fdt.property_u32("cache-size", l3_cache_size)?;
-        fdt.property_u32("cache-line-size", l3_cache_line_size)?;
-        fdt.property_u32("cache-sets", l3_cache_sets)?;
-        fdt.property_u32("cache-level", 3)?;
-        fdt.end_node(l3_cache_node)?;
-    }
-
     if let Some(topology) = vcpu_topology {
         let (threads_per_core, cores_per_package, packages) = topology;
         let cpu_map_node = fdt.begin_node("cpu-map")?;
 
         // Create device tree nodes with regard of above mapping.
-        for package_idx in 0..packages {
-            let package_name = format!("socket{package_idx:x}");
-            let package_node = fdt.begin_node(&package_name)?;
-
-            // Cluster is the container of cores, and it is mandatory in the CPU topology.
-            // Add a default "cluster0" in each socket/package.
-            let cluster_node = fdt.begin_node("cluster0")?;
+        for cluster_idx in 0..packages {
+            let cluster_name = format!("cluster{:x}", cluster_idx);
+            let cluster_node = fdt.begin_node(&cluster_name)?;
 
             for core_idx in 0..cores_per_package {
-                let core_name = format!("core{core_idx:x}");
+                let core_name = format!("core{:x}", core_idx);
                 let core_node = fdt.begin_node(&core_name)?;
 
                 for thread_idx in 0..threads_per_core {
-                    let thread_name = format!("thread{thread_idx:x}");
+                    let thread_name = format!("thread{:x}", thread_idx);
                     let thread_node = fdt.begin_node(&thread_name)?;
-                    let cpu_idx = threads_per_core * cores_per_package * package_idx
+                    let cpu_idx = threads_per_core * cores_per_package * cluster_idx
                         + threads_per_core * core_idx
                         + thread_idx;
                     fdt.property_u32("cpu", cpu_idx as u32 + FIRST_VCPU_PHANDLE)?;
@@ -406,7 +215,6 @@ fn create_cpu_nodes(
                 fdt.end_node(core_node)?;
             }
             fdt.end_node(cluster_node)?;
-            fdt.end_node(package_node)?;
         }
         fdt.end_node(cpu_map_node)?;
     } else {
@@ -423,118 +231,40 @@ fn create_memory_node(
     guest_mem: &GuestMemoryMmap,
     numa_nodes: &NumaNodes,
 ) -> FdtWriterResult<()> {
-    // See https://github.com/torvalds/linux/blob/58ae0b51506802713aa0e9956d1853ba4c722c98/Documentation/devicetree/bindings/numa.txt
-    // for NUMA setting in memory node.
     if numa_nodes.len() > 1 {
         for numa_node_idx in 0..numa_nodes.len() {
             let numa_node = numa_nodes.get(&(numa_node_idx as u32));
-            let mut mem_reg_prop: Vec<u64> = Vec::new();
-            let mut node_memory_addr: u64 = 0;
             // Each memory zone of numa will have its own memory node, but
             // different numa nodes should not share same memory zones.
             for memory_region in numa_node.unwrap().memory_regions.iter() {
                 let memory_region_start_addr: u64 = memory_region.start_addr().raw_value();
                 let memory_region_size: u64 = memory_region.size() as u64;
-                mem_reg_prop.push(memory_region_start_addr);
-                mem_reg_prop.push(memory_region_size);
-                // Set the node address the first non-zero regison address
-                if node_memory_addr == 0 {
-                    node_memory_addr = memory_region_start_addr;
+                let mem_reg_prop = [memory_region_start_addr, memory_region_size];
+                // With feature `acpi` enabled, RAM at 0-4M is for edk2 only
+                // and should be hidden to the guest.
+                #[cfg(feature = "acpi")]
+                if memory_region_start_addr == 0 {
+                    continue;
                 }
+
+                let memory_node_name = format!("memory@{:x}", memory_region_start_addr);
+                let memory_node = fdt.begin_node(&memory_node_name)?;
+                fdt.property_string("device_type", "memory")?;
+                fdt.property_array_u64("reg", &mem_reg_prop)?;
+                fdt.property_u32("numa-node-id", numa_node_idx as u32)?;
+                fdt.end_node(memory_node)?;
             }
-            let memory_node_name = format!("memory@{node_memory_addr:x}");
-            let memory_node = fdt.begin_node(&memory_node_name)?;
-            fdt.property_string("device_type", "memory")?;
-            fdt.property_array_u64("reg", &mem_reg_prop)?;
-            fdt.property_u32("numa-node-id", numa_node_idx as u32)?;
-            fdt.end_node(memory_node)?;
         }
     } else {
-        // Note: memory regions from "GuestMemory" are sorted and non-zero sized.
-        let ram_regions = {
-            let mut ram_regions = Vec::new();
-            let mut current_start = guest_mem
-                .iter()
-                .next()
-                .map(GuestMemoryRegion::start_addr)
-                .expect("GuestMemory must have one memory region at least")
-                .raw_value();
-            let mut current_end = current_start;
+        let mem_size = guest_mem.last_addr().raw_value() - super::layout::RAM_64BIT_START + 1;
+        // See https://github.com/torvalds/linux/blob/master/Documentation/devicetree/booting-without-of.txt#L960
+        // for an explanation of this.
+        let mem_reg_prop = [super::layout::RAM_64BIT_START as u64, mem_size as u64];
+        let memory_node = fdt.begin_node("memory")?;
 
-            for (start, size) in guest_mem
-                .iter()
-                .map(|m| (m.start_addr().raw_value(), m.len()))
-            {
-                if current_end == start {
-                    // This zone is continuous with the previous one.
-                    current_end += size;
-                } else {
-                    ram_regions.push((current_start, current_end));
-
-                    current_start = start;
-                    current_end = start + size;
-                }
-            }
-
-            ram_regions.push((current_start, current_end));
-
-            ram_regions
-        };
-
-        if ram_regions.len() > 2 {
-            panic!(
-                "There should be up to two non-continuous regions, devidided by the
-                    gap at the end of 32bit address space."
-            );
-        }
-
-        // Create the memory node for memory region before the gap
-        {
-            let (first_region_start, first_region_end) = ram_regions
-                .first()
-                .expect("There should be at last one memory region");
-            let ram_start = super::layout::RAM_START.raw_value();
-            let mem_32bit_reserved_start = super::layout::MEM_32BIT_RESERVED_START.raw_value();
-
-            if !((first_region_start <= &ram_start)
-                && (first_region_end > &ram_start)
-                && (first_region_end <= &mem_32bit_reserved_start))
-            {
-                panic!(
-                    "Unexpected first memory region layout: (start: 0x{:08x}, end: 0x{:08x}).
-                    ram_start: 0x{:08x}, mem_32bit_reserved_start: 0x{:08x}",
-                    first_region_start, first_region_end, ram_start, mem_32bit_reserved_start
-                );
-            }
-
-            let mem_size = first_region_end - ram_start;
-            let mem_reg_prop = [ram_start, mem_size];
-            let memory_node_name = format!("memory@{:x}", ram_start);
-            let memory_node = fdt.begin_node(&memory_node_name)?;
-            fdt.property_string("device_type", "memory")?;
-            fdt.property_array_u64("reg", &mem_reg_prop)?;
-            fdt.end_node(memory_node)?;
-        }
-
-        // Create the memory map entry for memory region after the gap if any
-        if let Some((second_region_start, second_region_end)) = ram_regions.get(1) {
-            let ram_64bit_start = super::layout::RAM_64BIT_START.raw_value();
-
-            if second_region_start != &ram_64bit_start {
-                panic!(
-                    "Unexpected second memory region layout: start: 0x{:08x}, ram_64bit_start: 0x{:08x}",
-                    second_region_start, ram_64bit_start
-                );
-            }
-
-            let mem_size = second_region_end - ram_64bit_start;
-            let mem_reg_prop = [ram_64bit_start, mem_size];
-            let memory_node_name = format!("memory@{:x}", ram_64bit_start);
-            let memory_node = fdt.begin_node(&memory_node_name)?;
-            fdt.property_string("device_type", "memory")?;
-            fdt.property_array_u64("reg", &mem_reg_prop)?;
-            fdt.end_node(memory_node)?;
-        }
+        fdt.property_string("device_type", "memory")?;
+        fdt.property_array_u64("reg", &mem_reg_prop)?;
+        fdt.end_node(memory_node)?;
     }
 
     Ok(())
@@ -549,7 +279,7 @@ fn create_chosen_node(
     fdt.property_string("bootargs", cmdline)?;
 
     if let Some(initrd_config) = initrd {
-        let initrd_start = initrd_config.address.raw_value();
+        let initrd_start = initrd_config.address.raw_value() as u64;
         let initrd_end = initrd_config.address.raw_value() + initrd_config.size as u64;
         fdt.property_u64("linux,initrd-start", initrd_start)?;
         fdt.property_u64("linux,initrd-end", initrd_end)?;
@@ -560,18 +290,18 @@ fn create_chosen_node(
     Ok(())
 }
 
-fn create_gic_node(fdt: &mut FdtWriter, gic_device: &Arc<Mutex<dyn Vgic>>) -> FdtWriterResult<()> {
-    let gic_reg_prop = gic_device.lock().unwrap().device_properties();
+fn create_gic_node(fdt: &mut FdtWriter, gic_device: &dyn GicDevice) -> FdtWriterResult<()> {
+    let gic_reg_prop = gic_device.device_properties();
 
     let intc_node = fdt.begin_node("intc")?;
 
-    fdt.property_string("compatible", gic_device.lock().unwrap().fdt_compatibility())?;
+    fdt.property_string("compatible", gic_device.fdt_compatibility())?;
     fdt.property_null("interrupt-controller")?;
     // "interrupt-cells" field specifies the number of cells needed to encode an
     // interrupt source. The type shall be a <u32> and the value shall be 3 if no PPI affinity description
     // is required.
     fdt.property_u32("#interrupt-cells", 3)?;
-    fdt.property_array_u64("reg", &gic_reg_prop)?;
+    fdt.property_array_u64("reg", gic_reg_prop)?;
     fdt.property_u32("phandle", GIC_PHANDLE)?;
     fdt.property_u32("#address-cells", 2)?;
     fdt.property_u32("#size-cells", 2)?;
@@ -579,18 +309,18 @@ fn create_gic_node(fdt: &mut FdtWriter, gic_device: &Arc<Mutex<dyn Vgic>>) -> Fd
 
     let gic_intr_prop = [
         GIC_FDT_IRQ_TYPE_PPI,
-        gic_device.lock().unwrap().fdt_maint_irq(),
+        gic_device.fdt_maint_irq(),
         IRQ_TYPE_LEVEL_HI,
     ];
     fdt.property_array_u32("interrupts", &gic_intr_prop)?;
 
-    if gic_device.lock().unwrap().msi_compatible() {
+    if gic_device.msi_compatible() {
         let msic_node = fdt.begin_node("msic")?;
-        fdt.property_string("compatible", gic_device.lock().unwrap().msi_compatibility())?;
+        fdt.property_string("compatible", gic_device.msi_compatibility())?;
         fdt.property_null("msi-controller")?;
         fdt.property_u32("phandle", MSI_PHANDLE)?;
-        let msi_reg_prop = gic_device.lock().unwrap().msi_properties();
-        fdt.property_array_u64("reg", &msi_reg_prop)?;
+        let msi_reg_prop = gic_device.msi_properties();
+        fdt.property_array_u64("reg", msi_reg_prop)?;
         fdt.end_node(msic_node)?;
     }
 
@@ -785,9 +515,16 @@ fn create_devices_node<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::Buil
     Ok(())
 }
 
-fn create_pmu_node(fdt: &mut FdtWriter) -> FdtWriterResult<()> {
+fn create_pmu_node(fdt: &mut FdtWriter, cpu_nums: usize) -> FdtWriterResult<()> {
+    let num_cpus = cpu_nums as u64 as u32;
     let compatible = "arm,armv8-pmuv3";
-    let irq = [GIC_FDT_IRQ_TYPE_PPI, AARCH64_PMU_IRQ, IRQ_TYPE_LEVEL_HI];
+    let cpu_mask: u32 =
+        (((1 << num_cpus) - 1) << GIC_FDT_IRQ_PPI_CPU_SHIFT) & GIC_FDT_IRQ_PPI_CPU_MASK;
+    let irq = [
+        GIC_FDT_IRQ_TYPE_PPI,
+        AARCH64_PMU_IRQ,
+        cpu_mask | IRQ_TYPE_LEVEL_HI,
+    ];
 
     let pmu_node = fdt.begin_node("pmu")?;
     fdt.property_string("compatible", compatible)?;
@@ -811,19 +548,20 @@ fn create_pci_nodes(
         // could fall bellow 4G.
         // Here we cut off PCI device space below 8G in FDT to workaround the EDK2 check.
         // But the address written in ACPI is not impacted.
-        let (pci_device_base_64bit, pci_device_size_64bit) =
-            if pci_device_info_elem.pci_device_space_start < PCI_HIGH_BASE.raw_value() {
-                (
-                    PCI_HIGH_BASE.raw_value(),
-                    pci_device_info_elem.pci_device_space_size
-                        - (PCI_HIGH_BASE.raw_value() - pci_device_info_elem.pci_device_space_start),
-                )
-            } else {
-                (
-                    pci_device_info_elem.pci_device_space_start,
-                    pci_device_info_elem.pci_device_space_size,
-                )
-            };
+        let (pci_device_base_64bit, pci_device_size_64bit) = if cfg!(feature = "acpi")
+            && (pci_device_info_elem.pci_device_space_start < PCI_HIGH_BASE)
+        {
+            (
+                PCI_HIGH_BASE,
+                pci_device_info_elem.pci_device_space_size
+                    - (PCI_HIGH_BASE - pci_device_info_elem.pci_device_space_start),
+            )
+        } else {
+            (
+                pci_device_info_elem.pci_device_space_start,
+                pci_device_info_elem.pci_device_space_size,
+            )
+        };
         // There is no specific requirement of the 32bit MMIO range, and
         // therefore at least we can make these ranges 4K aligned.
         let pci_device_size_32bit: u64 =
@@ -916,7 +654,7 @@ fn create_pci_nodes(
 
                 // See kernel document Documentation/devicetree/bindings/virtio/iommu.txt
                 // for virtio-iommu node settings.
-                let virtio_iommu_node_name = format!("virtio_iommu@{virtio_iommu_bdf:x}");
+                let virtio_iommu_node_name = format!("virtio_iommu@{:x}", virtio_iommu_bdf);
                 let virtio_iommu_node = fdt.begin_node(&virtio_iommu_node_name)?;
                 fdt.property_u32("#iommu-cells", 1)?;
                 fdt.property_string("compatible", "virtio,pci-iommu")?;

arch/src/aarch64/gic/dist_regs.rs

@@ -1,12 +1,12 @@
 // Copyright 2020 Amazon.com, Inc. or its affiliates. All Rights Reserved.
 // SPDX-License-Identifier: Apache-2.0
 
-use crate::arch::aarch64::gic::{Error, Result};
-use crate::device::HypervisorDeviceError;
-use crate::kvm::kvm_bindings::{
+use super::{Error, Result};
+use crate::layout::IRQ_BASE;
+use hypervisor::kvm::kvm_bindings::{
     kvm_device_attr, KVM_DEV_ARM_VGIC_GRP_DIST_REGS, KVM_DEV_ARM_VGIC_GRP_NR_IRQS,
 };
-use kvm_ioctls::DeviceFd;
+use std::sync::Arc;
 
 /*
  Distributor registers as detailed at page 456 from
@@ -77,7 +77,12 @@ static VGIC_DIST_REGS: &[DistReg] = &[
     VGIC_DIST_REG!(GICD_IPRIORITYR, 8, 0),
 ];
 
-fn dist_attr_access(gic: &DeviceFd, offset: u32, val: &u32, set: bool) -> Result<()> {
+fn dist_attr_access(
+    gic: &Arc<dyn hypervisor::Device>,
+    offset: u32,
+    val: &u32,
+    set: bool,
+) -> Result<()> {
     let mut gic_dist_attr = kvm_device_attr {
         group: KVM_DEV_ARM_VGIC_GRP_DIST_REGS,
         attr: offset as u64,
@@ -85,30 +90,28 @@ fn dist_attr_access(gic: &DeviceFd, offset: u32, val: &u32, set: bool) -> Result
         flags: 0,
     };
     if set {
-        gic.set_device_attr(&gic_dist_attr).map_err(|e| {
-            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
-        })?;
+        gic.set_device_attr(&gic_dist_attr)
+            .map_err(Error::SetDeviceAttribute)?;
     } else {
-        gic.get_device_attr(&mut gic_dist_attr).map_err(|e| {
-            Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
-        })?;
+        gic.get_device_attr(&mut gic_dist_attr)
+            .map_err(Error::GetDeviceAttribute)?;
     }
     Ok(())
 }
 
 /// Get the distributor control register.
-pub fn read_ctlr(gic: &DeviceFd) -> Result<u32> {
+pub fn read_ctlr(gic: &Arc<dyn hypervisor::Device>) -> Result<u32> {
     let val: u32 = 0;
     dist_attr_access(gic, GICD_CTLR, &val, false)?;
     Ok(val)
 }
 
 /// Set the distributor control register.
-pub fn write_ctlr(gic: &DeviceFd, val: u32) -> Result<()> {
+pub fn write_ctlr(gic: &Arc<dyn hypervisor::Device>, val: u32) -> Result<()> {
     dist_attr_access(gic, GICD_CTLR, &val, true)
 }
 
-fn get_interrupts_num(gic: &DeviceFd) -> Result<u32> {
+fn get_interrupts_num(gic: &Arc<dyn hypervisor::Device>) -> Result<u32> {
     let num_irq = 0;
 
     let mut nr_irqs_attr = kvm_device_attr {
@@ -117,18 +120,12 @@ fn get_interrupts_num(gic: &DeviceFd) -> Result<u32> {
         addr: &num_irq as *const u32 as u64,
         flags: 0,
     };
-    gic.get_device_attr(&mut nr_irqs_attr).map_err(|e| {
-        Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
-    })?;
+    gic.get_device_attr(&mut nr_irqs_attr)
+        .map_err(Error::GetDeviceAttribute)?;
     Ok(num_irq)
 }
 
-fn compute_reg_len(gic: &DeviceFd, reg: &DistReg, base: u32) -> Result<u32> {
-    // FIXME:
-    // Redefine some GIC constants to avoid the dependency on `layout` crate.
-    // This is temporary solution, will be fixed in future refactoring.
-    const LAYOUT_IRQ_BASE: u32 = 32;
-
+fn compute_reg_len(gic: &Arc<dyn hypervisor::Device>, reg: &DistReg, base: u32) -> Result<u32> {
     let mut end = base;
     let num_irq = get_interrupts_num(gic)?;
     if reg.length > 0 {
@@ -141,8 +138,8 @@ fn compute_reg_len(gic: &DeviceFd, reg: &DistReg, base: u32) -> Result<u32> {
         // This is the type of register that takes into account the number of interrupts
         // that the model has. It is also the type of register where
         // a register relates to multiple interrupts.
-        end = base + (reg.bpi as u32 * (num_irq - LAYOUT_IRQ_BASE) / 8);
-        if reg.bpi as u32 * (num_irq - LAYOUT_IRQ_BASE) % 8 > 0 {
+        end = base + (reg.bpi as u32 * (num_irq - IRQ_BASE) / 8);
+        if reg.bpi as u32 * (num_irq - IRQ_BASE) % 8 > 0 {
             end += REG_SIZE as u32;
         }
     }
@@ -150,7 +147,7 @@ fn compute_reg_len(gic: &DeviceFd, reg: &DistReg, base: u32) -> Result<u32> {
 }
 
 /// Set distributor registers of the GIC.
-pub fn set_dist_regs(gic: &DeviceFd, state: &[u32]) -> Result<()> {
+pub fn set_dist_regs(gic: &Arc<dyn hypervisor::Device>, state: &[u32]) -> Result<()> {
     let mut idx = 0;
 
     for dreg in VGIC_DIST_REGS {
@@ -167,7 +164,7 @@ pub fn set_dist_regs(gic: &DeviceFd, state: &[u32]) -> Result<()> {
     Ok(())
 }
 /// Get distributor registers of the GIC.
-pub fn get_dist_regs(gic: &DeviceFd) -> Result<Vec<u32>> {
+pub fn get_dist_regs(gic: &Arc<dyn hypervisor::Device>) -> Result<Vec<u32>> {
     let mut state = Vec::new();
 
     for dreg in VGIC_DIST_REGS {

arch/src/aarch64/gic/gicv3.rs

@@ -0,0 +1,261 @@
+// Copyright 2021 Arm Limited (or its affiliates). All rights reserved.
+// Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+// SPDX-License-Identifier: Apache-2.0
+//
+// This file implements the GicV3 device.
+
+pub mod kvm {
+    use crate::aarch64::gic::dist_regs::{get_dist_regs, read_ctlr, set_dist_regs, write_ctlr};
+    use crate::aarch64::gic::icc_regs::{get_icc_regs, set_icc_regs};
+    use crate::aarch64::gic::kvm::{save_pending_tables, KvmGicDevice};
+    use crate::aarch64::gic::redist_regs::{
+        construct_gicr_typers, get_redist_regs, set_redist_regs,
+    };
+    use crate::aarch64::gic::GicDevice;
+    use crate::layout;
+    use anyhow::anyhow;
+    use hypervisor::kvm::kvm_bindings;
+    use hypervisor::CpuState;
+    use std::any::Any;
+    use std::convert::TryInto;
+    use std::sync::Arc;
+    use std::{boxed::Box, result};
+    use versionize::{VersionMap, Versionize, VersionizeResult};
+    use versionize_derive::Versionize;
+    use vm_migration::{
+        Migratable, MigratableError, Pausable, Snapshot, Snapshottable, Transportable,
+        VersionMapped,
+    };
+
+    /// Errors thrown while saving/restoring the GICv3.
+    #[derive(Debug)]
+    pub enum Error {
+        /// Error in saving RDIST pending tables into guest RAM.
+        SavePendingTables(crate::aarch64::gic::Error),
+        /// Error in saving GIC distributor registers.
+        SaveDistributorRegisters(crate::aarch64::gic::Error),
+        /// Error in restoring GIC distributor registers.
+        RestoreDistributorRegisters(crate::aarch64::gic::Error),
+        /// Error in saving GIC distributor control registers.
+        SaveDistributorCtrlRegisters(crate::aarch64::gic::Error),
+        /// Error in restoring GIC distributor control registers.
+        RestoreDistributorCtrlRegisters(crate::aarch64::gic::Error),
+        /// Error in saving GIC redistributor registers.
+        SaveRedistributorRegisters(crate::aarch64::gic::Error),
+        /// Error in restoring GIC redistributor registers.
+        RestoreRedistributorRegisters(crate::aarch64::gic::Error),
+        /// Error in saving GIC CPU interface registers.
+        SaveIccRegisters(crate::aarch64::gic::Error),
+        /// Error in restoring GIC CPU interface registers.
+        RestoreIccRegisters(crate::aarch64::gic::Error),
+    }
+
+    type Result<T> = result::Result<T, Error>;
+
+    pub struct KvmGicV3 {
+        /// The hypervisor agnostic device for the GicV3
+        device: Arc<dyn hypervisor::Device>,
+
+        /// Vector holding values of GICR_TYPER for each vCPU
+        gicr_typers: Vec<u64>,
+
+        /// GIC device properties, to be used for setting up the fdt entry
+        properties: [u64; 4],
+
+        /// Number of CPUs handled by the device
+        vcpu_count: u64,
+    }
+
+    #[derive(Versionize)]
+    pub struct Gicv3State {
+        dist: Vec<u32>,
+        rdist: Vec<u32>,
+        icc: Vec<u32>,
+        // special register that enables interrupts and affinity routing
+        gicd_ctlr: u32,
+    }
+
+    impl VersionMapped for Gicv3State {}
+
+    impl KvmGicV3 {
+        // Device trees specific constants
+        pub const ARCH_GIC_V3_MAINT_IRQ: u32 = 9;
+
+        /// Get the address of the GIC distributor.
+        pub fn get_dist_addr() -> u64 {
+            layout::GIC_V3_DIST_START
+        }
+
+        /// Get the size of the GIC distributor.
+        pub fn get_dist_size() -> u64 {
+            layout::GIC_V3_DIST_SIZE
+        }
+
+        /// Get the address of the GIC redistributors.
+        pub fn get_redists_addr(vcpu_count: u64) -> u64 {
+            KvmGicV3::get_dist_addr() - KvmGicV3::get_redists_size(vcpu_count)
+        }
+
+        /// Get the size of the GIC redistributors.
+        pub fn get_redists_size(vcpu_count: u64) -> u64 {
+            vcpu_count * layout::GIC_V3_REDIST_SIZE
+        }
+
+        /// Save the state of GIC.
+        fn state(&self, gicr_typers: &[u64]) -> Result<Gicv3State> {
+            let gicd_ctlr =
+                read_ctlr(self.device()).map_err(Error::SaveDistributorCtrlRegisters)?;
+
+            let dist_state =
+                get_dist_regs(self.device()).map_err(Error::SaveDistributorRegisters)?;
+
+            let rdist_state = get_redist_regs(self.device(), gicr_typers)
+                .map_err(Error::SaveRedistributorRegisters)?;
+
+            let icc_state =
+                get_icc_regs(self.device(), gicr_typers).map_err(Error::SaveIccRegisters)?;
+
+            Ok(Gicv3State {
+                dist: dist_state,
+                rdist: rdist_state,
+                icc: icc_state,
+                gicd_ctlr,
+            })
+        }
+
+        /// Restore the state of GIC.
+        fn set_state(&mut self, gicr_typers: &[u64], state: &Gicv3State) -> Result<()> {
+            write_ctlr(self.device(), state.gicd_ctlr)
+                .map_err(Error::RestoreDistributorCtrlRegisters)?;
+
+            set_dist_regs(self.device(), &state.dist)
+                .map_err(Error::RestoreDistributorRegisters)?;
+
+            set_redist_regs(self.device(), gicr_typers, &state.rdist)
+                .map_err(Error::RestoreRedistributorRegisters)?;
+
+            set_icc_regs(self.device(), gicr_typers, &state.icc)
+                .map_err(Error::RestoreIccRegisters)?;
+
+            Ok(())
+        }
+    }
+
+    impl GicDevice for KvmGicV3 {
+        fn device(&self) -> &Arc<dyn hypervisor::Device> {
+            &self.device
+        }
+
+        fn fdt_compatibility(&self) -> &str {
+            "arm,gic-v3"
+        }
+
+        fn fdt_maint_irq(&self) -> u32 {
+            KvmGicV3::ARCH_GIC_V3_MAINT_IRQ
+        }
+
+        fn device_properties(&self) -> &[u64] {
+            &self.properties
+        }
+
+        fn vcpu_count(&self) -> u64 {
+            self.vcpu_count
+        }
+
+        fn set_its_device(&mut self, _its_device: Option<Arc<dyn hypervisor::Device>>) {}
+
+        fn set_gicr_typers(&mut self, vcpu_states: &[CpuState]) {
+            let gicr_typers = construct_gicr_typers(vcpu_states);
+            self.gicr_typers = gicr_typers;
+        }
+
+        fn as_any_concrete_mut(&mut self) -> &mut dyn Any {
+            self
+        }
+    }
+
+    impl KvmGicDevice for KvmGicV3 {
+        fn version() -> u32 {
+            kvm_bindings::kvm_device_type_KVM_DEV_TYPE_ARM_VGIC_V3
+        }
+
+        fn create_device(
+            device: Arc<dyn hypervisor::Device>,
+            vcpu_count: u64,
+        ) -> Box<dyn GicDevice> {
+            Box::new(KvmGicV3 {
+                device,
+                gicr_typers: vec![0; vcpu_count.try_into().unwrap()],
+                properties: [
+                    KvmGicV3::get_dist_addr(),
+                    KvmGicV3::get_dist_size(),
+                    KvmGicV3::get_redists_addr(vcpu_count),
+                    KvmGicV3::get_redists_size(vcpu_count),
+                ],
+                vcpu_count,
+            })
+        }
+
+        fn init_device_attributes(
+            _vm: &Arc<dyn hypervisor::Vm>,
+            gic_device: &mut dyn GicDevice,
+        ) -> crate::aarch64::gic::Result<()> {
+            /* Setting up the distributor attribute.
+             We are placing the GIC below 1GB so we need to substract the size of the distributor.
+            */
+            Self::set_device_attribute(
+                gic_device.device(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ADDR,
+                u64::from(kvm_bindings::KVM_VGIC_V3_ADDR_TYPE_DIST),
+                &KvmGicV3::get_dist_addr() as *const u64 as u64,
+                0,
+            )?;
+
+            /* Setting up the redistributors' attribute.
+            We are calculating here the start of the redistributors address. We have one per CPU.
+            */
+            Self::set_device_attribute(
+                gic_device.device(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ADDR,
+                u64::from(kvm_bindings::KVM_VGIC_V3_ADDR_TYPE_REDIST),
+                &KvmGicV3::get_redists_addr(gic_device.vcpu_count()) as *const u64 as u64,
+                0,
+            )?;
+
+            Ok(())
+        }
+    }
+
+    pub const GIC_V3_SNAPSHOT_ID: &str = "gic-v3";
+    impl Snapshottable for KvmGicV3 {
+        fn id(&self) -> String {
+            GIC_V3_SNAPSHOT_ID.to_string()
+        }
+
+        fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
+            let gicr_typers = self.gicr_typers.clone();
+            Snapshot::new_from_versioned_state(&self.id(), &self.state(&gicr_typers).unwrap())
+        }
+
+        fn restore(&mut self, snapshot: Snapshot) -> std::result::Result<(), MigratableError> {
+            let gicr_typers = self.gicr_typers.clone();
+            self.set_state(&gicr_typers, &snapshot.to_versioned_state(&self.id())?)
+                .map_err(|e| {
+                    MigratableError::Restore(anyhow!("Could not restore GICv3 state {:?}", e))
+                })
+        }
+    }
+
+    impl Pausable for KvmGicV3 {
+        fn pause(&mut self) -> std::result::Result<(), MigratableError> {
+            // Flush redistributors pending tables to guest RAM.
+            save_pending_tables(self.device()).map_err(|e| {
+                MigratableError::Pause(anyhow!("Could not save GICv3 GIC pending tables {:?}", e))
+            })?;
+
+            Ok(())
+        }
+    }
+    impl Transportable for KvmGicV3 {}
+    impl Migratable for KvmGicV3 {}
+}

arch/src/aarch64/gic/gicv3_its.rs

@@ -0,0 +1,514 @@
+// Copyright 2021 Arm Limited (or its affiliates). All rights reserved.
+// SPDX-License-Identifier: Apache-2.0
+//
+// This file implements the GicV3 device with ITS (Virtual Interrupt Translation Service).
+
+pub mod kvm {
+    use crate::aarch64::gic::dist_regs::{get_dist_regs, read_ctlr, set_dist_regs, write_ctlr};
+    use crate::aarch64::gic::icc_regs::{get_icc_regs, set_icc_regs};
+    use crate::aarch64::gic::redist_regs::{
+        construct_gicr_typers, get_redist_regs, set_redist_regs,
+    };
+
+    use crate::aarch64::gic::gicv3::kvm::KvmGicV3;
+    use crate::aarch64::gic::kvm::{save_pending_tables, KvmGicDevice};
+    use crate::aarch64::gic::GicDevice;
+    use crate::layout;
+    use anyhow::anyhow;
+    use hypervisor::kvm::kvm_bindings;
+    use hypervisor::CpuState;
+    use std::any::Any;
+    use std::convert::TryInto;
+    use std::sync::Arc;
+    use std::{boxed::Box, result};
+    use versionize::{VersionMap, Versionize, VersionizeResult};
+    use versionize_derive::Versionize;
+    use vm_migration::{
+        Migratable, MigratableError, Pausable, Snapshot, Snapshottable, Transportable,
+        VersionMapped,
+    };
+
+    const GITS_CTLR: u32 = 0x0000;
+    const GITS_IIDR: u32 = 0x0004;
+    const GITS_CBASER: u32 = 0x0080;
+    const GITS_CWRITER: u32 = 0x0088;
+    const GITS_CREADR: u32 = 0x0090;
+    const GITS_BASER: u32 = 0x0100;
+
+    /// Errors thrown while saving/restoring the GICv3ITS.
+    #[derive(Debug)]
+    pub enum Error {
+        /// Error in saving RDIST pending tables into guest RAM.
+        SavePendingTables(crate::aarch64::gic::Error),
+        /// Error in saving GIC distributor registers.
+        SaveDistributorRegisters(crate::aarch64::gic::Error),
+        /// Error in restoring GIC distributor registers.
+        RestoreDistributorRegisters(crate::aarch64::gic::Error),
+        /// Error in saving GIC distributor control registers.
+        SaveDistributorCtrlRegisters(crate::aarch64::gic::Error),
+        /// Error in restoring GIC distributor control registers.
+        RestoreDistributorCtrlRegisters(crate::aarch64::gic::Error),
+        /// Error in saving GIC redistributor registers.
+        SaveRedistributorRegisters(crate::aarch64::gic::Error),
+        /// Error in restoring GIC redistributor registers.
+        RestoreRedistributorRegisters(crate::aarch64::gic::Error),
+        /// Error in saving GIC CPU interface registers.
+        SaveIccRegisters(crate::aarch64::gic::Error),
+        /// Error in restoring GIC CPU interface registers.
+        RestoreIccRegisters(crate::aarch64::gic::Error),
+        /// Error in saving GICv3ITS IIDR register.
+        SaveITSIIDR(crate::aarch64::gic::Error),
+        /// Error in restoring GICv3ITS IIDR register.
+        RestoreITSIIDR(crate::aarch64::gic::Error),
+        /// Error in saving GICv3ITS CBASER register.
+        SaveITSCBASER(crate::aarch64::gic::Error),
+        /// Error in restoring GICv3ITS CBASER register.
+        RestoreITSCBASER(crate::aarch64::gic::Error),
+        /// Error in saving GICv3ITS CREADR register.
+        SaveITSCREADR(crate::aarch64::gic::Error),
+        /// Error in restoring GICv3ITS CREADR register.
+        RestoreITSCREADR(crate::aarch64::gic::Error),
+        /// Error in saving GICv3ITS CWRITER register.
+        SaveITSCWRITER(crate::aarch64::gic::Error),
+        /// Error in restoring GICv3ITS CWRITER register.
+        RestoreITSCWRITER(crate::aarch64::gic::Error),
+        /// Error in saving GICv3ITS BASER register.
+        SaveITSBASER(crate::aarch64::gic::Error),
+        /// Error in restoring GICv3ITS BASER register.
+        RestoreITSBASER(crate::aarch64::gic::Error),
+        /// Error in saving GICv3ITS CTLR register.
+        SaveITSCTLR(crate::aarch64::gic::Error),
+        /// Error in restoring GICv3ITS CTLR register.
+        RestoreITSCTLR(crate::aarch64::gic::Error),
+        /// Error in saving GICv3ITS restore tables.
+        SaveITSTables(crate::aarch64::gic::Error),
+        /// Error in restoring GICv3ITS restore tables.
+        RestoreITSTables(crate::aarch64::gic::Error),
+    }
+
+    type Result<T> = result::Result<T, Error>;
+
+    /// Access an ITS device attribute.
+    ///
+    /// This is a helper function to get/set the ITS device attribute depending
+    /// the bool parameter `set` provided.
+    pub fn gicv3_its_attr_access(
+        its_device: &Arc<dyn hypervisor::Device>,
+        group: u32,
+        attr: u32,
+        val: &u64,
+        set: bool,
+    ) -> crate::aarch64::gic::Result<()> {
+        let mut gicv3_its_attr = kvm_bindings::kvm_device_attr {
+            group,
+            attr: attr as u64,
+            addr: val as *const u64 as u64,
+            flags: 0,
+        };
+        if set {
+            its_device
+                .set_device_attr(&gicv3_its_attr)
+                .map_err(crate::aarch64::gic::Error::SetDeviceAttribute)?;
+        } else {
+            its_device
+                .get_device_attr(&mut gicv3_its_attr)
+                .map_err(crate::aarch64::gic::Error::GetDeviceAttribute)?;
+        }
+        Ok(())
+    }
+
+    /// Function that saves/restores ITS tables into guest RAM.
+    ///
+    /// The tables get flushed to guest RAM whenever the VM gets stopped.
+    pub fn gicv3_its_tables_access(
+        its_device: &Arc<dyn hypervisor::Device>,
+        save: bool,
+    ) -> crate::aarch64::gic::Result<()> {
+        let attr = if save {
+            u64::from(kvm_bindings::KVM_DEV_ARM_ITS_SAVE_TABLES)
+        } else {
+            u64::from(kvm_bindings::KVM_DEV_ARM_ITS_RESTORE_TABLES)
+        };
+
+        let init_gic_attr = kvm_bindings::kvm_device_attr {
+            group: kvm_bindings::KVM_DEV_ARM_VGIC_GRP_CTRL,
+            attr,
+            addr: 0,
+            flags: 0,
+        };
+        its_device
+            .set_device_attr(&init_gic_attr)
+            .map_err(crate::aarch64::gic::Error::SetDeviceAttribute)
+    }
+
+    pub struct KvmGicV3Its {
+        /// The hypervisor agnostic device for the GicV3
+        device: Arc<dyn hypervisor::Device>,
+
+        /// The hypervisor agnostic device for the Its device
+        its_device: Option<Arc<dyn hypervisor::Device>>,
+
+        /// Vector holding values of GICR_TYPER for each vCPU
+        gicr_typers: Vec<u64>,
+
+        /// GIC device properties, to be used for setting up the fdt entry
+        gic_properties: [u64; 4],
+
+        /// MSI device properties, to be used for setting up the fdt entry
+        msi_properties: [u64; 2],
+
+        /// Number of CPUs handled by the device
+        vcpu_count: u64,
+    }
+
+    #[derive(Versionize)]
+    pub struct Gicv3ItsState {
+        dist: Vec<u32>,
+        rdist: Vec<u32>,
+        icc: Vec<u32>,
+        // special register that enables interrupts and affinity routing
+        gicd_ctlr: u32,
+        its_ctlr: u64,
+        its_iidr: u64,
+        its_cbaser: u64,
+        its_cwriter: u64,
+        its_creadr: u64,
+        its_baser: [u64; 8],
+    }
+
+    impl VersionMapped for Gicv3ItsState {}
+
+    impl KvmGicV3Its {
+        fn get_msi_size() -> u64 {
+            layout::GIC_V3_ITS_SIZE
+        }
+
+        fn get_msi_addr(vcpu_count: u64) -> u64 {
+            KvmGicV3::get_redists_addr(vcpu_count) - KvmGicV3Its::get_msi_size()
+        }
+
+        /// Save the state of GICv3ITS.
+        fn state(&self, gicr_typers: &[u64]) -> Result<Gicv3ItsState> {
+            let gicd_ctlr =
+                read_ctlr(self.device()).map_err(Error::SaveDistributorCtrlRegisters)?;
+
+            let dist_state =
+                get_dist_regs(self.device()).map_err(Error::SaveDistributorRegisters)?;
+
+            let rdist_state = get_redist_regs(self.device(), gicr_typers)
+                .map_err(Error::SaveRedistributorRegisters)?;
+
+            let icc_state =
+                get_icc_regs(self.device(), gicr_typers).map_err(Error::SaveIccRegisters)?;
+
+            let its_baser_state: [u64; 8] = [0; 8];
+            for i in 0..8 {
+                gicv3_its_attr_access(
+                    self.its_device().unwrap(),
+                    kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                    GITS_BASER + i * 8,
+                    &its_baser_state[i as usize],
+                    false,
+                )
+                .map_err(Error::SaveITSBASER)?;
+            }
+
+            let its_ctlr_state: u64 = 0;
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_CTLR,
+                &its_ctlr_state,
+                false,
+            )
+            .map_err(Error::SaveITSCTLR)?;
+
+            let its_cbaser_state: u64 = 0;
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_CBASER,
+                &its_cbaser_state,
+                false,
+            )
+            .map_err(Error::SaveITSCBASER)?;
+
+            let its_creadr_state: u64 = 0;
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_CREADR,
+                &its_creadr_state,
+                false,
+            )
+            .map_err(Error::SaveITSCREADR)?;
+
+            let its_cwriter_state: u64 = 0;
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_CWRITER,
+                &its_cwriter_state,
+                false,
+            )
+            .map_err(Error::SaveITSCWRITER)?;
+
+            let its_iidr_state: u64 = 0;
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_IIDR,
+                &its_iidr_state,
+                false,
+            )
+            .map_err(Error::SaveITSIIDR)?;
+
+            Ok(Gicv3ItsState {
+                dist: dist_state,
+                rdist: rdist_state,
+                icc: icc_state,
+                gicd_ctlr,
+                its_ctlr: its_ctlr_state,
+                its_iidr: its_iidr_state,
+                its_cbaser: its_cbaser_state,
+                its_cwriter: its_cwriter_state,
+                its_creadr: its_creadr_state,
+                its_baser: its_baser_state,
+            })
+        }
+
+        /// Restore the state of GICv3ITS.
+        fn set_state(&mut self, gicr_typers: &[u64], state: &Gicv3ItsState) -> Result<()> {
+            write_ctlr(self.device(), state.gicd_ctlr)
+                .map_err(Error::RestoreDistributorCtrlRegisters)?;
+
+            set_dist_regs(self.device(), &state.dist)
+                .map_err(Error::RestoreDistributorRegisters)?;
+
+            set_redist_regs(self.device(), gicr_typers, &state.rdist)
+                .map_err(Error::RestoreRedistributorRegisters)?;
+
+            set_icc_regs(self.device(), gicr_typers, &state.icc)
+                .map_err(Error::RestoreIccRegisters)?;
+
+            //Restore GICv3ITS registers
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_IIDR,
+                &state.its_iidr,
+                true,
+            )
+            .map_err(Error::RestoreITSIIDR)?;
+
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_CBASER,
+                &state.its_cbaser,
+                true,
+            )
+            .map_err(Error::RestoreITSCBASER)?;
+
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_CREADR,
+                &state.its_creadr,
+                true,
+            )
+            .map_err(Error::RestoreITSCREADR)?;
+
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_CWRITER,
+                &state.its_cwriter,
+                true,
+            )
+            .map_err(Error::RestoreITSCWRITER)?;
+
+            for i in 0..8 {
+                gicv3_its_attr_access(
+                    self.its_device().unwrap(),
+                    kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                    GITS_BASER + i * 8,
+                    &state.its_baser[i as usize],
+                    true,
+                )
+                .map_err(Error::RestoreITSBASER)?;
+            }
+
+            // Restore ITS tables
+            gicv3_its_tables_access(self.its_device().unwrap(), false)
+                .map_err(Error::RestoreITSTables)?;
+
+            gicv3_its_attr_access(
+                self.its_device().unwrap(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
+                GITS_CTLR,
+                &state.its_ctlr,
+                true,
+            )
+            .map_err(Error::RestoreITSCTLR)?;
+
+            Ok(())
+        }
+    }
+
+    impl GicDevice for KvmGicV3Its {
+        fn device(&self) -> &Arc<dyn hypervisor::Device> {
+            &self.device
+        }
+
+        fn its_device(&self) -> Option<&Arc<dyn hypervisor::Device>> {
+            self.its_device.as_ref()
+        }
+
+        fn fdt_compatibility(&self) -> &str {
+            "arm,gic-v3"
+        }
+
+        fn msi_compatible(&self) -> bool {
+            true
+        }
+
+        fn msi_compatibility(&self) -> &str {
+            "arm,gic-v3-its"
+        }
+
+        fn fdt_maint_irq(&self) -> u32 {
+            KvmGicV3::ARCH_GIC_V3_MAINT_IRQ
+        }
+
+        fn msi_properties(&self) -> &[u64] {
+            &self.msi_properties
+        }
+
+        fn device_properties(&self) -> &[u64] {
+            &self.gic_properties
+        }
+
+        fn vcpu_count(&self) -> u64 {
+            self.vcpu_count
+        }
+
+        fn set_its_device(&mut self, its_device: Option<Arc<dyn hypervisor::Device>>) {
+            self.its_device = its_device;
+        }
+
+        fn set_gicr_typers(&mut self, vcpu_states: &[CpuState]) {
+            let gicr_typers = construct_gicr_typers(vcpu_states);
+            self.gicr_typers = gicr_typers;
+        }
+
+        fn as_any_concrete_mut(&mut self) -> &mut dyn Any {
+            self
+        }
+    }
+
+    impl KvmGicDevice for KvmGicV3Its {
+        fn version() -> u32 {
+            KvmGicV3::version()
+        }
+
+        fn create_device(
+            device: Arc<dyn hypervisor::Device>,
+            vcpu_count: u64,
+        ) -> Box<dyn GicDevice> {
+            Box::new(KvmGicV3Its {
+                device,
+                its_device: None,
+                gicr_typers: vec![0; vcpu_count.try_into().unwrap()],
+                gic_properties: [
+                    KvmGicV3::get_dist_addr(),
+                    KvmGicV3::get_dist_size(),
+                    KvmGicV3::get_redists_addr(vcpu_count),
+                    KvmGicV3::get_redists_size(vcpu_count),
+                ],
+                msi_properties: [
+                    KvmGicV3Its::get_msi_addr(vcpu_count),
+                    KvmGicV3Its::get_msi_size(),
+                ],
+                vcpu_count,
+            })
+        }
+
+        fn init_device_attributes(
+            vm: &Arc<dyn hypervisor::Vm>,
+            gic_device: &mut dyn GicDevice,
+        ) -> crate::aarch64::gic::Result<()> {
+            KvmGicV3::init_device_attributes(vm, gic_device)?;
+
+            let mut its_device = kvm_bindings::kvm_create_device {
+                type_: kvm_bindings::kvm_device_type_KVM_DEV_TYPE_ARM_VGIC_ITS,
+                fd: 0,
+                flags: 0,
+            };
+
+            let its_fd = vm
+                .create_device(&mut its_device)
+                .map_err(crate::aarch64::gic::Error::CreateGic)?;
+
+            Self::set_device_attribute(
+                &its_fd,
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ADDR,
+                u64::from(kvm_bindings::KVM_VGIC_ITS_ADDR_TYPE),
+                &KvmGicV3Its::get_msi_addr(gic_device.vcpu_count()) as *const u64 as u64,
+                0,
+            )?;
+
+            Self::set_device_attribute(
+                &its_fd,
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_CTRL,
+                u64::from(kvm_bindings::KVM_DEV_ARM_VGIC_CTRL_INIT),
+                0,
+                0,
+            )?;
+
+            gic_device.set_its_device(Some(its_fd));
+
+            Ok(())
+        }
+    }
+
+    pub const GIC_V3_ITS_SNAPSHOT_ID: &str = "gic-v3-its";
+    impl Snapshottable for KvmGicV3Its {
+        fn id(&self) -> String {
+            GIC_V3_ITS_SNAPSHOT_ID.to_string()
+        }
+
+        fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
+            let gicr_typers = self.gicr_typers.clone();
+            Snapshot::new_from_versioned_state(&self.id(), &self.state(&gicr_typers).unwrap())
+        }
+
+        fn restore(&mut self, snapshot: Snapshot) -> std::result::Result<(), MigratableError> {
+            let gicr_typers = self.gicr_typers.clone();
+            self.set_state(&gicr_typers, &snapshot.to_versioned_state(&self.id())?)
+                .map_err(|e| {
+                    MigratableError::Restore(anyhow!("Could not restore GICv3ITS state {:?}", e))
+                })
+        }
+    }
+
+    impl Pausable for KvmGicV3Its {
+        fn pause(&mut self) -> std::result::Result<(), MigratableError> {
+            // Flush redistributors pending tables to guest RAM.
+            save_pending_tables(self.device()).map_err(|e| {
+                MigratableError::Pause(anyhow!(
+                    "Could not save GICv3ITS GIC pending tables {:?}",
+                    e
+                ))
+            })?;
+            // Flush ITS tables to guest RAM.
+            gicv3_its_tables_access(self.its_device().unwrap(), true).map_err(|e| {
+                MigratableError::Pause(anyhow!("Could not save GICv3ITS ITS tables {:?}", e))
+            })?;
+
+            Ok(())
+        }
+    }
+    impl Transportable for KvmGicV3Its {}
+    impl Migratable for KvmGicV3Its {}
+}

arch/src/aarch64/gic/icc_regs.rs

@@ -1,16 +1,14 @@
-// Copyright 2022 Arm Limited (or its affiliates). All rights reserved.
 // Copyright 2020 Amazon.com, Inc. or its affiliates. All Rights Reserved.
 // SPDX-License-Identifier: Apache-2.0
 
-use crate::arch::aarch64::gic::{Error, Result};
-use crate::device::HypervisorDeviceError;
-use crate::kvm::kvm_bindings::{
+use super::{Error, Result};
+use hypervisor::kvm::kvm_bindings::{
     kvm_device_attr, KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS, KVM_REG_ARM64_SYSREG_CRM_MASK,
     KVM_REG_ARM64_SYSREG_CRM_SHIFT, KVM_REG_ARM64_SYSREG_CRN_MASK, KVM_REG_ARM64_SYSREG_CRN_SHIFT,
     KVM_REG_ARM64_SYSREG_OP0_MASK, KVM_REG_ARM64_SYSREG_OP0_SHIFT, KVM_REG_ARM64_SYSREG_OP1_MASK,
     KVM_REG_ARM64_SYSREG_OP1_SHIFT, KVM_REG_ARM64_SYSREG_OP2_MASK, KVM_REG_ARM64_SYSREG_OP2_SHIFT,
 };
-use kvm_ioctls::DeviceFd;
+use std::sync::Arc;
 
 const KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT: u32 = 32;
 const KVM_DEV_ARM_VGIC_V3_MPIDR_MASK: u64 = 0xffffffff << KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT as u64;
@@ -79,7 +77,13 @@ static VGIC_ICC_REGS: &[u64] = &[
     SYS_ICC_AP1R3_EL1,
 ];
 
-fn icc_attr_access(gic: &DeviceFd, offset: u64, typer: u64, val: &u32, set: bool) -> Result<()> {
+fn icc_attr_access(
+    gic: &Arc<dyn hypervisor::Device>,
+    offset: u64,
+    typer: u64,
+    val: &u32,
+    set: bool,
+) -> Result<()> {
     let mut gic_icc_attr = kvm_device_attr {
         group: KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS,
         attr: ((typer & KVM_DEV_ARM_VGIC_V3_MPIDR_MASK) | offset), // this needs the mpidr
@@ -87,19 +91,17 @@ fn icc_attr_access(gic: &DeviceFd, offset: u64, typer: u64, val: &u32, set: bool
         flags: 0,
     };
     if set {
-        gic.set_device_attr(&gic_icc_attr).map_err(|e| {
-            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
-        })?;
+        gic.set_device_attr(&gic_icc_attr)
+            .map_err(Error::SetDeviceAttribute)?;
     } else {
-        gic.get_device_attr(&mut gic_icc_attr).map_err(|e| {
-            Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
-        })?;
+        gic.get_device_attr(&mut gic_icc_attr)
+            .map_err(Error::GetDeviceAttribute)?;
     }
     Ok(())
 }
 
 /// Get ICC registers.
-pub fn get_icc_regs(gic: &DeviceFd, gicr_typer: &[u64]) -> Result<Vec<u32>> {
+pub fn get_icc_regs(gic: &Arc<dyn hypervisor::Device>, gicr_typer: &[u64]) -> Result<Vec<u32>> {
     let mut state: Vec<u32> = Vec::new();
     // We need this for the ICC_AP<m>R<n>_EL1 registers.
     let mut num_priority_bits = 0;
@@ -152,7 +154,11 @@ pub fn get_icc_regs(gic: &DeviceFd, gicr_typer: &[u64]) -> Result<Vec<u32>> {
 }
 
 /// Set ICC registers.
-pub fn set_icc_regs(gic: &DeviceFd, gicr_typer: &[u64], state: &[u32]) -> Result<()> {
+pub fn set_icc_regs(
+    gic: &Arc<dyn hypervisor::Device>,
+    gicr_typer: &[u64],
+    state: &[u32],
+) -> Result<()> {
     let mut num_priority_bits = 0;
     let mut idx = 0;
     for ix in gicr_typer {

arch/src/aarch64/gic/mod.rs

@@ -0,0 +1,220 @@
+// Copyright 2021 Arm Limited (or its affiliates). All rights reserved.
+// Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+// SPDX-License-Identifier: Apache-2.0
+
+pub mod dist_regs;
+pub mod gicv3;
+pub mod gicv3_its;
+pub mod icc_regs;
+pub mod redist_regs;
+
+pub use self::dist_regs::{get_dist_regs, read_ctlr, set_dist_regs, write_ctlr};
+pub use self::icc_regs::{get_icc_regs, set_icc_regs};
+pub use self::redist_regs::{get_redist_regs, set_redist_regs};
+use hypervisor::CpuState;
+use std::any::Any;
+use std::result;
+use std::sync::Arc;
+
+/// Errors thrown while setting up the GIC.
+#[derive(Debug)]
+pub enum Error {
+    /// Error while calling KVM ioctl for setting up the global interrupt controller.
+    CreateGic(hypervisor::HypervisorVmError),
+    /// Error while setting device attributes for the GIC.
+    SetDeviceAttribute(hypervisor::HypervisorDeviceError),
+    /// Error while getting device attributes for the GIC.
+    GetDeviceAttribute(hypervisor::HypervisorDeviceError),
+}
+type Result<T> = result::Result<T, Error>;
+
+pub trait GicDevice: Send {
+    /// Returns the hypervisor agnostic Device of the GIC device
+    fn device(&self) -> &Arc<dyn hypervisor::Device>;
+
+    /// Returns the hypervisor agnostic Device of the ITS device
+    fn its_device(&self) -> Option<&Arc<dyn hypervisor::Device>> {
+        None
+    }
+
+    /// Returns the fdt compatibility property of the device
+    fn fdt_compatibility(&self) -> &str;
+
+    /// Returns the maint_irq fdt property of the device
+    fn fdt_maint_irq(&self) -> u32;
+
+    /// Returns an array with GIC device properties
+    fn device_properties(&self) -> &[u64];
+
+    /// Returns the number of vCPUs this GIC handles
+    fn vcpu_count(&self) -> u64;
+
+    /// Returns whether the GIC device is MSI compatible or not
+    fn msi_compatible(&self) -> bool {
+        false
+    }
+
+    /// Returns the MSI compatibility property of the device
+    fn msi_compatibility(&self) -> &str {
+        ""
+    }
+
+    /// Returns the MSI reg property of the device
+    fn msi_properties(&self) -> &[u64] {
+        &[]
+    }
+
+    fn set_its_device(&mut self, its_device: Option<Arc<dyn hypervisor::Device>>);
+
+    /// Get the values of GICR_TYPER for each vCPU.
+    fn set_gicr_typers(&mut self, vcpu_states: &[CpuState]);
+
+    /// Downcast the trait object to its concrete type.
+    fn as_any_concrete_mut(&mut self) -> &mut dyn Any;
+}
+
+pub mod kvm {
+    use super::GicDevice;
+    use super::Result;
+    use crate::aarch64::gic::gicv3_its::kvm::KvmGicV3Its;
+    use crate::layout;
+    use hypervisor::kvm::kvm_bindings;
+    use std::boxed::Box;
+    use std::sync::Arc;
+
+    /// Trait for GIC devices.
+    pub trait KvmGicDevice: Send + Sync + GicDevice {
+        /// Returns the GIC version of the device
+        fn version() -> u32;
+
+        /// Create the GIC device object
+        fn create_device(
+            device: Arc<dyn hypervisor::Device>,
+            vcpu_count: u64,
+        ) -> Box<dyn GicDevice>;
+
+        /// Setup the device-specific attributes
+        fn init_device_attributes(
+            vm: &Arc<dyn hypervisor::Vm>,
+            gic_device: &mut dyn GicDevice,
+        ) -> Result<()>;
+
+        /// Initialize a GIC device
+        fn init_device(vm: &Arc<dyn hypervisor::Vm>) -> Result<Arc<dyn hypervisor::Device>> {
+            let mut gic_device = kvm_bindings::kvm_create_device {
+                type_: Self::version(),
+                fd: 0,
+                flags: 0,
+            };
+
+            vm.create_device(&mut gic_device)
+                .map_err(super::Error::CreateGic)
+        }
+
+        /// Set a GIC device attribute
+        fn set_device_attribute(
+            device: &Arc<dyn hypervisor::Device>,
+            group: u32,
+            attr: u64,
+            addr: u64,
+            flags: u32,
+        ) -> Result<()> {
+            let attr = kvm_bindings::kvm_device_attr {
+                flags,
+                group,
+                attr,
+                addr,
+            };
+            device
+                .set_device_attr(&attr)
+                .map_err(super::Error::SetDeviceAttribute)?;
+
+            Ok(())
+        }
+
+        /// Get a GIC device attribute
+        fn get_device_attribute(
+            device: &Arc<dyn hypervisor::Device>,
+            group: u32,
+            attr: u64,
+            addr: u64,
+            flags: u32,
+        ) -> Result<()> {
+            let mut attr = kvm_bindings::kvm_device_attr {
+                flags,
+                group,
+                attr,
+                addr,
+            };
+            device
+                .get_device_attr(&mut attr)
+                .map_err(super::Error::GetDeviceAttribute)?;
+
+            Ok(())
+        }
+
+        /// Finalize the setup of a GIC device
+        fn finalize_device(gic_device: &dyn GicDevice) -> Result<()> {
+            /* We need to tell the kernel how many irqs to support with this vgic.
+             * See the `layout` module for details.
+             */
+            let nr_irqs: u32 = layout::IRQ_NUM;
+            let nr_irqs_ptr = &nr_irqs as *const u32;
+            Self::set_device_attribute(
+                gic_device.device(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_NR_IRQS,
+                0,
+                nr_irqs_ptr as u64,
+                0,
+            )?;
+
+            /* Finalize the GIC.
+             * See https://code.woboq.org/linux/linux/virt/kvm/arm/vgic/vgic-kvm-device.c.html#211.
+             */
+            Self::set_device_attribute(
+                gic_device.device(),
+                kvm_bindings::KVM_DEV_ARM_VGIC_GRP_CTRL,
+                u64::from(kvm_bindings::KVM_DEV_ARM_VGIC_CTRL_INIT),
+                0,
+                0,
+            )?;
+
+            Ok(())
+        }
+
+        /// Method to initialize the GIC device
+        #[allow(clippy::new_ret_no_self)]
+        fn new(vm: &Arc<dyn hypervisor::Vm>, vcpu_count: u64) -> Result<Box<dyn GicDevice>> {
+            let vgic_fd = Self::init_device(vm)?;
+
+            let mut device = Self::create_device(vgic_fd, vcpu_count);
+
+            Self::init_device_attributes(vm, &mut *device)?;
+
+            Self::finalize_device(&*device)?;
+
+            Ok(device)
+        }
+    }
+
+    /// Create a GICv3-ITS device.
+    ///
+    pub fn create_gic(vm: &Arc<dyn hypervisor::Vm>, vcpu_count: u64) -> Result<Box<dyn GicDevice>> {
+        debug!("creating a GICv3-ITS");
+        KvmGicV3Its::new(vm, vcpu_count)
+    }
+
+    /// Function that saves RDIST pending tables into guest RAM.
+    ///
+    /// The tables get flushed to guest RAM whenever the VM gets stopped.
+    pub fn save_pending_tables(gic: &Arc<dyn hypervisor::Device>) -> Result<()> {
+        let init_gic_attr = kvm_bindings::kvm_device_attr {
+            group: kvm_bindings::KVM_DEV_ARM_VGIC_GRP_CTRL,
+            attr: u64::from(kvm_bindings::KVM_DEV_ARM_VGIC_SAVE_PENDING_TABLES),
+            addr: 0,
+            flags: 0,
+        };
+        gic.set_device_attr(&init_gic_attr)
+            .map_err(super::Error::SetDeviceAttribute)
+    }
+}

arch/src/aarch64/gic/redist_regs.rs

@@ -1,18 +1,10 @@
-// Copyright 2022 Arm Limited (or its affiliates). All rights reserved.
 // Copyright 2020 Amazon.com, Inc. or its affiliates. All Rights Reserved.
 // SPDX-License-Identifier: Apache-2.0
 
-use crate::arch::aarch64::gic::{Error, Result};
-use crate::device::HypervisorDeviceError;
-use crate::kvm::kvm_bindings::{
-    kvm_device_attr, KVM_DEV_ARM_VGIC_GRP_REDIST_REGS, KVM_REG_ARM64, KVM_REG_ARM64_SYSREG,
-    KVM_REG_ARM64_SYSREG_OP0_MASK, KVM_REG_ARM64_SYSREG_OP0_SHIFT, KVM_REG_ARM64_SYSREG_OP2_MASK,
-    KVM_REG_ARM64_SYSREG_OP2_SHIFT, KVM_REG_SIZE_U64,
-};
-use crate::kvm::Register;
-use crate::kvm::VcpuKvmState;
-use crate::CpuState;
-use kvm_ioctls::DeviceFd;
+use super::{Error, Result};
+use hypervisor::kvm::kvm_bindings::{kvm_device_attr, KVM_DEV_ARM_VGIC_GRP_REDIST_REGS};
+use hypervisor::CpuState;
+use std::sync::Arc;
 
 // Relevant redistributor registers that we want to save/restore.
 const GICR_CTLR: u32 = 0x0000;
@@ -39,12 +31,6 @@ const GICR_ICFGR0: u32 = GICR_SGI_OFFSET + 0x0C00;
 const KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT: u32 = 32;
 const KVM_DEV_ARM_VGIC_V3_MPIDR_MASK: u64 = 0xffffffff << KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT as u64;
 
-const KVM_ARM64_SYSREG_MPIDR_EL1: u64 = KVM_REG_ARM64
-    | KVM_REG_SIZE_U64
-    | KVM_REG_ARM64_SYSREG as u64
-    | (((3_u64) << KVM_REG_ARM64_SYSREG_OP0_SHIFT) & KVM_REG_ARM64_SYSREG_OP0_MASK as u64)
-    | (((5_u64) << KVM_REG_ARM64_SYSREG_OP2_SHIFT) & KVM_REG_ARM64_SYSREG_OP2_MASK as u64);
-
 /// This is how we represent the registers of a distributor.
 /// It is relrvant their offset from the base address of the
 /// distributor.
@@ -96,27 +82,31 @@ static VGIC_SGI_REGS: &[RdistReg] = &[
     VGIC_RDIST_REG!(GICR_IPRIORITYR0, 32),
 ];
 
-fn redist_attr_access(gic: &DeviceFd, offset: u32, typer: u64, val: &u32, set: bool) -> Result<()> {
-    let mut gic_redist_attr = kvm_device_attr {
+fn redist_attr_access(
+    gic: &Arc<dyn hypervisor::Device>,
+    offset: u32,
+    typer: u64,
+    val: &u32,
+    set: bool,
+) -> Result<()> {
+    let mut gic_dist_attr = kvm_device_attr {
         group: KVM_DEV_ARM_VGIC_GRP_REDIST_REGS,
         attr: (typer & KVM_DEV_ARM_VGIC_V3_MPIDR_MASK) | (offset as u64), // this needs the mpidr
         addr: val as *const u32 as u64,
         flags: 0,
     };
     if set {
-        gic.set_device_attr(&gic_redist_attr).map_err(|e| {
-            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
-        })?;
+        gic.set_device_attr(&gic_dist_attr)
+            .map_err(Error::SetDeviceAttribute)?;
     } else {
-        gic.get_device_attr(&mut gic_redist_attr).map_err(|e| {
-            Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
-        })?;
+        gic.get_device_attr(&mut gic_dist_attr)
+            .map_err(Error::GetDeviceAttribute)?;
     }
     Ok(())
 }
 
 fn access_redists_aux(
-    gic: &DeviceFd,
+    gic: &Arc<dyn hypervisor::Device>,
     gicr_typer: &[u64],
     state: &mut Vec<u32>,
     reg_list: &[RdistReg],
@@ -146,7 +136,7 @@ fn access_redists_aux(
 }
 
 /// Get redistributor registers.
-pub fn get_redist_regs(gic: &DeviceFd, gicr_typer: &[u64]) -> Result<Vec<u32>> {
+pub fn get_redist_regs(gic: &Arc<dyn hypervisor::Device>, gicr_typer: &[u64]) -> Result<Vec<u32>> {
     let mut state = Vec::new();
     let mut idx: usize = 0;
     access_redists_aux(
@@ -163,7 +153,11 @@ pub fn get_redist_regs(gic: &DeviceFd, gicr_typer: &[u64]) -> Result<Vec<u32>> {
 }
 
 /// Set redistributor registers.
-pub fn set_redist_regs(gic: &DeviceFd, gicr_typer: &[u64], state: &[u32]) -> Result<()> {
+pub fn set_redist_regs(
+    gic: &Arc<dyn hypervisor::Device>,
+    gicr_typer: &[u64],
+    state: &[u32],
+) -> Result<()> {
     let mut idx: usize = 0;
     let mut mut_state = state.to_owned();
     access_redists_aux(
@@ -199,16 +193,15 @@ pub fn construct_gicr_typers(vcpu_states: &[CpuState]) -> Vec<u64> {
      */
     let mut gicr_typers: Vec<u64> = Vec::new();
     for (index, state) in vcpu_states.iter().enumerate() {
-        let state: VcpuKvmState = state.clone().into();
-        let last = (index == vcpu_states.len() - 1) as u64;
-        // state.sys_regs is a big collection of system registers, including MIPDR_EL1
-        let mpidr: Vec<Register> = state
-            .sys_regs
-            .into_iter()
-            .filter(|reg| reg.id == KVM_ARM64_SYSREG_MPIDR_EL1)
-            .collect();
+        let last = {
+            if index == vcpu_states.len() - 1 {
+                1
+            } else {
+                0
+            }
+        };
         //calculate affinity
-        let mut cpu_affid = mpidr[0].addr & 1095233437695;
+        let mut cpu_affid = state.mpidr & 1095233437695;
         cpu_affid = ((cpu_affid & 0xFF00000000) >> 8) | (cpu_affid & 0xFFFFFF);
         gicr_typers.push((cpu_affid << 32) | (1 << 24) | (index as u64) << 8 | (last << 4));
     }

arch/src/aarch64/layout.rs

@@ -3,7 +3,7 @@
 // SPDX-License-Identifier: Apache-2.0
 
 //
-// Memory layout of AArch64 guest:
+// Memory layout of Aarch64 guest:
 //
 // Physical  +---------------------------------------------------------------+
 // address   |                                                               |
@@ -20,13 +20,6 @@
 //           |                            DRAM                               |
 //           |                                                               |
 //           |                                                               |
-// 4GB       +---------------------------------------------------------------+
-//           |                      32-bit devices hole                      |
-// 4GB-64M   +---------------------------------------------------------------+
-//           |                                                               |
-//           |                                                               |
-//           |                            DRAM                               |
-//           |                                                               |
 //           |                                                               |
 // 1GB       +---------------------------------------------------------------+
 //           |                                                               |
@@ -42,11 +35,11 @@
 //           |                        Legacy devices space                   |
 //           |                                                               |
 // 144 M     +---------------------------------------------------------------|
-//           |                                                               |
 //           |                    Reserved (now GIC is here)                 |
+// 64  M     +---------------------------------------------------------------+
+//           |                                                               |
+//           |                          UEFI space                           |
 //           |                                                               |
-//  4  M     +---------------------------------------------------------------+
-//           |                          UEFI flash                           |
 // 0GB       +---------------------------------------------------------------+
 //
 //
@@ -55,16 +48,17 @@ use vm_memory::GuestAddress;
 
 /// 0x0 ~ 0x40_0000 (4 MiB) is reserved to UEFI
 /// UEFI binary size is required less than 3 MiB, reserving 4 MiB is enough.
-pub const UEFI_START: GuestAddress = GuestAddress(0);
+pub const UEFI_START: u64 = 0x0;
+pub const MEM_UEFI_START: GuestAddress = GuestAddress(0);
 pub const UEFI_SIZE: u64 = 0x040_0000;
 
 /// Below this address will reside the GIC, above this address will reside the MMIO devices.
-const MAPPED_IO_START: GuestAddress = GuestAddress(0x0900_0000);
+pub const MAPPED_IO_START: u64 = 0x0900_0000;
 
 /// See kernel file arch/arm64/include/uapi/asm/kvm.h for the GIC related definitions.
 /// 0x08ff_0000 ~ 0x0900_0000 is reserved for GICv3 Distributor
 pub const GIC_V3_DIST_SIZE: u64 = 0x01_0000;
-pub const GIC_V3_DIST_START: GuestAddress = GuestAddress(MAPPED_IO_START.0 - GIC_V3_DIST_SIZE);
+pub const GIC_V3_DIST_START: u64 = MAPPED_IO_START - GIC_V3_DIST_SIZE;
 /// Below 0x08ff_0000 is reserved for GICv3 Redistributor.
 /// The size defined here is for each vcpu.
 /// The total size is 'number_of_vcpu * GIC_V3_REDIST_SIZE'
@@ -73,9 +67,9 @@ pub const GIC_V3_REDIST_SIZE: u64 = 0x02_0000;
 pub const GIC_V3_ITS_SIZE: u64 = 0x02_0000;
 
 /// Space 0x0900_0000 ~ 0x0905_0000 is reserved for legacy devices.
-pub const LEGACY_SERIAL_MAPPED_IO_START: GuestAddress = MAPPED_IO_START;
-pub const LEGACY_RTC_MAPPED_IO_START: GuestAddress = GuestAddress(0x0901_0000);
-pub const LEGACY_GPIO_MAPPED_IO_START: GuestAddress = GuestAddress(0x0902_0000);
+pub const LEGACY_SERIAL_MAPPED_IO_START: u64 = 0x0900_0000;
+pub const LEGACY_RTC_MAPPED_IO_START: u64 = 0x0901_0000;
+pub const LEGACY_GPIO_MAPPED_IO_START: u64 = 0x0902_0000;
 
 /// Space 0x0905_0000 ~ 0x0906_0000 is reserved for pcie io address
 pub const MEM_PCI_IO_START: GuestAddress = GuestAddress(0x0905_0000);
@@ -91,20 +85,8 @@ pub const PCI_MMCONFIG_SIZE: u64 = 256 << 20;
 // One bus with potentially 256 devices (32 slots x 8 functions).
 pub const PCI_MMIO_CONFIG_SIZE_PER_SEGMENT: u64 = 4096 * 256;
 
-/// Start of RAM.
-pub const RAM_START: GuestAddress = GuestAddress(0x4000_0000);
-
-/// 32-bit reserved area: 64MiB before 4GiB
-pub const MEM_32BIT_RESERVED_START: GuestAddress = GuestAddress(0xfc00_0000);
-pub const MEM_32BIT_RESERVED_SIZE: u64 = 0x0400_0000;
-
-/// TPM Address Range
-/// This Address range is specific to CRB Interface
-pub const TPM_START: GuestAddress = GuestAddress(0xfed4_0000);
-pub const TPM_SIZE: u64 = 0x1000;
-
-/// Start of 64-bit RAM.
-pub const RAM_64BIT_START: GuestAddress = GuestAddress(0x1_0000_0000);
+/// Start of RAM on 64 bit ARM.
+pub const RAM_64BIT_START: u64 = 0x4000_0000;
 
 /// Kernel command line maximum size.
 /// As per `arch/arm64/include/uapi/asm/setup.h`.
@@ -112,19 +94,19 @@ pub const CMDLINE_MAX_SIZE: usize = 2048;
 
 /// FDT is at the beginning of RAM.
 /// Maximum size of the device tree blob as specified in https://www.kernel.org/doc/Documentation/arm64/booting.txt.
-pub const FDT_START: GuestAddress = RAM_START;
-pub const FDT_MAX_SIZE: u64 = 0x20_0000;
+pub const FDT_START: u64 = RAM_64BIT_START;
+pub const FDT_MAX_SIZE: usize = 0x20_0000;
 
 /// Put ACPI table above dtb
-pub const ACPI_START: GuestAddress = GuestAddress(RAM_START.0 + FDT_MAX_SIZE);
-pub const ACPI_MAX_SIZE: u64 = 0x20_0000;
-pub const RSDP_POINTER: GuestAddress = ACPI_START;
+pub const ACPI_START: u64 = RAM_64BIT_START + FDT_MAX_SIZE as u64;
+pub const ACPI_MAX_SIZE: usize = 0x20_0000;
+pub const RSDP_POINTER: GuestAddress = GuestAddress(ACPI_START);
 
 /// Kernel start after FDT and ACPI
-pub const KERNEL_START: GuestAddress = GuestAddress(ACPI_START.0 + ACPI_MAX_SIZE);
+pub const KERNEL_START: u64 = ACPI_START + ACPI_MAX_SIZE as u64;
 
 /// Pci high memory base
-pub const PCI_HIGH_BASE: GuestAddress = GuestAddress(0x2_0000_0000);
+pub const PCI_HIGH_BASE: u64 = 0x2_0000_0000_u64;
 
 // As per virt/kvm/arm/vgic/vgic-kvm-device.c we need
 // the number of interrupts our GIC will support to be:

arch/src/aarch64/mod.rs

@@ -4,24 +4,24 @@
 
 /// Module for the flattened device tree.
 pub mod fdt;
+/// Module for the global interrupt controller configuration.
+pub mod gic;
 /// Layout for this aarch64 system.
 pub mod layout;
-/// Module for system registers definition
+/// Logic for configuring aarch64 registers.
 pub mod regs;
 /// Module for loading UEFI binary.
 pub mod uefi;
 
 pub use self::fdt::DeviceInfoForFdt;
 use crate::{DeviceType, GuestMemoryMmap, NumaNodes, PciSpaceInfo, RegionType};
-use hypervisor::arch::aarch64::gic::Vgic;
+use gic::GicDevice;
 use log::{log_enabled, Level};
 use std::collections::HashMap;
 use std::convert::TryInto;
 use std::fmt::Debug;
-use std::sync::{Arc, Mutex};
-use vm_memory::{Address, GuestAddress, GuestMemory, GuestMemoryAtomic};
-
-pub const _NSIG: i32 = 65;
+use std::sync::Arc;
+use vm_memory::{Address, GuestAddress, GuestMemory, GuestUsize};
 
 /// Errors thrown while configuring aarch64 system.
 #[derive(Debug)]
@@ -33,13 +33,13 @@ pub enum Error {
     WriteFdtToMemory(fdt::Error),
 
     /// Failed to create a GIC.
-    SetupGic,
+    SetupGic(gic::Error),
 
     /// Failed to compute the initramfs address.
     InitramfsAddress,
 
     /// Error configuring the general purpose registers
-    RegsConfiguration(hypervisor::HypervisorCpuError),
+    RegsConfiguration(regs::Error),
 
     /// Error configuring the MPIDR register
     VcpuRegMpidr(hypervisor::HypervisorCpuError),
@@ -50,7 +50,7 @@ pub enum Error {
 
 impl From<Error> for super::Error {
     fn from(e: Error) -> super::Error {
-        super::Error::PlatformSpecific(e)
+        super::Error::AArch64Setup(e)
     }
 }
 
@@ -64,31 +64,45 @@ pub struct EntryPoint {
 
 /// Configure the specified VCPU, and return its MPIDR.
 pub fn configure_vcpu(
-    vcpu: &Arc<dyn hypervisor::Vcpu>,
+    fd: &Arc<dyn hypervisor::Vcpu>,
     id: u8,
-    boot_setup: Option<(EntryPoint, &GuestMemoryAtomic<GuestMemoryMmap>)>,
+    kernel_entry_point: Option<EntryPoint>,
 ) -> super::Result<u64> {
-    if let Some((kernel_entry_point, _guest_memory)) = boot_setup {
-        vcpu.setup_regs(
-            id,
-            kernel_entry_point.entry_addr.raw_value(),
-            super::layout::FDT_START.raw_value(),
-        )
-        .map_err(Error::RegsConfiguration)?;
+    if let Some(kernel_entry_point) = kernel_entry_point {
+        regs::setup_regs(fd, id, kernel_entry_point.entry_addr.raw_value())
+            .map_err(Error::RegsConfiguration)?;
     }
 
-    let mpidr = vcpu
-        .get_sys_reg(regs::MPIDR_EL1)
-        .map_err(Error::VcpuRegMpidr)?;
+    let mpidr = fd.read_mpidr().map_err(Error::VcpuRegMpidr)?;
     Ok(mpidr)
 }
 
-pub fn arch_memory_regions() -> Vec<(GuestAddress, usize, RegionType)> {
+pub fn arch_memory_regions(size: GuestUsize) -> Vec<(GuestAddress, usize, RegionType)> {
+    // Normally UEFI should be loaded to a flash area at the beginning of memory.
+    // But now flash memory type is not supported.
+    // As a workaround, we take 4 MiB memory from the main RAM for UEFI.
+    // As a result, the RAM that the guest can see is less than what has been
+    // assigned in command line, when ACPI and UEFI is enabled.
+    let ram_deduction = if cfg!(feature = "acpi") {
+        layout::UEFI_SIZE
+    } else {
+        0
+    };
+
     vec![
-        // 0 MiB ~ 256 MiB: UEFI, GIC and legacy devices
+        // 0 ~ 4 MiB: Reserved for UEFI space
+        #[cfg(feature = "acpi")]
+        (GuestAddress(0), layout::UEFI_SIZE as usize, RegionType::Ram),
+        #[cfg(not(feature = "acpi"))]
         (
             GuestAddress(0),
-            layout::MEM_32BIT_DEVICES_START.0 as usize,
+            layout::UEFI_SIZE as usize,
+            RegionType::Reserved,
+        ),
+        // 4 MiB ~ 256 MiB: Gic and legacy devices
+        (
+            GuestAddress(layout::UEFI_SIZE),
+            (layout::MEM_32BIT_DEVICES_START.0 - layout::UEFI_SIZE) as usize,
             RegionType::Reserved,
         ),
         // 256 MiB ~ 768 MiB: MMIO space
@@ -103,20 +117,12 @@ pub fn arch_memory_regions() -> Vec<(GuestAddress, usize, RegionType)> {
             layout::PCI_MMCONFIG_SIZE as usize,
             RegionType::Reserved,
         ),
-        // 1GiB ~ 4032 MiB: RAM before the gap
+        // 1 GiB ~ : Ram
         (
-            layout::RAM_START,
-            layout::MEM_32BIT_RESERVED_START.unchecked_offset_from(layout::RAM_START) as usize,
+            GuestAddress(layout::RAM_64BIT_START),
+            (size - ram_deduction) as usize,
             RegionType::Ram,
         ),
-        // 4GiB ~ inf: RAM after the gap
-        (layout::RAM_64BIT_START, usize::MAX, RegionType::Ram),
-        // Add the 32-bit reserved memory hole as a reserved region
-        (
-            layout::MEM_32BIT_RESERVED_START,
-            layout::MEM_32BIT_RESERVED_SIZE as usize,
-            RegionType::Reserved,
-        ),
     ]
 }
 
@@ -131,7 +137,7 @@ pub fn configure_system<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::Bui
     initrd: &Option<super::InitramfsConfig>,
     pci_space_info: &[PciSpaceInfo],
     virtio_iommu_bdf: Option<u32>,
-    gic_device: &Arc<Mutex<dyn Vgic>>,
+    gic_device: &dyn GicDevice,
     numa_nodes: &NumaNodes,
     pmu_supported: bool,
 ) -> super::Result<()> {
@@ -173,15 +179,33 @@ pub fn initramfs_load_addr(
             if guest_mem.address_in_range(offset) {
                 Ok(offset.raw_value())
             } else {
-                Err(super::Error::PlatformSpecific(Error::InitramfsAddress))
+                Err(super::Error::AArch64Setup(Error::InitramfsAddress))
             }
         }
-        None => Err(super::Error::PlatformSpecific(Error::InitramfsAddress)),
+        None => Err(super::Error::AArch64Setup(Error::InitramfsAddress)),
     }
 }
 
-pub fn get_host_cpu_phys_bits(hypervisor: &Arc<dyn hypervisor::Hypervisor>) -> u8 {
-    let host_cpu_phys_bits = hypervisor.get_host_ipa_limit().try_into().unwrap();
+/// Returns the memory address where the kernel could be loaded.
+pub fn get_kernel_start() -> u64 {
+    layout::KERNEL_START
+}
+
+///Return guest memory address where the uefi should be loaded.
+pub fn get_uefi_start() -> u64 {
+    layout::UEFI_START
+}
+
+// Auxiliary function to get the address where the device tree blob is loaded.
+fn get_fdt_addr() -> u64 {
+    layout::FDT_START
+}
+
+pub fn get_host_cpu_phys_bits() -> u8 {
+    // A dummy hypervisor created only for querying the host IPA size and will
+    // be freed after the query.
+    let hv = hypervisor::new().unwrap();
+    let host_cpu_phys_bits = hv.get_host_ipa_limit().try_into().unwrap();
     if host_cpu_phys_bits == 0 {
         // Host kernel does not support `get_host_ipa_limit`,
         // we return the default value 40 here.
@@ -197,11 +221,10 @@ mod tests {
 
     #[test]
     fn test_arch_memory_regions_dram() {
-        let regions = arch_memory_regions();
-        assert_eq!(6, regions.len());
-        assert_eq!(layout::RAM_START, regions[3].0);
-        assert_eq!(RegionType::Ram, regions[3].2);
-        assert_eq!(RegionType::Reserved, regions[5].2);
+        let regions = arch_memory_regions((1usize << 32) as u64); //4GB
+        assert_eq!(5, regions.len());
+        assert_eq!(GuestAddress(layout::RAM_64BIT_START), regions[4].0);
+        assert_eq!(1usize << 32, regions[4].1);
         assert_eq!(RegionType::Ram, regions[4].2);
     }
 }

arch/src/aarch64/regs.rs

@@ -1,43 +1,75 @@
-// Copyright 2022 Arm Limited (or its affiliates). All rights reserved.
+// Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
 // SPDX-License-Identifier: Apache-2.0
-
-// AArch64 system register encoding:
-// See https://developer.arm.com/documentation/ddi0487 (chapter D12)
 //
-//   31      22  21 20 19 18 16 15 12 11  8 7   5 4  0
-//  +----------+---+-----+-----+-----+-----+-----+----+
-//  |1101010100| L | op0 | op1 | CRn | CRm | op2 | Rt |
-//  +----------+---+-----+-----+-----+-----+-----+----+
-//
-// Notes:
-// - L and Rt are reserved as implementation defined fields, ignored.
+// Portions Copyright 2017 The Chromium OS Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the THIRD-PARTY file.
 
-const SYSREG_HEAD: u32 = 0b1101010100u32 << 22;
-const SYSREG_OP0_SHIFT: u32 = 19;
-const SYSREG_OP0_MASK: u32 = 0b11u32 << 19;
-const SYSREG_OP1_SHIFT: u32 = 16;
-const SYSREG_OP1_MASK: u32 = 0b111u32 << 16;
-const SYSREG_CRN_SHIFT: u32 = 12;
-const SYSREG_CRN_MASK: u32 = 0b1111u32 << 12;
-const SYSREG_CRM_SHIFT: u32 = 8;
-const SYSREG_CRM_MASK: u32 = 0b1111u32 << 8;
-const SYSREG_OP2_SHIFT: u32 = 5;
-const SYSREG_OP2_MASK: u32 = 0b111u32 << 5;
+use super::get_fdt_addr;
+use hypervisor::kvm::kvm_bindings::{
+    kvm_regs, user_pt_regs, KVM_REG_ARM64, KVM_REG_ARM_CORE, KVM_REG_SIZE_U64,
+};
+use hypervisor::{arm64_core_reg_id, offset__of};
+use std::sync::Arc;
+use std::{mem, result};
 
-/// Define the ID of system registers
-#[macro_export]
-macro_rules! arm64_sys_reg {
-    ($name: tt, $op0: tt, $op1: tt, $crn: tt, $crm: tt, $op2: tt) => {
-        pub const $name: u32 = SYSREG_HEAD
-            | ((($op0 as u32) << SYSREG_OP0_SHIFT) & SYSREG_OP0_MASK as u32)
-            | ((($op1 as u32) << SYSREG_OP1_SHIFT) & SYSREG_OP1_MASK as u32)
-            | ((($crn as u32) << SYSREG_CRN_SHIFT) & SYSREG_CRN_MASK as u32)
-            | ((($crm as u32) << SYSREG_CRM_SHIFT) & SYSREG_CRM_MASK as u32)
-            | ((($op2 as u32) << SYSREG_OP2_SHIFT) & SYSREG_OP2_MASK as u32);
-    };
+/// Errors thrown while setting aarch64 registers.
+#[derive(Debug)]
+pub enum Error {
+    /// Failed to set core register (PC, PSTATE or general purpose ones).
+    SetCoreRegister(hypervisor::HypervisorCpuError),
+    /// Failed to get a system register.
+    GetSysRegister(hypervisor::HypervisorCpuError),
 }
+type Result<T> = result::Result<T, Error>;
 
-arm64_sys_reg!(MPIDR_EL1, 3, 0, 0, 0, 5);
-arm64_sys_reg!(ID_AA64MMFR0_EL1, 3, 0, 0, 7, 0);
-arm64_sys_reg!(TTBR1_EL1, 3, 0, 2, 0, 1);
-arm64_sys_reg!(TCR_EL1, 3, 0, 2, 0, 2);
+#[allow(non_upper_case_globals)]
+// PSR (Processor State Register) bits.
+// Taken from arch/arm64/include/uapi/asm/ptrace.h.
+const PSR_MODE_EL1h: u64 = 0x0000_0005;
+const PSR_F_BIT: u64 = 0x0000_0040;
+const PSR_I_BIT: u64 = 0x0000_0080;
+const PSR_A_BIT: u64 = 0x0000_0100;
+const PSR_D_BIT: u64 = 0x0000_0200;
+// Taken from arch/arm64/kvm/inject_fault.c.
+const PSTATE_FAULT_BITS_64: u64 = PSR_MODE_EL1h | PSR_A_BIT | PSR_F_BIT | PSR_I_BIT | PSR_D_BIT;
+
+/// Configure core registers for a given CPU.
+///
+/// # Arguments
+///
+/// * `vcpu` - Structure for the VCPU that holds the VCPU's fd.
+/// * `cpu_id` - Index of current vcpu.
+/// * `boot_ip` - Starting instruction pointer.
+/// * `mem` - Reserved DRAM for current VM.
+pub fn setup_regs(vcpu: &Arc<dyn hypervisor::Vcpu>, cpu_id: u8, boot_ip: u64) -> Result<()> {
+    let kreg_off = offset__of!(kvm_regs, regs);
+
+    // Get the register index of the PSTATE (Processor State) register.
+    let pstate = offset__of!(user_pt_regs, pstate) + kreg_off;
+    vcpu.set_reg(
+        arm64_core_reg_id!(KVM_REG_SIZE_U64, pstate),
+        PSTATE_FAULT_BITS_64,
+    )
+    .map_err(Error::SetCoreRegister)?;
+
+    // Other vCPUs are powered off initially awaiting PSCI wakeup.
+    if cpu_id == 0 {
+        // Setting the PC (Processor Counter) to the current program address (kernel address).
+        let pc = offset__of!(user_pt_regs, pc) + kreg_off;
+        vcpu.set_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, pc), boot_ip as u64)
+            .map_err(Error::SetCoreRegister)?;
+
+        // Last mandatory thing to set -> the address pointing to the FDT (also called DTB).
+        // "The device tree blob (dtb) must be placed on an 8-byte boundary and must
+        // not exceed 2 megabytes in size." -> https://www.kernel.org/doc/Documentation/arm64/booting.txt.
+        // We are choosing to place it the end of DRAM. See `get_fdt_addr`.
+        let regs0 = offset__of!(user_pt_regs, regs) + kreg_off;
+        vcpu.set_reg(
+            arm64_core_reg_id!(KVM_REG_SIZE_U64, regs0),
+            get_fdt_addr() as u64,
+        )
+        .map_err(Error::SetCoreRegister)?;
+    }
+    Ok(())
+}

arch/src/aarch64/uefi.rs

@@ -34,7 +34,9 @@ where
     if uefi_size > 0x300000 {
         return Err(Error::UefiTooBig);
     }
-    uefi_image.rewind().map_err(|_| Error::SeekUefiStart)?;
+    uefi_image
+        .seek(SeekFrom::Start(0))
+        .map_err(|_| Error::SeekUefiStart)?;
     guest_mem
         .read_exact_from(guest_addr, uefi_image, uefi_size)
         .map_err(|_| Error::ReadUefiImage)

arch/src/lib.rs

@@ -9,15 +9,15 @@
 
 #[macro_use]
 extern crate log;
+#[macro_use]
+extern crate serde_derive;
 
 #[cfg(target_arch = "x86_64")]
 use crate::x86_64::SgxEpcSection;
-use serde::{Deserialize, Serialize};
 use std::collections::BTreeMap;
 use std::fmt;
 use std::result;
 use std::sync::Arc;
-use thiserror::Error;
 use versionize::{VersionMap, Versionize, VersionizeError, VersionizeResult};
 use versionize_derive::Versionize;
 use vm_migration::VersionMapped;
@@ -26,27 +26,31 @@ type GuestMemoryMmap = vm_memory::GuestMemoryMmap<vm_memory::bitmap::AtomicBitma
 type GuestRegionMmap = vm_memory::GuestRegionMmap<vm_memory::bitmap::AtomicBitmap>;
 
 /// Type for returning error code.
-#[derive(Debug, Error)]
+#[derive(Debug)]
 pub enum Error {
     #[cfg(target_arch = "x86_64")]
-    #[error("Platform specific error (x86_64): {0:?}")]
-    PlatformSpecific(x86_64::Error),
+    /// X86_64 specific error triggered during system configuration.
+    X86_64Setup(x86_64::Error),
     #[cfg(target_arch = "aarch64")]
-    #[error("Platform specific error (aarch64): {0:?}")]
-    PlatformSpecific(aarch64::Error),
-    #[error("The memory map table extends past the end of guest memory")]
+    /// AArch64 specific error triggered during system configuration.
+    AArch64Setup(aarch64::Error),
+    /// The zero page extends past the end of guest_mem.
+    ZeroPagePastRamEnd,
+    /// Error writing the zero page of guest memory.
+    ZeroPageSetup(vm_memory::GuestMemoryError),
+    /// The memory map table extends past the end of guest memory.
     MemmapTablePastRamEnd,
-    #[error("Error writing memory map table to guest memory")]
+    /// Error writing memory map table to guest memory.
     MemmapTableSetup,
-    #[error("The hvm_start_info structure extends past the end of guest memory")]
+    /// The hvm_start_info structure extends past the end of guest memory.
     StartInfoPastRamEnd,
-    #[error("Error writing hvm_start_info to guest memory")]
+    /// Error writing hvm_start_info to guest memory.
     StartInfoSetup,
-    #[error("Failed to compute initramfs address")]
+    /// Failed to compute initramfs address.
     InitramfsAddress,
-    #[error("Error writing module entry to guest memory: {0}")]
-    ModlistSetup(#[source] vm_memory::GuestMemoryError),
-    #[error("RSDP extends past the end of guest memory")]
+    /// Error writing module entry to guest memory.
+    ModlistSetup(vm_memory::GuestMemoryError),
+    /// RSDP Beyond Guest Memory
     RsdpPastRamEnd,
 }
 
@@ -54,7 +58,7 @@ pub enum Error {
 pub type Result<T> = result::Result<T, Error>;
 
 /// Type for memory region types.
-#[derive(Clone, Copy, PartialEq, Eq, Debug, Serialize, Deserialize, Versionize)]
+#[derive(Clone, Copy, PartialEq, Debug, Serialize, Deserialize, Versionize)]
 pub enum RegionType {
     /// RAM type
     Ram,
@@ -81,8 +85,8 @@ pub mod aarch64;
 #[cfg(target_arch = "aarch64")]
 pub use aarch64::{
     arch_memory_regions, configure_system, configure_vcpu, fdt::DeviceInfoForFdt,
-    get_host_cpu_phys_bits, initramfs_load_addr, layout, layout::CMDLINE_MAX_SIZE,
-    layout::IRQ_BASE, uefi, EntryPoint, _NSIG,
+    get_host_cpu_phys_bits, get_kernel_start, get_uefi_start, initramfs_load_addr, layout,
+    layout::CMDLINE_MAX_SIZE, layout::IRQ_BASE, uefi, EntryPoint,
 };
 
 #[cfg(target_arch = "x86_64")]
@@ -92,14 +96,14 @@ pub mod x86_64;
 pub use x86_64::{
     arch_memory_regions, configure_system, configure_vcpu, generate_common_cpuid,
     get_host_cpu_phys_bits, initramfs_load_addr, layout, layout::CMDLINE_MAX_SIZE,
-    layout::CMDLINE_START, regs, CpuidFeatureEntry, EntryPoint, _NSIG,
+    layout::CMDLINE_START, regs, CpuidFeatureEntry, EntryPoint,
 };
 
 /// Safe wrapper for `sysconf(_SC_PAGESIZE)`.
 #[cfg(target_arch = "x86_64")]
 #[inline(always)]
 fn pagesize() -> usize {
-    // SAFETY: Trivially safe
+    // Trivially safe
     unsafe { libc::sysconf(libc::_SC_PAGESIZE) as usize }
 }
 
@@ -145,7 +149,7 @@ pub const PAGE_SIZE: usize = 4096;
 
 impl fmt::Display for DeviceType {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        write!(f, "{self:?}")
+        write!(f, "{:?}", self)
     }
 }
 
@@ -154,7 +158,6 @@ impl fmt::Display for DeviceType {
 #[cfg(target_arch = "aarch64")]
 pub struct MmioDeviceInfo {
     pub addr: u64,
-    pub len: u64,
     pub irq: u32,
 }
 
@@ -177,6 +180,6 @@ impl DeviceInfoForFdt for MmioDeviceInfo {
         self.irq
     }
     fn length(&self) -> u64 {
-        self.len
+        4096
     }
 }

arch/src/x86_64/interrupts.rs

@@ -5,6 +5,10 @@
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE-BSD-3-Clause file.
 
+use byteorder::{LittleEndian, ReadBytesExt, WriteBytesExt};
+use hypervisor::x86_64::LapicState;
+use std::io::Cursor;
+use std::mem;
 use std::result;
 use std::sync::Arc;
 
@@ -16,6 +20,32 @@ pub const APIC_LVT1: usize = 0x360;
 pub const APIC_MODE_NMI: u32 = 0x4;
 pub const APIC_MODE_EXTINT: u32 = 0x7;
 
+pub fn get_klapic_reg(klapic: &LapicState, reg_offset: usize) -> u32 {
+    let sliceu8 = unsafe {
+        // This array is only accessed as parts of a u32 word, so interpret it as a u8 array.
+        // Cursors are only readable on arrays of u8, not i8(c_char).
+        mem::transmute::<&[i8], &[u8]>(&klapic.regs[reg_offset..])
+    };
+    let mut reader = Cursor::new(sliceu8);
+    // Following call can't fail if the offsets defined above are correct.
+    reader
+        .read_u32::<LittleEndian>()
+        .expect("Failed to read klapic register")
+}
+
+pub fn set_klapic_reg(klapic: &mut LapicState, reg_offset: usize, value: u32) {
+    let sliceu8 = unsafe {
+        // This array is only accessed as parts of a u32 word, so interpret it as a u8 array.
+        // Cursors are only readable on arrays of u8, not i8(c_char).
+        mem::transmute::<&mut [i8], &mut [u8]>(&mut klapic.regs[reg_offset..])
+    };
+    let mut writer = Cursor::new(sliceu8);
+    // Following call can't fail if the offsets defined above are correct.
+    writer
+        .write_u32::<LittleEndian>(value)
+        .expect("Failed to write klapic register")
+}
+
 pub fn set_apic_delivery_mode(reg: u32, mode: u32) -> u32 {
     ((reg) & !0x700) | ((mode) << 8)
 }
@@ -27,13 +57,42 @@ pub fn set_apic_delivery_mode(reg: u32, mode: u32) -> u32 {
 pub fn set_lint(vcpu: &Arc<dyn hypervisor::Vcpu>) -> Result<()> {
     let mut klapic = vcpu.get_lapic()?;
 
-    let lvt_lint0 = klapic.get_klapic_reg(APIC_LVT0);
-    klapic.set_klapic_reg(
+    let lvt_lint0 = get_klapic_reg(&klapic, APIC_LVT0);
+    set_klapic_reg(
+        &mut klapic,
         APIC_LVT0,
         set_apic_delivery_mode(lvt_lint0, APIC_MODE_EXTINT),
     );
-    let lvt_lint1 = klapic.get_klapic_reg(APIC_LVT1);
-    klapic.set_klapic_reg(APIC_LVT1, set_apic_delivery_mode(lvt_lint1, APIC_MODE_NMI));
+    let lvt_lint1 = get_klapic_reg(&klapic, APIC_LVT1);
+    set_klapic_reg(
+        &mut klapic,
+        APIC_LVT1,
+        set_apic_delivery_mode(lvt_lint1, APIC_MODE_NMI),
+    );
 
     vcpu.set_lapic(&klapic)
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    const KVM_APIC_REG_SIZE: usize = 0x400;
+
+    #[test]
+    fn test_set_and_get_klapic_reg() {
+        let reg_offset = 0x340;
+        let mut klapic = LapicState::default();
+        set_klapic_reg(&mut klapic, reg_offset, 3);
+        let value = get_klapic_reg(&klapic, reg_offset);
+        assert_eq!(value, 3);
+    }
+
+    #[test]
+    #[should_panic]
+    fn test_set_and_get_klapic_out_of_bounds() {
+        let reg_offset = KVM_APIC_REG_SIZE + 10;
+        let mut klapic = LapicState::default();
+        set_klapic_reg(&mut klapic, reg_offset, 3);
+    }
+}

arch/src/x86_64/layout.rs

@@ -24,9 +24,6 @@ pub const LOW_RAM_START: GuestAddress = GuestAddress(0x0);
 
 // == Fixed addresses within the "Low RAM" range: ==
 
-// Location of EBDA address
-pub const EBDA_POINTER: GuestAddress = GuestAddress(0x40e);
-
 // Initial GDT/IDT needed to boot kernel
 pub const BOOT_GDT_START: GuestAddress = GuestAddress(0x500);
 pub const BOOT_IDT_START: GuestAddress = GuestAddress(0x520);
@@ -107,11 +104,6 @@ pub const KVM_TSS_SIZE: u64 = (3 * 4) << 10;
 pub const KVM_IDENTITY_MAP_START: GuestAddress = GuestAddress(KVM_TSS_START.0 + KVM_TSS_SIZE);
 pub const KVM_IDENTITY_MAP_SIZE: u64 = 4 << 10;
 
-/// TPM Address Range
-/// This Address range is specific to CRB Interface
-pub const TPM_START: GuestAddress = GuestAddress(0xfed4_0000);
-pub const TPM_SIZE: u64 = 0x1000;
-
 // IOAPIC
 pub const IOAPIC_START: GuestAddress = GuestAddress(0xfec0_0000);
 pub const IOAPIC_SIZE: u64 = 0x20;

arch/src/x86_64/mod.rs

@@ -15,8 +15,7 @@ pub mod regs;
 use crate::GuestMemoryMmap;
 use crate::InitramfsConfig;
 use crate::RegionType;
-use hypervisor::arch::x86::{CpuIdEntry, CPUID_FLAG_VALID_INDEX};
-use hypervisor::{CpuVendor, HypervisorCpuError, HypervisorError};
+use hypervisor::{CpuId, CpuIdEntry, HypervisorError, CPUID_FLAG_VALID_INDEX};
 use linux_loader::loader::bootparam::boot_params;
 use linux_loader::loader::elf::start_info::{
     hvm_memmap_table_entry, hvm_modlist_entry, hvm_start_info,
@@ -36,9 +35,9 @@ pub mod tdx;
 const TSC_DEADLINE_TIMER_ECX_BIT: u8 = 24; // tsc deadline timer ecx bit.
 const HYPERVISOR_ECX_BIT: u8 = 31; // Hypervisor ecx bit.
 const MTRR_EDX_BIT: u8 = 12; // Hypervisor ecx bit.
-const INVARIANT_TSC_EDX_BIT: u8 = 8; // Invariant TSC bit on 0x8000_0007 EDX
 
 // KVM feature bits
+const KVM_FEATURE_ASYNC_PF_INT_BIT: u8 = 14;
 #[cfg(feature = "tdx")]
 const KVM_FEATURE_CLOCKSOURCE_BIT: u8 = 0;
 #[cfg(feature = "tdx")]
@@ -52,8 +51,6 @@ const KVM_FEATURE_ASYNC_PF_VMEXIT_BIT: u8 = 10;
 #[cfg(feature = "tdx")]
 const KVM_FEATURE_STEAL_TIME_BIT: u8 = 5;
 
-pub const _NSIG: i32 = 65;
-
 #[derive(Debug, Copy, Clone)]
 /// Specifies the entry point address where the guest must start
 /// executing code, as well as which of the supported boot protocols
@@ -127,11 +124,9 @@ struct MemmapTableEntryWrapper(hvm_memmap_table_entry);
 #[derive(Copy, Clone, Default)]
 struct ModlistEntryWrapper(hvm_modlist_entry);
 
-// SAFETY: data structure only contain a series of integers
+// SAFETY: These data structures only contain a series of integers
 unsafe impl ByteValued for StartInfoWrapper {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for MemmapTableEntryWrapper {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for ModlistEntryWrapper {}
 
 // This is a workaround to the Rust enforcement specifying that any implementation of a foreign
@@ -191,24 +186,15 @@ pub enum Error {
 
     /// Error checking CPUID compatibility
     CpuidCheckCompatibility,
-
-    // Error writing EBDA address
-    EbdaSetup(vm_memory::GuestMemoryError),
-
-    // Error getting CPU TSC frequency
-    GetTscFrequency(HypervisorCpuError),
-
-    /// Error retrieving TDX capabilities through the hypervisor (kvm/mshv) API
-    #[cfg(feature = "tdx")]
-    TdxCapabilities(HypervisorError),
 }
 
 impl From<Error> for super::Error {
     fn from(e: Error) -> super::Error {
-        super::Error::PlatformSpecific(e)
+        super::Error::X86_64Setup(e)
     }
 }
 
+#[allow(dead_code, clippy::upper_case_acronyms)]
 #[derive(Copy, Clone, Debug)]
 pub enum CpuidReg {
     EAX,
@@ -229,15 +215,17 @@ pub struct CpuidPatch {
 
 impl CpuidPatch {
     pub fn set_cpuid_reg(
-        cpuid: &mut Vec<CpuIdEntry>,
+        cpuid: &mut CpuId,
         function: u32,
         index: Option<u32>,
         reg: CpuidReg,
         value: u32,
     ) {
+        let entries = cpuid.as_mut_slice();
+
         let mut entry_found = false;
-        for entry in cpuid.iter_mut() {
-            if entry.function == function && (index.is_none() || index.unwrap() == entry.index) {
+        for entry in entries.iter_mut() {
+            if entry.function == function && (index == None || index.unwrap() == entry.index) {
                 entry_found = true;
                 match reg {
                     CpuidReg::EAX => {
@@ -283,12 +271,16 @@ impl CpuidPatch {
                 }
             }
 
-            cpuid.push(entry);
+            if let Err(e) = cpuid.push(entry) {
+                error!("Failed adding new CPUID entry: {:?}", e);
+            }
         }
     }
 
-    pub fn patch_cpuid(cpuid: &mut [CpuIdEntry], patches: Vec<CpuidPatch>) {
-        for entry in cpuid {
+    pub fn patch_cpuid(cpuid: &mut CpuId, patches: Vec<CpuidPatch>) {
+        let entries = cpuid.as_mut_slice();
+
+        for entry in entries.iter_mut() {
             for patch in patches.iter() {
                 if entry.function == patch.function && entry.index == patch.index {
                     if let Some(flags_bit) = patch.flags_bit {
@@ -312,15 +304,16 @@ impl CpuidPatch {
     }
 
     pub fn is_feature_enabled(
-        cpuid: &[CpuIdEntry],
+        cpuid: &CpuId,
         function: u32,
         index: u32,
         reg: CpuidReg,
         feature_bit: usize,
     ) -> bool {
+        let entries = cpuid.as_slice();
         let mask = 1 << feature_bit;
 
-        for entry in cpuid {
+        for entry in entries.iter() {
             if entry.function == function && entry.index == index {
                 let reg_val = match reg {
                     CpuidReg::EAX => entry.eax,
@@ -469,12 +462,12 @@ impl CpuidFeatureEntry {
     }
 
     fn get_features_from_cpuid(
-        cpuid: &[CpuIdEntry],
+        cpuid: &CpuId,
         feature_entry_list: &[CpuidFeatureEntry],
     ) -> Vec<u32> {
         let mut features = vec![0; feature_entry_list.len()];
         for (i, feature_entry) in feature_entry_list.iter().enumerate() {
-            for cpuid_entry in cpuid {
+            for cpuid_entry in cpuid.as_slice().iter() {
                 if cpuid_entry.function == feature_entry.function
                     && cpuid_entry.index == feature_entry.index
                 {
@@ -504,8 +497,8 @@ impl CpuidFeatureEntry {
     // The function returns `Error` (a.k.a. "incompatible"), when the CPUID features from `src_vm_cpuid`
     // is not a subset of those of the `dest_vm_cpuid`.
     pub fn check_cpuid_compatibility(
-        src_vm_cpuid: &[CpuIdEntry],
-        dest_vm_cpuid: &[CpuIdEntry],
+        src_vm_cpuid: &CpuId,
+        dest_vm_cpuid: &CpuId,
     ) -> Result<(), Error> {
         let feature_entry_list = &Self::checked_feature_entry_list();
         let src_vm_features = Self::get_features_from_cpuid(src_vm_cpuid, feature_entry_list);
@@ -551,30 +544,13 @@ impl CpuidFeatureEntry {
 }
 
 pub fn generate_common_cpuid(
-    hypervisor: &Arc<dyn hypervisor::Hypervisor>,
+    hypervisor: Arc<dyn hypervisor::Hypervisor>,
     topology: Option<(u8, u8, u8)>,
     sgx_epc_sections: Option<Vec<SgxEpcSection>>,
     phys_bits: u8,
     kvm_hyperv: bool,
     #[cfg(feature = "tdx")] tdx_enabled: bool,
-) -> super::Result<Vec<CpuIdEntry>> {
-    // SAFETY: cpuid called with valid leaves
-    if unsafe { x86_64::__cpuid(1) }.ecx & 1 << HYPERVISOR_ECX_BIT == 1 << HYPERVISOR_ECX_BIT {
-        // SAFETY: cpuid called with valid leaves
-        let hypervisor_cpuid = unsafe { x86_64::__cpuid(0x4000_0000) };
-
-        let mut identifier: [u8; 12] = [0; 12];
-        identifier[0..4].copy_from_slice(&hypervisor_cpuid.ebx.to_le_bytes()[..]);
-        identifier[4..8].copy_from_slice(&hypervisor_cpuid.ecx.to_le_bytes()[..]);
-        identifier[8..12].copy_from_slice(&hypervisor_cpuid.edx.to_le_bytes()[..]);
-
-        info!(
-            "Running under nested virtualisation. Hypervisor string: {}",
-            String::from_utf8_lossy(&identifier)
-        );
-    }
-
-    info!("Generating guest CPUID for with physical address size: {phys_bits}");
+) -> super::Result<CpuId> {
     let cpuid_patches = vec![
         // Patch tsc deadline timer bit
         CpuidPatch {
@@ -609,9 +585,7 @@ pub fn generate_common_cpuid(
     ];
 
     // Supported CPUID
-    let mut cpuid = hypervisor
-        .get_supported_cpuid()
-        .map_err(Error::CpuidGetSupported)?;
+    let mut cpuid = hypervisor.get_cpuid().map_err(Error::CpuidGetSupported)?;
 
     CpuidPatch::patch_cpuid(&mut cpuid, cpuid_patches);
 
@@ -623,58 +597,21 @@ pub fn generate_common_cpuid(
         update_cpuid_sgx(&mut cpuid, sgx_epc_sections)?;
     }
 
-    #[cfg(feature = "tdx")]
-    let tdx_capabilities = if tdx_enabled {
-        let caps = hypervisor
-            .tdx_capabilities()
-            .map_err(Error::TdxCapabilities)?;
-        info!("TDX capabilities {:#?}", caps);
-        Some(caps)
-    } else {
-        None
-    };
-
     // Update some existing CPUID
     for entry in cpuid.as_mut_slice().iter_mut() {
         match entry.function {
-            0xd =>
-            {
-                #[cfg(feature = "tdx")]
-                if let Some(caps) = &tdx_capabilities {
-                    let xcr0_mask: u64 = 0x82ff;
-                    let xss_mask: u64 = !xcr0_mask;
-                    if entry.index == 0 {
-                        entry.eax &= (caps.xfam_fixed0 as u32) & (xcr0_mask as u32);
-                        entry.eax |= (caps.xfam_fixed1 as u32) & (xcr0_mask as u32);
-                        entry.edx &= ((caps.xfam_fixed0 & xcr0_mask) >> 32) as u32;
-                        entry.edx |= ((caps.xfam_fixed1 & xcr0_mask) >> 32) as u32;
-                    } else if entry.index == 1 {
-                        entry.ecx &= (caps.xfam_fixed0 as u32) & (xss_mask as u32);
-                        entry.ecx |= (caps.xfam_fixed1 as u32) & (xss_mask as u32);
-                        entry.edx &= ((caps.xfam_fixed0 & xss_mask) >> 32) as u32;
-                        entry.edx |= ((caps.xfam_fixed1 & xss_mask) >> 32) as u32;
-                    }
-                }
-            }
-            // Copy host L2 cache details if not populated by KVM
-            0x8000_0006 => {
-                if entry.eax == 0 && entry.ebx == 0 && entry.ecx == 0 && entry.edx == 0 {
-                    // SAFETY: cpuid called with valid leaves
-                    if unsafe { std::arch::x86_64::__cpuid(0x8000_0000).eax } >= 0x8000_0006 {
-                        // SAFETY: cpuid called with valid leaves
-                        let leaf = unsafe { std::arch::x86_64::__cpuid(0x8000_0006) };
-                        entry.eax = leaf.eax;
-                        entry.ebx = leaf.ebx;
-                        entry.ecx = leaf.ecx;
-                        entry.edx = leaf.edx;
-                    }
-                }
-            }
             // Set CPU physical bits
             0x8000_0008 => {
                 entry.eax = (entry.eax & 0xffff_ff00) | (phys_bits as u32 & 0xff);
             }
+            // Disable KVM_FEATURE_ASYNC_PF_INT
+            // This is required until we find out why the asynchronous page
+            // fault is generating unexpected behavior when using interrupt
+            // mechanism.
+            // TODO: Re-enable KVM_FEATURE_ASYNC_PF_INT (#2277)
             0x4000_0001 => {
+                entry.eax &= !(1 << KVM_FEATURE_ASYNC_PF_INT_BIT);
+
                 // These features are not supported by TDX
                 #[cfg(feature = "tdx")]
                 if tdx_enabled {
@@ -693,16 +630,17 @@ pub fn generate_common_cpuid(
     // Copy CPU identification string
     for i in 0x8000_0002..=0x8000_0004 {
         cpuid.retain(|c| c.function != i);
-        // SAFETY: call cpuid with valid leaves
         let leaf = unsafe { std::arch::x86_64::__cpuid(i) };
-        cpuid.push(CpuIdEntry {
-            function: i,
-            eax: leaf.eax,
-            ebx: leaf.ebx,
-            ecx: leaf.ecx,
-            edx: leaf.edx,
-            ..Default::default()
-        });
+        cpuid
+            .push(CpuIdEntry {
+                function: i,
+                eax: leaf.eax,
+                ebx: leaf.ebx,
+                ecx: leaf.ecx,
+                edx: leaf.edx,
+                ..Default::default()
+            })
+            .map_err(Error::CpuidIdentification)?;
     }
 
     if kvm_hyperv {
@@ -711,44 +649,56 @@ pub fn generate_common_cpuid(
         cpuid.retain(|c| c.function != 0x4000_0001);
         // See "Hypervisor Top Level Functional Specification" for details
         // Compliance with "Hv#1" requires leaves up to 0x4000_000a
-        cpuid.push(CpuIdEntry {
-            function: 0x40000000,
-            eax: 0x4000000a, // Maximum cpuid leaf
-            ebx: 0x756e694c, // "Linu"
-            ecx: 0x564b2078, // "x KV"
-            edx: 0x7648204d, // "M Hv"
-            ..Default::default()
-        });
-        cpuid.push(CpuIdEntry {
-            function: 0x40000001,
-            eax: 0x31237648, // "Hv#1"
-            ..Default::default()
-        });
-        cpuid.push(CpuIdEntry {
-            function: 0x40000002,
-            eax: 0x3839,  // "Build number"
-            ebx: 0xa0000, // "Version"
-            ..Default::default()
-        });
-        cpuid.push(CpuIdEntry {
-            function: 0x4000_0003,
-            eax: 1 << 1 // AccessPartitionReferenceCounter
+        cpuid
+            .push(CpuIdEntry {
+                function: 0x40000000,
+                eax: 0x4000000a, // Maximum cpuid leaf
+                ebx: 0x756e694c, // "Linu"
+                ecx: 0x564b2078, // "x KV"
+                edx: 0x7648204d, // "M Hv"
+                ..Default::default()
+            })
+            .map_err(Error::CpuidKvmHyperV)?;
+        cpuid
+            .push(CpuIdEntry {
+                function: 0x40000001,
+                eax: 0x31237648, // "Hv#1"
+                ..Default::default()
+            })
+            .map_err(Error::CpuidKvmHyperV)?;
+        cpuid
+            .push(CpuIdEntry {
+                function: 0x40000002,
+                eax: 0x3839,  // "Build number"
+                ebx: 0xa0000, // "Version"
+                ..Default::default()
+            })
+            .map_err(Error::CpuidKvmHyperV)?;
+        cpuid
+            .push(CpuIdEntry {
+                function: 0x4000_0003,
+                eax: 1 << 1 // AccessPartitionReferenceCounter
                    | 1 << 2 // AccessSynicRegs
                    | 1 << 3 // AccessSyntheticTimerRegs
                    | 1 << 9, // AccessPartitionReferenceTsc
-            edx: 1 << 3, // CPU dynamic partitioning
-            ..Default::default()
-        });
-        cpuid.push(CpuIdEntry {
-            function: 0x4000_0004,
-            eax: 1 << 5, // Recommend relaxed timing
-            ..Default::default()
-        });
-        for i in 0x4000_0005..=0x4000_000a {
-            cpuid.push(CpuIdEntry {
-                function: i,
+                edx: 1 << 3, // CPU dynamic partitioning
                 ..Default::default()
-            });
+            })
+            .map_err(Error::CpuidKvmHyperV)?;
+        cpuid
+            .push(CpuIdEntry {
+                function: 0x4000_0004,
+                eax: 1 << 5, // Recommend relaxed timing
+                ..Default::default()
+            })
+            .map_err(Error::CpuidKvmHyperV)?;
+        for i in 0x4000_0005..=0x4000_000a {
+            cpuid
+                .push(CpuIdEntry {
+                    function: i,
+                    ..Default::default()
+                })
+                .map_err(Error::CpuidKvmHyperV)?;
         }
     }
 
@@ -756,10 +706,11 @@ pub fn generate_common_cpuid(
 }
 
 pub fn configure_vcpu(
-    vcpu: &Arc<dyn hypervisor::Vcpu>,
+    fd: &Arc<dyn hypervisor::Vcpu>,
     id: u8,
-    boot_setup: Option<(EntryPoint, &GuestMemoryAtomic<GuestMemoryMmap>)>,
-    cpuid: Vec<CpuIdEntry>,
+    kernel_entry_point: Option<EntryPoint>,
+    vm_memory: &GuestMemoryAtomic<GuestMemoryMmap>,
+    cpuid: CpuId,
     kvm_hyperv: bool,
 ) -> super::Result<()> {
     // Per vCPU CPUID changes; common are handled via generate_common_cpuid()
@@ -767,58 +718,23 @@ pub fn configure_vcpu(
     CpuidPatch::set_cpuid_reg(&mut cpuid, 0xb, None, CpuidReg::EDX, u32::from(id));
     CpuidPatch::set_cpuid_reg(&mut cpuid, 0x1f, None, CpuidReg::EDX, u32::from(id));
 
-    // Set ApicId in cpuid for each vcpu
-    // SAFETY: get host cpuid when eax=1
-    let mut cpu_ebx = unsafe { core::arch::x86_64::__cpuid(1) }.ebx;
-    cpu_ebx &= 0xffffff;
-    cpu_ebx |= (id as u32) << 24;
-    CpuidPatch::set_cpuid_reg(&mut cpuid, 0x1, None, CpuidReg::EBX, cpu_ebx);
-
-    // The TSC frequency CPUID leaf should not be included when running with HyperV emulation
-    if !kvm_hyperv {
-        if let Some(tsc_khz) = vcpu.tsc_khz().map_err(Error::GetTscFrequency)? {
-            // Need to check that the TSC doesn't vary with dynamic frequency
-            // SAFETY: cpuid called with valid leaves
-            if unsafe { std::arch::x86_64::__cpuid(0x8000_0007) }.edx
-                & (1u32 << INVARIANT_TSC_EDX_BIT)
-                > 0
-            {
-                CpuidPatch::set_cpuid_reg(
-                    &mut cpuid,
-                    0x4000_0000,
-                    None,
-                    CpuidReg::EAX,
-                    0x4000_0010,
-                );
-                cpuid.retain(|c| c.function != 0x4000_0010);
-                cpuid.push(CpuIdEntry {
-                    function: 0x4000_0010,
-                    eax: tsc_khz,
-                    ebx: 1000000, /* LAPIC resolution of 1ns (freq: 1GHz) is hardcoded in KVM's
-                                   * APIC_BUS_CYCLE_NS */
-                    ..Default::default()
-                });
-            };
-        }
-    }
-
-    vcpu.set_cpuid2(&cpuid)
+    fd.set_cpuid2(&cpuid)
         .map_err(|e| Error::SetSupportedCpusFailed(e.into()))?;
 
     if kvm_hyperv {
-        vcpu.enable_hyperv_synic().unwrap();
+        fd.enable_hyperv_synic().unwrap();
     }
 
-    regs::setup_msrs(vcpu).map_err(Error::MsrsConfiguration)?;
-    if let Some((kernel_entry_point, guest_memory)) = boot_setup {
+    regs::setup_msrs(fd).map_err(Error::MsrsConfiguration)?;
+    if let Some(kernel_entry_point) = kernel_entry_point {
         if let Some(entry_addr) = kernel_entry_point.entry_addr {
             // Safe to unwrap because this method is called after the VM is configured
-            regs::setup_regs(vcpu, entry_addr.raw_value()).map_err(Error::RegsConfiguration)?;
-            regs::setup_fpu(vcpu).map_err(Error::FpuConfiguration)?;
-            regs::setup_sregs(&guest_memory.memory(), vcpu).map_err(Error::SregsConfiguration)?;
+            regs::setup_regs(fd, entry_addr.raw_value()).map_err(Error::RegsConfiguration)?;
+            regs::setup_fpu(fd).map_err(Error::FpuConfiguration)?;
+            regs::setup_sregs(&vm_memory.memory(), fd).map_err(Error::SregsConfiguration)?;
         }
     }
-    interrupts::set_lint(vcpu).map_err(|e| Error::LocalIntConfiguration(e.into()))?;
+    interrupts::set_lint(fd).map_err(|e| Error::LocalIntConfiguration(e.into()))?;
     Ok(())
 }
 
@@ -826,29 +742,47 @@ pub fn configure_vcpu(
 /// These should be used to configure the GuestMemory structure for the platform.
 /// For x86_64 all addresses are valid from the start of the kernel except a
 /// carve out at the end of 32bit address space.
-pub fn arch_memory_regions() -> Vec<(GuestAddress, usize, RegionType)> {
-    vec![
-        // 0 GiB ~ 3GiB: memory before the gap
-        (
+pub fn arch_memory_regions(size: GuestUsize) -> Vec<(GuestAddress, usize, RegionType)> {
+    let reserved_memory_gap_start = layout::MEM_32BIT_RESERVED_START
+        .checked_add(layout::MEM_32BIT_DEVICES_SIZE)
+        .expect("32-bit reserved region is too large");
+
+    let requested_memory_size = GuestAddress(size as u64);
+    let mut regions = Vec::new();
+
+    // case1: guest memory fits before the gap
+    if size as u64 <= layout::MEM_32BIT_RESERVED_START.raw_value() {
+        regions.push((GuestAddress(0), size as usize, RegionType::Ram));
+    // case2: guest memory extends beyond the gap
+    } else {
+        // push memory before the gap
+        regions.push((
             GuestAddress(0),
             layout::MEM_32BIT_RESERVED_START.raw_value() as usize,
             RegionType::Ram,
-        ),
-        // 4 GiB ~ inf: memory after the gap
-        (layout::RAM_64BIT_START, usize::MAX, RegionType::Ram),
-        // 3 GiB ~ 3712 MiB: 32-bit device memory hole
-        (
-            layout::MEM_32BIT_RESERVED_START,
-            layout::MEM_32BIT_DEVICES_SIZE as usize,
-            RegionType::SubRegion,
-        ),
-        // 3712 MiB ~ 3968 MiB: 32-bit reserved memory hole
-        (
-            layout::MEM_32BIT_RESERVED_START.unchecked_add(layout::MEM_32BIT_DEVICES_SIZE),
-            (layout::MEM_32BIT_RESERVED_SIZE - layout::MEM_32BIT_DEVICES_SIZE) as usize,
-            RegionType::Reserved,
-        ),
-    ]
+        ));
+        regions.push((
+            layout::RAM_64BIT_START,
+            requested_memory_size.unchecked_offset_from(layout::MEM_32BIT_RESERVED_START) as usize,
+            RegionType::Ram,
+        ));
+    }
+
+    // Add the 32-bit device memory hole as a sub region.
+    regions.push((
+        layout::MEM_32BIT_RESERVED_START,
+        layout::MEM_32BIT_DEVICES_SIZE as usize,
+        RegionType::SubRegion,
+    ));
+
+    // Add the 32-bit reserved memory hole as a sub region.
+    regions.push((
+        reserved_memory_gap_start,
+        (layout::MEM_32BIT_RESERVED_SIZE - layout::MEM_32BIT_DEVICES_SIZE) as usize,
+        RegionType::Reserved,
+    ));
+
+    regions
 }
 
 /// Configures the system and should be called once per vm before starting vcpu threads.
@@ -867,17 +801,8 @@ pub fn configure_system(
     _num_cpus: u8,
     rsdp_addr: Option<GuestAddress>,
     sgx_epc_region: Option<SgxEpcRegion>,
-    serial_number: Option<&str>,
-    uuid: Option<&str>,
-    oem_strings: Option<&[&str]>,
 ) -> super::Result<()> {
-    // Write EBDA address to location where ACPICA expects to find it
-    guest_mem
-        .write_obj((layout::EBDA_START.0 >> 4) as u16, layout::EBDA_POINTER)
-        .map_err(Error::EbdaSetup)?;
-
-    let size = smbios::setup_smbios(guest_mem, serial_number, uuid, oem_strings)
-        .map_err(Error::SmbiosSetup)?;
+    let size = smbios::setup_smbios(guest_mem).map_err(Error::SmbiosSetup)?;
 
     // Place the MP table after the SMIOS table aligned to 16 bytes
     let offset = GuestAddress(layout::SMBIOS_START).unchecked_add(size);
@@ -914,7 +839,7 @@ fn configure_pvh(
     start_info.0.magic = XEN_HVM_START_MAGIC_VALUE;
     start_info.0.version = 1; // pvh has version 1
     start_info.0.nr_modules = 0;
-    start_info.0.cmdline_paddr = cmdline_addr.raw_value();
+    start_info.0.cmdline_paddr = cmdline_addr.raw_value() as u64;
     start_info.0.memmap_paddr = layout::MEMMAP_START.raw_value();
 
     if let Some(rsdp_addr) = rsdp_addr {
@@ -946,102 +871,30 @@ fn configure_pvh(
     // Create the memory map entries.
     add_memmap_entry(&mut memmap, 0, layout::EBDA_START.raw_value(), E820_RAM);
 
-    // Merge continuous memory regions into one region.
-    // Note: memory regions from "GuestMemory" are sorted and non-zero sized.
-    let ram_regions = {
-        let mut ram_regions = Vec::new();
-        let mut current_start = guest_mem
-            .iter()
-            .next()
-            .map(GuestMemoryRegion::start_addr)
-            .expect("GuestMemory must have one memory region at least")
-            .raw_value();
-        let mut current_end = current_start;
-
-        for (start, size) in guest_mem
-            .iter()
-            .map(|m| (m.start_addr().raw_value(), m.len()))
-        {
-            if current_end == start {
-                // This zone is continuous with the previous one.
-                current_end += size;
-            } else {
-                ram_regions.push((current_start, current_end));
-
-                current_start = start;
-                current_end = start + size;
-            }
-        }
-
-        ram_regions.push((current_start, current_end));
-
-        ram_regions
-    };
-
-    if ram_regions.len() > 2 {
-        error!(
-            "There should be up to two non-continuous regions, devidided by the
-            gap at the end of 32bit address space (e.g. between 3G and 4G)."
-        );
-        return Err(super::Error::MemmapTableSetup);
-    }
-
-    // Create the memory map entry for memory region before the gap
-    {
-        let (first_region_start, first_region_end) =
-            ram_regions.first().ok_or(super::Error::MemmapTableSetup)?;
-        let high_ram_start = layout::HIGH_RAM_START.raw_value();
-        let mem_32bit_reserved_start = layout::MEM_32BIT_RESERVED_START.raw_value();
-
-        if !((first_region_start <= &high_ram_start)
-            && (first_region_end > &high_ram_start)
-            && (first_region_end <= &mem_32bit_reserved_start))
-        {
-            error!(
-                "Unexpected first memory region layout: (start: 0x{:08x}, end: 0x{:08x}).
-                high_ram_start: 0x{:08x}, mem_32bit_reserved_start: 0x{:08x}",
-                first_region_start, first_region_end, high_ram_start, mem_32bit_reserved_start
-            );
-
-            return Err(super::Error::MemmapTableSetup);
-        }
-
-        info!(
-            "create_memmap_entry, start: 0x{:08x}, end: 0x{:08x}",
-            high_ram_start, first_region_end
-        );
+    let mem_end = guest_mem.last_addr();
 
+    if mem_end < layout::MEM_32BIT_RESERVED_START {
         add_memmap_entry(
             &mut memmap,
-            high_ram_start,
-            first_region_end - high_ram_start,
+            layout::HIGH_RAM_START.raw_value(),
+            mem_end.unchecked_offset_from(layout::HIGH_RAM_START) + 1,
             E820_RAM,
         );
-    }
-
-    // Create the memory map entry for memory region after the gap if any
-    if let Some((second_region_start, second_region_end)) = ram_regions.get(1) {
-        let ram_64bit_start = layout::RAM_64BIT_START.raw_value();
-
-        if second_region_start != &ram_64bit_start {
-            error!(
-                "Unexpected second memory region layout: start: 0x{:08x}, ram_64bit_start: 0x{:08x}",
-                second_region_start, ram_64bit_start
-            );
-
-            return Err(super::Error::MemmapTableSetup);
-        }
-
-        info!(
-            "create_memmap_entry, start: 0x{:08x}, end: 0x{:08x}",
-            ram_64bit_start, second_region_end
-        );
+    } else {
         add_memmap_entry(
             &mut memmap,
-            ram_64bit_start,
-            second_region_end - ram_64bit_start,
+            layout::HIGH_RAM_START.raw_value(),
+            layout::MEM_32BIT_RESERVED_START.unchecked_offset_from(layout::HIGH_RAM_START),
             E820_RAM,
         );
+        if mem_end > layout::RAM_64BIT_START {
+            add_memmap_entry(
+                &mut memmap,
+                layout::RAM_64BIT_START.raw_value(),
+                mem_end.unchecked_offset_from(layout::RAM_64BIT_START) + 1,
+                E820_RAM,
+            );
+        }
     }
 
     add_memmap_entry(
@@ -1055,7 +908,7 @@ fn configure_pvh(
         add_memmap_entry(
             &mut memmap,
             sgx_epc_region.start().raw_value(),
-            sgx_epc_region.size(),
+            sgx_epc_region.size() as u64,
             E820_RESERVED,
         );
     }
@@ -1131,8 +984,7 @@ pub fn initramfs_load_addr(
     Ok(aligned_addr)
 }
 
-pub fn get_host_cpu_phys_bits(hypervisor: &Arc<dyn hypervisor::Hypervisor>) -> u8 {
-    // SAFETY: call cpuid with valid leaves
+pub fn get_host_cpu_phys_bits() -> u8 {
     unsafe {
         let leaf = x86_64::__cpuid(0x8000_0000);
 
@@ -1140,7 +992,9 @@ pub fn get_host_cpu_phys_bits(hypervisor: &Arc<dyn hypervisor::Hypervisor>) -> u
         // Some physical address bits may become reserved when the feature is enabled.
         // See AMD64 Architecture Programmer's Manual Volume 2, Section 7.10.1
         let reduced = if leaf.eax >= 0x8000_001f
-            && matches!(hypervisor.get_cpu_vendor(), CpuVendor::AMD)
+            && leaf.ebx == 0x6874_7541    // Vendor ID: AuthenticAMD
+            && leaf.ecx == 0x444d_4163
+            && leaf.edx == 0x6974_6e65
             && x86_64::__cpuid(0x8000_001f).eax & 0x1 != 0
         {
             (x86_64::__cpuid(0x8000_001f).ebx >> 6) & 0x3f
@@ -1158,7 +1012,7 @@ pub fn get_host_cpu_phys_bits(hypervisor: &Arc<dyn hypervisor::Hypervisor>) -> u
 }
 
 fn update_cpuid_topology(
-    cpuid: &mut Vec<CpuIdEntry>,
+    cpuid: &mut CpuId,
     threads_per_core: u8,
     cores_per_die: u8,
     dies_per_package: u8,
@@ -1222,10 +1076,7 @@ fn update_cpuid_topology(
 
 // The goal is to update the CPUID sub-leaves to reflect the number of EPC
 // sections exposed to the guest.
-fn update_cpuid_sgx(
-    cpuid: &mut Vec<CpuIdEntry>,
-    epc_sections: Vec<SgxEpcSection>,
-) -> Result<(), Error> {
+fn update_cpuid_sgx(cpuid: &mut CpuId, epc_sections: Vec<SgxEpcSection>) -> Result<(), Error> {
     // Something's wrong if there's no EPC section.
     if epc_sections.is_empty() {
         return Err(Error::NoSgxEpcSection);
@@ -1241,13 +1092,12 @@ fn update_cpuid_sgx(
 
     // Get host CPUID for leaf 0x12, subleaf 0x2. This is to retrieve EPC
     // properties such as confidentiality and integrity.
-    // SAFETY: call cpuid with valid leaves
     let leaf = unsafe { std::arch::x86_64::__cpuid_count(0x12, 0x2) };
 
     for (i, epc_section) in epc_sections.iter().enumerate() {
         let subleaf_idx = i + 2;
         let start = epc_section.start().raw_value();
-        let size = epc_section.size();
+        let size = epc_section.size() as u64;
         let eax = (start & 0xffff_f000) as u32 | 0x1;
         let ebx = (start >> 32) as u32;
         let ecx = (size & 0xffff_f000) as u32 | (leaf.ecx & 0xf);
@@ -1275,8 +1125,16 @@ mod tests {
     use super::*;
 
     #[test]
-    fn regions_base_addr() {
-        let regions = arch_memory_regions();
+    fn regions_lt_4gb() {
+        let regions = arch_memory_regions(1 << 29);
+        assert_eq!(3, regions.len());
+        assert_eq!(GuestAddress(0), regions[0].0);
+        assert_eq!(1usize << 29, regions[0].1);
+    }
+
+    #[test]
+    fn regions_gt_4gb() {
+        let regions = arch_memory_regions((1 << 32) + 0x8000);
         assert_eq!(4, regions.len());
         assert_eq!(GuestAddress(0), regions[0].0);
         assert_eq!(GuestAddress(1 << 32), regions[1].0);
@@ -1293,73 +1151,46 @@ mod tests {
             1,
             Some(layout::RSDP_POINTER),
             None,
-            None,
-            None,
-            None,
         );
         assert!(config_err.is_err());
 
         // Now assigning some memory that falls before the 32bit memory hole.
-        let arch_mem_regions = arch_memory_regions();
+        let mem_size = 128 << 20;
+        let arch_mem_regions = arch_memory_regions(mem_size);
         let ram_regions: Vec<(GuestAddress, usize)> = arch_mem_regions
             .iter()
-            .filter(|r| r.2 == RegionType::Ram && r.1 != usize::MAX)
+            .filter(|r| r.2 == RegionType::Ram)
             .map(|r| (r.0, r.1))
             .collect();
         let gm = GuestMemoryMmap::from_ranges(&ram_regions).unwrap();
 
-        configure_system(
-            &gm,
-            GuestAddress(0),
-            &None,
-            no_vcpus,
-            None,
-            None,
-            None,
-            None,
-            None,
-        )
-        .unwrap();
+        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None).unwrap();
 
-        // Now assigning some memory that falls after the 32bit memory hole.
-        let arch_mem_regions = arch_memory_regions();
+        // Now assigning some memory that is equal to the start of the 32bit memory hole.
+        let mem_size = 3328 << 20;
+        let arch_mem_regions = arch_memory_regions(mem_size);
         let ram_regions: Vec<(GuestAddress, usize)> = arch_mem_regions
             .iter()
             .filter(|r| r.2 == RegionType::Ram)
-            .map(|r| {
-                if r.1 == usize::MAX {
-                    (r.0, 128 << 20)
-                } else {
-                    (r.0, r.1)
-                }
-            })
+            .map(|r| (r.0, r.1))
             .collect();
         let gm = GuestMemoryMmap::from_ranges(&ram_regions).unwrap();
-        configure_system(
-            &gm,
-            GuestAddress(0),
-            &None,
-            no_vcpus,
-            None,
-            None,
-            None,
-            None,
-            None,
-        )
-        .unwrap();
+        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None).unwrap();
 
-        configure_system(
-            &gm,
-            GuestAddress(0),
-            &None,
-            no_vcpus,
-            None,
-            None,
-            None,
-            None,
-            None,
-        )
-        .unwrap();
+        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None).unwrap();
+
+        // Now assigning some memory that falls after the 32bit memory hole.
+        let mem_size = 3330 << 20;
+        let arch_mem_regions = arch_memory_regions(mem_size);
+        let ram_regions: Vec<(GuestAddress, usize)> = arch_mem_regions
+            .iter()
+            .filter(|r| r.2 == RegionType::Ram)
+            .map(|r| (r.0, r.1))
+            .collect();
+        let gm = GuestMemoryMmap::from_ranges(&ram_regions).unwrap();
+        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None).unwrap();
+
+        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None).unwrap();
     }
 
     #[test]
@@ -1384,6 +1215,6 @@ mod tests {
         add_memmap_entry(&mut memmap, 0, 0x1000, E820_RAM);
         add_memmap_entry(&mut memmap, 0x10000, 0xa000, E820_RESERVED);
 
-        assert_eq!(format!("{memmap:?}"), format!("{expected_memmap:?}"));
+        assert_eq!(format!("{:?}", memmap), format!("{:?}", expected_memmap));
     }
 }

arch/src/x86_64/mptable.rs

@@ -37,17 +37,11 @@ struct MpfIntelWrapper(mpspec::mpf_intel);
 
 // SAFETY: These `mpspec` wrapper types are only data, reading them from data is a safe initialization.
 unsafe impl ByteValued for MpcBusWrapper {}
-// SAFETY: see above
 unsafe impl ByteValued for MpcCpuWrapper {}
-// SAFETY: see above
 unsafe impl ByteValued for MpcIntsrcWrapper {}
-// SAFETY: see above
 unsafe impl ByteValued for MpcIoapicWrapper {}
-// SAFETY: see above
 unsafe impl ByteValued for MpcTableWrapper {}
-// SAFETY: see above
 unsafe impl ByteValued for MpcLintsrcWrapper {}
-// SAFETY: see above
 unsafe impl ByteValued for MpfIntelWrapper {}
 
 #[derive(Debug)]
@@ -101,7 +95,7 @@ const CPU_FEATURE_APIC: u32 = 0x200;
 const CPU_FEATURE_FPU: u32 = 0x001;
 
 fn compute_checksum<T: Copy>(v: &T) -> u8 {
-    // SAFETY: we are only reading the bytes within the size of the `T` reference `v`.
+    // Safe because we are only reading the bytes within the size of the `T` reference `v`.
     let v_slice = unsafe { slice::from_raw_parts(v as *const T as *const u8, mem::size_of::<T>()) };
     let mut checksum: u8 = 0;
     for i in v_slice.iter() {
@@ -300,7 +294,7 @@ mod tests {
             mpspec::MP_IOAPIC => mem::size_of::<MpcIoapicWrapper>(),
             mpspec::MP_INTSRC => mem::size_of::<MpcIntsrcWrapper>(),
             mpspec::MP_LINTSRC => mem::size_of::<MpcLintsrcWrapper>(),
-            _ => panic!("unrecognized mpc table entry type: {type_}"),
+            _ => panic!("unrecognized mpc table entry type: {}", type_),
         }
     }
 

arch/src/x86_64/regs.rs

@@ -9,8 +9,8 @@
 use crate::layout::{BOOT_GDT_START, BOOT_IDT_START, PVH_INFO_START};
 use crate::GuestMemoryMmap;
 use hypervisor::arch::x86::gdt::{gdt_entry, segment_from_gdt};
-use hypervisor::arch::x86::regs::CR0_PE;
-use hypervisor::arch::x86::{FpuState, SpecialRegisters, StandardRegisters};
+use hypervisor::arch::x86::regs::*;
+use hypervisor::x86_64::{FpuState, SpecialRegisters, StandardRegisters};
 use std::sync::Arc;
 use std::{mem, result};
 use vm_memory::{Address, Bytes, GuestMemory, GuestMemoryError};
@@ -66,7 +66,7 @@ pub fn setup_fpu(vcpu: &Arc<dyn hypervisor::Vcpu>) -> Result<()> {
 ///
 /// * `vcpu` - Structure for the VCPU that holds the VCPU's fd.
 pub fn setup_msrs(vcpu: &Arc<dyn hypervisor::Vcpu>) -> Result<()> {
-    vcpu.set_msrs(&vcpu.boot_msr_entries())
+    vcpu.set_msrs(&hypervisor::x86_64::boot_msr_entries())
         .map_err(Error::SetModelSpecificRegisters)?;
 
     Ok(())
@@ -124,7 +124,7 @@ pub fn configure_segments_and_sregs(
     mem: &GuestMemoryMmap,
     sregs: &mut SpecialRegisters,
 ) -> Result<()> {
-    let gdt_table: [u64; BOOT_GDT_MAX] = {
+    let gdt_table: [u64; BOOT_GDT_MAX as usize] = {
         // Configure GDT entries as specified by PVH boot protocol
         [
             gdt_entry(0, 0, 0),               // NULL

arch/src/x86_64/smbios.rs

@@ -12,10 +12,10 @@ use std::fmt::{self, Display};
 use std::mem;
 use std::result;
 use std::slice;
-use uuid::Uuid;
 use vm_memory::ByteValued;
 use vm_memory::{Address, Bytes, GuestAddress};
 
+#[allow(unused_variables)]
 #[derive(Debug)]
 pub enum Error {
     /// There was too little guest memory to store the entire SMBIOS table.
@@ -28,8 +28,6 @@ pub enum Error {
     WriteSmbiosEp,
     /// Failure to write additional data to memory
     WriteData,
-    /// Failure to parse uuid, uuid format may be error
-    ParseUuid(uuid::Error),
 }
 
 impl std::error::Error for Error {}
@@ -39,19 +37,14 @@ impl Display for Error {
         use self::Error::*;
 
         let description = match self {
-            NotEnoughMemory => {
-                "There was too little guest memory to store the SMBIOS table".to_string()
-            }
-            AddressOverflow => {
-                "The SMBIOS table has too little address space to be stored".to_string()
-            }
-            Clear => "Failure while zeroing out the memory for the SMBIOS table".to_string(),
-            WriteSmbiosEp => "Failure to write SMBIOS entrypoint structure".to_string(),
-            WriteData => "Failure to write additional data to memory".to_string(),
-            ParseUuid(e) => format!("Failure to parse uuid: {e}"),
+            NotEnoughMemory => "There was too little guest memory to store the SMBIOS table",
+            AddressOverflow => "The SMBIOS table has too little address space to be stored",
+            Clear => "Failure while zeroing out the memory for the SMBIOS table",
+            WriteSmbiosEp => "Failure to write SMBIOS entrypoint structure",
+            WriteData => "Failure to write additional data to memory",
         };
 
-        write!(f, "SMBIOS error: {description}")
+        write!(f, "SMBIOS error: {}", description)
     }
 }
 
@@ -61,13 +54,12 @@ pub type Result<T> = result::Result<T, Error>;
 const SM3_MAGIC_IDENT: &[u8; 5usize] = b"_SM3_";
 const BIOS_INFORMATION: u8 = 0;
 const SYSTEM_INFORMATION: u8 = 1;
-const OEM_STRINGS: u8 = 11;
 const END_OF_TABLE: u8 = 127;
 const PCI_SUPPORTED: u64 = 1 << 7;
 const IS_VIRTUAL_MACHINE: u8 = 1 << 4;
 
 fn compute_checksum<T: Copy>(v: &T) -> u8 {
-    // SAFETY: we are only reading the bytes within the size of the `T` reference `v`.
+    // Safe because we are only reading the bytes within the size of the `T` reference `v`.
     let v_slice = unsafe { slice::from_raw_parts(v as *const T as *const u8, mem::size_of::<T>()) };
     let mut checksum: u8 = 0;
     for i in v_slice.iter() {
@@ -76,85 +68,75 @@ fn compute_checksum<T: Copy>(v: &T) -> u8 {
     (!checksum).wrapping_add(1)
 }
 
-#[repr(C)]
 #[repr(packed)]
-#[derive(Default, Copy, Clone)]
-struct Smbios30Entrypoint {
-    signature: [u8; 5usize],
-    checksum: u8,
-    length: u8,
-    majorver: u8,
-    minorver: u8,
-    docrev: u8,
-    revision: u8,
-    reserved: u8,
-    max_size: u32,
-    physptr: u64,
+#[derive(Default, Copy)]
+pub struct Smbios30Entrypoint {
+    pub signature: [u8; 5usize],
+    pub checksum: u8,
+    pub length: u8,
+    pub majorver: u8,
+    pub minorver: u8,
+    pub docrev: u8,
+    pub revision: u8,
+    pub reserved: u8,
+    pub max_size: u32,
+    pub physptr: u64,
 }
 
-#[repr(C)]
-#[repr(packed)]
-#[derive(Default, Copy, Clone)]
-struct SmbiosBiosInfo {
-    r#type: u8,
-    length: u8,
-    handle: u16,
-    vendor: u8,
-    version: u8,
-    start_addr: u16,
-    release_date: u8,
-    rom_size: u8,
-    characteristics: u64,
-    characteristics_ext1: u8,
-    characteristics_ext2: u8,
+impl Clone for Smbios30Entrypoint {
+    fn clone(&self) -> Self {
+        *self
+    }
 }
 
-#[repr(C)]
 #[repr(packed)]
-#[derive(Default, Copy, Clone)]
-struct SmbiosSysInfo {
-    r#type: u8,
-    length: u8,
-    handle: u16,
-    manufacturer: u8,
-    product_name: u8,
-    version: u8,
-    serial_number: u8,
-    uuid: [u8; 16usize],
-    wake_up_type: u8,
-    sku: u8,
-    family: u8,
+#[derive(Default, Copy)]
+pub struct SmbiosBiosInfo {
+    pub typ: u8,
+    pub length: u8,
+    pub handle: u16,
+    pub vendor: u8,
+    pub version: u8,
+    pub start_addr: u16,
+    pub release_date: u8,
+    pub rom_size: u8,
+    pub characteristics: u64,
+    pub characteristics_ext1: u8,
+    pub characteristics_ext2: u8,
 }
 
-#[repr(C)]
-#[repr(packed)]
-#[derive(Default, Copy, Clone)]
-struct SmbiosOemStrings {
-    r#type: u8,
-    length: u8,
-    handle: u16,
-    count: u8,
+impl Clone for SmbiosBiosInfo {
+    fn clone(&self) -> Self {
+        *self
+    }
 }
 
-#[repr(C)]
 #[repr(packed)]
-#[derive(Default, Copy, Clone)]
-struct SmbiosEndOfTable {
-    r#type: u8,
-    length: u8,
-    handle: u16,
+#[derive(Default, Copy)]
+pub struct SmbiosSysInfo {
+    pub typ: u8,
+    pub length: u8,
+    pub handle: u16,
+    pub manufacturer: u8,
+    pub product_name: u8,
+    pub version: u8,
+    pub serial_number: u8,
+    pub uuid: [u8; 16usize],
+    pub wake_up_type: u8,
+    pub sku: u8,
+    pub family: u8,
 }
 
-// SAFETY: data structure only contain a series of integers
+impl Clone for SmbiosSysInfo {
+    fn clone(&self) -> Self {
+        *self
+    }
+}
+
+// SAFETY: These data structures only contain a series of integers
 unsafe impl ByteValued for Smbios30Entrypoint {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for SmbiosBiosInfo {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for SmbiosSysInfo {}
-// SAFETY: data structure only contain a series of integers
-unsafe impl ByteValued for SmbiosOemStrings {}
-// SAFETY: data structure only contain a series of integers
-unsafe impl ByteValued for SmbiosEndOfTable {}
 
 fn write_and_incr<T: ByteValued>(
     mem: &GuestMemoryMmap,
@@ -180,12 +162,7 @@ fn write_string(
     Ok(curptr)
 }
 
-pub fn setup_smbios(
-    mem: &GuestMemoryMmap,
-    serial_number: Option<&str>,
-    uuid: Option<&str>,
-    oem_strings: Option<&[&str]>,
-) -> Result<u64> {
+pub fn setup_smbios(mem: &GuestMemoryMmap) -> Result<u64> {
     let physptr = GuestAddress(SMBIOS_START)
         .checked_add(mem::size_of::<Smbios30Entrypoint>() as u64)
         .ok_or(Error::NotEnoughMemory)?;
@@ -195,7 +172,7 @@ pub fn setup_smbios(
     {
         handle += 1;
         let smbios_biosinfo = SmbiosBiosInfo {
-            r#type: BIOS_INFORMATION,
+            typ: BIOS_INFORMATION,
             length: mem::size_of::<SmbiosBiosInfo>() as u8,
             handle,
             vendor: 1,  // First string written in this section
@@ -212,59 +189,29 @@ pub fn setup_smbios(
 
     {
         handle += 1;
-
-        let uuid_number = uuid
-            .map(Uuid::parse_str)
-            .transpose()
-            .map_err(Error::ParseUuid)?
-            .unwrap_or(Uuid::nil());
         let smbios_sysinfo = SmbiosSysInfo {
-            r#type: SYSTEM_INFORMATION,
+            typ: SYSTEM_INFORMATION,
             length: mem::size_of::<SmbiosSysInfo>() as u8,
             handle,
             manufacturer: 1, // First string written in this section
             product_name: 2, // Second string written in this section
-            serial_number: serial_number.map(|_| 3).unwrap_or_default(), // 3rd string
-            uuid: uuid_number.to_bytes_le(), // set uuid
             ..Default::default()
         };
         curptr = write_and_incr(mem, smbios_sysinfo, curptr)?;
         curptr = write_string(mem, "Cloud Hypervisor", curptr)?;
         curptr = write_string(mem, "cloud-hypervisor", curptr)?;
-        if let Some(serial_number) = serial_number {
-            curptr = write_string(mem, serial_number, curptr)?;
-        }
-        curptr = write_and_incr(mem, 0u8, curptr)?;
-    }
-
-    if let Some(oem_strings) = oem_strings {
-        handle += 1;
-
-        let smbios_oemstrings = SmbiosOemStrings {
-            r#type: OEM_STRINGS,
-            length: mem::size_of::<SmbiosOemStrings>() as u8,
-            handle,
-            count: oem_strings.len() as u8,
-        };
-
-        curptr = write_and_incr(mem, smbios_oemstrings, curptr)?;
-
-        for s in oem_strings {
-            curptr = write_string(mem, s, curptr)?;
-        }
-
         curptr = write_and_incr(mem, 0u8, curptr)?;
     }
 
     {
         handle += 1;
-        let smbios_end = SmbiosEndOfTable {
-            r#type: END_OF_TABLE,
-            length: mem::size_of::<SmbiosEndOfTable>() as u8,
+        let smbios_sysinfo = SmbiosSysInfo {
+            typ: END_OF_TABLE,
+            length: mem::size_of::<SmbiosSysInfo>() as u8,
             handle,
+            ..Default::default()
         };
-        curptr = write_and_incr(mem, smbios_end, curptr)?;
-        curptr = write_and_incr(mem, 0u8, curptr)?;
+        curptr = write_and_incr(mem, smbios_sysinfo, curptr)?;
         curptr = write_and_incr(mem, 0u8, curptr)?;
     }
 
@@ -286,7 +233,7 @@ pub fn setup_smbios(
             .map_err(|_| Error::WriteSmbiosEp)?;
     }
 
-    Ok(curptr.unchecked_offset_from(physptr) + std::mem::size_of::<Smbios30Entrypoint>() as u64)
+    Ok(curptr.unchecked_offset_from(physptr))
 }
 
 #[cfg(test)]
@@ -316,7 +263,7 @@ mod tests {
     fn entrypoint_checksum() {
         let mem = GuestMemoryMmap::from_ranges(&[(GuestAddress(SMBIOS_START), 4096)]).unwrap();
 
-        setup_smbios(&mem, None, None, None).unwrap();
+        setup_smbios(&mem).unwrap();
 
         let smbios_ep: Smbios30Entrypoint = mem.read_obj(GuestAddress(SMBIOS_START)).unwrap();
 

arch/src/x86_64/tdx/mod.rs

@@ -4,9 +4,7 @@
 use crate::GuestMemoryMmap;
 use std::fs::File;
 use std::io::{Read, Seek, SeekFrom};
-use std::str::FromStr;
 use thiserror::Error;
-use uuid::Uuid;
 use vm_memory::{ByteValued, Bytes, GuestAddress, GuestMemoryError};
 
 #[derive(Error, Debug)]
@@ -15,8 +13,6 @@ pub enum TdvfError {
     ReadDescriptor(#[source] std::io::Error),
     #[error("Failed read TDVF descriptor offset: {0}")]
     ReadDescriptorOffset(#[source] std::io::Error),
-    #[error("Failed read GUID table: {0}")]
-    ReadGuidTable(#[source] std::io::Error),
     #[error("Invalid descriptor signature")]
     InvalidDescriptorSignature,
     #[error("Invalid descriptor size")]
@@ -25,13 +21,8 @@ pub enum TdvfError {
     InvalidDescriptorVersion,
     #[error("Failed to write HOB details to guest memory: {0}")]
     GuestMemoryWriteHob(#[source] GuestMemoryError),
-    #[error("Failed to create Uuid: {0}")]
-    UuidCreation(#[source] uuid::Error),
 }
 
-const TABLE_FOOTER_GUID: &str = "96b582de-1fb2-45f7-baea-a366c55a082d";
-const TDVF_METADATA_OFFSET_GUID: &str = "e47a6535-984a-4798-865e-4685a7bf8ec2";
-
 // TDVF_DESCRIPTOR
 #[repr(packed)]
 #[derive(Default)]
@@ -55,7 +46,7 @@ pub struct TdvfSection {
 }
 
 #[repr(u32)]
-#[derive(Clone, Copy, Debug, Default)]
+#[derive(Clone, Copy, Debug)]
 pub enum TdvfSectionType {
     Bfv,
     Cfv,
@@ -64,76 +55,16 @@ pub enum TdvfSectionType {
     PermMem,
     Payload,
     PayloadParam,
-    #[default]
     Reserved = 0xffffffff,
 }
 
-fn tdvf_descriptor_offset(file: &mut File) -> Result<(SeekFrom, bool), TdvfError> {
-    // Let's first try to identify the presence of the table footer GUID
-    file.seek(SeekFrom::End(-0x30))
-        .map_err(TdvfError::ReadGuidTable)?;
-    let mut table_footer_guid: [u8; 16] = [0; 16];
-    file.read_exact(&mut table_footer_guid)
-        .map_err(TdvfError::ReadGuidTable)?;
-    let uuid =
-        Uuid::from_slice_le(table_footer_guid.as_slice()).map_err(TdvfError::UuidCreation)?;
-    let expected_uuid = Uuid::from_str(TABLE_FOOTER_GUID).map_err(TdvfError::UuidCreation)?;
-    if uuid == expected_uuid {
-        // Retrieve the table size
-        file.seek(SeekFrom::End(-0x32))
-            .map_err(TdvfError::ReadGuidTable)?;
-        let mut table_size: [u8; 2] = [0; 2];
-        file.read_exact(&mut table_size)
-            .map_err(TdvfError::ReadGuidTable)?;
-        let table_size = u16::from_le_bytes(table_size) as usize;
-        let mut table: Vec<u8> = vec![0; table_size];
-
-        // Read the entire table
-        file.seek(SeekFrom::End(-(table_size as i64 + 0x20)))
-            .map_err(TdvfError::ReadGuidTable)?;
-        file.read_exact(table.as_mut_slice())
-            .map_err(TdvfError::ReadGuidTable)?;
-
-        // Let's start from the top and go backward down the table.
-        // We start after the footer GUID and the table length.
-        let mut offset = table_size - 18;
-
-        debug!("Parsing GUIDed structure");
-        while offset >= 18 {
-            let entry_uuid = Uuid::from_slice_le(&table[offset - 16..offset])
-                .map_err(TdvfError::UuidCreation)?;
-            let entry_size =
-                u16::from_le_bytes(table[offset - 18..offset - 16].try_into().unwrap()) as usize;
-            debug!(
-                "Entry GUID = {}, size = {}",
-                entry_uuid.hyphenated().to_string(),
-                entry_size
-            );
-
-            // Avoid going through an infinite loop if the entry size is 0
-            if entry_size == 0 {
-                break;
-            }
-
-            offset -= entry_size;
-
-            let expected_uuid =
-                Uuid::from_str(TDVF_METADATA_OFFSET_GUID).map_err(TdvfError::UuidCreation)?;
-            if entry_uuid == expected_uuid && entry_size == 22 {
-                return Ok((
-                    SeekFrom::End(
-                        -(u32::from_le_bytes(table[offset..offset + 4].try_into().unwrap()) as i64),
-                    ),
-                    true,
-                ));
-            }
-        }
+impl Default for TdvfSectionType {
+    fn default() -> Self {
+        TdvfSectionType::Reserved
     }
+}
 
-    // If we end up here, this means the firmware doesn't support the new way
-    // of exposing the TDVF descriptor offset through the table of GUIDs.
-    // That's why we fallback onto the deprecated method.
-
+pub fn parse_tdvf_sections(file: &mut File) -> Result<Vec<TdvfSection>, TdvfError> {
     // The 32-bit offset to the TDVF metadata is located 32 bytes from
     // the end of the file.
     // See "TDVF Metadata Pointer" in "TDX Virtual Firmware Design Guide
@@ -143,21 +74,13 @@ fn tdvf_descriptor_offset(file: &mut File) -> Result<(SeekFrom, bool), TdvfError
     let mut descriptor_offset: [u8; 4] = [0; 4];
     file.read_exact(&mut descriptor_offset)
         .map_err(TdvfError::ReadDescriptorOffset)?;
+    let descriptor_offset = u32::from_le_bytes(descriptor_offset) as u64;
 
-    Ok((
-        SeekFrom::Start(u32::from_le_bytes(descriptor_offset) as u64),
-        false,
-    ))
-}
-
-pub fn parse_tdvf_sections(file: &mut File) -> Result<(Vec<TdvfSection>, bool), TdvfError> {
-    let (descriptor_offset, guid_found) = tdvf_descriptor_offset(file)?;
-
-    file.seek(descriptor_offset)
+    file.seek(SeekFrom::Start(descriptor_offset))
         .map_err(TdvfError::ReadDescriptor)?;
 
     let mut descriptor: TdvfDescriptor = Default::default();
-    // SAFETY: we read exactly the size of the descriptor header
+    // Safe as we read exactly the size of the descriptor header
     file.read_exact(unsafe {
         std::slice::from_raw_parts_mut(
             &mut descriptor as *mut _ as *mut u8,
@@ -184,7 +107,7 @@ pub fn parse_tdvf_sections(file: &mut File) -> Result<(Vec<TdvfSection>, bool),
     let mut sections = Vec::new();
     sections.resize_with(descriptor.num_sections as usize, TdvfSection::default);
 
-    // SAFETY: we read exactly the advertised sections
+    // Safe as we read exactly the advertised sections
     file.read_exact(unsafe {
         std::slice::from_raw_parts_mut(
             sections.as_mut_ptr() as *mut u8,
@@ -193,20 +116,25 @@ pub fn parse_tdvf_sections(file: &mut File) -> Result<(Vec<TdvfSection>, bool),
     })
     .map_err(TdvfError::ReadDescriptor)?;
 
-    Ok((sections, guid_found))
+    Ok(sections)
 }
 
 #[repr(u16)]
-#[derive(Copy, Clone, Debug, Default)]
+#[derive(Copy, Clone, Debug)]
 enum HobType {
     Handoff = 0x1,
     ResourceDescriptor = 0x3,
     GuidExtension = 0x4,
-    #[default]
     Unused = 0xfffe,
     EndOfHobList = 0xffff,
 }
 
+impl Default for HobType {
+    fn default() -> Self {
+        HobType::Unused
+    }
+}
+
 #[repr(C)]
 #[derive(Copy, Clone, Default, Debug)]
 struct HobHeader {
@@ -256,14 +184,19 @@ struct HobGuidType {
 }
 
 #[repr(u32)]
-#[derive(Clone, Copy, Debug, Default)]
+#[derive(Clone, Copy, Debug)]
 pub enum PayloadImageType {
-    #[default]
     ExecutablePayload,
     BzImage,
     RawVmLinux,
 }
 
+impl Default for PayloadImageType {
+    fn default() -> Self {
+        PayloadImageType::ExecutablePayload
+    }
+}
+
 #[repr(C)]
 #[derive(Copy, Clone, Default, Debug)]
 pub struct PayloadInfo {
@@ -278,17 +211,12 @@ struct TdPayload {
     payload_info: PayloadInfo,
 }
 
-// SAFETY: data structure only contain a series of integers
+// SAFETY: These data structures only contain a series of integers
 unsafe impl ByteValued for HobHeader {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for HobHandoffInfoTable {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for HobResourceDescriptor {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for HobGuidType {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for PayloadInfo {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for TdPayload {}
 
 pub struct TdHob {
@@ -384,29 +312,22 @@ impl TdHob {
         physical_start: u64,
         resource_length: u64,
         ram: bool,
-        guid_found: bool,
     ) -> Result<(), TdvfError> {
         self.add_resource(
             mem,
             physical_start,
             resource_length,
             if ram {
-                if guid_found {
-                    0x7 /* EFI_RESOURCE_MEMORY_UNACCEPTED */
-                } else {
-                    0 /* EFI_RESOURCE_SYSTEM_MEMORY */
-                }
-            } else if guid_found {
                 0 /* EFI_RESOURCE_SYSTEM_MEMORY */
             } else {
                 0x5 /*EFI_RESOURCE_MEMORY_RESERVED */
             },
             /* TODO:
              * QEMU currently fills it in like this:
-             * EFI_RESOURCE_ATTRIBUTE_PRESENT | EFI_RESOURCE_ATTRIBUTE_INITIALIZED | EFI_RESOURCE_ATTRIBUTE_TESTED
+             * EFI_RESOURCE_ATTRIBUTE_PRESENT | EFI_RESOURCE_ATTRIBUTE_INITIALIZED | EFI_RESOURCE_ATTRIBUTE_ENCRYPTED | EFI_RESOURCE_ATTRIBUTE_TESTED
              * which differs from the spec (due to TDVF implementation issue?)
              */
-            0x7,
+            0x04000007,
         )
     }
 
@@ -524,9 +445,9 @@ mod tests {
     #[ignore]
     fn test_parse_tdvf_sections() {
         let mut f = std::fs::File::open("tdvf.fd").unwrap();
-        let (sections, _) = parse_tdvf_sections(&mut f).unwrap();
+        let sections = parse_tdvf_sections(&mut f).unwrap();
         for section in sections {
-            eprintln!("{section:x?}")
+            eprintln!("{:x?}", section)
         }
     }
 }

block/Cargo.toml

@@ -1,27 +0,0 @@
-[package]
-name = "block"
-version = "0.1.0"
-edition = "2021"
-authors = ["The Cloud Hypervisor Authors", "The Chromium OS Authors"]
-
-[features]
-default = []
-io_uring = ["dep:io-uring"]
-
-[dependencies]
-byteorder = "1.4.3"
-crc32c = "0.6.3"
-io-uring = { version = "0.6.0", optional = true }
-libc = "0.2.139"
-log = "0.4.17"
-remain = "0.2.11"
-smallvec = "1.10.0"
-thiserror = "1.0.40"
-uuid = { version = "1.3.4", features = ["v4"] }
-versionize = "0.1.10"
-versionize_derive = "0.1.4"
-virtio-bindings = { version = "0.2.0", features = ["virtio-v5_0_0"] }
-virtio-queue = "0.8.0"
-vm-memory = { version = "0.11.0", features = ["backend-mmap", "backend-atomic", "backend-bitmap"] }
-vm-virtio = { path = "../vm-virtio" }
-vmm-sys-util = "0.11.0"

block/src/fixed_vhd.rs

@@ -1,90 +0,0 @@
-// Copyright © 2021 Intel Corporation
-//
-// SPDX-License-Identifier: Apache-2.0
-
-use crate::vhd::VhdFooter;
-use crate::BlockBackend;
-use std::fs::File;
-use std::io::{Read, Seek, SeekFrom, Write};
-use std::os::unix::io::{AsRawFd, RawFd};
-
-#[derive(Debug)]
-pub struct FixedVhd {
-    file: File,
-    size: u64,
-    position: u64,
-}
-
-impl FixedVhd {
-    pub fn new(mut file: File) -> std::io::Result<Self> {
-        let footer = VhdFooter::new(&mut file)?;
-
-        Ok(Self {
-            file,
-            size: footer.current_size(),
-            position: 0,
-        })
-    }
-}
-
-impl AsRawFd for FixedVhd {
-    fn as_raw_fd(&self) -> RawFd {
-        self.file.as_raw_fd()
-    }
-}
-
-impl Read for FixedVhd {
-    fn read(&mut self, buf: &mut [u8]) -> std::io::Result<usize> {
-        match self.file.read(buf) {
-            Ok(r) => {
-                self.position = self.position.checked_add(r.try_into().unwrap()).unwrap();
-                Ok(r)
-            }
-            Err(e) => Err(e),
-        }
-    }
-}
-
-impl Write for FixedVhd {
-    fn write(&mut self, buf: &[u8]) -> std::io::Result<usize> {
-        match self.file.write(buf) {
-            Ok(r) => {
-                self.position = self.position.checked_add(r.try_into().unwrap()).unwrap();
-                Ok(r)
-            }
-            Err(e) => Err(e),
-        }
-    }
-
-    fn flush(&mut self) -> std::io::Result<()> {
-        self.file.sync_all()
-    }
-}
-
-impl Seek for FixedVhd {
-    fn seek(&mut self, newpos: SeekFrom) -> std::io::Result<u64> {
-        match self.file.seek(newpos) {
-            Ok(pos) => {
-                self.position = pos;
-                Ok(pos)
-            }
-            Err(e) => Err(e),
-        }
-    }
-}
-
-impl BlockBackend for FixedVhd {
-    fn size(&self) -> std::result::Result<u64, crate::Error> {
-        Ok(self.size)
-    }
-}
-
-impl Clone for FixedVhd {
-    fn clone(&self) -> Self {
-        Self {
-            file: self.file.try_clone().expect("FixedVhd cloning failed"),
-            size: self.size,
-            position: self.position,
-        }
-    }
-}

block_util/Cargo.toml

@@ -0,0 +1,24 @@
+[package]
+name = "block_util"
+version = "0.1.0"
+authors = ["The Cloud Hypervisor Authors"]
+edition = "2018"
+
+[features]
+default = []
+
+[dependencies]
+io-uring = "0.5.2"
+libc = "0.2.119"
+log = "0.4.14"
+qcow = { path = "../qcow" }
+thiserror = "1.0.30"
+versionize = "0.1.6"
+versionize_derive = "0.1.4"
+vhdx = { path = "../vhdx" }
+virtio-bindings = { version = "0.1.0", features = ["virtio-v5_0_0"] }
+virtio-queue = { git = "https://github.com/rust-vmm/vm-virtio", branch = "main" }
+vm-memory = { version = "0.7.0", features = ["backend-mmap", "backend-atomic", "backend-bitmap"] }
+vm-virtio = { path = "../vm-virtio" }
+vmm-sys-util = "0.9.0"
+

block_util/src/async_io.rs

@@ -3,11 +3,12 @@
 // SPDX-License-Identifier: Apache-2.0 AND BSD-3-Clause
 
 use libc::{ioctl, S_IFBLK, S_IFMT};
+use std::convert::TryInto;
 use std::fs::File;
 use std::os::unix::io::AsRawFd;
 use thiserror::Error;
 use vmm_sys_util::eventfd::EventFd;
-use vmm_sys_util::{ioctl_io_nr, ioctl_ioc_nr};
+use vmm_sys_util::{ioctl_expr, ioctl_io_nr, ioctl_ioc_nr};
 
 #[derive(Error, Debug)]
 pub enum DiskFileError {
@@ -53,21 +54,19 @@ enum BlockSize {
 impl DiskTopology {
     fn is_block_device(f: &mut File) -> std::io::Result<bool> {
         let mut stat = std::mem::MaybeUninit::<libc::stat>::uninit();
-        // SAFETY: FFI call with a valid fd and buffer
         let ret = unsafe { libc::fstat(f.as_raw_fd(), stat.as_mut_ptr()) };
         if ret != 0 {
             return Err(std::io::Error::last_os_error());
         }
 
-        // SAFETY: stat is valid at this point
         let is_block = unsafe { (*stat.as_ptr()).st_mode & S_IFMT == S_IFBLK };
         Ok(is_block)
     }
 
     // libc::ioctl() takes different types on different architectures
+    #[allow(clippy::useless_conversion)]
     fn query_block_size(f: &mut File, block_size_type: BlockSize) -> std::io::Result<u64> {
         let mut block_size = 0;
-        // SAFETY: FFI call with correct arguments
         let ret = unsafe {
             ioctl(
                 f.as_raw_fd(),
@@ -76,7 +75,9 @@ impl DiskTopology {
                     BlockSize::PhysicalBlock => BLKPBSZGET(),
                     BlockSize::MinimumIo => BLKIOMIN(),
                     BlockSize::OptimalIo => BLKIOOPT(),
-                } as _,
+                }
+                .try_into()
+                .unwrap(),
                 &mut block_size,
             )
         };
@@ -103,7 +104,7 @@ impl DiskTopology {
 
 pub type DiskFileResult<T> = std::result::Result<T, DiskFileError>;
 
-pub trait DiskFile: Send {
+pub trait DiskFile: Send + Sync {
     fn size(&mut self) -> DiskFileResult<u64>;
     fn new_async_io(&self, ring_depth: u32) -> DiskFileResult<Box<dyn AsyncIo>>;
     fn topology(&mut self) -> DiskTopology {
@@ -126,20 +127,20 @@ pub enum AsyncIoError {
 
 pub type AsyncIoResult<T> = std::result::Result<T, AsyncIoError>;
 
-pub trait AsyncIo: Send {
+pub trait AsyncIo: Send + Sync {
     fn notifier(&self) -> &EventFd;
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()>;
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()>;
     fn fsync(&mut self, user_data: Option<u64>) -> AsyncIoResult<()>;
-    fn next_completed_request(&mut self) -> Option<(u64, i32)>;
+    fn complete(&mut self) -> Vec<(u64, i32)>;
 }

block_util/src/fixed_vhd_async.rs

@@ -5,29 +5,36 @@
 use crate::async_io::{
     AsyncIo, AsyncIoError, AsyncIoResult, DiskFile, DiskFileError, DiskFileResult,
 };
-use crate::fixed_vhd::FixedVhd;
 use crate::raw_async::RawFileAsync;
-use crate::BlockBackend;
+use crate::vhd::VhdFooter;
 use std::fs::File;
 use std::os::unix::io::{AsRawFd, RawFd};
 use vmm_sys_util::eventfd::EventFd;
 
-pub struct FixedVhdDiskAsync(FixedVhd);
+pub struct FixedVhdDiskAsync {
+    file: File,
+    size: u64,
+}
 
 impl FixedVhdDiskAsync {
-    pub fn new(file: File) -> std::io::Result<Self> {
-        Ok(Self(FixedVhd::new(file)?))
+    pub fn new(mut file: File) -> std::io::Result<Self> {
+        let footer = VhdFooter::new(&mut file)?;
+
+        Ok(FixedVhdDiskAsync {
+            file,
+            size: footer.current_size(),
+        })
     }
 }
 
 impl DiskFile for FixedVhdDiskAsync {
     fn size(&mut self) -> DiskFileResult<u64> {
-        Ok(self.0.size().unwrap())
+        Ok(self.size)
     }
 
     fn new_async_io(&self, ring_depth: u32) -> DiskFileResult<Box<dyn AsyncIo>> {
         Ok(Box::new(
-            FixedVhdAsync::new(self.0.as_raw_fd(), ring_depth, self.0.size().unwrap())
+            FixedVhdAsync::new(self.file.as_raw_fd(), ring_depth, self.size)
                 .map_err(DiskFileError::NewAsyncIo)?,
         ) as Box<dyn AsyncIo>)
     }
@@ -57,7 +64,7 @@ impl AsyncIo for FixedVhdAsync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         if offset as u64 >= self.size {
@@ -76,7 +83,7 @@ impl AsyncIo for FixedVhdAsync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         if offset as u64 >= self.size {
@@ -97,7 +104,7 @@ impl AsyncIo for FixedVhdAsync {
         self.raw_file_async.fsync(user_data)
     }
 
-    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
-        self.raw_file_async.next_completed_request()
+    fn complete(&mut self) -> Vec<(u64, i32)> {
+        self.raw_file_async.complete()
     }
 }

block_util/src/fixed_vhd_sync.rs

@@ -5,29 +5,36 @@
 use crate::async_io::{
     AsyncIo, AsyncIoError, AsyncIoResult, DiskFile, DiskFileError, DiskFileResult,
 };
-use crate::fixed_vhd::FixedVhd;
 use crate::raw_sync::RawFileSync;
-use crate::BlockBackend;
+use crate::vhd::VhdFooter;
 use std::fs::File;
 use std::os::unix::io::{AsRawFd, RawFd};
 use vmm_sys_util::eventfd::EventFd;
 
-pub struct FixedVhdDiskSync(FixedVhd);
+pub struct FixedVhdDiskSync {
+    file: File,
+    size: u64,
+}
 
 impl FixedVhdDiskSync {
-    pub fn new(file: File) -> std::io::Result<Self> {
-        Ok(Self(FixedVhd::new(file)?))
+    pub fn new(mut file: File) -> std::io::Result<Self> {
+        let footer = VhdFooter::new(&mut file)?;
+
+        Ok(FixedVhdDiskSync {
+            file,
+            size: footer.current_size(),
+        })
     }
 }
 
 impl DiskFile for FixedVhdDiskSync {
     fn size(&mut self) -> DiskFileResult<u64> {
-        Ok(self.0.size().unwrap())
+        Ok(self.size)
     }
 
     fn new_async_io(&self, _ring_depth: u32) -> DiskFileResult<Box<dyn AsyncIo>> {
         Ok(Box::new(
-            FixedVhdSync::new(self.0.as_raw_fd(), self.0.size().unwrap())
+            FixedVhdSync::new(self.file.as_raw_fd(), self.size)
                 .map_err(DiskFileError::NewAsyncIo)?,
         ) as Box<dyn AsyncIo>)
     }
@@ -55,7 +62,7 @@ impl AsyncIo for FixedVhdSync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         if offset as u64 >= self.size {
@@ -74,7 +81,7 @@ impl AsyncIo for FixedVhdSync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         if offset as u64 >= self.size {
@@ -94,7 +101,7 @@ impl AsyncIo for FixedVhdSync {
         self.raw_file_sync.fsync(user_data)
     }
 
-    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
-        self.raw_file_sync.next_completed_request()
+    fn complete(&mut self) -> Vec<(u64, i32)> {
+        self.raw_file_sync.complete()
     }
 }

block_util/src/lib.rs

@@ -12,35 +12,19 @@
 extern crate log;
 
 pub mod async_io;
-pub mod fixed_vhd;
-#[cfg(feature = "io_uring")]
-/// Enabled with the `"io_uring"` feature
 pub mod fixed_vhd_async;
 pub mod fixed_vhd_sync;
-pub mod qcow;
 pub mod qcow_sync;
-#[cfg(feature = "io_uring")]
-/// Async primitives based on `io-uring`
-///
-/// Enabled with the `"io_uring"` feature
 pub mod raw_async;
 pub mod raw_sync;
 pub mod vhd;
-pub mod vhdx;
 pub mod vhdx_sync;
 
 use crate::async_io::{AsyncIo, AsyncIoError, AsyncIoResult};
-use crate::fixed_vhd::FixedVhd;
-use crate::qcow::{QcowFile, RawFile};
-use crate::vhdx::{Vhdx, VhdxError};
-#[cfg(feature = "io_uring")]
 use io_uring::{opcode, IoUring, Probe};
-use smallvec::SmallVec;
 use std::alloc::{alloc_zeroed, dealloc, Layout};
 use std::cmp;
-use std::collections::VecDeque;
 use std::convert::TryInto;
-use std::fmt::Debug;
 use std::fs::File;
 use std::io::{self, IoSlice, IoSliceMut, Read, Seek, SeekFrom, Write};
 use std::os::linux::fs::MetadataExt;
@@ -48,11 +32,9 @@ use std::path::Path;
 use std::result;
 use std::sync::Arc;
 use std::sync::MutexGuard;
-use std::time::Instant;
-use thiserror::Error;
 use versionize::{VersionMap, Versionize, VersionizeResult};
 use versionize_derive::Versionize;
-use virtio_bindings::virtio_blk::*;
+use virtio_bindings::bindings::virtio_blk::*;
 use virtio_queue::DescriptorChain;
 use vm_memory::{
     bitmap::AtomicBitmap, bitmap::Bitmap, ByteValued, Bytes, GuestAddress, GuestMemory,
@@ -66,36 +48,26 @@ type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
 const SECTOR_SHIFT: u8 = 9;
 pub const SECTOR_SIZE: u64 = 0x01 << SECTOR_SHIFT;
 
-#[derive(Error, Debug)]
+#[derive(Debug)]
 pub enum Error {
-    #[error("Guest gave us bad memory addresses")]
+    /// Guest gave us bad memory addresses.
     GuestMemory(GuestMemoryError),
-    #[error("Guest gave us offsets that would have overflowed a usize")]
+    /// Guest gave us offsets that would have overflowed a usize.
     CheckedOffset(GuestAddress, usize),
-    #[error("Guest gave us a write only descriptor that protocol says to read from")]
+    /// Guest gave us a write only descriptor that protocol says to read from.
     UnexpectedWriteOnlyDescriptor,
-    #[error("Guest gave us a read only descriptor that protocol says to write to")]
+    /// Guest gave us a read only descriptor that protocol says to write to.
     UnexpectedReadOnlyDescriptor,
-    #[error("Guest gave us too few descriptors in a descriptor chain")]
+    /// Guest gave us too few descriptors in a descriptor chain.
     DescriptorChainTooShort,
-    #[error("Guest gave us a descriptor that was too short to use")]
+    /// Guest gave us a descriptor that was too short to use.
     DescriptorLengthTooSmall,
-    #[error("Failed to detect image type: {0}")]
-    DetectImageType(std::io::Error),
-    #[error("Failure in fixed vhd: {0}")]
-    FixedVhdError(std::io::Error),
-    #[error("Getting a block's metadata fails for any reason")]
+    /// Getting a block's metadata fails for any reason.
     GetFileMetadata,
-    #[error("The requested operation would cause a seek beyond disk end")]
+    /// The requested operation would cause a seek beyond disk end.
     InvalidOffset,
-    #[error("Failure in qcow: {0}")]
-    QcowError(qcow::Error),
-    #[error("Failure in raw file: {0}")]
-    RawFileError(std::io::Error),
-    #[error("The requested operation does not support multiple descriptors")]
+    /// The requested operation does not support multiple descriptors.
     TooManyDescriptors,
-    #[error("Failure in vhdx: {0}")]
-    VhdxError(VhdxError),
 }
 
 fn build_device_id(disk_path: &Path) -> result::Result<String, Error> {
@@ -130,31 +102,20 @@ pub fn build_disk_image_id(disk_path: &Path) -> Vec<u8> {
     default_disk_image_id
 }
 
-#[derive(Error, Debug)]
+#[derive(Debug)]
 pub enum ExecuteError {
-    #[error("Bad request: {0}")]
     BadRequest(Error),
-    #[error("Falied to flush: {0}")]
     Flush(io::Error),
-    #[error("Failed to read: {0}")]
     Read(GuestMemoryError),
-    #[error("Failed to seek: {0}")]
     Seek(io::Error),
-    #[error("Failed to write: {0}")]
     Write(GuestMemoryError),
-    #[error("Unsupported request: {0}")]
     Unsupported(u32),
-    #[error("Failed to submit io uring: {0}")]
     SubmitIoUring(io::Error),
-    #[error("Failed to get guest address: {0}")]
     GetHostAddress(GuestMemoryError),
-    #[error("Failed to async read: {0}")]
     AsyncRead(AsyncIoError),
-    #[error("Failed to async write: {0}")]
     AsyncWrite(AsyncIoError),
-    #[error("failed to async flush: {0}")]
     AsyncFlush(AsyncIoError),
-    #[error("Failed allocating a temporary buffer: {0}")]
+    /// Failed allocating a temporary buffer.
     TemporaryBufferAllocation(io::Error),
 }
 
@@ -177,7 +138,7 @@ impl ExecuteError {
     }
 }
 
-#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+#[derive(Clone, Copy, Debug, PartialEq)]
 pub enum RequestType {
     In,
     Out,
@@ -222,11 +183,10 @@ pub struct AlignedOperation {
 pub struct Request {
     pub request_type: RequestType,
     pub sector: u64,
-    pub data_descriptors: SmallVec<[(GuestAddress, u32); 1]>,
+    pub data_descriptors: Vec<(GuestAddress, u32)>,
     pub status_addr: GuestAddress,
     pub writeback: bool,
-    pub aligned_operations: SmallVec<[AlignedOperation; 1]>,
-    pub start: Instant,
+    pub aligned_operations: Vec<AlignedOperation>,
 }
 
 impl Request {
@@ -249,16 +209,15 @@ impl Request {
 
         let hdr_desc_addr = hdr_desc
             .addr()
-            .translate_gva(access_platform, hdr_desc.len() as usize);
+            .translate(access_platform, hdr_desc.len() as usize);
 
         let mut req = Request {
             request_type: request_type(desc_chain.memory(), hdr_desc_addr)?,
             sector: sector(desc_chain.memory(), hdr_desc_addr)?,
-            data_descriptors: SmallVec::with_capacity(1),
+            data_descriptors: Vec::new(),
             status_addr: GuestAddress(0),
             writeback: true,
-            aligned_operations: SmallVec::with_capacity(1),
-            start: Instant::now(),
+            aligned_operations: Vec::new(),
         };
 
         let status_desc;
@@ -278,7 +237,6 @@ impl Request {
                 return Err(Error::DescriptorChainTooShort);
             }
         } else {
-            req.data_descriptors.reserve_exact(1);
             while desc.has_next() {
                 if desc.is_write_only() && req.request_type == RequestType::Out {
                     return Err(Error::UnexpectedWriteOnlyDescriptor);
@@ -291,8 +249,7 @@ impl Request {
                 }
 
                 req.data_descriptors.push((
-                    desc.addr()
-                        .translate_gva(access_platform, desc.len() as usize),
+                    desc.addr().translate(access_platform, desc.len() as usize),
                     desc.len(),
                 ));
                 desc = desc_chain
@@ -317,7 +274,7 @@ impl Request {
 
         req.status_addr = status_desc
             .addr()
-            .translate_gva(access_platform, status_desc.len() as usize);
+            .translate(access_platform, status_desc.len() as usize);
 
         Ok(req)
     }
@@ -383,8 +340,7 @@ impl Request {
         let request_type = self.request_type;
         let offset = (sector << SECTOR_SHIFT) as libc::off_t;
 
-        let mut iovecs: SmallVec<[libc::iovec; 1]> =
-            SmallVec::with_capacity(self.data_descriptors.len());
+        let mut iovecs = Vec::new();
         for (data_addr, data_len) in &self.data_descriptors {
             if *data_len == 0 {
                 continue;
@@ -412,7 +368,7 @@ impl Request {
             let iov_base = if (origin_ptr as u64) % SECTOR_SIZE != 0 {
                 let layout =
                     Layout::from_size_align(*data_len as usize, SECTOR_SIZE as usize).unwrap();
-                // SAFETY: layout has non-zero size
+                // Safe because layout has non-zero size
                 let aligned_ptr = unsafe { alloc_zeroed(layout) };
                 if aligned_ptr.is_null() {
                     return Err(ExecuteError::TemporaryBufferAllocation(
@@ -423,7 +379,7 @@ impl Request {
                 // We need to perform the copy beforehand in case we're writing
                 // data out.
                 if request_type == RequestType::Out {
-                    // SAFETY: destination buffer has been allocated with
+                    // Safe because destination buffer has been allocated with
                     // the proper size.
                     unsafe {
                         std::ptr::copy(origin_ptr as *const u8, aligned_ptr, *data_len as usize)
@@ -461,12 +417,12 @@ impl Request {
                         .mark_dirty(0, *data_len as usize);
                 }
                 disk_image
-                    .read_vectored(offset, &iovecs, user_data)
+                    .read_vectored(offset, iovecs, user_data)
                     .map_err(ExecuteError::AsyncRead)?;
             }
             RequestType::Out => {
                 disk_image
-                    .write_vectored(offset, &iovecs, user_data)
+                    .write_vectored(offset, iovecs, user_data)
                     .map_err(ExecuteError::AsyncWrite)?;
             }
             RequestType::Flush => {
@@ -498,7 +454,7 @@ impl Request {
             // We need to perform the copy after the data has been read inside
             // the aligned buffer in case we're reading data in.
             if self.request_type == RequestType::In {
-                // SAFETY: origin buffer has been allocated with the
+                // Safe because origin buffer has been allocated with the
                 // proper size.
                 unsafe {
                     std::ptr::copy(
@@ -510,7 +466,7 @@ impl Request {
             }
 
             // Free the temporary aligned buffer.
-            // SAFETY: aligned_ptr was allocated by alloc_zeroed with the same
+            // Safe because aligned_ptr was allocated by alloc_zeroed with the same
             // layout
             unsafe {
                 dealloc(
@@ -559,67 +515,57 @@ pub struct VirtioBlockGeometry {
     pub sectors: u8,
 }
 
-// SAFETY: data structure only contain a series of integers
+// SAFETY: these data structures only contain a series of integers
 unsafe impl ByteValued for VirtioBlockConfig {}
-// SAFETY: data structure only contain a series of integers
 unsafe impl ByteValued for VirtioBlockGeometry {}
 
 /// Check if io_uring for block device can be used on the current system, as
 /// it correctly supports the expected io_uring features.
 pub fn block_io_uring_is_supported() -> bool {
-    #[cfg(not(feature = "io_uring"))]
-    {
-        info!("io_uring is disabled by crate features");
-        false
+    let error_msg = "io_uring not supported:";
+
+    // Check we can create an io_uring instance, which effectively verifies
+    // that io_uring_setup() syscall is supported.
+    let io_uring = match IoUring::new(1) {
+        Ok(io_uring) => io_uring,
+        Err(e) => {
+            info!("{} failed to create io_uring instance: {}", error_msg, e);
+            return false;
+        }
+    };
+
+    let submitter = io_uring.submitter();
+
+    let mut probe = Probe::new();
+
+    // Check we can register a probe to validate supported operations.
+    match submitter.register_probe(&mut probe) {
+        Ok(_) => {}
+        Err(e) => {
+            info!("{} failed to register a probe: {}", error_msg, e);
+            return false;
+        }
     }
 
-    #[cfg(feature = "io_uring")]
-    {
-        let error_msg = "io_uring not supported:";
-
-        // Check we can create an io_uring instance, which effectively verifies
-        // that io_uring_setup() syscall is supported.
-        let io_uring = match IoUring::new(1) {
-            Ok(io_uring) => io_uring,
-            Err(e) => {
-                info!("{} failed to create io_uring instance: {}", error_msg, e);
-                return false;
-            }
-        };
-
-        let submitter = io_uring.submitter();
-
-        let mut probe = Probe::new();
-
-        // Check we can register a probe to validate supported operations.
-        match submitter.register_probe(&mut probe) {
-            Ok(_) => {}
-            Err(e) => {
-                info!("{} failed to register a probe: {}", error_msg, e);
-                return false;
-            }
-        }
-
-        // Check IORING_OP_FSYNC is supported
-        if !probe.is_supported(opcode::Fsync::CODE) {
-            info!("{} IORING_OP_FSYNC operation not supported", error_msg);
-            return false;
-        }
-
-        // Check IORING_OP_READV is supported
-        if !probe.is_supported(opcode::Readv::CODE) {
-            info!("{} IORING_OP_READV operation not supported", error_msg);
-            return false;
-        }
-
-        // Check IORING_OP_WRITEV is supported
-        if !probe.is_supported(opcode::Writev::CODE) {
-            info!("{} IORING_OP_WRITEV operation not supported", error_msg);
-            return false;
-        }
-
-        true
+    // Check IORING_OP_FSYNC is supported
+    if !probe.is_supported(opcode::Fsync::CODE) {
+        info!("{} IORING_OP_FSYNC operation not supported", error_msg);
+        return false;
     }
+
+    // Check IORING_OP_READ is supported
+    if !probe.is_supported(opcode::Read::CODE) {
+        info!("{} IORING_OP_READ operation not supported", error_msg);
+        return false;
+    }
+
+    // Check IORING_OP_WRITE is supported
+    if !probe.is_supported(opcode::Write::CODE) {
+        info!("{} IORING_OP_WRITE operation not supported", error_msg);
+        return false;
+    }
+
+    true
 }
 
 pub trait AsyncAdaptor<F>
@@ -629,15 +575,14 @@ where
     fn read_vectored_sync(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
         eventfd: &EventFd,
-        completion_list: &mut VecDeque<(u64, i32)>,
+        completion_list: &mut Vec<(u64, i32)>,
     ) -> AsyncIoResult<()> {
         // Convert libc::iovec into IoSliceMut
-        let mut slices: SmallVec<[IoSliceMut; 1]> = SmallVec::with_capacity(iovecs.len());
+        let mut slices = Vec::new();
         for iovec in iovecs.iter() {
-            // SAFETY: on Linux IoSliceMut wraps around libc::iovec
             slices.push(IoSliceMut::new(unsafe { std::mem::transmute(*iovec) }));
         }
 
@@ -653,7 +598,7 @@ where
                 .map_err(AsyncIoError::ReadVectored)?
         };
 
-        completion_list.push_back((user_data, result as i32));
+        completion_list.push((user_data, result as i32));
         eventfd.write(1).unwrap();
 
         Ok(())
@@ -662,15 +607,14 @@ where
     fn write_vectored_sync(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
         eventfd: &EventFd,
-        completion_list: &mut VecDeque<(u64, i32)>,
+        completion_list: &mut Vec<(u64, i32)>,
     ) -> AsyncIoResult<()> {
         // Convert libc::iovec into IoSlice
-        let mut slices: SmallVec<[IoSlice; 1]> = SmallVec::with_capacity(iovecs.len());
+        let mut slices = Vec::new();
         for iovec in iovecs.iter() {
-            // SAFETY: on Linux IoSlice wraps around libc::iovec
             slices.push(IoSlice::new(unsafe { std::mem::transmute(*iovec) }));
         }
 
@@ -686,7 +630,7 @@ where
                 .map_err(AsyncIoError::WriteVectored)?
         };
 
-        completion_list.push_back((user_data, result as i32));
+        completion_list.push((user_data, result as i32));
         eventfd.write(1).unwrap();
 
         Ok(())
@@ -696,7 +640,7 @@ where
         &mut self,
         user_data: Option<u64>,
         eventfd: &EventFd,
-        completion_list: &mut VecDeque<(u64, i32)>,
+        completion_list: &mut Vec<(u64, i32)>,
     ) -> AsyncIoResult<()> {
         let result: i32 = {
             let mut file = self.file();
@@ -708,7 +652,7 @@ where
         };
 
         if let Some(user_data) = user_data {
-            completion_list.push_back((user_data, result));
+            completion_list.push((user_data, result));
             eventfd.write(1).unwrap();
         }
 
@@ -753,26 +697,3 @@ pub fn detect_image_type(f: &mut File) -> std::io::Result<ImageType> {
 
     Ok(image_type)
 }
-
-pub trait BlockBackend: Read + Write + Seek + Send + Debug {
-    fn size(&self) -> Result<u64, Error>;
-}
-
-/// Inspect the image file type and create an appropriate disk file to match it.
-pub fn create_disk_file(mut file: File, direct_io: bool) -> Result<Box<dyn BlockBackend>, Error> {
-    let image_type = detect_image_type(&mut file).map_err(Error::DetectImageType)?;
-
-    Ok(match image_type {
-        ImageType::Qcow2 => {
-            Box::new(QcowFile::from(RawFile::new(file, direct_io)).map_err(Error::QcowError)?)
-                as Box<dyn BlockBackend>
-        }
-        ImageType::FixedVhd => {
-            Box::new(FixedVhd::new(file).map_err(Error::FixedVhdError)?) as Box<dyn BlockBackend>
-        }
-        ImageType::Vhdx => {
-            Box::new(Vhdx::new(file).map_err(Error::VhdxError)?) as Box<dyn BlockBackend>
-        }
-        ImageType::Raw => Box::new(RawFile::new(file, direct_io)) as Box<dyn BlockBackend>,
-    })
-}

block_util/src/qcow_sync.rs

@@ -3,9 +3,8 @@
 // SPDX-License-Identifier: Apache-2.0 AND BSD-3-Clause
 
 use crate::async_io::{AsyncIo, AsyncIoResult, DiskFile, DiskFileError, DiskFileResult};
-use crate::qcow::{QcowFile, RawFile, Result as QcowResult};
 use crate::AsyncAdaptor;
-use std::collections::VecDeque;
+use qcow::{QcowFile, RawFile, Result as QcowResult};
 use std::fs::File;
 use std::io::{Seek, SeekFrom};
 use std::sync::{Arc, Mutex, MutexGuard};
@@ -27,7 +26,7 @@ impl DiskFile for QcowDiskSync {
     fn size(&mut self) -> DiskFileResult<u64> {
         let mut file = self.qcow_file.lock().unwrap();
 
-        file.seek(SeekFrom::End(0)).map_err(DiskFileError::Size)
+        Ok(file.seek(SeekFrom::End(0)).map_err(DiskFileError::Size)? as u64)
     }
 
     fn new_async_io(&self, _ring_depth: u32) -> DiskFileResult<Box<dyn AsyncIo>> {
@@ -38,7 +37,7 @@ impl DiskFile for QcowDiskSync {
 pub struct QcowSync {
     qcow_file: Arc<Mutex<QcowFile>>,
     eventfd: EventFd,
-    completion_list: VecDeque<(u64, i32)>,
+    completion_list: Vec<(u64, i32)>,
 }
 
 impl QcowSync {
@@ -47,7 +46,7 @@ impl QcowSync {
             qcow_file,
             eventfd: EventFd::new(libc::EFD_NONBLOCK)
                 .expect("Failed creating EventFd for QcowSync"),
-            completion_list: VecDeque::new(),
+            completion_list: Vec::new(),
         }
     }
 }
@@ -66,7 +65,7 @@ impl AsyncIo for QcowSync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         self.qcow_file.read_vectored_sync(
@@ -81,7 +80,7 @@ impl AsyncIo for QcowSync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         self.qcow_file.write_vectored_sync(
@@ -98,7 +97,7 @@ impl AsyncIo for QcowSync {
             .fsync_sync(user_data, &self.eventfd, &mut self.completion_list)
     }
 
-    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
-        self.completion_list.pop_front()
+    fn complete(&mut self) -> Vec<(u64, i32)> {
+        self.completion_list.drain(..).collect()
     }
 }

block_util/src/raw_async.rs

@@ -23,9 +23,10 @@ impl RawFileDisk {
 
 impl DiskFile for RawFileDisk {
     fn size(&mut self) -> DiskFileResult<u64> {
-        self.file
+        Ok(self
+            .file
             .seek(SeekFrom::End(0))
-            .map_err(DiskFileError::Size)
+            .map_err(DiskFileError::Size)? as u64)
     }
 
     fn new_async_io(&self, ring_depth: u32) -> DiskFileResult<Box<dyn AsyncIo>> {
@@ -76,17 +77,17 @@ impl AsyncIo for RawFileAsync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         let (submitter, mut sq, _) = self.io_uring.split();
 
-        // SAFETY: we know the file descriptor is valid and we
+        // Safe because we know the file descriptor is valid and we
         // relied on vm-memory to provide the buffer address.
         let _ = unsafe {
             sq.push(
                 &opcode::Readv::new(types::Fd(self.fd), iovecs.as_ptr(), iovecs.len() as u32)
-                    .offset(offset.try_into().unwrap())
+                    .offset(offset)
                     .build()
                     .flags(squeue::Flags::ASYNC)
                     .user_data(user_data),
@@ -104,17 +105,17 @@ impl AsyncIo for RawFileAsync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         let (submitter, mut sq, _) = self.io_uring.split();
 
-        // SAFETY: we know the file descriptor is valid and we
+        // Safe because we know the file descriptor is valid and we
         // relied on vm-memory to provide the buffer address.
         let _ = unsafe {
             sq.push(
                 &opcode::Writev::new(types::Fd(self.fd), iovecs.as_ptr(), iovecs.len() as u32)
-                    .offset(offset.try_into().unwrap())
+                    .offset(offset)
                     .build()
                     .flags(squeue::Flags::ASYNC)
                     .user_data(user_data),
@@ -133,7 +134,7 @@ impl AsyncIo for RawFileAsync {
         if let Some(user_data) = user_data {
             let (submitter, mut sq, _) = self.io_uring.split();
 
-            // SAFETY: we know the file descriptor is valid.
+            // Safe because we know the file descriptor is valid.
             let _ = unsafe {
                 sq.push(
                     &opcode::Fsync::new(types::Fd(self.fd))
@@ -148,17 +149,20 @@ impl AsyncIo for RawFileAsync {
             sq.sync();
             submitter.submit().map_err(AsyncIoError::Fsync)?;
         } else {
-            // SAFETY: FFI call with a valid fd
             unsafe { libc::fsync(self.fd) };
         }
 
         Ok(())
     }
 
-    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
-        self.io_uring
-            .completion()
-            .next()
-            .map(|entry| (entry.user_data(), entry.result()))
+    fn complete(&mut self) -> Vec<(u64, i32)> {
+        let mut completion_list = Vec::new();
+
+        let cq = self.io_uring.completion();
+        for cq_entry in cq {
+            completion_list.push((cq_entry.user_data(), cq_entry.result()));
+        }
+
+        completion_list
     }
 }

block_util/src/raw_sync.rs

@@ -5,7 +5,6 @@
 use crate::async_io::{
     AsyncIo, AsyncIoError, AsyncIoResult, DiskFile, DiskFileError, DiskFileResult, DiskTopology,
 };
-use std::collections::VecDeque;
 use std::fs::File;
 use std::io::{Seek, SeekFrom};
 use std::os::unix::io::{AsRawFd, RawFd};
@@ -23,9 +22,10 @@ impl RawFileDiskSync {
 
 impl DiskFile for RawFileDiskSync {
     fn size(&mut self) -> DiskFileResult<u64> {
-        self.file
+        Ok(self
+            .file
             .seek(SeekFrom::End(0))
-            .map_err(DiskFileError::Size)
+            .map_err(DiskFileError::Size)? as u64)
     }
 
     fn new_async_io(&self, _ring_depth: u32) -> DiskFileResult<Box<dyn AsyncIo>> {
@@ -45,7 +45,7 @@ impl DiskFile for RawFileDiskSync {
 pub struct RawFileSync {
     fd: RawFd,
     eventfd: EventFd,
-    completion_list: VecDeque<(u64, i32)>,
+    completion_list: Vec<(u64, i32)>,
 }
 
 impl RawFileSync {
@@ -53,7 +53,7 @@ impl RawFileSync {
         RawFileSync {
             fd,
             eventfd: EventFd::new(libc::EFD_NONBLOCK).expect("Failed creating EventFd for RawFile"),
-            completion_list: VecDeque::new(),
+            completion_list: Vec::new(),
         }
     }
 }
@@ -66,14 +66,13 @@ impl AsyncIo for RawFileSync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
-        // SAFETY: FFI call with valid arguments
         let result = unsafe {
             libc::preadv(
                 self.fd as libc::c_int,
-                iovecs.as_ptr(),
+                iovecs.as_ptr() as *const libc::iovec,
                 iovecs.len() as libc::c_int,
                 offset,
             )
@@ -82,7 +81,7 @@ impl AsyncIo for RawFileSync {
             return Err(AsyncIoError::ReadVectored(std::io::Error::last_os_error()));
         }
 
-        self.completion_list.push_back((user_data, result as i32));
+        self.completion_list.push((user_data, result as i32));
         self.eventfd.write(1).unwrap();
 
         Ok(())
@@ -91,14 +90,13 @@ impl AsyncIo for RawFileSync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
-        // SAFETY: FFI call with valid arguments
         let result = unsafe {
             libc::pwritev(
                 self.fd as libc::c_int,
-                iovecs.as_ptr(),
+                iovecs.as_ptr() as *const libc::iovec,
                 iovecs.len() as libc::c_int,
                 offset,
             )
@@ -107,28 +105,27 @@ impl AsyncIo for RawFileSync {
             return Err(AsyncIoError::WriteVectored(std::io::Error::last_os_error()));
         }
 
-        self.completion_list.push_back((user_data, result as i32));
+        self.completion_list.push((user_data, result as i32));
         self.eventfd.write(1).unwrap();
 
         Ok(())
     }
 
     fn fsync(&mut self, user_data: Option<u64>) -> AsyncIoResult<()> {
-        // SAFETY: FFI call
         let result = unsafe { libc::fsync(self.fd as libc::c_int) };
         if result < 0 {
             return Err(AsyncIoError::Fsync(std::io::Error::last_os_error()));
         }
 
         if let Some(user_data) = user_data {
-            self.completion_list.push_back((user_data, result));
+            self.completion_list.push((user_data, result as i32));
             self.eventfd.write(1).unwrap();
         }
 
         Ok(())
     }
 
-    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
-        self.completion_list.pop_front()
+    fn complete(&mut self) -> Vec<(u64, i32)> {
+        self.completion_list.drain(..).collect()
     }
 }

block_util/src/vhdx_sync.rs

@@ -3,11 +3,10 @@
 // SPDX-License-Identifier: Apache-2.0
 
 use crate::async_io::{AsyncIo, AsyncIoResult, DiskFile, DiskFileError, DiskFileResult};
-use crate::vhdx::{Result as VhdxResult, Vhdx};
 use crate::AsyncAdaptor;
-use std::collections::VecDeque;
 use std::fs::File;
 use std::sync::{Arc, Mutex, MutexGuard};
+use vhdx::vhdx::{Result as VhdxResult, Vhdx};
 use vmm_sys_util::eventfd::EventFd;
 
 pub struct VhdxDiskSync {
@@ -38,7 +37,7 @@ impl DiskFile for VhdxDiskSync {
 pub struct VhdxSync {
     vhdx_file: Arc<Mutex<Vhdx>>,
     eventfd: EventFd,
-    completion_list: VecDeque<(u64, i32)>,
+    completion_list: Vec<(u64, i32)>,
 }
 
 impl VhdxSync {
@@ -46,7 +45,7 @@ impl VhdxSync {
         Ok(VhdxSync {
             vhdx_file,
             eventfd: EventFd::new(libc::EFD_NONBLOCK)?,
-            completion_list: VecDeque::new(),
+            completion_list: Vec::new(),
         })
     }
 }
@@ -65,7 +64,7 @@ impl AsyncIo for VhdxSync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         self.vhdx_file.read_vectored_sync(
@@ -80,7 +79,7 @@ impl AsyncIo for VhdxSync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: &[libc::iovec],
+        iovecs: Vec<libc::iovec>,
         user_data: u64,
     ) -> AsyncIoResult<()> {
         self.vhdx_file.write_vectored_sync(
@@ -97,7 +96,7 @@ impl AsyncIo for VhdxSync {
             .fsync_sync(user_data, &self.eventfd, &mut self.completion_list)
     }
 
-    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
-        self.completion_list.pop_front()
+    fn complete(&mut self) -> Vec<(u64, i32)> {
+        self.completion_list.drain(..).collect()
     }
 }

build.rs

@@ -3,12 +3,15 @@
 // SPDX-License-Identifier: Apache-2.0
 //
 
+#[macro_use(crate_version)]
+extern crate clap;
+
 use std::process::Command;
 
 fn main() {
-    let mut version = "v".to_owned() + env!("CARGO_PKG_VERSION");
+    let mut version = "v".to_owned() + crate_version!();
 
-    if let Ok(git_out) = Command::new("git").args(["describe", "--dirty"]).output() {
+    if let Ok(git_out) = Command::new("git").args(&["describe", "--dirty"]).output() {
         if git_out.status.success() {
             if let Ok(git_out_str) = String::from_utf8(git_out.stdout) {
                 version = git_out_str;
@@ -17,8 +20,8 @@ fn main() {
     }
 
     // This println!() has a special behavior, as it will set the environment
-    // variable BUILD_VERSION, so that it can be reused from the binary.
+    // variable BUILT_VERSION, so that it can be reused from the binary.
     // Particularly, this is used from src/main.rs to display the exact
     // version.
-    println!("cargo:rustc-env=BUILD_VERSION={version}");
+    println!("cargo:rustc-env=BUILT_VERSION={}", version);
 }

devices/Cargo.toml

@@ -2,31 +2,26 @@
 name = "devices"
 version = "0.1.0"
 authors = ["The Chromium OS Authors"]
-edition = "2021"
+edition = "2018"
 
 [dependencies]
-acpi_tables = { git = "https://github.com/rust-vmm/acpi_tables", branch = "main"  }
-anyhow = "1.0.71"
+acpi_tables = { path = "../acpi_tables", optional = true }
+anyhow = "1.0.55"
 arch = { path = "../arch" }
-bitflags = "2.3.3"
+bitflags = "1.3.2"
 byteorder = "1.4.3"
-event_monitor = { path = "../event_monitor" }
-hypervisor = { path = "../hypervisor" }
-libc = "0.2.139"
-log = "0.4.17"
-pci = { path = "../pci" }
-thiserror = "1.0.40"
-tpm = { path = "../tpm" }
-versionize = "0.1.10"
+epoll = "4.3.1"
+libc = "0.2.119"
+log = "0.4.14"
+versionize = "0.1.6"
 versionize_derive = "0.1.4"
-vm-allocator = { path = "../vm-allocator" }
 vm-device = { path = "../vm-device" }
-vm-memory = "0.11.0"
+vm-memory = "0.7.0"
 vm-migration = { path = "../vm-migration" }
-vmm-sys-util = "0.11.0"
-
-[target.'cfg(target_arch = "aarch64")'.dependencies]
-arch = { path = "../arch" }
+vmm-sys-util = "0.9.0"
 
 [features]
 default = []
+acpi = ["acpi_tables"]
+cmos = []
+fwdebug = []

devices/src/acpi.rs

@@ -4,10 +4,8 @@
 //
 
 use super::AcpiNotificationFlags;
-use acpi_tables::{aml, Aml, AmlSink};
-use std::sync::atomic::{AtomicBool, Ordering};
+use acpi_tables::{aml, aml::Aml};
 use std::sync::{Arc, Barrier};
-use std::thread;
 use std::time::Instant;
 use vm_device::interrupt::InterruptSourceGroup;
 use vm_device::BusDevice;
@@ -20,20 +18,14 @@ pub const GED_DEVICE_ACPI_SIZE: usize = 0x1;
 pub struct AcpiShutdownDevice {
     exit_evt: EventFd,
     reset_evt: EventFd,
-    vcpus_kill_signalled: Arc<AtomicBool>,
 }
 
 impl AcpiShutdownDevice {
     /// Constructs a device that will signal the given event when the guest requests it.
-    pub fn new(
-        exit_evt: EventFd,
-        reset_evt: EventFd,
-        vcpus_kill_signalled: Arc<AtomicBool>,
-    ) -> AcpiShutdownDevice {
+    pub fn new(exit_evt: EventFd, reset_evt: EventFd) -> AcpiShutdownDevice {
         AcpiShutdownDevice {
             exit_evt,
             reset_evt,
-            vcpus_kill_signalled,
         }
     }
 }
@@ -51,13 +43,6 @@ impl BusDevice for AcpiShutdownDevice {
             if let Err(e) = self.reset_evt.write(1) {
                 error!("Error triggering ACPI reset event: {}", e);
             }
-            // Spin until we are sure the reset_evt has been handled and that when
-            // we return from the KVM_RUN we will exit rather than re-enter the guest.
-            while !self.vcpus_kill_signalled.load(Ordering::SeqCst) {
-                // This is more effective than thread::yield_now() at
-                // avoiding a priority inversion with the VMM thread
-                thread::sleep(std::time::Duration::from_millis(1));
-            }
         }
         // The ACPI DSDT table specifies the S5 sleep state (shutdown) as value 5
         const S5_SLEEP_VALUE: u8 = 5;
@@ -68,13 +53,6 @@ impl BusDevice for AcpiShutdownDevice {
             if let Err(e) = self.exit_evt.write(1) {
                 error!("Error triggering ACPI shutdown event: {}", e);
             }
-            // Spin until we are sure the reset_evt has been handled and that when
-            // we return from the KVM_RUN we will exit rather than re-enter the guest.
-            while !self.vcpus_kill_signalled.load(Ordering::SeqCst) {
-                // This is more effective than thread::yield_now() at
-                // avoiding a priority inversion with the VMM thread
-                thread::sleep(std::time::Duration::from_millis(1));
-            }
         }
         None
     }
@@ -124,39 +102,39 @@ impl BusDevice for AcpiGedDevice {
     }
 }
 
+#[cfg(feature = "acpi")]
 impl Aml for AcpiGedDevice {
-    fn to_aml_bytes(&self, sink: &mut dyn AmlSink) {
+    fn append_aml_bytes(&self, bytes: &mut Vec<u8>) {
         aml::Device::new(
-            "_SB_.GEC_".into(),
+            "_SB_.GED_".into(),
             vec![
-                &aml::Name::new("_HID".into(), &aml::EISAName::new("PNP0A06")),
-                &aml::Name::new("_UID".into(), &"Generic Event Controller"),
+                &aml::Name::new("_HID".into(), &"ACPI0013"),
+                &aml::Name::new("_UID".into(), &aml::ZERO),
                 &aml::Name::new(
                     "_CRS".into(),
-                    &aml::ResourceTemplate::new(vec![&aml::AddressSpace::new_memory(
-                        aml::AddressSpaceCachable::NotCacheable,
+                    &aml::ResourceTemplate::new(vec![&aml::Interrupt::new(
                         true,
-                        self.address.0,
-                        self.address.0 + GED_DEVICE_ACPI_SIZE as u64 - 1,
-                        None,
+                        true,
+                        false,
+                        false,
+                        self.ged_irq,
                     )]),
                 ),
                 &aml::OpRegion::new(
                     "GDST".into(),
                     aml::OpRegionSpace::SystemMemory,
-                    &(self.address.0 as usize),
-                    &GED_DEVICE_ACPI_SIZE,
+                    self.address.0 as usize,
+                    GED_DEVICE_ACPI_SIZE,
                 ),
                 &aml::Field::new(
                     "GDST".into(),
                     aml::FieldAccessType::Byte,
-                    aml::FieldLockRule::NoLock,
                     aml::FieldUpdateRule::WriteAsZeroes,
                     vec![aml::FieldEntry::Named(*b"GDAT", 8)],
                 ),
                 &aml::Method::new(
-                    "ESCN".into(),
-                    0,
+                    "_EVT".into(),
+                    1,
                     true,
                     vec![
                         &aml::Store::new(&aml::Local(0), &aml::Path::new("GDAT")),
@@ -187,31 +165,7 @@ impl Aml for AcpiGedDevice {
                 ),
             ],
         )
-        .to_aml_bytes(sink);
-        aml::Device::new(
-            "_SB_.GED_".into(),
-            vec![
-                &aml::Name::new("_HID".into(), &"ACPI0013"),
-                &aml::Name::new("_UID".into(), &aml::ZERO),
-                &aml::Name::new(
-                    "_CRS".into(),
-                    &aml::ResourceTemplate::new(vec![&aml::Interrupt::new(
-                        true,
-                        true,
-                        false,
-                        false,
-                        self.ged_irq,
-                    )]),
-                ),
-                &aml::Method::new(
-                    "_EVT".into(),
-                    1,
-                    true,
-                    vec![&aml::MethodCall::new("\\_SB_.GEC_.ESCN".into(), vec![])],
-                ),
-            ],
-        )
-        .to_aml_bytes(sink)
+        .append_aml_bytes(bytes)
     }
 }
 

devices/src/gic.rs

@@ -4,45 +4,36 @@
 
 use super::interrupt_controller::{Error, InterruptController};
 extern crate arch;
-use anyhow::anyhow;
-use arch::layout;
-use hypervisor::{
-    arch::aarch64::gic::{Vgic, VgicConfig},
-    CpuState, GicState,
-};
+use arch::aarch64::gic::GicDevice;
 use std::result;
 use std::sync::{Arc, Mutex};
 use vm_device::interrupt::{
     InterruptIndex, InterruptManager, InterruptSourceConfig, InterruptSourceGroup,
     LegacyIrqSourceConfig, MsiIrqGroupConfig,
 };
-use vm_memory::address::Address;
-use vm_migration::{Migratable, MigratableError, Pausable, Snapshot, Snapshottable, Transportable};
 use vmm_sys_util::eventfd::EventFd;
 
 type Result<T> = result::Result<T, Error>;
 
-// Reserve 32 IRQs for legacy devices.
-pub const IRQ_LEGACY_BASE: usize = layout::IRQ_BASE as usize;
+// Reserve 32 IRQs for legacy device.
+pub const IRQ_LEGACY_BASE: usize = arch::layout::IRQ_BASE as usize;
 pub const IRQ_LEGACY_COUNT: usize = 32;
-pub const GIC_SNAPSHOT_ID: &str = "gic-v3-its";
 
-// Gic (Generic Interupt Controller) struct provides all the functionality of a
-// GIC device. It wraps a hypervisor-emulated GIC device (Vgic) provided by the
-// `hypervisor` crate.
-// Gic struct also implements InterruptController to provide interrupt delivery
-// service.
+// This Gic struct implements InterruptController to provide interrupt delivery service.
+// The Gic source files in arch/ folder maintain the Aarch64 specific Gic device.
+// The 2 Gic instances could be merged together.
+// Leave this refactoring to future. Two options may be considered:
+//   1. Move Gic*.rs from arch/ folder here.
+//   2. Move this file and ioapic.rs to arch/, as they are architecture specific.
 pub struct Gic {
     interrupt_source_group: Arc<dyn InterruptSourceGroup>,
-    // The hypervisor agnostic virtual GIC
-    vgic: Option<Arc<Mutex<dyn Vgic>>>,
+    gic_device: Option<Arc<Mutex<Box<dyn GicDevice>>>>,
 }
 
 impl Gic {
     pub fn new(
-        vcpu_count: u8,
+        _vcpu_count: u8,
         interrupt_manager: Arc<dyn InterruptManager<GroupConfig = MsiIrqGroupConfig>>,
-        vm: Arc<dyn hypervisor::Vm>,
     ) -> Result<Gic> {
         let interrupt_source_group = interrupt_manager
             .create_group(MsiIrqGroupConfig {
@@ -51,34 +42,22 @@ impl Gic {
             })
             .map_err(Error::CreateInterruptSourceGroup)?;
 
-        let vgic = vm
-            .create_vgic(Gic::create_default_config(vcpu_count as u64))
-            .map_err(Error::CreateGic)?;
-
-        let gic = Gic {
+        Ok(Gic {
             interrupt_source_group,
-            vgic: Some(vgic),
-        };
-        gic.enable()?;
-
-        Ok(gic)
+            gic_device: None,
+        })
     }
 
-    pub fn restore_vgic(
-        &mut self,
-        state: Option<GicState>,
-        saved_vcpu_states: &[CpuState],
-    ) -> Result<()> {
-        self.set_gicr_typers(saved_vcpu_states);
-        self.vgic
-            .clone()
-            .unwrap()
-            .lock()
-            .unwrap()
-            .set_state(&state.unwrap())
-            .map_err(Error::RestoreGic)
+    pub fn set_gic_device(&mut self, gic_device: Arc<Mutex<Box<dyn GicDevice>>>) {
+        self.gic_device = Some(gic_device);
     }
 
+    pub fn get_gic_device(&self) -> Option<&Arc<Mutex<Box<dyn GicDevice>>>> {
+        self.gic_device.as_ref()
+    }
+}
+
+impl InterruptController for Gic {
     fn enable(&self) -> Result<()> {
         // Set irqfd for legacy interrupts
         self.interrupt_source_group
@@ -97,45 +76,12 @@ impl Gic {
                 .update(
                     i as InterruptIndex,
                     InterruptSourceConfig::LegacyIrq(config),
-                    false,
-                    false,
                 )
                 .map_err(Error::EnableInterrupt)?;
         }
-
-        self.interrupt_source_group
-            .set_gsi()
-            .map_err(Error::EnableInterrupt)?;
         Ok(())
     }
 
-    /// Default config implied by arch::layout
-    pub fn create_default_config(vcpu_count: u64) -> VgicConfig {
-        let redists_size = layout::GIC_V3_REDIST_SIZE * vcpu_count;
-        let redists_addr = layout::GIC_V3_DIST_START.raw_value() - redists_size;
-        VgicConfig {
-            vcpu_count,
-            dist_addr: layout::GIC_V3_DIST_START.raw_value(),
-            dist_size: layout::GIC_V3_DIST_SIZE,
-            redists_addr,
-            redists_size,
-            msi_addr: redists_addr - layout::GIC_V3_ITS_SIZE,
-            msi_size: layout::GIC_V3_ITS_SIZE,
-            nr_irqs: layout::IRQ_NUM,
-        }
-    }
-
-    pub fn get_vgic(&mut self) -> Result<Arc<Mutex<dyn Vgic>>> {
-        Ok(self.vgic.clone().unwrap())
-    }
-
-    pub fn set_gicr_typers(&mut self, vcpu_states: &[CpuState]) {
-        let vgic = self.vgic.as_ref().unwrap().clone();
-        vgic.lock().unwrap().set_gicr_typers(vcpu_states);
-    }
-}
-
-impl InterruptController for Gic {
     // This should be called anytime an interrupt needs to be injected into the
     // running guest.
     fn service_irq(&mut self, irq: usize) -> Result<()> {
@@ -150,31 +96,3 @@ impl InterruptController for Gic {
         self.interrupt_source_group.notifier(irq as InterruptIndex)
     }
 }
-
-impl Snapshottable for Gic {
-    fn id(&self) -> String {
-        GIC_SNAPSHOT_ID.to_string()
-    }
-
-    fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        let vgic = self.vgic.as_ref().unwrap().clone();
-        let state = vgic.lock().unwrap().state().unwrap();
-        Snapshot::new_from_state(&state)
-    }
-}
-
-impl Pausable for Gic {
-    fn pause(&mut self) -> std::result::Result<(), MigratableError> {
-        // Flush tables to guest RAM
-        let vgic = self.vgic.as_ref().unwrap().clone();
-        vgic.lock().unwrap().save_data_tables().map_err(|e| {
-            MigratableError::Pause(anyhow!(
-                "Could not save GICv3ITS GIC pending tables {:?}",
-                e
-            ))
-        })?;
-        Ok(())
-    }
-}
-impl Transportable for Gic {}
-impl Migratable for Gic {}

devices/src/interrupt_controller.rs

@@ -24,12 +24,6 @@ pub enum Error {
     UpdateInterrupt(io::Error),
     /// Failed enabling the interrupt.
     EnableInterrupt(io::Error),
-    #[cfg(target_arch = "aarch64")]
-    /// Failed creating GIC device.
-    CreateGic(hypervisor::HypervisorVmError),
-    #[cfg(target_arch = "aarch64")]
-    /// Failed restoring GIC device.
-    RestoreGic(hypervisor::arch::aarch64::gic::Error),
 }
 
 type Result<T> = result::Result<T, Error>;
@@ -59,6 +53,8 @@ pub struct MsiMessage {
 // IOAPIC (X86) or GIC (Arm).
 pub trait InterruptController: Send {
     fn service_irq(&mut self, irq: usize) -> Result<()>;
+    #[cfg(target_arch = "aarch64")]
+    fn enable(&self) -> Result<()>;
     #[cfg(target_arch = "x86_64")]
     fn end_of_interrupt(&mut self, vec: u8);
     fn notifier(&self, irq: usize) -> Option<EventFd>;

devices/src/ioapic.rs

@@ -10,6 +10,7 @@
 // See https://pdos.csail.mit.edu/6.828/2016/readings/ia32/ioapic.pdf for a specification.
 
 use super::interrupt_controller::{Error, InterruptController};
+use anyhow::anyhow;
 use byteorder::{ByteOrder, LittleEndian};
 use std::result;
 use std::sync::{Arc, Barrier};
@@ -193,7 +194,6 @@ impl Ioapic {
         id: String,
         apic_address: GuestAddress,
         interrupt_manager: Arc<dyn InterruptManager<GroupConfig = MsiIrqGroupConfig>>,
-        state: Option<IoapicState>,
     ) -> Result<Ioapic> {
         let interrupt_source_group = interrupt_manager
             .create_group(MsiIrqGroupConfig {
@@ -202,52 +202,17 @@ impl Ioapic {
             })
             .map_err(Error::CreateInterruptSourceGroup)?;
 
-        let (id_reg, reg_sel, reg_entries, used_entries, apic_address) = if let Some(state) = &state
-        {
-            (
-                state.id_reg,
-                state.reg_sel,
-                state.reg_entries,
-                state.used_entries,
-                GuestAddress(state.apic_address),
-            )
-        } else {
-            (
-                0,
-                0,
-                [0x10000; NUM_IOAPIC_PINS],
-                [false; NUM_IOAPIC_PINS],
-                apic_address,
-            )
-        };
-
         // The IOAPIC is created with entries already masked. The guest will be
         // in charge of unmasking them if/when necessary.
-        let ioapic = Ioapic {
+        Ok(Ioapic {
             id,
-            id_reg,
-            reg_sel,
-            reg_entries,
-            used_entries,
+            id_reg: 0,
+            reg_sel: 0,
+            reg_entries: [0x10000; NUM_IOAPIC_PINS],
+            used_entries: [false; NUM_IOAPIC_PINS],
             apic_address,
             interrupt_source_group,
-        };
-
-        // When restoring the Ioapic, we must enable used entries.
-        if state.is_some() {
-            for (irq, entry) in ioapic.used_entries.iter().enumerate() {
-                if *entry {
-                    ioapic.update_entry(irq, false)?;
-                }
-            }
-
-            ioapic
-                .interrupt_source_group
-                .set_gsi()
-                .map_err(Error::UpdateInterrupt)?;
-        }
-
-        Ok(ioapic)
+        })
     }
 
     fn ioapic_write(&mut self, val: u32) {
@@ -283,7 +248,7 @@ impl Ioapic {
                 }
                 // The entry must be updated through the interrupt source
                 // group.
-                if let Err(e) = self.update_entry(index, true) {
+                if let Err(e) = self.update_entry(index) {
                     error!("Failed updating IOAPIC entry: {:?}", e);
                 }
                 // Store the information this IRQ is now being used.
@@ -334,7 +299,22 @@ impl Ioapic {
         }
     }
 
-    fn update_entry(&self, irq: usize, set_gsi: bool) -> Result<()> {
+    fn set_state(&mut self, state: &IoapicState) -> Result<()> {
+        self.id_reg = state.id_reg;
+        self.reg_sel = state.reg_sel;
+        self.reg_entries = state.reg_entries;
+        self.used_entries = state.used_entries;
+        self.apic_address = GuestAddress(state.apic_address);
+        for (irq, entry) in self.used_entries.iter().enumerate() {
+            if *entry {
+                self.update_entry(irq)?;
+            }
+        }
+
+        Ok(())
+    }
+
+    fn update_entry(&self, irq: usize) -> Result<()> {
         let entry = self.reg_entries[irq];
 
         // Validate Destination Mode value, and retrieve Destination ID
@@ -387,14 +367,19 @@ impl Ioapic {
         };
 
         self.interrupt_source_group
-            .update(
-                irq as InterruptIndex,
-                InterruptSourceConfig::MsiIrq(config),
-                interrupt_mask(entry) == 1,
-                set_gsi,
-            )
+            .update(irq as InterruptIndex, InterruptSourceConfig::MsiIrq(config))
             .map_err(Error::UpdateInterrupt)?;
 
+        if interrupt_mask(entry) == 1 {
+            self.interrupt_source_group
+                .mask(irq as InterruptIndex)
+                .map_err(Error::MaskInterrupt)?;
+        } else {
+            self.interrupt_source_group
+                .unmask(irq as InterruptIndex)
+                .map_err(Error::UnmaskInterrupt)?;
+        }
+
         Ok(())
     }
 }
@@ -444,7 +429,18 @@ impl Snapshottable for Ioapic {
     }
 
     fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        Snapshot::new_from_versioned_state(&self.state())
+        Snapshot::new_from_versioned_state(&self.id, &self.state())
+    }
+
+    fn restore(&mut self, snapshot: Snapshot) -> std::result::Result<(), MigratableError> {
+        self.set_state(&snapshot.to_versioned_state(&self.id)?)
+            .map_err(|e| {
+                MigratableError::Restore(anyhow!(
+                    "Could not restore state for {}: {:?}",
+                    self.id,
+                    e
+                ))
+            })
     }
 }
 

devices/src/legacy/cmos.rs

@@ -5,11 +5,8 @@
 use libc::{clock_gettime, gmtime_r, timespec, tm, CLOCK_REALTIME};
 use std::cmp::min;
 use std::mem;
-use std::sync::atomic::{AtomicBool, Ordering};
 use std::sync::{Arc, Barrier};
-use std::thread;
 use vm_device::BusDevice;
-use vmm_sys_util::eventfd::EventFd;
 
 // https://github.com/rust-lang/libc/issues/1848
 #[cfg_attr(target_env = "musl", allow(deprecated))]
@@ -24,20 +21,13 @@ const DATA_LEN: usize = 128;
 pub struct Cmos {
     index: u8,
     data: [u8; DATA_LEN],
-    reset_evt: EventFd,
-    vcpus_kill_signalled: Option<Arc<AtomicBool>>,
 }
 
 impl Cmos {
     /// Constructs a CMOS/RTC device with initial data.
     /// `mem_below_4g` is the size of memory in bytes below the 32-bit gap.
     /// `mem_above_4g` is the size of memory in bytes above the 32-bit gap.
-    pub fn new(
-        mem_below_4g: u64,
-        mem_above_4g: u64,
-        reset_evt: EventFd,
-        vcpus_kill_signalled: Option<Arc<AtomicBool>>,
-    ) -> Cmos {
+    pub fn new(mem_below_4g: u64, mem_above_4g: u64) -> Cmos {
         let mut data = [0u8; DATA_LEN];
 
         // Extended memory from 16 MB to 4 GB in units of 64 KB
@@ -54,12 +44,7 @@ impl Cmos {
         data[0x5c] = (high_mem >> 8) as u8;
         data[0x5d] = (high_mem >> 16) as u8;
 
-        Cmos {
-            index: 0,
-            data,
-            reset_evt,
-            vcpus_kill_signalled,
-        }
+        Cmos { index: 0, data }
     }
 }
 
@@ -71,24 +56,8 @@ impl BusDevice for Cmos {
         }
 
         match offset {
-            INDEX_OFFSET => self.index = data[0],
-            DATA_OFFSET => {
-                if self.index == 0x8f && data[0] == 0 {
-                    info!("CMOS reset");
-                    self.reset_evt.write(1).unwrap();
-                    if let Some(vcpus_kill_signalled) = self.vcpus_kill_signalled.take() {
-                        // Spin until we are sure the reset_evt has been handled and that when
-                        // we return from the KVM_RUN we will exit rather than re-enter the guest.
-                        while !vcpus_kill_signalled.load(Ordering::SeqCst) {
-                            // This is more effective than thread::yield_now() at
-                            // avoiding a priority inversion with the VMM thread
-                            thread::sleep(std::time::Duration::from_millis(1));
-                        }
-                    }
-                } else {
-                    self.data[(self.index & INDEX_MASK) as usize] = data[0]
-                }
-            }
+            INDEX_OFFSET => self.index = data[0] & INDEX_MASK,
+            DATA_OFFSET => self.data[self.index as usize] = data[0],
             o => warn!("bad write offset on CMOS device: {}", o),
         };
         None
@@ -115,7 +84,7 @@ impl BusDevice for Cmos {
                 let day;
                 let month;
                 let year;
-                // SAFETY: The clock_gettime and gmtime_r calls are safe as long as the structs they are
+                // The clock_gettime and gmtime_r calls are safe as long as the structs they are
                 // given are large enough, and neither of them fail. It is safe to zero initialize
                 // the tm and timespec struct because it contains only plain data.
                 let update_in_progress = unsafe {

devices/src/legacy/debug_port.rs

@@ -1,86 +0,0 @@
-// Copyright © 2022 Intel Corporation
-//
-// SPDX-License-Identifier: Apache-2.0
-//
-
-use std::fmt;
-use std::time::Instant;
-use vm_device::BusDevice;
-
-/// Debug I/O port, see:
-/// https://www.intel.com/content/www/us/en/support/articles/000005500/boards-and-kits.html
-///
-/// Since we're not a physical platform, we can freely assign code ranges for
-/// debugging specific parts of our virtual platform.
-pub enum DebugIoPortRange {
-    Firmware,
-    Bootloader,
-    Kernel,
-    Userspace,
-    Custom,
-}
-
-#[cfg(target_arch = "x86_64")]
-const DEBUG_IOPORT_PREFIX: &str = "Debug I/O port";
-
-#[cfg(target_arch = "x86_64")]
-impl DebugIoPortRange {
-    fn from_u8(value: u8) -> DebugIoPortRange {
-        match value {
-            0x00..=0x1f => DebugIoPortRange::Firmware,
-            0x20..=0x3f => DebugIoPortRange::Bootloader,
-            0x40..=0x5f => DebugIoPortRange::Kernel,
-            0x60..=0x7f => DebugIoPortRange::Userspace,
-            _ => DebugIoPortRange::Custom,
-        }
-    }
-}
-
-#[cfg(target_arch = "x86_64")]
-impl fmt::Display for DebugIoPortRange {
-    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        match self {
-            DebugIoPortRange::Firmware => write!(f, "{DEBUG_IOPORT_PREFIX}: Firmware"),
-            DebugIoPortRange::Bootloader => write!(f, "{DEBUG_IOPORT_PREFIX}: Bootloader"),
-            DebugIoPortRange::Kernel => write!(f, "{DEBUG_IOPORT_PREFIX}: Kernel"),
-            DebugIoPortRange::Userspace => write!(f, "{DEBUG_IOPORT_PREFIX}: Userspace"),
-            DebugIoPortRange::Custom => write!(f, "{DEBUG_IOPORT_PREFIX}: Custom"),
-        }
-    }
-}
-
-pub struct DebugPort {
-    timestamp: Instant,
-}
-
-impl DebugPort {
-    pub fn new(timestamp: Instant) -> Self {
-        Self { timestamp }
-    }
-}
-
-impl BusDevice for DebugPort {
-    fn read(&mut self, _base: u64, _offset: u64, _data: &mut [u8]) {
-        error!("Invalid read to debug port")
-    }
-
-    fn write(
-        &mut self,
-        _base: u64,
-        _offset: u64,
-        data: &[u8],
-    ) -> Option<std::sync::Arc<std::sync::Barrier>> {
-        let elapsed = self.timestamp.elapsed();
-
-        let code = data[0];
-        warn!(
-            "[{} code 0x{:x}] {}.{:>06} seconds",
-            DebugIoPortRange::from_u8(code),
-            code,
-            elapsed.as_secs(),
-            elapsed.as_micros()
-        );
-
-        None
-    }
-}

devices/src/legacy/gpio_pl061.rs

@@ -51,13 +51,13 @@ pub enum Error {
 impl fmt::Display for Error {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         match self {
-            Error::BadWriteOffset(offset) => write!(f, "Bad Write Offset: {offset}"),
+            Error::BadWriteOffset(offset) => write!(f, "Bad Write Offset: {}", offset),
             Error::GpioInterruptDisabled => write!(f, "GPIO interrupt disabled by guest driver.",),
             Error::GpioInterruptFailure(ref e) => {
-                write!(f, "Could not trigger GPIO interrupt: {e}.")
+                write!(f, "Could not trigger GPIO interrupt: {}.", e)
             }
             Error::GpioTriggerKeyFailure(key) => {
-                write!(f, "Invalid GPIO Input key triggerd: {key}.")
+                write!(f, "Invalid GPIO Input key triggerd: {}.", key)
             }
         }
     }
@@ -106,39 +106,18 @@ impl VersionMapped for GpioState {}
 
 impl Gpio {
     /// Constructs an PL061 GPIO device.
-    pub fn new(
-        id: String,
-        interrupt: Arc<dyn InterruptSourceGroup>,
-        state: Option<GpioState>,
-    ) -> Self {
-        let (data, old_in_data, dir, isense, ibe, iev, im, istate, afsel) =
-            if let Some(state) = state {
-                (
-                    state.data,
-                    state.old_in_data,
-                    state.dir,
-                    state.isense,
-                    state.ibe,
-                    state.iev,
-                    state.im,
-                    state.istate,
-                    state.afsel,
-                )
-            } else {
-                (0, 0, 0, 0, 0, 0, 0, 0, 0)
-            };
-
+    pub fn new(id: String, interrupt: Arc<dyn InterruptSourceGroup>) -> Self {
         Self {
             id,
-            data,
-            old_in_data,
-            dir,
-            isense,
-            ibe,
-            iev,
-            im,
-            istate,
-            afsel,
+            data: 0,
+            old_in_data: 0,
+            dir: 0,
+            isense: 0,
+            ibe: 0,
+            iev: 0,
+            im: 0,
+            istate: 0,
+            afsel: 0,
             interrupt,
         }
     }
@@ -157,12 +136,24 @@ impl Gpio {
         }
     }
 
+    fn set_state(&mut self, state: &GpioState) {
+        self.data = state.data;
+        self.old_in_data = state.old_in_data;
+        self.dir = state.dir;
+        self.isense = state.isense;
+        self.ibe = state.ibe;
+        self.iev = state.iev;
+        self.im = state.im;
+        self.istate = state.istate;
+        self.afsel = state.afsel;
+    }
+
     fn pl061_internal_update(&mut self) {
         // FIXME:
         //  Missing Output Interrupt Emulation.
 
         // Input Edging Interrupt Emulation.
-        let changed = (self.old_in_data ^ self.data) & !self.dir;
+        let changed = ((self.old_in_data ^ self.data) & !self.dir) as u32;
         if changed > 0 {
             self.old_in_data = self.data;
             for i in 0..N_GPIOS {
@@ -328,7 +319,12 @@ impl Snapshottable for Gpio {
     }
 
     fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        Snapshot::new_from_versioned_state(&self.state())
+        Snapshot::new_from_versioned_state(&self.id, &self.state())
+    }
+
+    fn restore(&mut self, snapshot: Snapshot) -> std::result::Result<(), MigratableError> {
+        self.set_state(&snapshot.to_versioned_state(&self.id)?);
+        Ok(())
     }
 }
 
@@ -360,16 +356,10 @@ mod tests {
             &self,
             _index: InterruptIndex,
             _config: InterruptSourceConfig,
-            _masked: bool,
-            _set_gsi: bool,
         ) -> result::Result<(), std::io::Error> {
             Ok(())
         }
 
-        fn set_gsi(&self) -> result::Result<(), std::io::Error> {
-            Ok(())
-        }
-
         fn notifier(&self, _index: InterruptIndex) -> Option<EventFd> {
             Some(self.event_fd.try_clone().unwrap())
         }
@@ -387,7 +377,6 @@ mod tests {
         let mut gpio = Gpio::new(
             String::from(GPIO_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
-            None,
         );
         let mut data = [0; 4];
 

devices/src/legacy/i8042.rs

@@ -2,27 +2,19 @@
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE-BSD-3-Clause file.
 
-use std::sync::{
-    atomic::{AtomicBool, Ordering},
-    Arc, Barrier,
-};
-use std::thread;
+use std::sync::{Arc, Barrier};
 use vm_device::BusDevice;
 use vmm_sys_util::eventfd::EventFd;
 
 /// A i8042 PS/2 controller that emulates just enough to shutdown the machine.
 pub struct I8042Device {
     reset_evt: EventFd,
-    vcpus_kill_signalled: Arc<AtomicBool>,
 }
 
 impl I8042Device {
     /// Constructs a i8042 device that will signal the given event when the guest requests it.
-    pub fn new(reset_evt: EventFd, vcpus_kill_signalled: Arc<AtomicBool>) -> I8042Device {
-        I8042Device {
-            reset_evt,
-            vcpus_kill_signalled,
-        }
+    pub fn new(reset_evt: EventFd) -> I8042Device {
+        I8042Device { reset_evt }
     }
 }
 
@@ -46,13 +38,6 @@ impl BusDevice for I8042Device {
             if let Err(e) = self.reset_evt.write(1) {
                 error!("Error triggering i8042 reset event: {}", e);
             }
-            // Spin until we are sure the reset_evt has been handled and that when
-            // we return from the KVM_RUN we will exit rather than re-enter the guest.
-            while !self.vcpus_kill_signalled.load(Ordering::SeqCst) {
-                // This is more effective than thread::yield_now() at
-                // avoiding a priority inversion with the VMM thread
-                thread::sleep(std::time::Duration::from_millis(1));
-            }
         }
 
         None

devices/src/legacy/mod.rs

@@ -5,10 +5,9 @@
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE-BSD-3-Clause file.
 
+#[cfg(feature = "cmos")]
 mod cmos;
-#[cfg(target_arch = "x86_64")]
-mod debug_port;
-#[cfg(target_arch = "x86_64")]
+#[cfg(feature = "fwdebug")]
 mod fwdebug;
 #[cfg(target_arch = "aarch64")]
 mod gpio_pl061;
@@ -19,10 +18,9 @@ mod serial;
 #[cfg(target_arch = "aarch64")]
 mod uart_pl011;
 
+#[cfg(feature = "cmos")]
 pub use self::cmos::Cmos;
-#[cfg(target_arch = "x86_64")]
-pub use self::debug_port::DebugPort;
-#[cfg(target_arch = "x86_64")]
+#[cfg(feature = "fwdebug")]
 pub use self::fwdebug::FwDebugDevice;
 pub use self::i8042::I8042Device;
 pub use self::serial::Serial;

devices/src/legacy/rtc_pl031.rs

@@ -48,8 +48,8 @@ pub enum Error {
 impl fmt::Display for Error {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         match self {
-            Error::BadWriteOffset(offset) => write!(f, "Bad Write Offset: {offset}"),
-            Error::InterruptFailure(e) => write!(f, "Failed to trigger interrupt: {e}"),
+            Error::BadWriteOffset(offset) => write!(f, "Bad Write Offset: {}", offset),
+            Error::InterruptFailure(e) => write!(f, "Failed to trigger interrupt: {}", e),
         }
     }
 }
@@ -61,10 +61,12 @@ pub enum ClockType {
     /// Equivalent to `libc::CLOCK_MONOTONIC`.
     Monotonic,
     /// Equivalent to `libc::CLOCK_REALTIME`.
+    #[allow(dead_code)]
     Real,
     /// Equivalent to `libc::CLOCK_PROCESS_CPUTIME_ID`.
     ProcessCpu,
     /// Equivalent to `libc::CLOCK_THREAD_CPUTIME_ID`.
+    #[allow(dead_code)]
     ThreadCpu,
 }
 
@@ -99,7 +101,7 @@ pub struct LocalTime {
 
 impl LocalTime {
     /// Returns the [LocalTime](struct.LocalTime.html) structure for the calling moment.
-    #[cfg(test)]
+    #[allow(dead_code)]
     pub fn now() -> LocalTime {
         let mut timespec = libc::timespec {
             tv_sec: 0,
@@ -119,7 +121,7 @@ impl LocalTime {
             tm_zone: std::ptr::null(),
         };
 
-        // SAFETY: the parameters are valid.
+        // Safe because the parameters are valid.
         unsafe {
             libc::clock_gettime(libc::CLOCK_REALTIME, &mut timespec);
             libc::localtime_r(&timespec.tv_sec, &mut tm);
@@ -171,6 +173,22 @@ impl Default for TimestampUs {
     }
 }
 
+/// Returns a timestamp in nanoseconds from a monotonic clock.
+///
+/// Uses `_rdstc` on `x86_64` and [`get_time`](fn.get_time.html) on other architectures.
+#[allow(dead_code)]
+pub fn timestamp_cycles() -> u64 {
+    #[cfg(target_arch = "x86_64")]
+    // Safe because there's nothing that can go wrong with this call.
+    unsafe {
+        std::arch::x86_64::_rdtsc() as u64
+    }
+    #[cfg(not(target_arch = "x86_64"))]
+    {
+        get_time(ClockType::Monotonic)
+    }
+}
+
 /// Returns a timestamp in nanoseconds based on the provided clock type.
 ///
 /// # Arguments
@@ -181,7 +199,7 @@ pub fn get_time(clock_type: ClockType) -> u64 {
         tv_sec: 0,
         tv_nsec: 0,
     };
-    // SAFETY: the parameters are valid.
+    // Safe because the parameters are valid.
     unsafe { libc::clock_gettime(clock_type.into(), &mut time_struct) };
     seconds_to_nanoseconds(time_struct.tv_sec).unwrap() as u64 + (time_struct.tv_nsec as u64)
 }
@@ -412,16 +430,10 @@ mod tests {
             &self,
             _index: InterruptIndex,
             _config: InterruptSourceConfig,
-            _masked: bool,
-            _set_gsi: bool,
         ) -> result::Result<(), std::io::Error> {
             Ok(())
         }
 
-        fn set_gsi(&self) -> result::Result<(), std::io::Error> {
-            Ok(())
-        }
-
         fn notifier(&self, _index: InterruptIndex) -> Option<EventFd> {
             Some(self.event_fd.try_clone().unwrap())
         }
@@ -512,6 +524,7 @@ mod tests {
         ($test_name: ident, $write_fn_name: ident, $read_fn_name: ident, $is_be: expr, $data_type: ty) => {
             #[test]
             fn $test_name() {
+                #[allow(overflowing_literals)]
                 let test_cases = [
                     (
                         0x0123_4567_89AB_CDEF as u64,

devices/src/legacy/serial.rs

@@ -63,6 +63,7 @@ pub struct Serial {
     id: String,
     interrupt_enable: u8,
     interrupt_identification: u8,
+    interrupt: Arc<dyn InterruptSourceGroup>,
     line_control: u8,
     line_status: u8,
     modem_control: u8,
@@ -70,7 +71,6 @@ pub struct Serial {
     scratch: u8,
     baud_divisor: u16,
     in_buffer: VecDeque<u8>,
-    interrupt: Arc<dyn InterruptSourceGroup>,
     out: Option<Box<dyn io::Write + Send>>,
 }
 
@@ -93,56 +93,19 @@ impl Serial {
         id: String,
         interrupt: Arc<dyn InterruptSourceGroup>,
         out: Option<Box<dyn io::Write + Send>>,
-        state: Option<SerialState>,
     ) -> Serial {
-        let (
-            interrupt_enable,
-            interrupt_identification,
-            line_control,
-            line_status,
-            modem_control,
-            modem_status,
-            scratch,
-            baud_divisor,
-            in_buffer,
-        ) = if let Some(state) = state {
-            (
-                state.interrupt_enable,
-                state.interrupt_identification,
-                state.line_control,
-                state.line_status,
-                state.modem_control,
-                state.modem_status,
-                state.scratch,
-                state.baud_divisor,
-                state.in_buffer.into(),
-            )
-        } else {
-            (
-                0,
-                DEFAULT_INTERRUPT_IDENTIFICATION,
-                DEFAULT_LINE_CONTROL,
-                DEFAULT_LINE_STATUS,
-                DEFAULT_MODEM_CONTROL,
-                DEFAULT_MODEM_STATUS,
-                0,
-                DEFAULT_BAUD_DIVISOR,
-                VecDeque::new(),
-            )
-        };
-
         Serial {
             id,
-            interrupt_enable,
-            interrupt_identification,
-            line_control,
-            line_status,
-            modem_control,
-            modem_status,
-            scratch,
-            baud_divisor,
-            in_buffer,
+            interrupt_enable: 0,
+            interrupt_identification: DEFAULT_INTERRUPT_IDENTIFICATION,
             interrupt,
+            line_control: DEFAULT_LINE_CONTROL,
+            line_status: DEFAULT_LINE_STATUS,
+            modem_control: DEFAULT_MODEM_CONTROL,
+            modem_status: DEFAULT_MODEM_STATUS,
+            scratch: 0,
+            baud_divisor: DEFAULT_BAUD_DIVISOR,
+            in_buffer: VecDeque::new(),
             out,
         }
     }
@@ -152,18 +115,13 @@ impl Serial {
         id: String,
         interrupt: Arc<dyn InterruptSourceGroup>,
         out: Box<dyn io::Write + Send>,
-        state: Option<SerialState>,
     ) -> Serial {
-        Self::new(id, interrupt, Some(out), state)
+        Self::new(id, interrupt, Some(out))
     }
 
     /// Constructs a Serial port with no connected output.
-    pub fn new_sink(
-        id: String,
-        interrupt: Arc<dyn InterruptSourceGroup>,
-        state: Option<SerialState>,
-    ) -> Serial {
-        Self::new(id, interrupt, None, state)
+    pub fn new_sink(id: String, interrupt: Arc<dyn InterruptSourceGroup>) -> Serial {
+        Self::new(id, interrupt, None)
     }
 
     pub fn set_out(&mut self, out: Box<dyn io::Write + Send>) {
@@ -241,7 +199,7 @@ impl Serial {
     }
 
     fn handle_write(&mut self, offset: u8, v: u8) -> Result<()> {
-        match offset {
+        match offset as u8 {
             DLAB_LOW if self.is_dlab_set() => {
                 self.baud_divisor = (self.baud_divisor & 0xff00) | u16::from(v)
             }
@@ -284,6 +242,18 @@ impl Serial {
             in_buffer: self.in_buffer.clone().into(),
         }
     }
+
+    fn set_state(&mut self, state: &SerialState) {
+        self.interrupt_enable = state.interrupt_enable;
+        self.interrupt_identification = state.interrupt_identification;
+        self.line_control = state.line_control;
+        self.line_status = state.line_status;
+        self.modem_control = state.modem_control;
+        self.modem_status = state.modem_status;
+        self.scratch = state.scratch;
+        self.baud_divisor = state.baud_divisor;
+        self.in_buffer = state.in_buffer.clone().into();
+    }
 }
 
 impl BusDevice for Serial {
@@ -334,7 +304,12 @@ impl Snapshottable for Serial {
     }
 
     fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        Snapshot::new_from_versioned_state(&self.state())
+        Snapshot::new_from_versioned_state(&self.id, &self.state())
+    }
+
+    fn restore(&mut self, snapshot: Snapshot) -> std::result::Result<(), MigratableError> {
+        self.set_state(&snapshot.to_versioned_state(&self.id)?);
+        Ok(())
     }
 }
 
@@ -364,14 +339,9 @@ mod tests {
             &self,
             _index: InterruptIndex,
             _config: InterruptSourceConfig,
-            _masked: bool,
-            _set_gsi: bool,
         ) -> result::Result<(), std::io::Error> {
             Ok(())
         }
-        fn set_gsi(&self) -> result::Result<(), std::io::Error> {
-            Ok(())
-        }
         fn notifier(&self, _index: InterruptIndex) -> Option<EventFd> {
             Some(self.event_fd.try_clone().unwrap())
         }
@@ -413,7 +383,6 @@ mod tests {
             String::from(SERIAL_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
             Box::new(serial_out.clone()),
-            None,
         );
 
         serial.write(0, DATA as u64, &[b'x', b'y']);
@@ -434,7 +403,6 @@ mod tests {
             String::from(SERIAL_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
             Box::new(serial_out),
-            None,
         );
 
         // write 1 to the interrupt event fd, so that read doesn't block in case the event fd
@@ -471,7 +439,6 @@ mod tests {
         let mut serial = Serial::new_sink(
             String::from(SERIAL_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
-            None,
         );
 
         // write 1 to the interrupt event fd, so that read doesn't block in case the event fd
@@ -494,7 +461,6 @@ mod tests {
         let mut serial = Serial::new_sink(
             String::from(SERIAL_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
-            None,
         );
 
         serial.write(0, LCR as u64, &[LCR_DLAB_BIT]);
@@ -516,7 +482,6 @@ mod tests {
         let mut serial = Serial::new_sink(
             String::from(SERIAL_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
-            None,
         );
 
         serial.write(0, MCR as u64, &[MCR_LOOP_BIT]);
@@ -543,7 +508,6 @@ mod tests {
         let mut serial = Serial::new_sink(
             String::from(SERIAL_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
-            None,
         );
 
         serial.write(0, SCR as u64, &[0x12]);

devices/src/legacy/uart_pl011.rs

@@ -10,7 +10,6 @@ use crate::{read_le_u32, write_le_u32};
 use std::collections::VecDeque;
 use std::fmt;
 use std::sync::{Arc, Barrier};
-use std::time::Instant;
 use std::{io, result};
 use versionize::{VersionMap, Versionize, VersionizeResult};
 use versionize_derive::Versionize;
@@ -35,7 +34,6 @@ const UARTRIS: u64 = 15;
 const UARTMIS: u64 = 16;
 const UARTICR: u64 = 17;
 const UARTDMACR: u64 = 18;
-const UARTDEBUG: u64 = 0x3c0;
 
 const PL011_INT_TX: u32 = 0x20;
 const PL011_INT_RX: u32 = 0x10;
@@ -60,11 +58,11 @@ pub enum Error {
 impl fmt::Display for Error {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         match self {
-            Error::BadWriteOffset(offset) => write!(f, "pl011_write: Bad Write Offset: {offset}"),
+            Error::BadWriteOffset(offset) => write!(f, "pl011_write: Bad Write Offset: {}", offset),
             Error::DmaNotImplemented => write!(f, "pl011: DMA not implemented."),
-            Error::InterruptFailure(e) => write!(f, "Failed to trigger interrupt: {e}"),
-            Error::WriteAllFailure(e) => write!(f, "Failed to write: {e}"),
-            Error::FlushFailure(e) => write!(f, "Failed to flush: {e}"),
+            Error::InterruptFailure(e) => write!(f, "Failed to trigger interrupt: {}", e),
+            Error::WriteAllFailure(e) => write!(f, "Failed to write: {}", e),
+            Error::FlushFailure(e) => write!(f, "Failed to flush: {}", e),
         }
     }
 }
@@ -79,7 +77,6 @@ pub struct Pl011 {
     rsr: u32,
     cr: u32,
     dmacr: u32,
-    debug: u32,
     int_enabled: u32,
     int_level: u32,
     read_fifo: VecDeque<u8>,
@@ -91,7 +88,6 @@ pub struct Pl011 {
     read_trigger: u32,
     irq: Arc<dyn InterruptSourceGroup>,
     out: Option<Box<dyn io::Write + Send>>,
-    timestamp: std::time::Instant,
 }
 
 #[derive(Versionize)]
@@ -101,7 +97,6 @@ pub struct Pl011State {
     rsr: u32,
     cr: u32,
     dmacr: u32,
-    debug: u32,
     int_enabled: u32,
     int_level: u32,
     read_fifo: Vec<u8>,
@@ -121,83 +116,25 @@ impl Pl011 {
         id: String,
         irq: Arc<dyn InterruptSourceGroup>,
         out: Option<Box<dyn io::Write + Send>>,
-        timestamp: Instant,
-        state: Option<Pl011State>,
     ) -> Self {
-        let (
-            flags,
-            lcr,
-            rsr,
-            cr,
-            dmacr,
-            debug,
-            int_enabled,
-            int_level,
-            read_fifo,
-            ilpr,
-            ibrd,
-            fbrd,
-            ifl,
-            read_count,
-            read_trigger,
-        ) = if let Some(state) = state {
-            (
-                state.flags,
-                state.lcr,
-                state.rsr,
-                state.cr,
-                state.dmacr,
-                state.debug,
-                state.int_enabled,
-                state.int_level,
-                state.read_fifo.into(),
-                state.ilpr,
-                state.ibrd,
-                state.fbrd,
-                state.ifl,
-                state.read_count,
-                state.read_trigger,
-            )
-        } else {
-            (
-                0x90,
-                0,
-                0,
-                0x300,
-                0,
-                0,
-                0,
-                0,
-                VecDeque::new(),
-                0,
-                0,
-                0,
-                0x12,
-                0,
-                1,
-            )
-        };
-
         Self {
             id,
-            flags,
-            lcr,
-            rsr,
-            cr,
-            dmacr,
-            debug,
-            int_enabled,
-            int_level,
-            read_fifo,
-            ilpr,
-            ibrd,
-            fbrd,
-            ifl,
-            read_count,
-            read_trigger,
+            flags: 0x90u32,
+            lcr: 0u32,
+            rsr: 0u32,
+            cr: 0x300u32,
+            dmacr: 0u32,
+            int_enabled: 0u32,
+            int_level: 0u32,
+            read_fifo: VecDeque::new(),
+            ilpr: 0u32,
+            ibrd: 0u32,
+            fbrd: 0u32,
+            ifl: 0x12u32,
+            read_count: 0u32,
+            read_trigger: 1u32,
             irq,
             out,
-            timestamp,
         }
     }
 
@@ -212,7 +149,6 @@ impl Pl011 {
             rsr: self.rsr,
             cr: self.cr,
             dmacr: self.dmacr,
-            debug: self.debug,
             int_enabled: self.int_enabled,
             int_level: self.int_level,
             read_fifo: self.read_fifo.clone().into(),
@@ -225,6 +161,23 @@ impl Pl011 {
         }
     }
 
+    fn set_state(&mut self, state: &Pl011State) {
+        self.flags = state.flags;
+        self.lcr = state.lcr;
+        self.rsr = state.rsr;
+        self.cr = state.cr;
+        self.dmacr = state.dmacr;
+        self.int_enabled = state.int_enabled;
+        self.int_level = state.int_level;
+        self.read_fifo = state.read_fifo.clone().into();
+        self.ilpr = state.ilpr;
+        self.ibrd = state.ibrd;
+        self.fbrd = state.fbrd;
+        self.ifl = state.ifl;
+        self.read_count = state.read_count;
+        self.read_trigger = state.read_trigger;
+    }
+
     /// Queues raw bytes for the guest to read and signals the interrupt
     pub fn queue_input_bytes(&mut self, c: &[u8]) -> vmm_sys_util::errno::Result<()> {
         self.read_fifo.extend(c);
@@ -327,50 +280,13 @@ impl Pl011 {
                     return Err(Error::DmaNotImplemented);
                 }
             }
-            UARTDEBUG => {
-                self.debug = val;
-                self.handle_debug();
-            }
             off => {
-                debug!("PL011: Bad write offset, offset: {}", off);
                 return Err(Error::BadWriteOffset(off));
             }
         }
         Ok(())
     }
 
-    fn handle_debug(&self) {
-        let elapsed = self.timestamp.elapsed();
-
-        match self.debug {
-            0x00..=0x1f => warn!(
-                "[Debug I/O port: Firmware code: 0x{:x}] {}.{:>06} seconds",
-                self.debug,
-                elapsed.as_secs(),
-                elapsed.as_micros()
-            ),
-            0x20..=0x3f => warn!(
-                "[Debug I/O port: Bootloader code: 0x{:x}] {}.{:>06} seconds",
-                self.debug,
-                elapsed.as_secs(),
-                elapsed.as_micros()
-            ),
-            0x40..=0x5f => warn!(
-                "[Debug I/O port: Kernel code: 0x{:x}] {}.{:>06} seconds",
-                self.debug,
-                elapsed.as_secs(),
-                elapsed.as_micros()
-            ),
-            0x60..=0x7f => warn!(
-                "[Debug I/O port: Userspace code: 0x{:x}] {}.{:>06} seconds",
-                self.debug,
-                elapsed.as_secs(),
-                elapsed.as_micros()
-            ),
-            _ => {}
-        }
-    }
-
     fn trigger_interrupt(&mut self) -> result::Result<(), io::Error> {
         self.irq.trigger(0)
     }
@@ -409,9 +325,8 @@ impl BusDevice for Pl011 {
                 UARTIFLS => self.ifl,
                 UARTIMSC => self.int_enabled,
                 UARTRIS => self.int_level,
-                UARTMIS => self.int_level & self.int_enabled,
+                UARTMIS => (self.int_level & self.int_enabled),
                 UARTDMACR => self.dmacr,
-                UARTDEBUG => self.debug,
                 _ => {
                     read_ok = false;
                     0
@@ -454,7 +369,12 @@ impl Snapshottable for Pl011 {
     }
 
     fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        Snapshot::new_from_versioned_state(&self.state())
+        Snapshot::new_from_versioned_state(&self.id, &self.state())
+    }
+
+    fn restore(&mut self, snapshot: Snapshot) -> std::result::Result<(), MigratableError> {
+        self.set_state(&snapshot.to_versioned_state(&self.id)?);
+        Ok(())
     }
 }
 
@@ -484,14 +404,9 @@ mod tests {
             &self,
             _index: InterruptIndex,
             _config: InterruptSourceConfig,
-            _masked: bool,
-            _set_gsi: bool,
         ) -> result::Result<(), std::io::Error> {
             Ok(())
         }
-        fn set_gsi(&self) -> result::Result<(), std::io::Error> {
-            Ok(())
-        }
         fn notifier(&self, _index: InterruptIndex) -> Option<EventFd> {
             Some(self.event_fd.try_clone().unwrap())
         }
@@ -533,14 +448,12 @@ mod tests {
             String::from(SERIAL_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
             Some(Box::new(pl011_out.clone())),
-            Instant::now(),
-            None,
         );
 
-        pl011.write(0, UARTDR, &[b'x', b'y']);
-        pl011.write(0, UARTDR, &[b'a']);
-        pl011.write(0, UARTDR, &[b'b']);
-        pl011.write(0, UARTDR, &[b'c']);
+        pl011.write(0, UARTDR as u64, &[b'x', b'y']);
+        pl011.write(0, UARTDR as u64, &[b'a']);
+        pl011.write(0, UARTDR as u64, &[b'b']);
+        pl011.write(0, UARTDR as u64, &[b'c']);
         assert_eq!(
             pl011_out.buf.lock().unwrap().as_slice(),
             &[b'x', b'a', b'b', b'c']
@@ -555,8 +468,6 @@ mod tests {
             String::from(SERIAL_NAME),
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
             Some(Box::new(pl011_out)),
-            Instant::now(),
-            None,
         );
 
         // write 1 to the interrupt event fd, so that read doesn't block in case the event fd
@@ -567,11 +478,11 @@ mod tests {
         assert_eq!(intr_evt.read().unwrap(), 2);
 
         let mut data = [0u8];
-        pl011.read(0, UARTDR, &mut data);
+        pl011.read(0, UARTDR as u64, &mut data);
         assert_eq!(data[0], b'a');
-        pl011.read(0, UARTDR, &mut data);
+        pl011.read(0, UARTDR as u64, &mut data);
         assert_eq!(data[0], b'b');
-        pl011.read(0, UARTDR, &mut data);
+        pl011.read(0, UARTDR as u64, &mut data);
         assert_eq!(data[0], b'c');
     }
 }

devices/src/lib.rs

@@ -10,10 +10,9 @@
 #[macro_use]
 extern crate bitflags;
 #[macro_use]
-extern crate event_monitor;
-#[macro_use]
 extern crate log;
 
+#[cfg(feature = "acpi")]
 pub mod acpi;
 #[cfg(target_arch = "aarch64")]
 pub mod gic;
@@ -21,11 +20,9 @@ pub mod interrupt_controller;
 #[cfg(target_arch = "x86_64")]
 pub mod ioapic;
 pub mod legacy;
-pub mod pvpanic;
-pub mod tpm;
 
+#[cfg(feature = "acpi")]
 pub use self::acpi::{AcpiGedDevice, AcpiPmTimerDevice, AcpiShutdownDevice};
-pub use self::pvpanic::{PvPanicDevice, PVPANIC_DEVICE_MMIO_SIZE};
 
 bitflags! {
     pub struct AcpiNotificationFlags: u8 {
@@ -37,9 +34,11 @@ bitflags! {
     }
 }
 
+#[allow(unused_macros)]
 #[cfg(target_arch = "aarch64")]
 macro_rules! generate_read_fn {
     ($fn_name: ident, $data_type: ty, $byte_type: ty, $type_size: expr, $endian_type: ident) => {
+        #[allow(dead_code)]
         pub fn $fn_name(input: &[$byte_type]) -> $data_type {
             assert!($type_size == std::mem::size_of::<$data_type>());
             let mut array = [0u8; $type_size];
@@ -51,9 +50,11 @@ macro_rules! generate_read_fn {
     };
 }
 
+#[allow(unused_macros)]
 #[cfg(target_arch = "aarch64")]
 macro_rules! generate_write_fn {
     ($fn_name: ident, $data_type: ty, $byte_type: ty, $endian_type: ident) => {
+        #[allow(dead_code)]
         pub fn $fn_name(buf: &mut [$byte_type], n: $data_type) {
             for (byte, read) in buf
                 .iter_mut()

devices/src/pvpanic.rs

@@ -1,272 +0,0 @@
-// Copyright © 2023 Tencent Corporation
-//
-// SPDX-License-Identifier: Apache-2.0
-//
-
-use anyhow::anyhow;
-use pci::{
-    BarReprogrammingParams, PciBarConfiguration, PciBarPrefetchable, PciBarRegionType,
-    PciClassCode, PciConfiguration, PciDevice, PciDeviceError, PciHeaderType, PciSubclass,
-    PCI_CONFIGURATION_ID,
-};
-use std::any::Any;
-use std::result;
-use std::sync::{Arc, Barrier, Mutex};
-use thiserror::Error;
-use versionize::{VersionMap, Versionize, VersionizeResult};
-use versionize_derive::Versionize;
-use vm_allocator::{AddressAllocator, SystemAllocator};
-use vm_device::{BusDevice, Resource};
-use vm_memory::{Address, GuestAddress};
-use vm_migration::{
-    Migratable, MigratableError, Pausable, Snapshot, Snapshottable, Transportable, VersionMapped,
-};
-
-const PVPANIC_VENDOR_ID: u16 = 0x1b36;
-const PVPANIC_DEVICE_ID: u16 = 0x0011;
-
-pub const PVPANIC_DEVICE_MMIO_SIZE: u64 = 0x2;
-
-const PVPANIC_PANICKED: u8 = 1 << 0;
-const PVPANIC_CRASH_LOADED: u8 = 1 << 1;
-
-#[derive(Debug, Error)]
-pub enum PvPanicError {
-    #[error("Failed creating PvPanicDevice: {0}")]
-    CreatePvPanicDevice(#[source] anyhow::Error),
-    #[error("Failed to retrieve PciConfigurationState: {0}")]
-    RetrievePciConfigurationState(#[source] anyhow::Error),
-}
-
-#[allow(dead_code)]
-#[derive(Copy, Clone)]
-pub enum PvPanicSubclass {
-    Other = 0x80,
-}
-
-impl PciSubclass for PvPanicSubclass {
-    fn get_register_value(&self) -> u8 {
-        *self as u8
-    }
-}
-
-/// A device for handling guest panic event
-pub struct PvPanicDevice {
-    id: String,
-    events: u8,
-
-    // PCI configuration registers.
-    configuration: PciConfiguration,
-    bar_regions: Vec<PciBarConfiguration>,
-}
-
-#[derive(Versionize)]
-pub struct PvPanicDeviceState {
-    events: u8,
-}
-
-impl VersionMapped for PvPanicDeviceState {}
-
-impl PvPanicDevice {
-    pub fn new(id: String, snapshot: Option<Snapshot>) -> Result<Self, PvPanicError> {
-        let pci_configuration_state =
-            vm_migration::versioned_state_from_id(snapshot.as_ref(), PCI_CONFIGURATION_ID)
-                .map_err(|e| {
-                    PvPanicError::RetrievePciConfigurationState(anyhow!(
-                        "Failed to get PciConfigurationState from Snapshot: {}",
-                        e
-                    ))
-                })?;
-
-        let mut configuration = PciConfiguration::new(
-            PVPANIC_VENDOR_ID,
-            PVPANIC_DEVICE_ID,
-            0x1, // modern pci devices
-            PciClassCode::BaseSystemPeripheral,
-            &PvPanicSubclass::Other,
-            None,
-            PciHeaderType::Device,
-            0,
-            0,
-            None,
-            pci_configuration_state,
-        );
-
-        let command: [u8; 2] = [0x03, 0x01];
-        configuration.write_config_register(1, 0, &command);
-
-        let state: Option<PvPanicDeviceState> = snapshot
-            .as_ref()
-            .map(|s| s.to_versioned_state())
-            .transpose()
-            .map_err(|e| {
-                PvPanicError::CreatePvPanicDevice(anyhow!(
-                    "Failed to get PvPanicDeviceState from Snapshot: {}",
-                    e
-                ))
-            })?;
-        let events = if let Some(state) = state {
-            state.events
-        } else {
-            PVPANIC_PANICKED | PVPANIC_CRASH_LOADED
-        };
-
-        let pvpanic_device = PvPanicDevice {
-            id,
-            events,
-            configuration,
-            bar_regions: vec![],
-        };
-
-        Ok(pvpanic_device)
-    }
-
-    pub fn event_to_string(&self, event: u8) -> String {
-        if event == PVPANIC_PANICKED {
-            "panic".to_string()
-        } else if event == PVPANIC_CRASH_LOADED {
-            "crash_loaded".to_string()
-        } else {
-            "unknown_event".to_string()
-        }
-    }
-
-    fn state(&self) -> PvPanicDeviceState {
-        PvPanicDeviceState {
-            events: self.events,
-        }
-    }
-
-    pub fn config_bar_addr(&self) -> u64 {
-        self.configuration.get_bar_addr(0)
-    }
-}
-
-impl BusDevice for PvPanicDevice {
-    fn read(&mut self, base: u64, offset: u64, data: &mut [u8]) {
-        self.read_bar(base, offset, data)
-    }
-
-    fn write(&mut self, _base: u64, _offset: u64, data: &[u8]) -> Option<Arc<Barrier>> {
-        let event = self.event_to_string(data[0]);
-        info!("pvpanic got guest event {}", event);
-        event!("guest", "panic", "event", &event);
-        None
-    }
-}
-
-impl PciDevice for PvPanicDevice {
-    fn write_config_register(
-        &mut self,
-        reg_idx: usize,
-        offset: u64,
-        data: &[u8],
-    ) -> Option<Arc<Barrier>> {
-        self.configuration
-            .write_config_register(reg_idx, offset, data);
-        None
-    }
-
-    fn read_config_register(&mut self, reg_idx: usize) -> u32 {
-        self.configuration.read_reg(reg_idx)
-    }
-
-    fn detect_bar_reprogramming(
-        &mut self,
-        reg_idx: usize,
-        data: &[u8],
-    ) -> Option<BarReprogrammingParams> {
-        self.configuration.detect_bar_reprogramming(reg_idx, data)
-    }
-
-    fn allocate_bars(
-        &mut self,
-        allocator: &Arc<Mutex<SystemAllocator>>,
-        _mmio_allocator: &mut AddressAllocator,
-        resources: Option<Vec<Resource>>,
-    ) -> std::result::Result<Vec<PciBarConfiguration>, PciDeviceError> {
-        let mut bars = Vec::new();
-        let region_type = PciBarRegionType::Memory32BitRegion;
-        let bar_id = 0;
-        let region_size = PVPANIC_DEVICE_MMIO_SIZE;
-        let restoring = resources.is_some();
-        let bar_addr = allocator
-            .lock()
-            .unwrap()
-            .allocate_mmio_hole_addresses(None, region_size, None)
-            .ok_or(PciDeviceError::IoAllocationFailed(region_size))?;
-
-        let bar = PciBarConfiguration::default()
-            .set_index(bar_id as usize)
-            .set_address(bar_addr.raw_value())
-            .set_size(region_size)
-            .set_region_type(region_type)
-            .set_prefetchable(PciBarPrefetchable::NotPrefetchable);
-
-        debug!("pvpanic bar address 0x{:x}", bar_addr.0);
-        if !restoring {
-            self.configuration
-                .add_pci_bar(&bar)
-                .map_err(|e| PciDeviceError::IoRegistrationFailed(bar_addr.raw_value(), e))?;
-        }
-
-        bars.push(bar);
-        self.bar_regions = bars.clone();
-
-        Ok(bars)
-    }
-
-    fn free_bars(
-        &mut self,
-        allocator: &mut SystemAllocator,
-        _mmio_allocator: &mut AddressAllocator,
-    ) -> std::result::Result<(), PciDeviceError> {
-        for bar in self.bar_regions.drain(..) {
-            allocator.free_mmio_hole_addresses(GuestAddress(bar.addr()), bar.size());
-        }
-
-        Ok(())
-    }
-
-    fn move_bar(&mut self, old_base: u64, new_base: u64) -> result::Result<(), std::io::Error> {
-        for bar in self.bar_regions.iter_mut() {
-            if bar.addr() == old_base {
-                *bar = bar.set_address(new_base);
-            }
-        }
-
-        Ok(())
-    }
-
-    fn read_bar(&mut self, _base: u64, _offset: u64, data: &mut [u8]) {
-        data[0] = self.events;
-    }
-
-    fn as_any(&mut self) -> &mut dyn Any {
-        self
-    }
-
-    fn id(&self) -> Option<String> {
-        Some(self.id.clone())
-    }
-}
-
-impl Pausable for PvPanicDevice {}
-
-impl Snapshottable for PvPanicDevice {
-    fn id(&self) -> String {
-        self.id.clone()
-    }
-
-    fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        let mut snapshot = Snapshot::new_from_versioned_state(&self.state())?;
-
-        // Snapshot PciConfiguration
-        snapshot.add_snapshot(self.configuration.id(), self.configuration.snapshot()?);
-
-        Ok(snapshot)
-    }
-}
-
-impl Transportable for PvPanicDevice {}
-impl Migratable for PvPanicDevice {}

devices/src/tpm.rs

@@ -1,551 +0,0 @@
-// Copyright © 2022, Microsoft Corporation
-//
-// SPDX-License-Identifier: Apache-2.0
-//
-
-use anyhow::anyhow;
-#[cfg(target_arch = "aarch64")]
-use arch::aarch64::layout::{TPM_SIZE, TPM_START};
-#[cfg(target_arch = "x86_64")]
-use arch::x86_64::layout::{TPM_SIZE, TPM_START};
-use std::cmp;
-use std::sync::{Arc, Barrier};
-use thiserror::Error;
-use tpm::emulator::{BackendCmd, Emulator};
-use tpm::TPM_CRB_BUFFER_MAX;
-use vm_device::BusDevice;
-
-#[derive(Error, Debug)]
-pub enum Error {
-    #[error("Emulator doesn't implement min required capabilities: {0}")]
-    CheckCaps(#[source] anyhow::Error),
-    #[error("Failed to initialize tpm: {0}")]
-    Init(#[source] anyhow::Error),
-}
-type Result<T> = anyhow::Result<T, Error>;
-
-#[allow(dead_code)]
-enum LocStateFields {
-    TpmEstablished,
-    LocAssigned,
-    ActiveLocality,
-    Reserved,
-    TpmRegValidSts,
-}
-
-enum LocStsFields {
-    Granted,
-    BeenSeized,
-}
-
-#[allow(dead_code)]
-enum IntfIdFields {
-    InterfaceType,
-    InterfaceVersion,
-    CapLocality,
-    CapCRBIdleBypass,
-    Reserved1,
-    CapDataXferSizeSupport,
-    CapFIFO,
-    CapCRB,
-    CapIFRes,
-    InterfaceSelector,
-    IntfSelLock,
-    Reserved2,
-    Rid,
-}
-
-#[allow(dead_code)]
-enum IntfId2Fields {
-    Vid,
-    Did,
-}
-
-enum CtrlStsFields {
-    TpmSts,
-    TpmIdle,
-}
-
-enum CrbRegister {
-    LocState(LocStateFields),
-    LocSts(LocStsFields),
-    IntfId(IntfIdFields),
-    IntfId2(IntfId2Fields),
-    CtrlSts(CtrlStsFields),
-}
-
-/* crb 32-bit registers */
-const CRB_LOC_STATE: u32 = 0x0;
-//Register Fields
-// Field => (base, offset, length)
-// base:   starting position of the register
-// offset: lowest bit in the bit field numbered from 0
-// length: length of the bit field
-const fn get_crb_loc_state_field(f: LocStateFields) -> (u32, u32, u32) {
-    let (offset, len) = match f {
-        LocStateFields::TpmEstablished => (0, 1),
-        LocStateFields::LocAssigned => (1, 1),
-        LocStateFields::ActiveLocality => (2, 3),
-        LocStateFields::Reserved => (5, 2),
-        LocStateFields::TpmRegValidSts => (7, 1),
-    };
-
-    (CRB_LOC_STATE, offset, len)
-}
-
-const CRB_LOC_CTRL: u32 = 0x08;
-const CRB_LOC_CTRL_REQUEST_ACCESS: u32 = 1 << 0;
-const CRB_LOC_CTRL_RELINQUISH: u32 = 1 << 1;
-const CRB_LOC_CTRL_RESET_ESTABLISHMENT_BIT: u32 = 1 << 3;
-const CRB_LOC_STS: u32 = 0x0C;
-const fn get_crb_loc_sts_field(f: LocStsFields) -> (u32, u32, u32) {
-    let (offset, len) = match f {
-        LocStsFields::Granted => (0, 1),
-        LocStsFields::BeenSeized => (1, 1),
-    };
-
-    (CRB_LOC_STS, offset, len)
-}
-
-const CRB_INTF_ID: u32 = 0x30;
-const fn get_crb_intf_id_field(f: IntfIdFields) -> (u32, u32, u32) {
-    let (offset, len) = match f {
-        IntfIdFields::InterfaceType => (0, 4),
-        IntfIdFields::InterfaceVersion => (4, 4),
-        IntfIdFields::CapLocality => (8, 1),
-        IntfIdFields::CapCRBIdleBypass => (9, 1),
-        IntfIdFields::Reserved1 => (10, 1),
-        IntfIdFields::CapDataXferSizeSupport => (11, 2),
-        IntfIdFields::CapFIFO => (13, 1),
-        IntfIdFields::CapCRB => (14, 1),
-        IntfIdFields::CapIFRes => (15, 2),
-        IntfIdFields::InterfaceSelector => (17, 2),
-        IntfIdFields::IntfSelLock => (19, 1),
-        IntfIdFields::Reserved2 => (20, 4),
-        IntfIdFields::Rid => (24, 8),
-    };
-
-    (CRB_INTF_ID, offset, len)
-}
-
-const CRB_INTF_ID2: u32 = 0x34;
-const fn get_crb_intf_id2_field(f: IntfId2Fields) -> (u32, u32, u32) {
-    let (offset, len) = match f {
-        IntfId2Fields::Vid => (0, 16),
-        IntfId2Fields::Did => (16, 16),
-    };
-
-    (CRB_INTF_ID2, offset, len)
-}
-
-const CRB_CTRL_REQ: u32 = 0x40;
-const CRB_CTRL_REQ_CMD_READY: u32 = 1 << 0;
-const CRB_CTRL_REQ_GO_IDLE: u32 = 1 << 1;
-const CRB_CTRL_STS: u32 = 0x44;
-const fn get_crb_ctrl_sts_field(f: CtrlStsFields) -> (u32, u32, u32) {
-    let (offset, len) = match f {
-        CtrlStsFields::TpmSts => (0, 1),
-        CtrlStsFields::TpmIdle => (1, 1),
-    };
-
-    (CRB_CTRL_STS, offset, len)
-}
-const CRB_CTRL_CANCEL: u32 = 0x48;
-const CRB_CANCEL_INVOKE: u32 = 1 << 0;
-const CRB_CTRL_START: u32 = 0x4C;
-const CRB_START_INVOKE: u32 = 1 << 0;
-const CRB_CTRL_CMD_LADDR: u32 = 0x5C;
-const CRB_CTRL_CMD_HADDR: u32 = 0x60;
-const CRB_CTRL_RSP_SIZE: u32 = 0x64;
-const CRB_CTRL_RSP_ADDR: u32 = 0x68;
-const CRB_DATA_BUFFER: u32 = 0x80;
-
-const TPM_CRB_NO_LOCALITY: u32 = 0xff;
-
-const TPM_CRB_ADDR_BASE: u32 = TPM_START.0 as u32;
-const TPM_CRB_ADDR_SIZE: usize = TPM_SIZE as usize;
-
-const TPM_CRB_R_MAX: usize = CRB_DATA_BUFFER as usize;
-
-// CRB Protocol details
-const CRB_INTF_TYPE_CRB_ACTIVE: u32 = 0b1;
-const CRB_INTF_VERSION_CRB: u32 = 0b1;
-const CRB_INTF_CAP_LOCALITY_0_ONLY: u32 = 0b0;
-const CRB_INTF_CAP_IDLE_FAST: u32 = 0b0;
-const CRB_INTF_CAP_XFER_SIZE_64: u32 = 0b11;
-const CRB_INTF_CAP_FIFO_NOT_SUPPORTED: u32 = 0b0;
-const CRB_INTF_CAP_CRB_SUPPORTED: u32 = 0b1;
-const CRB_INTF_IF_SELECTOR_CRB: u32 = 0b1;
-const PCI_VENDOR_ID_IBM: u32 = 0x1014;
-const CRB_CTRL_CMD_SIZE_REG: u32 = 0x58;
-const CRB_CTRL_CMD_SIZE: usize = TPM_CRB_ADDR_SIZE - CRB_DATA_BUFFER as usize;
-
-// Returns (register base, offset, len)
-const fn get_field(reg: CrbRegister) -> (u32, u32, u32) {
-    match reg {
-        CrbRegister::LocState(f) => get_crb_loc_state_field(f),
-        CrbRegister::LocSts(f) => get_crb_loc_sts_field(f),
-        CrbRegister::IntfId(f) => get_crb_intf_id_field(f),
-        CrbRegister::IntfId2(f) => get_crb_intf_id2_field(f),
-        CrbRegister::CtrlSts(f) => get_crb_ctrl_sts_field(f),
-    }
-}
-
-// Set a particular field in a Register
-fn set_reg_field(regs: &mut [u32; TPM_CRB_R_MAX], reg: CrbRegister, value: u32) {
-    let (base, offset, len) = get_field(reg);
-    let mask = (!(0_u32) >> (32 - len)) << offset;
-    regs[base as usize] = (regs[base as usize] & !mask) | ((value << offset) & mask);
-}
-
-// Get the value of a particular field in a Register
-const fn get_reg_field(regs: &[u32; TPM_CRB_R_MAX], reg: CrbRegister) -> u32 {
-    let (base, offset, len) = get_field(reg);
-    let mask = (!(0_u32) >> (32 - len)) << offset;
-    (regs[base as usize] & mask) >> offset
-}
-
-fn locality_from_addr(addr: u32) -> u8 {
-    (addr >> 12) as u8
-}
-
-pub struct Tpm {
-    emulator: Emulator,
-    regs: [u32; TPM_CRB_R_MAX],
-    backend_buff_size: usize,
-    data_buff: [u8; TPM_CRB_BUFFER_MAX],
-    data_buff_len: usize,
-}
-
-impl Tpm {
-    pub fn new(path: String) -> Result<Self> {
-        let emulator = Emulator::new(path)
-            .map_err(|e| Error::Init(anyhow!("Failed while initializing tpm Emulator: {:?}", e)))?;
-        let mut tpm = Tpm {
-            emulator,
-            regs: [0; TPM_CRB_R_MAX],
-            backend_buff_size: TPM_CRB_BUFFER_MAX,
-            data_buff: [0; TPM_CRB_BUFFER_MAX],
-            data_buff_len: 0,
-        };
-        tpm.reset()?;
-        Ok(tpm)
-    }
-
-    fn get_active_locality(&mut self) -> u32 {
-        if get_reg_field(
-            &self.regs,
-            CrbRegister::LocState(LocStateFields::LocAssigned),
-        ) == 0
-        {
-            return TPM_CRB_NO_LOCALITY;
-        }
-        get_reg_field(
-            &self.regs,
-            CrbRegister::LocState(LocStateFields::ActiveLocality),
-        )
-    }
-
-    fn request_completed(&mut self, success: bool) {
-        self.regs[CRB_CTRL_START as usize] = !CRB_START_INVOKE;
-        if !success {
-            set_reg_field(
-                &mut self.regs,
-                CrbRegister::CtrlSts(CtrlStsFields::TpmSts),
-                1,
-            );
-        }
-    }
-
-    fn reset(&mut self) -> Result<()> {
-        let cur_buff_size = self.emulator.get_buffer_size();
-        self.regs = [0; TPM_CRB_R_MAX];
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::LocState(LocStateFields::TpmRegValidSts),
-            1,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::CtrlSts(CtrlStsFields::TpmIdle),
-            1,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::InterfaceType),
-            CRB_INTF_TYPE_CRB_ACTIVE,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::InterfaceVersion),
-            CRB_INTF_VERSION_CRB,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::CapLocality),
-            CRB_INTF_CAP_LOCALITY_0_ONLY,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::CapCRBIdleBypass),
-            CRB_INTF_CAP_IDLE_FAST,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::CapDataXferSizeSupport),
-            CRB_INTF_CAP_XFER_SIZE_64,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::CapFIFO),
-            CRB_INTF_CAP_FIFO_NOT_SUPPORTED,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::CapCRB),
-            CRB_INTF_CAP_CRB_SUPPORTED,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::InterfaceSelector),
-            CRB_INTF_IF_SELECTOR_CRB,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId(IntfIdFields::Rid),
-            0b0000,
-        );
-        set_reg_field(
-            &mut self.regs,
-            CrbRegister::IntfId2(IntfId2Fields::Vid),
-            PCI_VENDOR_ID_IBM,
-        );
-
-        self.regs[CRB_CTRL_CMD_SIZE_REG as usize] = CRB_CTRL_CMD_SIZE as u32;
-        self.regs[CRB_CTRL_CMD_LADDR as usize] = TPM_CRB_ADDR_BASE + CRB_DATA_BUFFER;
-        self.regs[CRB_CTRL_RSP_SIZE as usize] = CRB_CTRL_CMD_SIZE as u32;
-        self.regs[CRB_CTRL_RSP_ADDR as usize] = TPM_CRB_ADDR_BASE + CRB_DATA_BUFFER;
-
-        self.backend_buff_size = cmp::min(cur_buff_size, TPM_CRB_BUFFER_MAX);
-
-        if let Err(e) = self.emulator.startup_tpm(self.backend_buff_size) {
-            return Err(Error::Init(anyhow!(
-                "Failed while running Startup TPM. Error: {:?}",
-                e
-            )));
-        }
-        Ok(())
-    }
-}
-
-impl BusDevice for Tpm {
-    fn read(&mut self, _base: u64, offset: u64, data: &mut [u8]) {
-        let mut offset: u32 = offset as u32;
-        let read_len: usize = data.len();
-
-        if offset >= CRB_DATA_BUFFER
-            && (offset + read_len as u32) < (CRB_DATA_BUFFER + self.data_buff.len() as u32)
-        {
-            // Read from Data Buffer
-            let start: usize = (offset as usize) - (CRB_DATA_BUFFER as usize);
-            let end: usize = start + read_len;
-            data[..].clone_from_slice(&self.data_buff[start..end]);
-        } else {
-            offset &= 0xff;
-            let mut val = self.regs[offset as usize];
-
-            if offset == CRB_LOC_STATE && !self.emulator.get_established_flag() {
-                val |= 0x1;
-            }
-
-            if data.len() <= 4 {
-                data.clone_from_slice(val.to_ne_bytes()[0..read_len].as_ref());
-            } else {
-                error!(
-                    "Invalid tpm read: offset {:#X}, data length {:?}",
-                    offset,
-                    data.len()
-                );
-            }
-        }
-        debug!(
-            "MMIO Read: offset {:#X} len {:?} val = {:02X?}  ",
-            offset,
-            data.len(),
-            data
-        );
-    }
-
-    fn write(&mut self, _base: u64, offset: u64, data: &[u8]) -> Option<Arc<Barrier>> {
-        debug!(
-            "MMIO Write: offset {:#X} len {:?} input data {:02X?}",
-            offset,
-            data.len(),
-            data
-        );
-        let mut offset: u32 = offset as u32;
-        if offset < CRB_DATA_BUFFER {
-            offset &= 0xff;
-        }
-        let locality = locality_from_addr(offset) as u32;
-        let write_len = data.len();
-
-        if offset >= CRB_DATA_BUFFER
-            && (offset + write_len as u32) < (CRB_DATA_BUFFER + self.data_buff.len() as u32)
-        {
-            let start: usize = (offset as usize) - (CRB_DATA_BUFFER as usize);
-            if start == 0 {
-                // If filling data_buff at index 0, reset length to 0
-                self.data_buff_len = 0;
-                self.data_buff.fill(0);
-            }
-            let end: usize = start + data.len();
-            self.data_buff[start..end].clone_from_slice(data);
-            self.data_buff_len += data.len();
-        } else {
-            // Ctrl Commands that take more than 4 bytes as input are not yet supported
-            // CTRL_RSP_ADDR usually gets 8 byte write request. Last 4 bytes are zeros.
-            if write_len > 4 && offset != CRB_CTRL_RSP_ADDR {
-                error!(
-                    "Invalid tpm write: offset {:#X}, data length {}",
-                    offset,
-                    data.len()
-                );
-                return None;
-            }
-
-            let mut input: [u8; 4] = [0; 4];
-            input.copy_from_slice(&data[0..4]);
-            let v = u32::from_le_bytes(input);
-
-            match offset {
-                CRB_CTRL_CMD_SIZE_REG => {
-                    self.regs[CRB_CTRL_CMD_SIZE_REG as usize] = v;
-                }
-                CRB_CTRL_CMD_LADDR => {
-                    self.regs[CRB_CTRL_CMD_LADDR as usize] = v;
-                }
-                CRB_CTRL_CMD_HADDR => {
-                    self.regs[CRB_CTRL_CMD_HADDR as usize] = v;
-                }
-                CRB_CTRL_RSP_SIZE => {
-                    self.regs[CRB_CTRL_RSP_SIZE as usize] = v;
-                }
-                CRB_CTRL_RSP_ADDR => {
-                    self.regs[CRB_CTRL_RSP_ADDR as usize] = v;
-                }
-                CRB_CTRL_REQ => match v {
-                    CRB_CTRL_REQ_CMD_READY => {
-                        set_reg_field(
-                            &mut self.regs,
-                            CrbRegister::CtrlSts(CtrlStsFields::TpmIdle),
-                            0,
-                        );
-                    }
-                    CRB_CTRL_REQ_GO_IDLE => {
-                        set_reg_field(
-                            &mut self.regs,
-                            CrbRegister::CtrlSts(CtrlStsFields::TpmIdle),
-                            1,
-                        );
-                    }
-                    _ => {
-                        error!("Invalid value passed to CRTL_REQ register");
-                        return None;
-                    }
-                },
-                CRB_CTRL_CANCEL => {
-                    if v == CRB_CANCEL_INVOKE
-                        && (self.regs[CRB_CTRL_START as usize] & CRB_START_INVOKE != 0)
-                    {
-                        if let Err(e) = self.emulator.cancel_cmd() {
-                            error!("Failed to run cancel command. Error: {:?}", e);
-                        }
-                    }
-                }
-                CRB_CTRL_START => {
-                    if v == CRB_START_INVOKE
-                        && ((self.regs[CRB_CTRL_START as usize] & CRB_START_INVOKE) == 0)
-                        && self.get_active_locality() == locality
-                    {
-                        self.regs[CRB_CTRL_START as usize] |= CRB_START_INVOKE;
-
-                        let mut cmd = BackendCmd {
-                            buffer: &mut self.data_buff,
-                            input_len: cmp::min(self.data_buff_len, TPM_CRB_BUFFER_MAX),
-                        };
-
-                        let status = self.emulator.deliver_request(&mut cmd).is_ok();
-
-                        self.request_completed(status);
-                    }
-                }
-                CRB_LOC_CTRL => {
-                    warn!(
-                        "CRB_LOC_CTRL locality to write = {:?} val = {:?}",
-                        locality, v
-                    );
-                    match v {
-                        CRB_LOC_CTRL_RESET_ESTABLISHMENT_BIT => {}
-                        CRB_LOC_CTRL_RELINQUISH => {
-                            set_reg_field(
-                                &mut self.regs,
-                                CrbRegister::LocState(LocStateFields::LocAssigned),
-                                0,
-                            );
-                            set_reg_field(
-                                &mut self.regs,
-                                CrbRegister::LocSts(LocStsFields::Granted),
-                                0,
-                            );
-                        }
-                        CRB_LOC_CTRL_REQUEST_ACCESS => {
-                            set_reg_field(
-                                &mut self.regs,
-                                CrbRegister::LocSts(LocStsFields::Granted),
-                                1,
-                            );
-                            set_reg_field(
-                                &mut self.regs,
-                                CrbRegister::LocSts(LocStsFields::BeenSeized),
-                                0,
-                            );
-                            set_reg_field(
-                                &mut self.regs,
-                                CrbRegister::LocState(LocStateFields::LocAssigned),
-                                1,
-                            );
-                        }
-                        _ => {
-                            error!("Invalid value to write in CRB_LOC_CTRL {:#X} ", v);
-                        }
-                    }
-                }
-                _ => {
-                    error!(
-                        "Invalid tpm write: offset {:#X}, data length {:?}",
-                        offset,
-                        data.len()
-                    );
-                }
-            }
-        }
-        None
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use super::*;
-
-    #[test]
-    fn test_set_get_reg_field() {
-        let mut regs: [u32; TPM_CRB_R_MAX] = [0; TPM_CRB_R_MAX];
-        set_reg_field(&mut regs, CrbRegister::IntfId(IntfIdFields::Rid), 0xAC);
-        assert_eq!(
-            get_reg_field(&regs, CrbRegister::IntfId(IntfIdFields::Rid)),
-            0xAC,
-            concat!("Test: ", stringify!(set_get_reg_field))
-        );
-    }
-}

docs/api.md

@@ -1,34 +1,29 @@
 - [Cloud Hypervisor API](#cloud-hypervisor-api)
-  - [External API](#external-api)
-    - [REST API](#rest-api)
-      - [REST API Location and availability](#rest-api-location-and-availability)
-      - [REST API Endpoints](#rest-api-endpoints)
-        - [Virtual Machine Manager (VMM) Actions](#virtual-machine-manager-vmm-actions)
-        - [Virtual Machine (VM) Actions](#virtual-machine-vm-actions)
+  * [External API](#external-api)
+    + [REST API](#rest-api)
+      - [Location and availability](#location-and-availability)
+      - [Endpoints](#endpoints)
+		* [Virtual Machine Manager (VMM) Actions](#virtual-machine-manager-vmm-actions)
+		* [Virtual Machine (VM) Actions](#virtual-machine-vm-actions)
       - [REST API Examples](#rest-api-examples)
-        - [Create a Virtual Machine](#create-a-virtual-machine)
-        - [Boot a Virtual Machine](#boot-a-virtual-machine)
-        - [Dump a Virtual Machine Information](#dump-a-virtual-machine-information)
-        - [Reboot a Virtual Machine](#reboot-a-virtual-machine)
-        - [Shut a Virtual Machine Down](#shut-a-virtual-machine-down)
-    - [D-Bus API](#d-bus-api)
-      - [D-Bus API Location and availability](#d-bus-api-location-and-availability)
-      - [D-Bus API Interface](#d-bus-api-interface)
-    - [Command Line Interface](#command-line-interface)
-    - [REST API, D-Bus API and CLI Architectural Relationship](#rest-api-and-cli-architectural-relationship)
-  - [Internal API](#internal-api)
-    - [Goals and Design](#goals-and-design)
-  - [End to End Example](#end-to-end-example)
+        * [Create a Virtual Machine](#create-a-virtual-machine)
+	    * [Boot a Virtual Machine](#boot-a-virtual-machine)
+        * [Dump a Virtual Machine Information](#dump-a-virtual-machine-information)
+        * [Reboot a Virtual Machine](#reboot-a-virtual-machine)
+        * [Shut a Virtual Machine Down](#shut-a-virtual-machine-down)
+    + [Command Line Interface](#command-line-interface)
+    + [REST API and CLI Architectural Relationship](#rest-api-and-cli-architectural-relationship)
+  * [Internal API](#internal-api)
+    + [Goals and Design](#goals-and-design)
+  * [End to End Example](#end-to-end-example)
 
 # Cloud Hypervisor API
 
 The Cloud Hypervisor API is made of 2 distinct interfaces:
 
-1. **The External API** This is the user facing API. Users and operators
-   can control and manage the Cloud Hypervisor through various options
-   including a REST API, a Command Line Interface (CLI) or a D-Bus based API,
-   which is not compiled into Cloud Hypervisor by default.
-
+1. **The external API**. This is the user facing API. Users and operators can
+   control and manage Cloud Hypervisor through either a REST API or a Command
+   Line Interface (CLI).
 1. **The internal API**, based on [rust's Multi-Producer, Single-Consumer (MPSC)](https://doc.rust-lang.org/std/sync/mpsc/)
    module. This API is used internally by the Cloud Hypervisor threads to
    communicate between each others.
@@ -45,10 +40,10 @@ API triggers VM and VMM specific actions, and as such it is designed as a
 collection of RPC-style, static methods.
 
 The API is [OpenAPI 3.0](https://github.com/OAI/OpenAPI-Specification/blob/master/versions/3.0.0.md)
-compliant. Please consult the [Cloud Hypervisor OpenAPI Document](https://raw.githubusercontent.com/cloud-hypervisor/cloud-hypervisor/master/vmm/src/api/openapi/cloud-hypervisor.yaml)
-for more details about the API payloads and responses.
+compliant. Please consult the [Cloud Hypervisor API](https://raw.githubusercontent.com/cloud-hypervisor/cloud-hypervisor/master/vmm/src/api/openapi/cloud-hypervisor.yaml)
+document for more details about the API payloads and responses.
 
-#### REST API Location and availability
+### Location and availability
 
 The REST API is available as soon as the Cloud Hypervisor binary is started,
 through a local UNIX socket.
@@ -70,55 +65,44 @@ Cloud Hypervisor Guest
     Disk(s): None
 ```
 
-#### REST API Endpoints
+### Endpoints
 
 The Cloud Hypervisor API exposes the following actions through its endpoints:
 
-##### Virtual Machine Manager (VMM) Actions
+#### Virtual Machine Manager (VMM) Actions
 
-| Action                              | Endpoint        | Request Body | Response Body              | Prerequisites      |
-| ----------------------------------- | --------------- | ------------ | -------------------------- | ------------------ |
-| Check for the REST API availability | `/vmm.ping`     | N/A          | `/schemas/VmmPingResponse` | N/A                |
-| Shut the VMM down                   | `/vmm.shutdown` | N/A          | N/A                        | The VMM is running |
+Action                              | Endpoint        | Request Body | Response Body              | Prerequisites
+------------------------------------|-----------------|--------------|----------------------------|---------------------------
+Check for the REST API availability | `/vmm.ping`     | N/A          | `/schemas/VmmPingResponse` | N/A
+Shut the VMM down                   | `/vmm.shutdown` | N/A          | N/A                        | The VMM is running
 
-##### Virtual Machine (VM) Actions
+#### Virtual Machine (VM) Actions
 
-| Action                             | Endpoint                | Request Body                    | Response Body            | Prerequisites                                          |
-| ---------------------------------- | ----------------------- | ------------------------------- | ------------------------ | ------------------------------------------------------ |
-| Create the VM                      | `/vm.create`            | `/schemas/VmConfig`             | N/A                      | The VM is not created yet                              |
-| Delete the VM                      | `/vm.delete`            | N/A                             | N/A                      | N/A                                                    |
-| Boot the VM                        | `/vm.boot`              | N/A                             | N/A                      | The VM is created but not booted                       |
-| Shut the VM down                   | `/vm.shutdown`          | N/A                             | N/A                      | The VM is booted                                       |
-| Reboot the VM                      | `/vm.reboot`            | N/A                             | N/A                      | The VM is booted                                       |
-| Trigger power button of the VM     | `/vm.power-button`      | N/A                             | N/A                      | The VM is booted                                       |
-| Pause the VM                       | `/vm.pause`             | N/A                             | N/A                      | The VM is booted                                       |
-| Resume the VM                      | `/vm.resume`            | N/A                             | N/A                      | The VM is paused                                       |
-| Task a snapshot of the VM          | `/vm.snapshot`          | `/schemas/VmSnapshotConfig`     | N/A                      | The VM is paused                                       |
-| Perform a coredump of the VM*      | `/vm.coredump`          | `/schemas/VmCoredumpData`       | N/A                      | The VM is paused                                       |
-| Restore the VM from a snapshot     | `/vm.restore`           | `/schemas/RestoreConfig`        | N/A                      | The VM is created but not booted                       |
-| Add/remove CPUs to/from the VM     | `/vm.resize`            | `/schemas/VmResize`             | N/A                      | The VM is booted                                       |
-| Add/remove memory from the VM      | `/vm.resize`            | `/schemas/VmResize`             | N/A                      | The VM is booted                                       |
-| Add/remove memory from a zone      | `/vm.resize-zone`       | `/schemas/VmResizeZone`         | N/A                      | The VM is booted                                       |
-| Dump the VM information            | `/vm.info`              | N/A                             | `/schemas/VmInfo`        | The VM is created                                      |
-| Add VFIO PCI device to the VM      | `/vm.add-device`        | `/schemas/VmAddDevice`          | `/schemas/PciDeviceInfo` | The VM is booted                                       |
-| Add disk device to the VM          | `/vm.add-disk`          | `/schemas/DiskConfig`           | `/schemas/PciDeviceInfo` | The VM is booted                                       |
-| Add fs device to the VM            | `/vm.add-fs`            | `/schemas/FsConfig`             | `/schemas/PciDeviceInfo` | The VM is booted                                       |
-| Add pmem device to the VM          | `/vm.add-pmem`          | `/schemas/PmemConfig`           | `/schemas/PciDeviceInfo` | The VM is booted                                       |
-| Add network device to the VM       | `/vm.add-net`           | `/schemas/NetConfig`            | `/schemas/PciDeviceInfo` | The VM is booted                                       |
-| Add userspace PCI device to the VM | `/vm.add-user-device`   | `/schemas/VmAddUserDevice`      | `/schemas/PciDeviceInfo` | The VM is booted                                       |
-| Add vdpa device to the VM          | `/vm.add-vdpa`          | `/schemas/VdpaConfig`           | `/schemas/PciDeviceInfo` | The VM is booted                                       |
-| Add vsock device to the VM         | `/vm.add-vsock`         | `/schemas/VsockConfig`          | `/schemas/PciDeviceInfo` | The VM is booted                                       |
-| Remove device from the VM          | `/vm.remove-device`     | `/schemas/VmRemoveDevice`       | N/A                      | The VM is booted                                       |
-| Dump the VM counters               | `/vm.counters`          | N/A                             | `/schemas/VmCounters`    | The VM is booted                                       |
-| Prepare to receive a migration     | `/vm.receive-migration` | `/schemas/ReceiveMigrationData` | N/A                      | N/A                                                    |
-| Start to send migration to target  | `/vm.send-migration`    | `/schemas/SendMigrationData`    | N/A                      | The VM is booted and (shared mem or hugepages enabled) |
+Action                             | Endpoint             | Request Body              | Response Body            | Prerequisites
+-----------------------------------|----------------------|---------------------------|--------------------------|---------------------------
+Create the VM                      | `/vm.create`         | `/schemas/VmConfig`       | N/A                      | The VM is not created yet
+Delete the VM                      | `/vm.delete`         | N/A                       | N/A                      | N/A
+Boot the VM                        | `/vm.boot`           | N/A                       | N/A                      | The VM is created but not booted
+Shut the VM down                   | `/vm.shutdown`       | N/A                       | N/A                      | The VM is booted
+Reboot the VM                      | `/vm.reboot`         | N/A                       | N/A                      | The VM is booted
+Trigger power button of the VM     | `/vm.power-button`   | N/A                       | N/A                      | The VM is booted
+Pause the VM                       | `/vm.pause`          | N/A                       | N/A                      | The VM is booted
+Resume the VM                      | `/vm.resume`         | N/A                       | N/A                      | The VM is paused
+Add/remove CPUs to/from the VM     | `/vm.resize`         | `/schemas/VmResize`       | N/A                      | The VM is booted
+Add/remove memory from the VM      | `/vm.resize`         | `/schemas/VmResize`       | N/A                      | The VM is booted
+Add/remove memory from a zone      | `/vm.resize-zone`    | `/schemas/VmResizeZone`   | N/A                      | The VM is booted
+Dump the VM information            | `/vm.info`           | N/A                       | `/schemas/VmInfo`        | The VM is created
+Add VFIO PCI device to the VM      | `/vm.add-device`     | `/schemas/VmAddDevice`    | `/schemas/PciDeviceInfo` | The VM is booted
+Add disk device to the VM          | `/vm.add-disk`       | `/schemas/DiskConfig`     | `/schemas/PciDeviceInfo` | The VM is booted
+Add fs device to the VM            | `/vm.add-fs`         | `/schemas/FsConfig`       | `/schemas/PciDeviceInfo` | The VM is booted
+Add pmem device to the VM          | `/vm.add-pmem`       | `/schemas/PmemConfig`     | `/schemas/PciDeviceInfo` | The VM is booted
+Add network device to the VM       | `/vm.add-net`        | `/schemas/NetConfig`      | `/schemas/PciDeviceInfo` | The VM is booted
+Add userspace PCI device to the VM | `/vm.add-user-device`| `/schemas/VmAddUserDevice`| `/schemas/PciDeviceInfo` | The VM is booted
+Add vsock device to the VM         | `/vm.add-vsock`      | `/schemas/VsockConfig`    | `/schemas/PciDeviceInfo` | The VM is booted
+Remove device from the VM          | `/vm.remove-device`  | `/schemas/VmRemoveDevice` | N/A                      | The VM is booted
+Dump the VM counters               | `/vm.counters`       | N/A                       | `/schemas/VmCounters`    | The VM is booted
 
-* The `vmcoredump` action is available exclusively for the `x86_64`
-architecture and can be executed only when the `guest_debug` feature is
-enabled. Without this feature, the corresponding [REST API](#rest-api) or
-[D-Bus API](#d-bus-api) endpoints are not available.
-
-#### REST API Examples
+### REST API Examples
 
 For the following set of examples, we assume Cloud Hypervisor is started with
 the REST API available at `/tmp/cloud-hypervisor.sock`:
@@ -134,7 +118,7 @@ Cloud Hypervisor Guest
     Disk(s): None
 ```
 
-##### Create a Virtual Machine
+#### Create a Virtual Machine
 
 We want to create a virtual machine with the following characteristics:
 
@@ -155,14 +139,15 @@ curl --unix-socket /tmp/cloud-hypervisor.sock -i \
      -H 'Content-Type: application/json'         \
      -d '{
          "cpus":{"boot_vcpus": 4, "max_vcpus": 4},
-         "payload":{"kernel":"/opt/clh/kernel/vmlinux-virtio-fs-virtio-iommu", "cmdline":"console=ttyS0 console=hvc0 root=/dev/vda1 rw"},
+         "kernel":{"path":"/opt/clh/kernel/vmlinux-virtio-fs-virtio-iommu"},
+         "cmdline":{"args":"console=ttyS0 console=hvc0 root=/dev/vda1 rw"},
          "disks":[{"path":"/opt/clh/images/focal-server-cloudimg-amd64.raw"}],
          "rng":{"src":"/dev/urandom"},
          "net":[{"ip":"192.168.10.10", "mask":"255.255.255.0", "mac":"12:34:56:78:90:01"}]
          }'
 ```
 
-##### Boot a Virtual Machine
+#### Boot a Virtual Machine
 
 Once the VM is created, we can boot it:
 
@@ -172,7 +157,7 @@ Once the VM is created, we can boot it:
 curl --unix-socket /tmp/cloud-hypervisor.sock -i -X PUT 'http://localhost/api/v1/vm.boot'
 ```
 
-##### Dump a Virtual Machine Information
+#### Dump a Virtual Machine Information
 
 We can fetch information about any VM, as soon as it's created:
 
@@ -184,7 +169,7 @@ curl --unix-socket /tmp/cloud-hypervisor.sock -i \
      -H 'Accept: application/json'
 ```
 
-##### Reboot a Virtual Machine
+#### Reboot a Virtual Machine
 
 We can reboot a VM that's already booted:
 
@@ -194,7 +179,7 @@ We can reboot a VM that's already booted:
 curl --unix-socket /tmp/cloud-hypervisor.sock -i -X PUT 'http://localhost/api/v1/vm.reboot'
 ```
 
-##### Shut a Virtual Machine Down
+#### Shut a Virtual Machine Down
 
 Once booted, we can shut a VM down from the REST API:
 
@@ -204,50 +189,6 @@ Once booted, we can shut a VM down from the REST API:
 curl --unix-socket /tmp/cloud-hypervisor.sock -i -X PUT 'http://localhost/api/v1/vm.shutdown'
 ```
 
-### D-Bus API
-
-Cloud Hypervisor offers a D-Bus API as an alternative to its REST API. As of
-writing this document, the D-Bus API mirrors the functionality of the REST
-API and shares the same set of endpoints, meaning that it supports every call
-that is supported by the REST API and can be a drop-in replacement since it
-also consumes/produces JSON.
-
-#### D-Bus API Location and availability
-
-This feature is not compiled into Cloud Hypervisor by default. Users who
-wish to use the D-Bus API, must explicitly enable it with the `dbus_api`
-feature flag when compiling Cloud Hypervisor.
-
-```sh
-$ ./scripts/dev_cli.sh build --release --libc musl -- --features dbus_api
-```
-
-Once this feature is enabled, it can be configured with the following
-CLI options:
-
-```
-  --dbus-service-name
-                    well known name of the service
-  --dbus-object-path
-                    object path to serve the dbus interface
-  --dbus-system-bus use the system bus instead of a session bus
-```
-
-Example invocation:
-
-```sh
-$ ./cloud-hypervisor --dbus-service-name "org.cloudhypervisor.DBusApi" \
-                     --dbus-object-path "/org/cloudhypervisor/DBusApi"
-```
-
-This will start serving a service with the name `org.cloudhypervisor.DBusApi1`
-which in turn can be used to control and manage Cloud Hypervisor.
-
-#### D-Bus API Interface
-
-Please refer to the [REST API](#rest-api) documentation. As previously
-mentioned, the D-Bus API currently mirrors the behaviour of the REST API.
-
 ### Command Line Interface
 
 The Cloud Hypervisor Command Line Interface (CLI) can only be used for launching
@@ -255,41 +196,33 @@ the Cloud Hypervisor binary, i.e. it can not be used for controlling the VMM or
 the launched VM once they're up and running.
 
 If you want to inspect the VMM, or control the VM after launching Cloud
-Hypervisor from the CLI, you must use either the [REST API](#rest-api)
-or the [D-Bus API](#d-bus-api).
+Hypervisor from the CLI, you must use the [REST API](#rest-api).
 
-From the CLI, one can:
+From the CLI, one can either:
 
 1. Create and boot a complete virtual machine by using the CLI options to build
    the VM config. Run `cloud-hypervisor --help` for a complete list of CLI
-   options. As soon as the `cloud-hypervisor` binary is launched, contrary
-   to the [D-Bus API](#d-bus-api), the [REST API](#rest-api) is available
-   for controlling and managing the VM. The [D-Bus API](#d-bus-api) doesn't start
-   automatically and needs to be explicitly configured in order to be run.
-1. Start either the REST API, D-Bus API or both simultaneously without passing
-   any VM configuration options. The VM can then be asynchronously created and
-   booted by calling API methods of choice. It should be noted that one external
-   API does not exclude another; it is possible to have both the REST and D-Bus
-   APIs running simultaneously.
+   options. As soon as the `cloud-hypervisor` binary is launched, the
+   [REST API](#rest-api) is available for controlling and managing the VM.
+1. Start the [REST API](#rest-api) server only, by not passing any VM
+   configuration options. The VM can then be asynchronously created and booted
+   by sending HTTP commands to the [REST API](#rest-api). Check the
+   [REST API examples](#rest-api-examples) section for more details.
 
-### REST API, D-Bus API and CLI Architectural Relationship
+### REST API and CLI Architectural Relationship
 
-The REST API, D-Bus API and the CLI all rely on a common, [internal API](#internal-api).
+The REST API and the CLI both rely on a common, [internal API](#internal-api).
 
 The CLI options are parsed by the
-[argh crate](https://docs.rs/argh/latest/argh/) and then translated into
+[clap crate](https://docs.rs/clap/2.33.0/clap/) and then translated into
 [internal API](#internal-api) commands.
 
 The REST API is processed by an HTTP thread using the
-[Firecracker's `micro_http`](https://github.com/firecracker-microvm/micro-http)
+[Firecracker's `micro_http`](https://github.com/firecracker-microvm/firecracker/tree/master/src/micro_http)
 crate. As with the CLI, the HTTP requests eventually get translated into
 [internal API](#internal-api) commands.
 
-The D-Bus API is implemented using the [zbus](https://github.com/dbus2/zbus)
-crate and runs in its own thread. Whenever it needs to call the [internal API](#internal-api),
-the [blocking](https://github.com/smol-rs/blocking) crate is used perform the call in zbus' async context.
-
-As a summary, the REST API, the D-Bus API and the CLI are essentially frontends for the
+As a summary, the REST API and the CLI are essentially frontends for the
 [internal API](#internal-api):
 
 ```
@@ -300,16 +233,16 @@ As a summary, the REST API, the D-Bus API and the CLI are essentially frontends
                        |          +------------------+        |
                        |                                      |      +------------------------+
                        |                                      |      |                        |
-+------------+         |            +----------+              |      |                        |
-|            |         |  D-Bus API |          |              |      | +--------------+       |
-|    User    +---------+----------->+   zbus   +--------------+------> | Internal API |       |
-|            |         |            |          |              |      | +--------------+       |
-+------------+         |            +----------+              |      |                        |
++------------+         |                                      |      |                        |
+|            |         |                                      |      | +--------------+       |
+|    User    +---------+                                      +------> | Internal API |       |
+|            |         |                                      |      | +--------------+       |
++------------+         |                                      |      |                        |
                        |                                      |      |                        |
                        |                                      |      +------------------------+
                        |            +----------+              |                 VMM
                        |     CLI    |          |              |
-                       +----------->+   argh   +--------------+
+                       +----------->+   clap   +--------------+
                                     |          |
                                     +----------+
 
@@ -319,23 +252,22 @@ As a summary, the REST API, the D-Bus API and the CLI are essentially frontends
 ## Internal API
 
 The Cloud Hypervisor internal API, as its name suggests, is used internally
-by the different Cloud Hypervisor threads (VMM, HTTP, D-Bus, control loop,
-etc) to send commands and responses to each others.
+by the different Cloud Hypervisor threads (VMM, HTTP, control loop, etc) to
+send commands and responses to each others.
 
 It is based on [rust's Multi-Producer, Single-Consumer (MPSC)](https://doc.rust-lang.org/std/sync/mpsc/),
 and the single consumer (a.k.a. the API receiver) is the Cloud Hypervisor
 control loop.
 
-API producers are the HTTP thread handling the [REST API](#rest-api), the
-D-Bus thread handling the [D-Bus API](#d-bus-api) and the main thread that
-initially parses the [CLI](#command-line-interface).
+API producers are the HTTP thread handling the [REST API](#rest-api) and the
+main thread that initially parses the [CLI](#command-line-interface).
 
 ### Goals and Design
 
 The internal API is designed for controlling, managing and inspecting a Cloud
 Hypervisor VMM and its guest. It is a backend for handling external, user
-visible requests through the [REST API](#rest-api), the [D-Bus API](#d-bus-api)
-or the [CLI](#command-line-interface) interfaces.
+visible requests through either the [REST API](#rest-api) or the
+[CLI](#command-line-interface) interfaces.
 
 The API follows a command-response scheme that closely maps the [REST API](#rest-api).
 Any command must be replied to with a response.
@@ -375,7 +307,8 @@ APIs work together, let's look at a complete VM creation flow, from the
 		-H 'Content-Type: application/json'         \
 		-d '{
 			"cpus":{"boot_vcpus": 4, "max_vcpus": 4},
-			"payload":{"kernel":"/opt/clh/kernel/vmlinux-virtio-fs-virtio-iommu", "cmdline":"console=ttyS0 console=hvc0 root=/dev/vda1 rw"},
+			"kernel":{"path":"/opt/clh/kernel/vmlinux-virtio-fs-virtio-iommu"},
+			"cmdline":{"args":"console=ttyS0 console=hvc0 root=/dev/vda1 rw"},
 			"disks":[{"path":"/opt/clh/images/focal-server-cloudimg-amd64.raw"}],
 			"rng":{"src":"/dev/urandom"},
 			"net":[{"ip":"192.168.10.10", "mask":"255.255.255.0", "mac":"12:34:56:78:90:01"}]
@@ -444,5 +377,5 @@ APIs work together, let's look at a complete VM creation flow, from the
    ```
 1. The Cloud Hypervisor HTTP thread sends the formed HTTP response back to the
    user. This is abstracted by the
-   [micro_http](https://github.com/firecracker-microvm/micro-http)
+   [micro_http](https://github.com/firecracker-microvm/firecracker/tree/master/src/micro_http)
    crate.

docs/arm64.md

@@ -0,0 +1,152 @@
+# How to build and test Cloud Hypervisor on AArch64
+
+This document introduces how to build and test Cloud Hypervisor on AArch64.
+Currently, Cloud Hypervisor supports 2 methods of booting on AArch64: UEFI
+booting and direct-kernel booting. The document covers both methods.
+
+All the steps are based on Ubuntu. We use the Ubuntu cloud image for guest VM
+disk.
+
+## Hardware requirements
+
+- AArch64 servers (recommended) or development boards equipped with the GICv3
+interrupt controller.
+
+- On development boards that have constrained RAM resources, if the creation of
+a VM consumes a large portion of the free memory on the host, it may be required
+to enable swap. For example, this was required on a board with 3 GB of RAM
+booting a 2 GB VM at a point in time when 2.8 GB were free. Without enabling
+swap the `cloud-hypervisor` process was terminated by the OOM killer. In this
+situation memory was allocated for the virtual machine using memfd while the
+page cache was filled, leading to a situation where the kernel could not even
+drop caches. Making a small section of swap available (observably, 1 to 15 MB),
+this situation can be resolved and the resulting memory footprint of
+`cloud-hypervisor` is as expected.
+
+## Getting started
+
+We create a folder to build and run Cloud Hypervisor at `$HOME/cloud-hypervisor`
+
+```shell
+$ export CLOUDH=$HOME/cloud-hypervisor
+$ mkdir $CLOUDH
+```
+
+## Prerequisites
+
+You need to install some prerequisite packages to build and test Cloud Hypervisor.
+
+### Tools
+
+```bash
+# Install rust tool chain
+$ curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
+# Install the tools used for building guest kernel, EDK2 and converting guest disk
+$ sudo apt-get update
+$ sudo apt-get install git build-essential m4 bison flex uuid-dev qemu-utils
+```
+
+### Building Cloud Hypervisor
+
+```bash
+$ pushd $CLOUDH
+$ git clone https://github.com/cloud-hypervisor/cloud-hypervisor.git
+$ cd cloud-hypervisor
+$ cargo build
+$ popd
+```
+
+### Disk image
+
+Download the Ubuntu cloud image and convert the image type.
+
+```bash
+$ pushd $CLOUDH
+$ wget https://cloud-images.ubuntu.com/focal/current/focal-server-cloudimg-arm64.img
+$ qemu-img convert -p -f qcow2 -O raw focal-server-cloudimg-arm64.img focal-server-cloudimg-arm64.raw
+$ popd
+```
+
+## UEFI booting
+
+This part introduces how to build EDK2 firmware and boot Cloud Hypervisor with it.
+
+### Building EDK2
+
+```bash
+$ pushd $CLOUDH
+
+# Clone source code repos
+$ git clone --depth 1 https://github.com/tianocore/edk2.git -b master
+$ cd edk2
+$ git submodule update --init
+$ cd ..
+$ git clone --depth 1 https://github.com/tianocore/edk2-platforms.git -b master
+$ git clone --depth 1 https://github.com/acpica/acpica.git -b master
+
+# Build tools
+$ export PACKAGES_PATH="$PWD/edk2:$PWD/edk2-platforms"
+$ export IASL_PREFIX="$PWD/acpica/generate/unix/bin/"
+$ make -C acpica
+$ cd edk2/
+$ . edksetup.sh
+$ cd ..
+$ make -C edk2/BaseTools
+
+# Build EDK2
+$ build -a AARCH64 -t GCC5 -p ArmVirtPkg/ArmVirtCloudHv.dsc -b RELEASE
+
+$ popd
+```
+
+If the build goes well, the EDK2 binary is available at
+`edk2/Build/ArmVirtCloudHv-AARCH64/RELEASE_GCC5/FV/CLOUDHV_EFI.fd`.
+
+### Booting the guest VM
+
+```bash
+$ pushd $CLOUDH
+$ sudo RUST_BACKTRACE=1 $CLOUDH/cloud-hypervisor/target/debug/cloud-hypervisor \
+           --api-socket /tmp/cloud-hypervisor.sock \
+           --kernel $CLOUDH/edk2/Build/ArmVirtCloudHv-AARCH64/RELEASE_GCC5/FV/CLOUDHV_EFI.fd \
+           --disk path=$CLOUDH/focal-server-cloudimg-arm64.raw \
+           --cpus boot=4 \
+           --memory size=4096M \
+           --net tap=,mac=12:34:56:78:90:01,ip=192.168.1.1,mask=255.255.255.0 \
+           --serial tty \
+           --console off
+$ popd
+```
+
+## Direct-kernel booting
+
+Alternativelly, you can build your own kernel for guest VM. This way, UEFI is
+not involved and ACPI cannot be enabled.
+
+### Building kernel
+
+```bash
+$ pushd $CLOUDH
+$ git clone --depth 1 "https://github.com/cloud-hypervisor/linux.git" -b ch-5.12
+$ cd linux
+$ cp $CLOUDH/cloud-hypervisor/resources/linux-config-aarch64 .config
+$ make -j `nproc`
+$ popd
+```
+
+### Booting the guest VM
+
+```bash
+$ pushd $CLOUDH
+$ sudo $CLOUDH/cloud-hypervisor/target/debug/cloud-hypervisor \
+           --api-socket /tmp/cloud-hypervisor.sock \
+           --kernel $CLOUDH/linux/arch/arm64/boot/Image \
+           --disk path=focal-server-cloudimg-arm64.raw \
+           --cmdline "keep_bootcon console=ttyAMA0 reboot=k panic=1 root=/dev/vda1 rw" \
+           --cpus boot=4 \
+           --memory size=4096M \
+           --net tap=,mac=12:34:56:78:90:01,ip=192.168.1.1,mask=255.255.255.0 \
+           --serial tty \
+           --console off
+$ popd
+```

docs/balloon.md

@@ -1,76 +0,0 @@
-# Balloon
-
-Cloud Hypervisor implements a balloon device based on the VIRTIO specification.
-Its main purpose is to provide the host a way to reclaim memory by controlling
-the amount of memory visible to the guest. But it also provides some interesting
-features related to guest memory management.
-
-## Parameters
-
-`BalloonConfig` (known as `--balloon` from the CLI perspective) contains the
-list of parameters available for the balloon device.
-
-```rust
-struct BalloonConfig {
-    pub size: u64,
-    pub deflate_on_oom: bool,
-    pub free_page_reporting: bool,
-}
-```
-
-```
---balloon <balloon>	Balloon parameters "size=<balloon_size>,deflate_on_oom=on|off,free_page_reporting=on|off"
-```
-
-### `size`
-
-Size of the balloon device. It is subtracted from the VM's total size. For
-instance, if creating a VM with 4GiB of RAM, along with a balloon of 1GiB, the
-guest will be able to use 3GiB of accessible memory. The guest sees all the RAM
-and unless it is balloon enlightened is entitled to all of it.
-
-This parameter is mandatory.
-
-Value is an unsigned integer of 64 bits corresponding to the balloon size in
-bytes.
-
-_Example_
-
-```
---balloon size=1G
-```
-
-### `deflate_on_oom`
-
-Allow the guest to deflate the balloon if running Out Of Memory (OOM). Assuming
-the balloon size is greater than 0, this means the guest is allowed to reduce
-the balloon size all the way down to 0 if this can help recover from the OOM
-event.
-
-This parameter is optional.
-
-Value is a boolean set to `off` by default.
-
-_Example_
-
-```
---ballloon size=2G,deflate_on_oom=on
-```
-
-### `free_page_reporting`
-
-Allow the guest to report lists of free pages. This feature doesn't require the
-balloon to be of any specific size as it doesn't impact the balloon size. The
-guest can let the VMM know about pages that are free after they have been used.
-Based on this information, the VMM can advise the host that it doesn't need
-these pages anymore.
-
-This parameter is optional.
-
-Value is a boolean set to `off` by default.
-
-_Example_
-
-```
---ballloon size=0,free_page_reporting=on
-```

docs/building.md

@@ -1,86 +0,0 @@
-- [Building Cloud Hypervisor](#building-cloud-hypervisor)
-  - [Preparation](#preparation)
-  - [Install prerequisites](#install-prerequisites)
-  - [Clone and build](#clone-and-build)
-    - [Containerized builds and tests](#containerized-builds-and-tests)
-
-# Building Cloud Hypervisor
-
-We recommend users use the pre-built binaries that are mentioned in the README.md file in the root of the repository. Building from source is only necessary if you wish to make modifications.
-
-## Preparation
-
-We create a folder to build and run `cloud-hypervisor` at `$HOME/cloud-hypervisor`
-
-```shell
-$ export CLOUDH=$HOME/cloud-hypervisor
-$ mkdir $CLOUDH
-```
-
-## Install prerequisites
-
-You need to install some prerequisite packages in order to build and test Cloud
-Hypervisor. Here, all the steps are based on Ubuntu, for other Linux
-distributions please replace the package manager and package name.
-
-```shell
-# Install basic packages needed. For a package list targeting for more
-# functionalities for example the test, please see resources/Dockerfile.
-$ sudo apt-get update
-$ sudo apt install git build-essential m4 bison flex uuid-dev qemu-utils musl-tools
-# Install rust tool chain
-$ curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
-# If you want to build statically linked binary please add musl target
-$ rustup target add x86_64-unknown-linux-musl  # x86-64
-$ rustup target add aarch64-unknown-linux-musl # AArch64
-```
-
-## Clone and build
-
-First you need to clone and build the Cloud Hypervisor repository:
-
-```shell
-$ pushd $CLOUDH
-$ git clone https://github.com/cloud-hypervisor/cloud-hypervisor.git
-$ cd cloud-hypervisor
-$ cargo build --release
-
-# We need to give the cloud-hypervisor binary the NET_ADMIN capabilities for it to set TAP interfaces up on the host.
-$ sudo setcap cap_net_admin+ep ./target/release/cloud-hypervisor
-
-# If you want to build statically linked binary
-$ cargo build --release --target=x86_64-unknown-linux-musl --all  # x86-64
-$ cargo build --release --target=aarch64-unknown-linux-musl --all # AArch64
-$ popd
-```
-
-This will build a `cloud-hypervisor` binary under
-`$CLOUDH/cloud-hypervisor/target/release/cloud-hypervisor`.
-
-### Containerized builds and tests
-
-If you want to build and test Cloud Hypervisor without having to install all the
-required dependencies (The rust toolchain, cargo tools, etc), you can also use
-Cloud Hypervisor's development script: `dev_cli.sh`. Please note that upon its
-first invocation, this script will pull a fairly large container image.
-
-For example, to build the Cloud Hypervisor release binary:
-
-```shell
-$ pushd $CLOUDH
-$ cd cloud-hypervisor
-$ ./scripts/dev_cli.sh build --release
-```
-
-With `dev_cli.sh`, one can also run the Cloud Hypervisor CI locally. This can be
-very convenient for debugging CI errors without having to fully rely on the
-Cloud Hypervisor CI infrastructure.
-
-For example, to run the Cloud Hypervisor unit tests:
-
-```shell
-$ ./scripts/dev_cli.sh tests --unit
-```
-
-Run the `./scripts/dev_cli.sh --help` command to view all the supported
-development script commands and their related options.

docs/cpu.md

@@ -17,12 +17,11 @@ struct CpusConfig {
     kvm_hyperv: bool,
     max_phys_bits: u8,
     affinity: Option<Vec<CpuAffinity>>,
-    features: CpuFeatures,
 }
 ```
 
 ```
---cpus boot=<boot_vcpus>,max=<max_vcpus>,topology=<threads_per_core>:<cores_per_die>:<dies_per_package>:<packages>,kvm_hyperv=on|off,max_phys_bits=<maximum_number_of_physical_bits>,affinity=<list_of_vcpus_with_their_associated_cpuset>,features=<list_of_features_to_enable>
+--cpus boot=<boot_vcpus>,max=<max_vcpus>,topology=<threads_per_core>:<cores_per_die>:<dies_per_package>:<packages>,kvm_hyperv=on|off,max_phys_bits=<maximum_number_of_physical_bits>,affinity=<list_of_vcpus_with_their_associated_cpuset>
 ```
 
 ### `boot`
@@ -188,24 +187,3 @@ _Example_
 In this example, assuming the host has 4 CPUs, vCPU 0 will run exclusively on
 host CPUs 2 and 3, while vCPU 1 will run exclusively on host CPUs 0 and 1.
 Because nothing is defined for vCPU 2, it can run on any of the 4 host CPUs.
-
-### `features`
-
-Set of CPU features to enable.
-
-This option allows the user to enable a set of CPU features that are disabled
-by default otherwise.
-
-The currently available feature set is: `amx`.
-
-The `amx` feature will enable the x86 extension adding hardware units for
-matrix operations (int and float dot products). The goal of the extension is to
-provide performance enhancements for these common operations.
-
-_Example_
-
-```
---cpus features=amx
-```
-
-In this example the amx CPU feature will be enabled for the VMM.

docs/custom-image.md

@@ -59,16 +59,6 @@ mkdir -p /mnt
 sudo mount -o loop,offset=$((227328 * 512)) focal-server-cloudimg-amd64.raw /mnt
 ```
 
-### Set up DNS
-
-The next step describes changing the root directory to the rootfs contained by 
-the cloud image. For DNS to work in the root directory, you will need to first bind-mount 
-the host `/etc/resolv.conf` onto the mounted linux partition of the cloud image.
-
-```bash
-sudo mount -o bind /etc/resolv.conf /mnt/etc/resolv.conf
-```
-
 ### Change root directory
 
 Changing the root directory will allow us to install new packages to the rootfs
@@ -88,7 +78,7 @@ Ubuntu distributions.
 
 ```bash
 apt update
-apt install fio iperf iperf3 socat stress cpuid tpm2-tools
+apt install fio iperf iperf3 socat stress
 ```
 
 ### Remove counterproductive packages
@@ -118,7 +108,6 @@ umount /dev/pts
 umount /proc
 history -c
 exit
-umount /mnt/etc/resolv.conf
 umount /mnt
 ```
 
@@ -158,172 +147,3 @@ as we might need to update the direct kernel boot command line, replacing
 `/dev/vda1` with the appropriate partition number.
 
 Update all references to the previous image name to the new one.
-
-## NVIDIA image for VFIO baremetal CI
-
-Here we are going to describe how to create a cloud image that contains the
-necessary NVIDIA drivers for our VFIO baremetal CI.
-
-### Download base image
-
-We usually start from one of the custom cloud image we have previously created
-but we can use a stock cloud image as well.
-
-```bash
-wget https://cloud-hypervisor.azureedge.net/jammy-server-cloudimg-amd64-custom-20230119-0.raw
-mv jammy-server-cloudimg-amd64-custom-20230119-0.raw jammy-server-cloudimg-amd64-nvidia.raw
-```
-
-### Extend the image size
-
-The NVIDIA drivers consume lots of space, which is why we must resize the image
-before we proceed any further.
-
-```bash
-qemu-img resize jammy-server-cloudimg-amd64-nvidia.raw 5G
-```
-
-### Resize the partition
-
-We use `parted` for fixing the GPT after the image was resized, as well as for
-resizing the `Linux` partition.
-
-```bash
-sudo parted jammy-server-cloudimg-amd64-nvidia.raw
-
-(parted) print
-Warning: Not all of the space available to jammy-server-cloudimg-amd64-nvidia.raw
-appears to be used, you can fix the GPT to use all of the space (an extra 5873664
-blocks) or continue with the current setting?
-Fix/Ignore? Fix
-Model:  (file)
-Disk jammy-server-cloudimg-amd64-nvidia.raw: 5369MB
-Sector size (logical/physical): 512B/512B
-Partition Table: gpt
-Disk Flags:
-
-Number  Start   End     Size    File system  Name  Flags
-14      1049kB  5243kB  4194kB                     bios_grub
-15      5243kB  116MB   111MB   fat32              boot, esp
- 1      116MB   2361MB  2245MB  ext4
-
-(parted) resizepart 1 5369MB
-(parted) print
-Model:  (file)
-Disk jammy-server-cloudimg-amd64-nvidia.raw: 5369MB
-Sector size (logical/physical): 512B/512B
-Partition Table: gpt
-Disk Flags:
-
-Number  Start   End     Size    File system  Name  Flags
-14      1049kB  5243kB  4194kB                     bios_grub
-15      5243kB  116MB   111MB   fat32              boot, esp
- 1      116MB   5369MB  5252MB  ext4
-
-(parted) quit
-```
-
-### Create a macvtap interface
-
-Rely on the following [documentation](macvtap-bridge.md) to set up a
-macvtap interface to provide your VM with proper connectivity.
-
-### Boot the image
-
-It is particularly important to boot with a `cloud-init` disk attached to the
-VM as it will automatically resize the Linux `ext4` filesystem based on the
-partition that we have previously resized.
-
-```bash
-./cloud-hypervisor \
-	--kernel hypervisor-fw  \
-	--disk path=focal-server-cloudimg-amd64-nvidia.raw path=/tmp/ubuntu-cloudinit.img \
-	--cpus boot=4 \
-	--memory size=4G \
-	--net fd=3,mac=$mac 3<>$"$tapdevice"
-```
-	
-### Bring up connectivity
-
-If your network has a DHCP server, run the following from your VM
-
-```bash
-sudo dhclient
-```
-
-But if that's not the case, let's give it an IP manually (the IP addresses
-depend on your actual network) and set the DNS server IP address as well.
-
-```bash
-sudo ip addr add 192.168.2.10/24 dev ens4
-sudo ip link set up dev ens4
-sudo ip route add default via 192.168.2.1
-sudo resolvectl dns ens4 8.8.8.8
-```
-
-#### Check connectivity and update the image
-
-```bash
-sudo apt update
-sudo apt upgrade
-```
-
-### Install NVIDIA drivers
-
-The following steps and commands are referenced from the
-[NVIDIA official documentation](https://docs.nvidia.com/datacenter/tesla/tesla-installation-notes/index.html#ubuntu-lts)
-about Tesla compute cards.
-
-```bash
-distribution=$(. /etc/os-release;echo $ID$VERSION_ID | sed -e 's/\.//g')
-wget https://developer.download.nvidia.com/compute/cuda/repos/$distribution/x86_64/cuda-keyring_1.0-1_all.deb
-sudo dpkg -i cuda-keyring_1.0-1_all.deb
-sudo apt-key del 7fa2af80
-sudo apt update
-sudo apt -y install cuda-drivers
-```
-
-### Check the `nvidia-smi` tool
-
-Quickly validate that you can find and run the `nvidia-smi` command from your
-VM. At this point it should fail given no NVIDIA card has been passed through
-the VM, therefore no NVIDIA driver is loaded.
-
-### Workaround LA57 reboot issue
-
-Add `reboot=a` to `GRUB_CMDLINE_LINUX` in `etc/default/grub` so that the VM
-will be booted with the ACPI reboot type. This resolves a reboot issue when
-running on 5-level paging systems.
-
-```bash
-sudo vim /etc/default/grub
-sudo update-grub
-sudo reboot
-```
-
-### Remove previous logins
-
-Since our integration tests rely on past logins to count the number of reboots,
-we must ensure to clear the list.
-
-```bash
->/var/log/lastlog
->/var/log/wtmp
->/var/log/btmp
-```
-
-### Clear history
-
-```
-history -c
-rm /home/cloud/.bash_history
-```
-
-### Reset cloud-init
-
-This is mandatory as we want `cloud-init` provisioning to work again when a new
-VM will be booted with this image.
-
-```
-sudo cloud-init clean
-```

docs/debug-port.md

@@ -36,7 +36,7 @@ Assuming parts of the guest software stack have been instrumented to use the
 `cloud-hypervisor` debug I/O port, we may want to gather the related logs.
 
 To do so we need to start `cloud-hypervisor` with the right debug level
-(`-v -v -v`). It is also recommended to have it log into a dedicated file in order
+(`-vvv`). It is also recommended to have it log into a dedicated file in order
 to easily grep for the tracing logs (e.g.
 `--log-file /tmp/cloud-hypervisor.log`):
 
@@ -48,7 +48,7 @@ to easily grep for the tracing logs (e.g.
     --memory size=1024M \
     --rng \
     --log-file /tmp/ch-fw.log \
-    -v -v -v
+    -vvv
 ```
 
 After booting the guest, we then have to grep for the debug I/O port traces in

docs/device_model.md

@@ -6,22 +6,22 @@ This document describes the device model supported by `cloud-hypervisor`.
 
 | Device | Build configurable | Enabled by default | Runtime configurable |
 | :----: | :----: | :----: | :----: |
-| Serial port | :x: | :x: | :heavy_check_mark: |
-| RTC/CMOS | :heavy_check_mark: | :heavy_check_mark: | :x: |
-| I/O APIC | :x: | :x: | :heavy_check_mark: |
-| i8042 shutdown/reboot | :x: | :x: | :x: |
-| ACPI shutdown/reboot | :x: | :heavy_check_mark: | :x: |
-| virtio-blk | :x: | :x: | :heavy_check_mark: |
-| virtio-console | :x: | :x: | :heavy_check_mark: |
-| virtio-iommu | :x: | :x: | :heavy_check_mark: |
-| virtio-net | :x: | :x: | :heavy_check_mark: |
-| virtio-pmem | :x: | :x: | :heavy_check_mark: |
-| virtio-rng | :x: | :x: | :heavy_check_mark: |
-| virtio-vsock | :x: | :x: | :heavy_check_mark: |
-| vhost-user-blk | :x: | :x: | :heavy_check_mark: |
-| vhost-user-fs | :x: | :x: | :heavy_check_mark: |
-| vhost-user-net | :x: | :x: | :heavy_check_mark: |
-| VFIO | :heavy_check_mark: | :x: | :heavy_check_mark: |
+| Serial port | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| RTC/CMOS | :heavy_check_mark: | :heavy_check_mark: | :negative_squared_cross_mark: |
+| I/O APIC | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| i8042 shutdown/reboot | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :negative_squared_cross_mark: |
+| ACPI shutdown/reboot | :negative_squared_cross_mark: | :heavy_check_mark: | :negative_squared_cross_mark: |
+| virtio-blk | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| virtio-console | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| virtio-iommu | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| virtio-net | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| virtio-pmem | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| virtio-rng | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| virtio-vsock | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| vhost-user-blk | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| vhost-user-fs | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| vhost-user-net | :negative_squared_cross_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
+| VFIO | :heavy_check_mark: | :negative_squared_cross_mark: | :heavy_check_mark: |
 
 ## Legacy devices
 

docs/fs.md

@@ -1,18 +1,14 @@
 # How to use virtio-fs
 
-In the context of virtualization, it is always convenient to be able to share a
-directory from the host with the guest.
+In the context of virtualization, it is always convenient to be able to share a directory from the host with the guest.
 
-__virtio-fs__, also known as __vhost-user-fs__ is a virtual device defined by
-the VIRTIO specification which allows any VMM to perform filesystem sharing.
+__virtio-fs__, also known as __vhost-user-fs__ is a virtual device defined by the VIRTIO specification which allows any VMM to perform filesystem sharing.
 
 ## Pre-requisites
 
 ### The daemon
 
-This virtual device relies on the _vhost-user_ protocol, which assumes the
-backend (device emulation) is handled by a dedicated process running on the
-host. This daemon is called __virtiofsd__ and needs to be present on the host.
+This virtual device relies on the _vhost-user_ protocol, which assumes the backend (device emulation) is handled by a dedicated process running on the host. This daemon is called __virtiofsd__ and needs to be present on the host.
 
 _Build virtiofsd_
 ```bash
@@ -29,46 +25,31 @@ mkdir /tmp/shared_dir
 _Run virtiofsd_
 ```bash
 ./virtiofsd \
-    --log-level debug \
+    -d \
     --socket-path=/tmp/virtiofs \
     --shared-dir=/tmp/shared_dir \
-    --cache=never \
-    --thread-pool-size=$N
+    --cache=never
 ```
 
-The `cache=never` option is the default when using `virtiofsd` with
-Cloud Hypervisor. This prevents from using the host page cache, reducing the
-overall footprint on host memory. This increases the maximum density of virtual
-machines that can be launched on a single host.
+The `cache=never` option should be the default when using `virtiofsd` with the __cloud-hypervisor__ VMM. This prevents from using the guest page cache, which reduces the memory footprint of the guest. When running multiple virtual machines on the same host, this will let the host deal with page cache, which will increase the density of virtual machines which can be launched.
 
-The `cache=always` option will allow the host page cache to be used, which can
-result in better performance for the guest's workload at the cost of increasing
-the footprint on host memory.
-
-The `thread-pool-size` option controls how many IO threads are spawned. For
-very fast storage like NVMe spawning enough worker threads is critical to
-getting an acceptable performance compared to native.
+The `cache=always` option will allow for the guest page cache to be used, which will increase the memory footprint of the guest. This option should be used only for specific use cases where a single VM is going to be running on a host.
 
 ### Kernel support
 
-Modern Linux kernels (at least v5.10) have support for virtio-fs. Use of older
-kernels, with additional patches, are not supported.
+Modern Linux kernels starting (at least v5.10) have support for virtio-fs. Use
+of older kernels, with additional patches, are not supported.
 
 ## How to share directories with cloud-hypervisor
 
 ### Start the VM
+Once the daemon is running, the option `--fs` from __cloud-hypervisor__ needs to be used.
 
-Once the daemon is running, the option `--fs` from Cloud Hypervisor needs
-to be used.
+Direct kernel boot is the preferred option, but we can boot from an EFI cloud image if it contains a recent enough kernel.
 
-Both direct kernel boot and EFI firmware can be used to boot a VM with
-virtio-fs, given that the cloud image contains a recent enough kernel.
+Because _vhost-user_ expects a dedicated process (__virtiofsd__ in this case) to be able to access the guest RAM to communicate through the _virtqueues_ with the driver running in the guest, `--memory` option needs to be slightly modified. It must specify `shared=on` to share the memory pages so that an external process can access them.
 
-Correct functioning of `--fs` requires `--memory shared=on` to facilitate
-interprocess memory sharing.
-
-Assuming you have `focal-server-cloudimg-amd64.raw` and `vmlinux` on your
-system, here is the Cloud Hypervisor command you need to run:
+Assuming you have `focal-server-cloudimg-amd64.raw` and `vmlinux` on your system, here is the __cloud-hypervisor__ command you need to run:
 ```bash
 ./cloud-hypervisor \
     --cpus boot=1 \
@@ -79,20 +60,26 @@ system, here is the Cloud Hypervisor command you need to run:
     --fs tag=myfs,socket=/tmp/virtiofs,num_queues=1,queue_size=512
 ```
 
-### Mount the shared directory
-
-The last step is to mount the shared directory inside the guest, using the
-`virtiofs` filesystem type.
+By default, DAX is enabled with a cache window of 8GiB. You can specify a custom size (let's say 4GiB for this example) for the cache by explicitly setting DAX and the cache size:
 
 ```bash
-mkdir mount_dir
-mount -t virtiofs myfs mount_dir/
+--fs tag=myfs,socket=/tmp/virtiofs,num_queues=1,queue_size=512,dax=on,cache_size=4G
+
 ```
 
-The `tag` needs to be consistent with what has been provided through the
-Cloud Hypervisor command line, which happens to be `myfs` in this example.
+In case you don't want to use a shared window of cache to pass the shared files content, this means you will have to explicitly disable DAX with `dax=off`. Note that in this case, the `cache_size` parameter will be ignored.
 
-## DAX feature
+```bash
+--fs tag=myfs,socket=/tmp/virtiofs,num_queues=1,queue_size=512,dax=off
 
-Given the DAX feature is not stable yet from a daemon standpoint, it is not
-available in Cloud Hypervisor.
+```
+
+### Mount the shared directory
+The last step is to mount the shared directory inside the guest, using the `virtiofs` filesystem type.
+```bash
+mkdir mount_dir
+mount -t virtiofs -o dax myfs mount_dir/
+```
+The `tag` needs to be consistent with what has been provided through the __cloud-hypervisor__ command line, which happens to be `myfs` in this example.
+
+The `-o dax` option must be removed in case the shared cache region is not enabled from the VMM.

docs/gdb.md

@@ -2,13 +2,13 @@
 
 This feature allows remote guest debugging using GDB. Note that this feature is only supported on x86_64/KVM.
 
-To enable debugging with GDB, build with the `guest_debug` feature enabled:
+To enable debugging with GDB, build with the `gdb` feature enabled:
 
 ```bash
-cargo build --features guest_debug
+cargo build --features gdb
 ```
 
-To use the `--gdb` option, specify the Unix Domain Socket with `path` that Cloud Hypervisor will use to communicate with the host's GDB:
+To use the `--gdb` option, specify the Unix Domain Socket with `--path` that Cloud Hypervisor will use to communicate with the host's GDB:
 
 ```bash
 ./cloud-hypervisor \
@@ -44,4 +44,4 @@ Continuing.
 
 Breakpoint 1, 0x00000000001121b7 in ?? ()
 (gdb)
-```
+```

docs/heap-profiling.md

@@ -1,41 +0,0 @@
-# Heap profiling
-
-Cloud Hypervisor supports generating a profile using
-[dhat](https://docs.rs/dhat/latest/dhat/) of the heap allocations made during
-the runtime of the process.
-
-## Building a suitable binary
-
-This adds the symbol information to the release binary but does not otherwise
-affect the performance.
-
-```
-$ cargo build --profile profiling --features "dhat-heap"
-```
-
-## Generating output
-
-Cloud Hypervisor can then be run as usual. However it is necessary to run with `--seccomp false` as the profiling requires extra syscalls.
-
-```
-$ target/profiling/cloud-hypervisor \
-        --kernel ~/src/linux/vmlinux \
-        --pmem file=~/workloads/focal.raw \
-        --cpus boot=1 --memory size=1G \
-        --cmdline "root=/dev/pmem0p1 console=ttyS0" \
-        --serial tty --console off \
-        --api-socket /tmp/api1 \
-        --seccomp false
-```
-
-When the VMM exits a message like the following will be shown:
-
-```
-dhat: Total:     384,582 bytes in 3,512 blocks
-dhat: At t-gmax: 133,885 bytes in 379 blocks
-dhat: At t-end:  12,160 bytes in 20 blocks
-dhat: The data has been saved to dhat-heap.json, and is viewable with dhat/dh_view.html
-```
-
-The JSON output can then be uploaded to [the dh_view tool](https://nnethercote.github.io/dh_view/dh_view.html) for analysis.
-

docs/hotplug.md

@@ -4,7 +4,7 @@ Currently Cloud Hypervisor supports hot plugging of CPUs devices (x86 only), PCI
 
 ## Kernel support
 
-For hotplug on Cloud Hypervisor ACPI GED support is needed. This can either be achieved by turning on `CONFIG_ACPI_REDUCED_HARDWARE_ONLY`
+For hotplug on Cloud Hypervisor ACPI GED support is needed. This can either be achieved by turning on `CONFIG_ACPI_REDUCED_HARDWARE_ONLY` 
 or by using this kernel patch (available in 5.5-rc1 and later): https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/patch/drivers/acpi/Makefile?id=ac36d37e943635fc072e9d4f47e40a48fbcdb3f0
 
 ## CPU Hot Plug
@@ -27,16 +27,16 @@ $ ./cloud-hypervisor/target/release/cloud-hypervisor \
 	--memory size=1024M \
 	--net "tap=,mac=,ip=,mask=" \
 	--rng \
-	--api-socket /tmp/ch-socket
+	--api-socket=/tmp/ch-socket
 $ popd
 ```
 
-Notice the addition of `--api-socket /tmp/ch-socket` and a `max` parameter on `--cpus boot=4,max=8`.
+Notice the addition of `--api-socket=/tmp/ch-socket` and a `max` parameter on `--cpus boot=4,max=8`.
 
 To ask the VMM to add additional vCPUs then use the resize API:
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket resize --cpus 8
+./ch-remote --api-socket=/tmp/ch-socket resize --cpus 8
 ```
 
 The extra vCPU threads will be created and advertised to the running kernel. The kernel does not bring up the CPUs immediately and instead the user must "online" them from inside the VM:
@@ -56,7 +56,7 @@ After a reboot the added CPUs will remain.
 Removing CPUs works similarly by reducing the number in the "desired_vcpus" field of the reisze API. The CPUs will be automatically offlined inside the guest so there is no need to run any commands inside the guest:
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket resize --cpus 2
+./ch-remote --api-socket=/tmp/ch-socket resize --cpus 2
 ```
 
 As per adding CPUs to the guest, after a reboot the VM will be running with the reduced number of vCPUs.
@@ -85,7 +85,7 @@ $ ./cloud-hypervisor/target/release/cloud-hypervisor \
 	--memory size=1024M,hotplug_size=8192M \
 	--net "tap=,mac=,ip=,mask=" \
 	--rng \
-	--api-socket /tmp/ch-socket
+	--api-socket=/tmp/ch-socket
 $ popd
 ```
 
@@ -98,7 +98,7 @@ root@ch-guest ~ # echo online | sudo tee /sys/devices/system/memory/auto_online_
 To ask the VMM to expand the RAM for the VM:
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket resize --memory 3G
+./ch-remote --api-socket=/tmp/ch-socket resize --memory 3G
 ```
 
 The new memory is now available to use inside the VM:
@@ -134,14 +134,14 @@ $ ./cloud-hypervisor/target/release/cloud-hypervisor \
 	--disk path=focal-server-cloudimg-amd64.raw \
 	--memory size=1024M,hotplug_size=8192M,hotplug_method=virtio-mem \
 	--net "tap=,mac=,ip=,mask=" \
-	--api-socket /tmp/ch-socket
+	--api-socket=/tmp/ch-socket
 $ popd
 ```
 
 To ask the VMM to expand the RAM for the VM (request is in bytes):
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket resize --memory 3G
+./ch-remote --api-socket=/tmp/ch-socket resize --memory 3G
 ```
 
 The new memory is now available to use inside the VM:
@@ -159,7 +159,7 @@ The same API can also be used to reduce the desired RAM for a VM. It is importan
 
 Extra PCI devices can be added and removed from a running `cloud-hypervisor` instance. This is controlled by making a HTTP API request to the VMM to ask for the additional device to be added, or for the existing device to be removed.
 
-Note: On AArch64 platform, PCI device hotplug can only be achieved using ACPI. Please refer to the [documentation](uefi.md#building-uefi-firmware-for-aarch64) for more information.
+Note: On AArch64 platform, PCI device hotplug can only be achieved using ACPI. Please refer to the [documentation](arm64.md#uefi-booting) for more information.
 
 To use PCI device hotplug start the VM with the HTTP server.
 
@@ -172,17 +172,17 @@ $ ./cloud-hypervisor/target/release/cloud-hypervisor \
 	--cpus boot=4 \
 	--memory size=1024M \
 	--net "tap=,mac=,ip=,mask=" \
-	--api-socket /tmp/ch-socket
+	--api-socket=/tmp/ch-socket
 ```
 
-Notice the addition of `--api-socket /tmp/ch-socket`.
+Notice the addition of `--api-socket=/tmp/ch-socket`.
 
 ### Add VFIO Device
 
 To ask the VMM to add additional VFIO device then use the `add-device` API.
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket add-device path=/sys/bus/pci/devices/0000:01:00.0/
+./ch-remote --api-socket=/tmp/ch-socket add-device path=/sys/bus/pci/devices/0000:01:00.0/
 ```
 
 ### Add Disk Device
@@ -190,7 +190,7 @@ To ask the VMM to add additional VFIO device then use the `add-device` API.
 To ask the VMM to add additional disk device then use the `add-disk` API.
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket add-disk path=/foo/bar/cloud.img
+./ch-remote --api-socket=/tmp/ch-socket add-disk path=/foo/bar/cloud.img
 ```
 
 ### Add Fs Device
@@ -198,7 +198,7 @@ To ask the VMM to add additional disk device then use the `add-disk` API.
 To ask the VMM to add additional fs device then use the `add-fs` API.
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket add-fs tag=myfs,socket=/foo/bar/virtiofs.sock
+./ch-remote --api-socket=/tmp/ch-socket add-fs tag=myfs,socket=/foo/bar/virtiofs.sock
 ```
 
 ### Add Net Device
@@ -206,7 +206,7 @@ To ask the VMM to add additional fs device then use the `add-fs` API.
 To ask the VMM to add additional network device then use the `add-net` API.
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket add-net tap=chtap0
+./ch-remote --api-socket=/tmp/ch-socket add-net tap=chtap0
 ```
 
 ### Add Pmem Device
@@ -214,7 +214,7 @@ To ask the VMM to add additional network device then use the `add-net` API.
 To ask the VMM to add additional PMEM device then use the `add-pmem` API.
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket add-pmem file=/foo/bar.cloud.img
+./ch-remote --api-socket=/tmp/ch-socket add-pmem file=/foo/bar.cloud.img
 ```
 
 ### Add Vsock Device
@@ -222,7 +222,7 @@ To ask the VMM to add additional PMEM device then use the `add-pmem` API.
 To ask the VMM to add additional vsock device then use the `add-vsock` API.
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket add-vsock cid=3,socket=/foo/bar/vsock.sock
+./ch-remote --api-socket=/tmp/ch-socket add-vsock cid=3,socket=/foo/bar/vsock.sock
 ```
 
 ### Common Across All PCI Devices
@@ -244,7 +244,7 @@ After a reboot the added PCI device will remain.
 Removing a PCI device works the same way for all kind of PCI devices. The unique identifier related to the device must be provided. This identifier can be provided by the user when adding the new device, or by default Cloud Hypervisor will assign one.
 
 ```shell
-./ch-remote --api-socket /tmp/ch-socket remove-device _disk0
+./ch-remote --api-socket=/tmp/ch-socket remove-device _disk0
 ```
 
 As per adding a PCI device to the guest, after a reboot the VM will be running without the removed PCI device.

docs/intel_tdx.md

@@ -2,64 +2,49 @@
 
 Intel® Trust Domain Extensions (Intel® TDX) is an Intel technology designed to
 isolate virtual machines from the VMM, hypervisor and any other software on the
-host platform. Here are some useful links:
+host platform.
 
-* [TDX Homepage](https://www.intel.com/content/www/us/en/developer/articles/technical/intel-trust-domain-extensions.html):
-more information about TDX technical aspects, design and specification
+For more information about TDX technical aspects, design and specification
+please refer to the
+[TDX Homepage](https://www.intel.com/content/www/us/en/developer/articles/technical/intel-trust-domain-extensions.html).
 
-* [KVM TDX tree](https://github.com/intel/tdx/tree/kvm): the required
-  Linux kernel changes for the host side
+The required Linux changes for the host side can be found in the
+[KVM TDX tree](https://github.com/intel/tdx/tree/kvm) while the changes for
+the guest side can be found in the [Guest TDX tree](https://github.com/intel/tdx/tree/guest).
 
-* [Guest TDX tree](https://github.com/intel/tdx/tree/guest): the Linux
-  kernel changes for the guest side
+The TDVF firmware can be found in the
+[EDK2 staging project](https://github.com/tianocore/edk2-staging/tree/TDVF).
 
-* [EDK2 project](https://github.com/tianocore/edk2): the TDVF firmware
-
-* [Confidential Containers project](https://github.com/confidential-containers/td-shim):
-  the TDShim firmware
-
-* [TDX Tools](https://github.com/intel/tdx-tools): a collection of tools
-  and scripts to setup TDX environment for testing purpose (such as
-  installing required packages on the host, creating guest images, and
-  building the custom Linux kernel for TDX host and guest)
+The TDShim firmware can be found in the
+[Confidential Containers project](https://github.com/confidential-containers/td-shim).
 
 ## Cloud Hypervisor support
 
-It is required to use a machine with TDX enabled in hardware and
+First, you must be running on a machine with TDX enabled in hardware, and
 with the host OS compiled from the [KVM TDX tree](https://github.com/intel/tdx/tree/kvm).
-The host environment can also be setup with the [TDX Tools](https://github.com/intel/tdx-tools).
 
-Cloud Hypervisor can run TDX VM (Trust Domain) by loading a TD firmware ([TDVF](https://github.com/tianocore/edk2)),
+Cloud Hypervisor can run TDX VM (Trust Domain) by loading a TD firmware,
 which will then load the guest kernel from the image. The image must be custom
 as it must include a kernel built from the [Guest TDX tree](https://github.com/intel/tdx/tree/guest).
-Cloud Hypervisor can also boot a TDX VM with direct kernel boot using [TDshim](https://github.com/confidential-containers/td-shim).
-The custom Linux kernel for the guest can be built with the [TDX Tools](https://github.com/intel/tdx-tools).
-
-> **Note**
-> The latest version of custom host and guest kernel being tested is
-> from [TDX Tools - 2023ww01](https://github.com/intel/tdx-tools/commits/2023ww01).
 
 ### TDVF
 
-> **Note**
-> The latest version of TDVF being tested is [_13b9773_](https://github.com/tianocore/edk2/commit/13b97736c876919b9786055829caaa4fa46984b7).
-
 The firmware can be built as follows:
 
 ```bash
-git clone https://github.com/tianocore/edk2.git
-cd edk2
-git checkout 13b97736c876919b9786055829caaa4fa46984b7
+git clone https://github.com/tianocore/edk2-staging.git
+cd edk2-staging
+git checkout origin/TDVF
 git submodule update --init --recursive
 make -C BaseTools
 source ./edksetup.sh
-build -p OvmfPkg/IntelTdx/IntelTdxX64.dsc -a X64 -t GCC5 -b RELEASE
+build -p OvmfPkg/OvmfCh.dsc -a X64 -t GCC5 -b RELEASE
 ```
 
 If debug logs are needed, here is the alternative command:
 
 ```bash
-build -p OvmfPkg/IntelTdx/IntelTdxX64.dsc -a X64 -t GCC5 -D DEBUG_ON_SERIAL_PORT=TRUE
+build -p OvmfPkg/OvmfCh.dsc -a X64 -t GCC5 -D DEBUG_ON_SERIAL_PORT=TRUE
 ```
 
 On the Cloud Hypervisor side, all you need is to build the project with the
@@ -70,108 +55,50 @@ cargo build --features tdx
 ```
 
 And run a TDX VM by providing the firmware previously built, along with the
-guest image containing the TDX enlightened kernel. The latest image
-`td-guest-rhel8.5.raw` contains `console=hvc0` on the kernel boot parameters,
-meaning it will be printing guest kernel logs to the `virtio-console` device.
+guest image containing the TDX enlightened kernel. Assuming the guest kernel
+command line contains `console=hvc0` (printing to the `virtio-console` device),
+run Cloud Hypervisor as follows:
 
 ```bash
 ./cloud-hypervisor \
-    --platform tdx=on
-    --firmware edk2/Build/IntelTdx/RELEASE_GCC5/FV/OVMF.fd \
+    --tdx firmware=edk2-staging/Build/OvmfCh/RELEASE_GCC5/FV/OVMF.fd \
     --cpus boot=1 \
     --memory size=1G \
     --disk path=tdx_guest_img
 ```
 
-And here is the alternative command when looking for debug logs from the
-firmware:
+And here is the alternative command when looking for debug logs (assuming the
+guest kernel command line contains `console=ttyS0`):
 
 ```bash
 ./cloud-hypervisor \
-    --platform tdx=on
-    --firmware edk2/Build/IntelTdx/DEBUG_GCC5/FV/OVMF.fd \
+    --tdx firmware=edk2-staging/Build/OvmfCh/DEBUG_GCC5/FV/OVMF.fd \
     --cpus boot=1 \
     --memory size=1G \
     --disk path=tdx_guest_img \
-    --serial file=/tmp/ch_serial \
-    --console tty
+    --serial tty \
+    --console off
 ```
 
 ### TDShim
 
-> **Note**
-> The latest version of TDShim being tested is [_66bb334_](https://github.com/confidential-containers/td-shim/tree/66bb33451befbf1291abe3cfea7ee9e99d922b0d).
-
 This is a lightweight version of the TDVF, written in Rust and designed for
 direct kernel boot, which is useful for containers use cases.
 
-To build TDShim from source, it is required to install `Rust`, `NASM`,
-and `LLVM` first. The TDshim can be build as follows:
-```bash
-git clone https://github.com/confidential-containers/td-shim
-cd td-shim
-git checkout 66bb33451befbf1291abe3cfea7ee9e99d922b0d
-cargo install cargo-xbuild
-export CC=clang
-export AR=llvm-ar
-export CC_x86_64_unknown_none=clang
-export AR_x86_64_unknown_none=llvm-ar
-git submodule update --init --recursive
-./sh_script/preparation.sh
-cargo xbuild -p td-shim --target x86_64-unknown-none --release --features=main,tdx
-cargo run -p td-shim-tools --bin td-shim-ld --features=linker -- target/x86_64-unknown-none/release/ResetVector.bin target/x86_64-unknown-none/release/td-shim -o target/release/final.bin
-```
-
-If debug logs from the TDShim is needed, here are the alternative
-commands:
-```bash
-cargo xbuild -p td-shim --target x86_64-unknown-none --features=main,tdx
-cargo run -p td-shim-tools --bin td-shim-ld --features=linker -- target/x86_64-unknown-none/debug/ResetVector.bin target/x86_64-unknown-none/debug/td-shim -o target/debug/final.bin
-```
+You can find the instructions for building the firmware directly from the
+project [documentation](https://github.com/confidential-containers/td-shim/tree/staging#how-to-build).
 
 And run a TDX VM by providing the firmware previously built, along with a guest
-kernel built from the [Guest TDX tree](https://github.com/intel/tdx/tree/guest)
-or the [TDX Tools](https://github.com/intel/tdx-tools).
+kernel built from the [Guest TDX tree](https://github.com/intel/tdx/tree/guest).
 The appropriate kernel boot options must be provided through the `--cmdline`
 option as well.
 
 ```bash
 ./cloud-hypervisor \
-    --platform tdx=on
-    --firmware td-shim/target/release/final.bin \
+    --tdx firmware=tdshim \
     --kernel bzImage \
-    --cmdline "root=/dev/vda3 console=hvc0 rw"
+    --cmdline "root=/dev/vda1 console=hvc0 rw tdx_allow_acpi=MCFG"
     --cpus boot=1 \
     --memory size=1G \
     --disk path=tdx_guest_img
-```
-
-And here is the alternative command when looking for debug logs from the
-TDShim:
-
-```bash
-./cloud-hypervisor \
-    --platform tdx=on
-    --firmware td-shim/target/debug/final.bin \
-    --kernel bzImage \
-    --cmdline "root=/dev/vda3 console=hvc0 rw"
-    --cpus boot=1 \
-    --memory size=1G \
-    --disk path=tdx_guest_img
-```
-
-### Guest kernel limitations
-
-#### Serial ports disabled
-
-The latest guest kernel that can be found in the latest image
-`td-guest-rhel8.5.raw` disabled the support for serial ports. This means adding
-`console=ttyS0` will have no effect and will not print any log from the guest.
-
-#### PCI hotplug through ACPI
-
-Unless you run the guest kernel with the parameter `tdx_disable_filter`, ACPI
-devices responsible for handling PCI hotplug (PCI hotplug controller, PCI
-Express Bus and Generic Event Device) will not be allowed, therefore the
-corresponding drivers will not be loaded and the PCI hotplug feature will not
-be supported.
+```

docs/io_throttling.md

@@ -24,8 +24,8 @@ bucket is unbounded in speed which allows for bursts bound in size by
 the amount of tokens available. Once the token bucket is empty,
 consumption speed is bound by the "refill-rate". Similarly, Cloud
 Hypervisor provides another three options for limiting I/O operations,
-i.e., `ops_size` (I/O operations), `ops_one_time_burst` (I/O operations),
-and `ops_refill_time` (ms).
+i.e., `ops_size` (I/O operations), `bw_one_time_burst` (I/O operations),
+and `bw_refill_time` (ms).
 
 One caveat in the I/O throttling is that every-time the bucket gets
 empty, it will stop I/O operations for a fixed amount of time

docs/iommu.md

@@ -231,38 +231,3 @@ Last thing is to start the L2 guest with the huge pages memory backend.
     --cmdline "console=ttyS0 console=hvc0 root=/dev/vda1 rw" \
     --device path=/sys/bus/pci/devices/0000:00:04.0
 ```
-
-### Dedicated IOMMU PCI segments
-
-To facilitate hotplug of devices that require being behind an IOMMU it is
-possible to mark entire PCI segments as behind the IOMMU.
-
-This is accomplished through `--platform
-num_pci_segments=<number_of_segments>,iommu_segments=<range of segments>` or
-via the equivalents in `PlatformConfig` for the API.
-
-e.g.
-
-```bash
-./cloud-hypervisor \
-    --api-socket /tmp/api \
-    --cpus boot=1 \
-    --memory size=4G,hugepages=on \
-    --disk path=focal-server-cloudimg-amd64.raw \
-    --kernel custom-vmlinux \
-    --cmdline "console=ttyS0 console=hvc0 root=/dev/vda1 rw" \
-    --platform num_pci_segments=2,iommu_segments=1
-```
-
-This adds a second PCI segment to the platform behind the IOMMU. A VFIO device
-requiring the IOMMU then may be hotplugged:
-
-e.g.
-
-```bash
-./ch-remote --api-socket /tmp/api add-device path=/sys/bus/pci/devices/0000:00:04.0,iommu=on,pci_segment=1
-```
-
-Devices that cannot be placed behind an IOMMU (e.g. lacking an `iommu=` option)
-cannot be placed on the IOMMU segments.
-

docs/live_migration.md

@@ -16,22 +16,22 @@ $ target/release/cloud-hypervisor
     --disk path=~/workloads/focal.raw \
     --cpus boot=1 --memory size=1G,shared=on \
     --cmdline "root=/dev/vda1 console=ttyS0"  \
-    --serial tty --console off --api-socket /tmp/api1
+    --serial tty --console off --api-socket=/tmp/api1
 ```
 
 Launch the destination VM from the same directory (on the host machine):
 ```bash
-$ target/release/cloud-hypervisor --api-socket /tmp/api2
+$ target/release/cloud-hypervisor --api-socket=/tmp/api2
 ```
 
 Get ready for receiving migration for the destination VM (on the host machine):
 ```bash
-$ target/release/ch-remote --api-socket /tmp/api2 receive-migration unix:/tmp/sock
+$ target/release/ch-remote --api-socket=/tmp/api2 receive-migration unix:/tmp/sock
 ```
 
 Start to send migration for the source VM (on the host machine):
 ```bash
-$ target/release/ch-remote --api-socket /tmp/api1 send-migration --local unix:/tmp/sock
+$ target/release/ch-remote --api-socket=/tmp/api1 send-migration --local unix:/tmp/sock
 ```
 
 When the above commands completed, the source VM should be successfully
@@ -51,7 +51,7 @@ $ sudo /target/release/cloud-hypervisor \
         --cpus boot=1 --memory size=512M \
         --kernel vmlinux \
         --cmdline "root=/dev/vda1 console=ttyS0"  \
-        --disk path=focal-1.raw path=focal-nested.raw --disk path=tmp.img\
+        --disk path=focal-1.raw path=focal-nested.raw path=tmp.img\
         --net ip=192.168.101.1
 ```
 
@@ -63,7 +63,7 @@ $ sudo /target/release/cloud-hypervisor \
         --cpus boot=1 --memory size=512M \
         --kernel vmlinux \
         --cmdline "root=/dev/vda1 console=ttyS0"  \
-        --disk path=focal-2.raw path=focal-nested.raw --disk path=tmp.img\
+        --disk path=focal-2.raw path=focal-nested.raw path=tmp.img\
         --net ip=192.168.102.1
 ```
 
@@ -74,8 +74,8 @@ vm-1:~$ sudo ./cloud-hypervisor \
         --memory size=128M \
         --kernel vmlinux \
         --cmdline "console=ttyS0 root=/dev/vda1" \
-        --disk path=/dev/vdb --disk path=/dev/vdc \
-        --api-socket /tmp/api1 \
+        --disk path=/dev/vdb path=/dev/vdc \
+        --api-socket=/tmp/api1 \
         --net ip=192.168.100.1
 vm-1:~$ # setup the guest network if needed
 vm-1:~$ sudo ip addr add 192.168.101.2/24 dev ens4
@@ -108,7 +108,7 @@ echo "tmp = $tmp"
 
 Launch the nested destination VM (inside the guest OS of the VM 2):
 ```bash
-vm-2:~$ sudo ./cloud-hypervisor --api-socket /tmp/api2
+vm-2:~$ sudo ./cloud-hypervisor --api-socket=/tmp/api2
 vm-2:~$ # setup the guest network with the following commands if needed
 vm-2:~$ sudo ip addr add 192.168.102.2/24 dev ens4
 vm-2:~$ sudo ip link set up dev ens4
@@ -122,7 +122,7 @@ vm-2:~$ ping 192.168.101.2 # This should succeed
 Get ready for receiving migration for the nested destination VM (inside
 the guest OS of the VM 2):
 ```bash
-vm-2:~$ sudo ./ch-remote --api-socket /tmp/api2 receive-migration unix:/tmp/sock2
+vm-2:~$ sudo ./ch-remote --api-socket=/tmp/api2 receive-migration unix:/tmp/sock2
 vm-2:~$ sudo socat TCP-LISTEN:6000,reuseaddr UNIX-CLIENT:/tmp/sock2
 ```
 
@@ -130,7 +130,7 @@ Start to send migration for the nested source VM (inside the guest OS of
 the VM 1):
 ```bash
 vm-1:~$ sudo socat UNIX-LISTEN:/tmp/sock1,reuseaddr TCP:192.168.102.2:6000
-vm-1:~$ sudo ./ch-remote --api-socket /tmp/api1 send-migration unix:/tmp/sock1
+vm-1:~$ sudo ./ch-remote --api-socket=/tmp/api1 send-migration unix:/tmp/sock1
 ```
 
 When the above commands completed, the source VM should be successfully

docs/logging.md

@@ -38,6 +38,6 @@ This level is for the benefit of developers. It should be used for sporadic and
 
 ### `debug!()`
 
-Use `-v -v` to enable.
+Use `-vv` to enable.
 
-For the most verbose of logging messages. It is acceptable to "spam" the log with repeated invocations of the same message. This level of logging would be combined with `--log-file`.
+For the most verbose of logging messages. It is acceptable to "spam" the log with repeated invocations of the same message. This level of logging would be combined with `--log-file`.

docs/memory.md

@@ -20,13 +20,12 @@ struct MemoryConfig {
     hugepages: bool,
     hugepage_size: Option<u64>,
     prefault: bool,
-    thp: bool
     zones: Option<Vec<MemoryZoneConfig>>,
 }
 ```
 
 ```
---memory <memory>	Memory parameters "size=<guest_memory_size>,mergeable=on|off,shared=on|off,hugepages=on|off,hugepage_size=<hugepage_size>,hotplug_method=acpi|virtio-mem,hotplug_size=<hotpluggable_memory_size>,hotplugged_size=<hotplugged_memory_size>,prefault=on|off,thp=on|off" [default: size=512M,thp=on]
+--memory <memory>	Memory parameters "size=<guest_memory_size>,mergeable=on|off,shared=on|off,hugepages=on|off,hugepage_size=<hugepage_size>,hotplug_method=acpi|virtio-mem,hotplug_size=<hotpluggable_memory_size>,hotplugged_size=<hotplugged_memory_size>,prefault=on|off" [default: size=512M]
 ```
 
 ### `size`
@@ -118,9 +117,6 @@ needing access to the guest RAM content.
 By default this option is turned off, which results in performing `mmap(2)`
 with `MAP_PRIVATE` flag.
 
-If `hugepages=on` then the value of this field is ignored as huge pages always
-requires `MAP_SHARED`.
-
 _Example_
 
 ```
@@ -144,9 +140,6 @@ behaviour, e.g. error with `ReadKernelImage` is common. If there is a strange
 error with `hugepages` enabled, just disable it or check whether there are enough
 huge pages.
 
-If `hugepages=on` then the value of `shared` is ignored as huge pages always
-requires `MAP_SHARED`.
-
 By default this option is turned off.
 
 _Example_
@@ -178,25 +171,6 @@ _Example_
 --memory size=1G,prefault=on
 ```
 
-### `thp`
-
-Specifies if private anonymous memory for the guest (i.e. `shared=off` and no
-backing file) should be labelled `MADV_HUGEPAGE` with `madvise(2)` indicating
-to the kernel that this memory may be backed with huge pages transparently.
-
-The use of transparent huge pages can improve the performance of the guest as
-there will fewer virtualisation related page faults. Unlike using
-`hugepages=on` a specific number of huge pages do not need to be allocated by
-the kernel.
-
-By default this option is turned on.
-
-_Example_
-
-```
---memory size=1G,thp=on
-```
-
 ## Advanced Parameters
 
 `MemoryZoneConfig` or what is known as `--memory-zone` from the CLI perspective
@@ -266,8 +240,11 @@ _Example_
 
 ### `file`
 
-Path to the file backing the memory zone. The file will be opened and used as
-the backing file for the `mmap(2)` operation.
+Path to the file backing the memory zone. This can be either a file or a
+directory. In case of a file, it will be opened and used as the backing file
+for the `mmap(2)` operation. In case of a directory, a temporary file with no
+hard link on the filesystem will be created. This file will be used as the
+backing file for the `mmap(2)` operation.
 
 This option can be particularly useful when trying to back a part of the guest
 RAM with a well known file. In the context of the snapshot/restore feature, and
@@ -294,9 +271,6 @@ other processes running on the host. One can use this option when running
 vhost-user devices as part of the VM device model, as they will be driven
 by standalone daemons needing access to the guest RAM content.
 
-If `hugepages=on` then the value of this field is ignored as huge pages always
-requires `MAP_SHARED`.
-
 By default this option is turned off, which result in performing `mmap(2)`
 with `MAP_PRIVATE` flag.
 
@@ -324,9 +298,6 @@ behaviour, e.g. error with `ReadKernelImage` is common. If there is a strange
 error with `hugepages` enabled, just disable it or check whether there are enough
 huge pages.
 
-If `hugepages=on` then the value of `shared` is ignored as huge pages always
-requires `MAP_SHARED`.
-
 By default this option is turned off.
 
 _Example_
@@ -516,7 +487,7 @@ different distances, it can be described with the following example.
 _Example_
 
 ```
---numa guest_numa_id=0,distances=[1@15,2@25] --numa guest_numa_id=1,distances=[0@15,2@20] guest_numa_id=2,distances=[0@25,1@20]
+--numa guest_numa_id=0,distances=[1@15,2@25] guest_numa_id=1,distances=[0@15,2@20] guest_numa_id=2,distances=[0@25,1@20]
 ```
 
 ### `memory_zones`
@@ -540,14 +511,14 @@ demarcate the list.
 
 Note that a memory zone must belong to a single NUMA node. The following
 configuration is incorrect, therefore not allowed:
-`--numa guest_numa_id=0,memory_zones=mem0 --numa guest_numa_id=1,memory_zones=mem0`
+`--numa guest_numa_id=0,memory_zones=mem0 guest_numa_id=1,memory_zones=mem0`
 
 _Example_
 
 ```
 --memory size=0
---memory-zone id=mem0,size=1G id=mem1,size=1G --memory-zone id=mem2,size=1G
---numa guest_numa_id=0,memory_zones=[mem0,mem2] --numa guest_numa_id=1,memory_zones=mem1
+--memory-zone id=mem0,size=1G id=mem1,size=1G id=mem2,size=1G
+--numa guest_numa_id=0,memory_zones=[mem0,mem2] guest_numa_id=1,memory_zones=mem1
 ```
 
 ### `sgx_epc_sections`
@@ -567,7 +538,7 @@ _Example_
 
 ```
 --sgx-epc id=epc0,size=32M id=epc1,size=64M id=epc2,size=32M
---numa guest_numa_id=0,sgx_epc_sections=epc1 --numa guest_numa_id=1,sgx_epc_sections=[epc0,epc2]
+--numa guest_numa_id=0,sgx_epc_sections=epc1 guest_numa_id=1,sgx_epc_sections=[epc0,epc2]
 ```
 
 ### PCI bus

docs/performance_metrics.md

@@ -1,29 +0,0 @@
-# Performance Metrics
-
-Cloud Hypervisor provides a [performance metrics](https://github.com/cloud-hypervisor/cloud-hypervisor/tree/main/performance-metrics)
-binary for users to generate metrics data from their own
-environment. This document describes how to generate metrics data
-quickly by using Cloud Hypervisor's development script,
-e.g. `dev_cli.sh`. The only prerequisite is [Docker installation](https://docs.docker.com/engine/install/).
-Please note that upon its first invocation, this script will pull a
-fairly large container image.
-
-To generate metrics data for all available performance tests (including
-boot time, block I/O throughput, and network throughput & latency) and
-output the result into a json file:
-
-```
-$ ./scripts/dev_cli.sh tests --metrics -- -- --report-file /tmp/metrics.json
-```
-
-To get a list of available performance tests:
-
-```
-$ ./scripts/dev_cli.sh tests --metrics -- -- --list-tests
-```
-
-To generate metrics data for selected performance tests, e.g. boot time only:
-
-```
-$ ./scripts/dev_cli.sh tests --metrics -- -- --report-file /tmp/metrics.json --test-filter boot_time
-```

docs/profiling.md

@@ -4,12 +4,20 @@
 
 ## Building a suitable binary
 
+Modify the `Cargo.toml` file to add `debug = 1` to the `[profile.release]` block. It should look like this:
+
+```
+[profile.release]
+lto = true
+debug = 1
+```
+
 This adds the symbol information to the release binary but does not otherwise affect the performance.
 
 The binary must also be built with frame pointers included so that the call graph can be captured by the profiler.
 
 ```
-$ cargo clean && RUSTFLAGS='-C force-frame-pointers=y' cargo build --profile profiling
+$ cargo clean && RUSTFLAGS='-C force-frame-pointers=y' cargo build --release
 ```
 
 ## Profiling
@@ -19,13 +27,13 @@ $ cargo clean && RUSTFLAGS='-C force-frame-pointers=y' cargo build --profile pro
 e.g.
 
 ```
-$ perf record -g target/profiling/cloud-hypervisor \
+$ perf record -g target/release/cloud-hypervisor \
         --kernel ~/src/linux/vmlinux \
         --pmem file=~/workloads/focal.raw \
         --cpus boot=1 --memory size=1G \
         --cmdline "root=/dev/pmem0p1 console=ttyS0" \
         --serial tty --console off \
-        --api-socket /tmp/api1
+        --api-socket=/tmp/api1
 ```
 
 For analysing the samples:
@@ -35,22 +43,3 @@ $ perf report -g
 ```
 
 If profiling with a network device attached either the TAP device must be already created and configured or the profiling must be done as root so that the TAP device can be created.
-
-## Userspace only profiling with LBR
-
-The use of LBR (Last Branch Record; available since Haswell) offers lower
-overhead if only userspace profiling is required. This lower overhead can allow
-a higher frequency of sampling. This also removes the requirement to compile
-with custom `RUSTFLAGS` however debug symbols should still be included:
-
-e.g.
-
-```
-$ perf record --call-graph lbr --all-user --user-callchains -g target/release/cloud-hypervisor \
-        --kernel ~/src/linux/vmlinux \
-        --pmem file=~/workloads/focal.raw \
-        --cpus boot=1 --memory size=1G \
-        --cmdline "root=/dev/pmem0p1 console=ttyS0" \
-        --serial tty --console off \
-        --api-socket /tmp/api1
-```

docs/releases.md

@@ -1,103 +0,0 @@
-# Release Documentation
-
-## Abstract
-
-This document provides guidance to users, downstream maintainers and
-any other consumers of the Cloud Hypervisor project, this document
-describes the release process, release cadence, stability expectations and
-related topics.
-
-## Basic Terms
-
-### Stability
-
-For Cloud Hypervisor the following areas are subject to stability guarantees:
-
-- [REST API](api.md#rest-api)
-- [Command line options](api.md#command-line-interface)
-- [Device Model](device_model.md)
-- Device tree, device list, ACPI, Hyper-V enlightenments and any other
-  features exposed to guest
-- KVM compatibility
-- Rust edition compatibility
-
-This list is incomplete but this document serves as a best effort guide to stability
-across releases.
-
-### Experimental features
-
-Experimental features are under active development and no guarantees are made about their stability.
-
-List of experimental features:
-
-- TDX
-- vfio-user
-- vDPA
-
-### Security
-
-Security fixes should be included in a new point release.
-
-For security issues an advisory will be published via the GitHub security advisory process along with the release. Watching the project on GitHub will notify you of those issues.
-
-## Releases
-
-### Versioning
-
-The versioning scheme uses `MAJOR.POINT` pattern:
-
-- `MAJOR` can introduce incompatible changes along with support for new features. Changes to the [API](api.md#rest-api),
-  [CLI options](api.md#command-line-interface) and [device model](device_model.md)
-  require a notice at least 2 releases in advance for the actual change to take
-  place.
-- `POINT` contains bug fixes and/or security fixes.
-
-### Major Release Cadence
-
-Cloud Hypervisor is under active development. A new major release is issued approximately
-every 6 weeks. Point releases are issued on demand, when important bug fixes are in
-the queue. A major release would receive bug fixes for the next two cycles (~12 weeks)
-and then be considered EOL.
-
-```
-+ - Active release support
-E - EOL
-
-        2021                2022                2023 
-         |    |    |    |    |    |    |    |    | 
-18.0     |    |    |  ++++++++E
-19.0     |    |    |    |++++++++E
-20.0     |    |    |    |   ++++++++E
-21.0     |    |    |    |    | ++++++++E
-22.0     |    |    |    |    |    +++++++++E
-23.0     |    |    |    |    |    |  +++++++++E
-
-```
-
-### Major Release Stability Considerations
-
-Snapshot/restore support is not compatible across `MAJOR` versions.
-Live migration support is not compatible across `MAJOR` versions.
-
-### LTS Release Cadence
-
-A regular release is promoted to LTS every 12 months. An LTS release is supported for 18 months. This gives a 6 months window for users to move to the new LTS.
-
-```
-+ - Active release support
-E - EOL
-
-        2022                2023               2024                 2025                2026
-         |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
-23.0     |    |+++++++++++++++++++++++++++++E
-43.0     |    |    |    |    |    |+++++++++++++++++++++++++++++E
-63.0     |    |    |    |    |    |    |    |    |    |+++++++++++++++++++++++++++++E
-
-```
-
-### LTS Stablity Considerations
-
-An LTS release is just a `MAJOR` release for which point releases are made for
-longer following the same rules for what can be backported to a `POINT` release.
-The focus lays on critical and security bug fixes which are pulled at the
-maintainer's discretion.

docs/seccomp.md

@@ -64,5 +64,5 @@ it will log every system call issued by the process. It is important to use
 `-f` option in order to trace each and every thread belonging to the process.
 
 ```
-strace --decode-pids=comm -f ./cloud-hypervisor ...
+strace -f ./cloud-hypervisor ...
 ```

docs/snapshot_restore.md

@@ -25,14 +25,14 @@ First thing, we must run a Cloud Hypervisor VM:
 At any point in time when the VM is running, one might choose to pause it:
 
 ```bash
-./ch-remote --api-socket /tmp/cloud-hypervisor.sock pause
+./ch-remote --api-socket=/tmp/cloud-hypervisor.sock pause
 ```
 
 Once paused, the VM can be safely snapshot into the specified directory and
 using the following command:
 
 ```bash
-./ch-remote --api-socket /tmp/cloud-hypervisor.sock snapshot file:///home/foo/snapshot
+./ch-remote --api-socket=/tmp/cloud-hypervisor.sock snapshot file:///home/foo/snapshot
 ```
 
 Given the directory was present on the system, the snapshot will succeed and
@@ -79,7 +79,7 @@ Or using two different commands from two terminals:
 ./cloud-hypervisor --api-socket /tmp/cloud-hypervisor.sock
 
 # Second terminal
-./ch-remote --api-socket /tmp/cloud-hypervisor.sock restore source_url=file:///home/foo/snapshot
+./ch-remote --api-socket=/tmp/cloud-hypervisor.sock restore source_url=file:///home/foo/snapshot
 ```
 
 Remember the VM is restored in a `paused` state, which was the VM's state when
@@ -87,7 +87,7 @@ it was snapshot. For this reason, one must explicitly `resume` the VM before to
 start using it.
 
 ```bash
-./ch-remote --api-socket /tmp/cloud-hypervisor.sock resume
+./ch-remote --api-socket=/tmp/cloud-hypervisor.sock resume
 ```
 
 At this point, the VM is fully restored and is identical to the VM which was

docs/tpm.md

@@ -1,122 +0,0 @@
-# TPM
-Tpm in Cloud-Hypervisor is emulated using `swtpm` as the backend. [swtpm](https://github.com/stefanberger/swtpm) is the link to swtpm project.
-
-Current implementation only supports TPM `2.0` version. At the moment only
-`CRB Interface` is implemented. This interface is described in
-[TCG PC Client Platform TPM Profile Specification for TPM 2.0, Revision 01.05 v4](https://trustedcomputinggroup.org/wp-content/uploads/PC-Client-Specific-Platform-TPM-Profile-for-TPM-2p0-v1p05p_r14_pub.pdf).
-
-
-## Usage
-`--tpm`, an optional argument, can be passed to enable tpm device.
-This argument takes an UNIX domain Socket as a `socket` value.
-
-_Example_
-
-An Example invocation with `--tpm` argument:
-
-```
- ./cloud-hypervisor/target/release/cloud-hypervisor \
-	--kernel ./hypervisor-fw \
-	--disk path=focal-server-cloudimg-amd64.raw \
-	--cpus boot=4 \
-	--memory size=1024M \
-	--net "tap=,mac=,ip=,mask=" \
-	--tpm socket="/var/run/swtpm.socket"
-```
-
-## swtpm
-Before invoking cloud-hypervisor with `--tpm` argument, a `swtpm`
-process should be started to listen at the input socket. Below is an
-example invocation of swtpm process.
-
-```
-swtpm socket --tpmstate dir=/var/run/swtpm \
-	--ctrl type=unixio,path="/var/run/swtpm.socket" \
-	--flags startup-clear \
-	--tpm2
-```
-
-## Guest
-After starting a guest with the above commands, ensure below listed modules are
-loaded in the guest:
-
-```
-# lsmod | grep tpm
-tpm_crb                20480  0
-tpm                    81920  1 tpm_crb
-```
-
-Below is the IO Memory map configured in the guest:
-
-```
-# cat /proc/iomem  | grep MSFT
-fed40000-fed40fff : MSFT0101:00
-  fed40000-fed40fff : MSFT0101:00
-```
-Below are the devices created in the guest:
-
-```
-# ls /dev/tpm*
-/dev/tpm0  /dev/tpmrm0
-```
-
-
-## Testing
-
-Inside the guest install `tpm2-tools` package. This package provides some
-commands to run against TPM that supports 2.0 version.
-
-_Examples_
-```
-// Run Self Test
-# tpm2_selftest -f
-# echo $?
-0
-
-
-# echo "hello" > input.txt
-// this command generates hash of the input file using all the algos supported by TPM
-
-# tpm2_pcrevent input.txt
-sha1: f572d396fae9206628714fb2ce00f72e94f2258f
-sha256: 5891b5b522d5df086d0ff0b110fbd9d21bb4fc7163af34d08286a2e846f6be03
-sha384: 1d0f284efe3edea4b9ca3bd514fa134b17eae361ccc7a1eefeff801b9bd6604e01f21f6bf249ef030599f0c
-218f2ba8c
-sha512: e7c22b994c59d9cf2b48e549b1e24666636045930d3da7c1acb299d1c3b7f931f94aae41edda2c2b207a36e
-10f8bcb8d45223e54878f5b316e7ce3b6bc019629
-
-// verify one of the hashes
-# sha256sum input.txt
-5891b5b522d5df086d0ff0b110fbd9d21bb4fc7163af34d08286a2e846f6be03  input.txt
-```
-
-### Bundled Functional Test
-
-Build time dependencies for `tpm2-tss` are captured in [INSTALL](https://github.com/tpm2-software/tpm2-tss/blob/master/INSTALL.md).
-
-```
-# git clone https://github.com/tpm2-software/tpm2-tss.git
-# cd tpm2-tss
-# ./configure --enable-integration --with-devicetests="mandatory,optional" --with-device=/dev/tpm0
-# sudo make check-device
-.
-.
-.
-.
-============================================================================
-Testsuite summary for tpm2-tss 3.2.0-74-ge03617d9
-============================================================================
-# TOTAL: 154
-# PASS:  88
-# SKIP:  7
-# XFAIL: 0
-# FAIL:  59
-# XPASS: 0
-# ERROR: 0
-============================================================================
-See ./test-suite.log
-Please report to https://github.com/tpm2-software/tpm2-tss/issues
-============================================================================
-```
-The same set of failures are noticed while running these tests on `Qemu` with
-its TPM implementation.

docs/tracing.md

@@ -1,42 +0,0 @@
-# Tracing
-
-Cloud Hypervisor has a basic tracing infrastucture, particularly focussed on
-the tracing of the initial VM setup.
-
-## Usage
-
-To enabling tracing, build with "tracing" feature. When compiled without the
-feature the tracing is compiled out.
-
-```bash
-cargo build --features "tracing"
-```
-
-And then run Cloud Hypervisor as you wish, the trace will be written to the current directory as `cloud-hypervisor-<pid>.trace`. This is JSON file which you can inspect yourself.
-
-Alternatively you can use the provided script in
-`scripts/ch-trace-visualiser.py` to generate an SVG:
-
-For example: 
-
-```bash
-scripts/ch-trace-visualiser.py cloud-hypervisor-39466.trace output.svg
-```
-
-## Tracing in the codebase
-
-There are existing tracepoints in the code base; extra ones can be added for
-more detailed tracing.
-
-The `tracer::trace_scoped!()` macro is used to add the current existing scope
-appears as a block in the trace. Other than providing a useful name for the
-event nothing else is required from the developer.
-
-The `tracer::start()` and `tracer::end()` functions are already in place for
-generating traces of the boot. These can be relocated for focus tracing on a
-narrow part of the code base.
-
-A `tracer::trace_point!()` macro is also provided for an instantaneous trace
-point however this is not in use in the code base currently nor is handled by
-the visualisation script due to the difficulty in representation in the SVG.
-

docs/uefi.md

@@ -2,19 +2,17 @@
 
 Cloud Hypervisor supports UEFI boot through the utilization of the EDK II based UEFI firmware. 
 
-## Building UEFI Firmware for x86-64
+## Building UEFI Firmware
 
 To avoid any unnecessary issues, it is recommended to use Ubuntu 18.04 and its default toolset. Any other compatible Linux distribution is otherwise suitable, however it is suggested to use a temporary Docker container with Ubuntu 18.04 for a quick build on an existing Linux machine.
 
-Please note that nasm-2.15 is required for the build to succeed.
-
 The commands below will compile an OVMF firmware suitable for Cloud Hypervisor.
 
 ```shell
 sudo apt-get update
 sudo apt-get install uuid-dev nasm iasl build-essential python3-distutils git
 
-git clone https://github.com/tianocore/edk2
+git clone https://github.com/cloud-hypervisor/edk2 -b ch
 cd edk2
 . edksetup.sh
 git submodule update --init
@@ -29,40 +27,11 @@ build
 
 After the successful build, the resulting firmware binaries are available under `Build/CloudHvX64/DEBUG_GCC5/FV` underneath the edk2 checkout.
 
-## Building UEFI Firmware for AArch64
-
-```shell
-# On an AArch64 machine:
-$ sudo apt-get update
-$ sudo apt-get install uuid-dev nasm iasl build-essential python3-distutils git
-$ git clone --depth 1 https://github.com/tianocore/edk2.git -b master
-$ cd edk2
-$ git submodule update --init
-$ cd ..
-$ git clone --depth 1 https://github.com/tianocore/edk2-platforms.git -b master
-$ git clone --depth 1 https://github.com/acpica/acpica.git -b master
-
-# Build tools
-$ export PACKAGES_PATH="$PWD/edk2:$PWD/edk2-platforms"
-$ export IASL_PREFIX="$PWD/acpica/generate/unix/bin/"
-$ make -C acpica
-$ cd edk2/
-$ . edksetup.sh
-$ cd ..
-$ make -C edk2/BaseTools
-
-# Build EDK2
-$ build -a AARCH64 -t GCC5 -p ArmVirtPkg/ArmVirtCloudHv.dsc -b RELEASE
-```
-
-If the build goes well, the EDK2 binary is available at
-`edk2/Build/ArmVirtCloudHv-AARCH64/RELEASE_GCC5/FV/CLOUDHV_EFI.fd`.
-
 ## Using OVMF Binaries
 
 Any UEFI capable image can be booted using the Cloud Hypervisor specific firmware. Windows guests under Cloud Hypervisor only support UEFI boot, therefore OVMF is mandatory there.
 
-To make Cloud Hypervisor use UEFI boot, pass the `CLOUDHV.fd` (for x86-64) / `CLOUDHV_EFI.fd` (for AArch64) file path as an argument to the `--kernel` option. The firmware file will be opened in read only mode.
+To make Cloud Hypervisor use UEFI boot, pass the `CLOUDHV.fd` file path as an argument to the `--kernel` option. The firmware file will be opened in read only mode.
 
 # Links
 

docs/vdpa.md

@@ -1,154 +0,0 @@
-# Virtio Data Path Acceleration
-
-vDPA aims at achieving bare-metal performance for devices passed into a virtual
-machine. It is an alternative to VFIO, as it provides a simpler solution for
-achieving migration.
-
-It is a kernel framework introduced recently to handle devices complying with
-the VIRTIO specification on their data-path, while the control path is vendor
-specific. In practice, virtqueues are accessed directly through DMA mechanism
-between the hardware and the guest. The control path is accessed through the
-vDPA framework, being exposed through the vhost interface as a vhost-vdpa
-device.
-
-Because DMA accesses between device and guest are going through virtqueues,
-migration can be achieved without requiring device's driver to implement any
-specific migration support. In case of VFIO, each vendor is expected to provide
-an implementation of the VFIO migration framework, complicating things as it
-must be done for each and every device's driver.
-
-The official [website](https://vdpa-dev.gitlab.io/) contains some extensive
-documentation on the topic.
-
-## Usage
-
-`VdpaConfig` (known as `--vdpa` from the CLI perspective) contains the list of
-parameters available for the vDPA device.
-
-```rust
-struct VdpaConfig {
-    path: PathBuf,
-    num_queues: usize,
-    id: Option<String>,
-    pci_segment: u16,
-}
-```
-
-```
---vdpa <vdpa>	vDPA device "path=<device_path>,num_queues=<number_of_queues>,iommu=on|off,id=<device_id>,pci_segment=<segment_id>"
-```
-
-### `path`
-
-Path of the vDPA device. Usually `/dev/vhost-vdpa-X`.
-
-This parameter is mandatory.
-
-Value is a string.
-
-_Example_
-
-```
---vdpa path=/dev/vhost-vdpa-0
-```
-
-### `num_queues`
-
-Number of virtqueues supported by the vDPA device.
-
-This parameter is optional.
-
-Value is an unsigned integer set to `1` by default.
-
-_Example_
-
-```
---vdpa path=/dev/vhost-vdpa-0,num_queues=2
-```
-
-### `id`
-
-Identifier of the vDPA device.
-
-This parameter is optional. If provided, it must be unique across the entire
-virtual machine.
-
-Value is a string.
-
-_Example_
-
-```
---vdpa path=/dev/vhost-vdpa-0,id=vdpa0
-```
-
-### `pci_segment`
-
-PCI segment number to which the vDPA device should be attached to.
-
-This parameter is optional.
-
-Value is an unsigned integer of 16 bits set to `0` by default.
-
-_Example_
-
-```
---vdpa path=/dev/vhost-vdpa-0,pci_segment=1
-```
-
-## Example with vDPA block simulator
-
-The vDPA framework provides a simulator with both `virtio-block` and
-`virtio-net` implementations. This is very useful for testing vDPA when we
-don't have access to the specific hardware.
-
-Given the host kernel has the appropriate modules available, let's load them
-all:
-
-```
-sudo modprobe vdpa
-sudo modprobe vhost_vdpa
-sudo modprobe vdpa_sim
-sudo modprobe vdpa_sim_blk
-```
-
-Given you have the `iproute2/vdpa` tool installed, let's now create the
-`virtio-block` vDPA device:
-
-```sh
-sudo vdpa dev add name vdpa-blk1 mgmtdev vdpasim_blk
-sudo chown $USER:$USER /dev/vhost-vdpa-0
-sudo chmod 660 /dev/vhost-vdpa-0
-```
-
-Increase the maximum locked memory to ensure setting up IOMMU mappings will
-succeed:
-
-```sh
-ulimit -l unlimited
-```
-
-Start Cloud Hypervisor:
-
-```sh
-cloud-hypervisor \
-    --cpus boot=1 \
-    --memory size=1G,hugepages=on \
-    --disk path=focal-server-cloudimg-amd64.raw \
-    --kernel vmlinux \
-    --cmdline "root=/dev/vda1 console=hvc0" \
-    --vdpa path=/dev/vhost-vdpa-0,num_queues=1
-```
-
-The `virtio-block` device backed by the vDPA simulator can be found as
-`/dev/vdb` in the guest:
-
-```
-cloud@cloud:~$ lsblk
-NAME    MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
-nullb0  252:0    0  250G  0 disk 
-vda     254:0    0  2.2G  0 disk 
-├─vda1  254:1    0  2.1G  0 part /
-├─vda14 254:14   0    4M  0 part 
-└─vda15 254:15   0  106M  0 part /boot/efi
-vdb     254:16   0  128M  0 disk
-```

docs/vhost-user-blk-testing.md

@@ -94,7 +94,7 @@ VMs run in client mode. They connect to the socket created by the `dpdkvhostuser
         --memory size=1024M,hugepages=on,shared=true \
         --kernel linux/arch/x86/boot/compressed/vmlinux.bin \
         --cmdline "console=ttyS0 root=/dev/vda1 rw iommu=off" \
-        --disk path=images/focal-server-cloudimg-amd64.raw --disk vhost_user=true,socket=/var/tmp/vhost.1,num_queues=4,queue_size=128 \
+        --disk path=images/focal-server-cloudimg-amd64.raw vhost_user=true,socket=/var/tmp/vhost.1,num_queues=4,queue_size=128 \
         --console off \
         --serial tty \
         --rng