.github: Re-order release steps to ensure binaries are available

Since we run "cargo clean" before running the aarch64 build we need to
create the release and upload the x86-64 assets before the clean.

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

.github/workflows/build.yaml

@@ -34,28 +34,22 @@ jobs:
           override: true
 
       - name: Build (default features)
-        run: cargo rustc --locked --bin cloud-hypervisor -- -D warnings
+        run: cargo rustc --locked --bin cloud-hypervisor -- -D warnings -D clippy::undocumented_unsafe_blocks
 
-      - name: Build (common + kvm)
-        run: cargo rustc --locked --bin cloud-hypervisor --no-default-features --features "common,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
 
       - name: Build (default features + tdx)
-        run: cargo rustc --locked --bin cloud-hypervisor --features "tdx" -- -D warnings
-
-      - name: Build (default features + amx)
-        run: cargo rustc --locked --bin cloud-hypervisor --features "amx" -- -D warnings
-
-      - name: Build (default features + gdb)
-        run: cargo rustc --locked --bin cloud-hypervisor --features "gdb" -- -D warnings
+        run: cargo rustc --locked --bin cloud-hypervisor --features "tdx" -- -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
+        run: cargo rustc --locked --bin cloud-hypervisor --features "guest_debug" -- -D warnings -D clippy::undocumented_unsafe_blocks
 
-      - name: Build (common + mshv)
-        run: cargo rustc --locked --bin cloud-hypervisor --no-default-features --features "common,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 (common + mshv + kvm)
-        run: cargo rustc --locked --bin cloud-hypervisor --no-default-features --features "common,mshv,kvm"  -- -D warnings
+      - name: Build (mshv + kvm)
+        run: cargo rustc --locked --bin cloud-hypervisor --no-default-features --features "mshv,kvm"  -- -D warnings -D clippy::undocumented_unsafe_blocks
 
       - name: Release Build (default features)
         run: cargo build --locked --all --release --target=${{ matrix.target }}

.github/workflows/quality-aarch64.yaml

@@ -1,53 +0,0 @@
-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: ${{ matrix.target }}
-          override: true
-          components: rustfmt, clippy
-
-      - name: Debug Check (default features)
-        run: |
-          git rev-list origin/main..$GITHUB_SHA | xargs -t -I % sh -c 'git checkout %; cargo check --tests --all --target=${{ matrix.target }}'
-          git checkout $GITHUB_SHA
-
-      - name: Formatting (rustfmt)
-        run: cargo fmt -- --check
-
-      - name: Clippy (common + kvm)
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: true
-          command: clippy
-          args: --target=${{ matrix.target }} --tests --all --no-default-features --features "common,kvm" -- -D warnings
-
-      - name: Clippy (default features)
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: true
-          command: clippy
-          args: --target=${{ matrix.target }} --tests --all -- -D warnings

.github/workflows/quality.yaml

@@ -13,15 +13,31 @@ 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@v2
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+
       - name: Install Rust toolchain (${{ matrix.rust }})
         uses: actions-rs/toolchain@v1
         with:
@@ -31,33 +47,67 @@ jobs:
           components: rustfmt, clippy
 
       - name: Debug Check (default features)
+        if: ${{ matrix.target == 'x86_64-unknown-linux-gnu' }}
         run: |
-          git rev-list origin/main..$GITHUB_SHA | xargs -t -I % sh -c 'git checkout %; cargo check --tests --all --target=${{ matrix.target }}'
-          git checkout $GITHUB_SHA
+          set -e
+          commits=$(git rev-list origin/${{ github.base_ref }}..${{ github.sha }})
+          for commit in $commits; do git checkout $commit; cargo check --tests --all --target=${{ matrix.target }}; done
+          git checkout ${{ github.sha }}
 
       - name: Formatting (rustfmt)
         run: cargo fmt -- --check
 
-      - name: Clippy (common + kvm)
-        run: cargo clippy --locked --all --all-targets --no-default-features --tests --features "common,kvm" -- -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 --features "kvm" -- -D warnings -D clippy::undocumented_unsafe_blocks
 
       - name: Clippy (default features)
-        run: cargo clippy --locked --all --all-targets --tests -- -D warnings
-
-      - name: Clippy (default features + amx)
-        run: cargo clippy --locked --all --all-targets --tests --features "amx" -- -D warnings
-
-      - name: Clippy (default features + gdb)
-        run: cargo clippy --locked --all --all-targets --tests --features "gdb" -- -D warnings
+        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 -- -D warnings -D clippy::undocumented_unsafe_blocks
 
       - name: Clippy (default features + guest_debug)
-        run: cargo clippy --locked --all --all-targets --tests --features "guest_debug" -- -D warnings
+        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 --features "guest_debug" -- -D warnings -D clippy::undocumented_unsafe_blocks
 
-      - name: Clippy (common + mshv)
-        run: cargo clippy --locked --all --all-targets --no-default-features --tests --features "common,mshv" -- -D warnings
+      - 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 --features "tracing" -- -D warnings -D clippy::undocumented_unsafe_blocks
 
-      - name: Clippy (common + mshv + kvm)
-        run: cargo clippy --locked --all --all-targets --no-default-features --tests --features "common,mshv,kvm" -- -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 --features "mshv" -- -D warnings -D clippy::undocumented_unsafe_blocks
+
+      - 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 --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 --features "tdx,kvm" -- -D warnings -D clippy::undocumented_unsafe_blocks
 
       - name: Check build did not modify any files
         run: test -z "$(git status --porcelain)"

.github/workflows/release.yaml

@@ -1,62 +1,47 @@
 name: Cloud Hypervisor Release
-on: [create]
+on: [pull_request, create]
 
 jobs:
   release:
-    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-${{ github.event.ref }}
       - name: Install Rust toolchain (x86_64-unknown-linux-gnu)
         uses: actions-rs/toolchain@v1
         with:
-            toolchain: "1.62"
-            target: x86_64-unknown-linux-gnu
+          toolchain: "1.62"
+          target: x86_64-unknown-linux-gnu
       - name: Install Rust toolchain (x86_64-unknown-linux-musl)
         uses: actions-rs/toolchain@v1
         with:
-            toolchain: "1.62"
-            target: x86_64-unknown-linux-musl
+          toolchain: "1.62"
+          target: x86_64-unknown-linux-musl
       - name: Build
         uses: actions-rs/cargo@v1
         with:
-            toolchain: "1.62"
-            command: build
-            args: --all --release --target=x86_64-unknown-linux-gnu
+          toolchain: "1.62"
+          command: build
+          args: --all --release --no-default-features --features "kvm,mshv" --target=x86_64-unknown-linux-gnu
       - name: Static Build
         uses: actions-rs/cargo@v1
         with:
-            toolchain: "1.62"
-            command: build
-            args: --all --release --target=x86_64-unknown-linux-musl
-      - name: Strip cloud-hypervisor binaries
-        run: strip target/*/release/cloud-hypervisor
+          toolchain: "1.62"
+          command: build
+          args: --all --release --no-default-features --features "kvm,mshv" --target=x86_64-unknown-linux-musl
       - name: Install Rust toolchain (aarch64-unknown-linux-musl)
         uses: actions-rs/toolchain@v1
         with:
-            toolchain: "1.62"
-            target: aarch64-unknown-linux-musl
-            override: true
-      - name: Static Build (AArch64)
-        uses: actions-rs/cargo@v1
-        with:
-            use-cross: true
-            command: build
-            args: --all --release --target=aarch64-unknown-linux-musl
-      - name: Vendor
-        working-directory: ../cloud-hypervisor-${{ github.event.ref }}
-        run: |
-          mkdir ../vendor-cargo-home
-          export CARGO_HOME=$(realpath ../vendor-cargo-home)
-          mkdir .cargo
-          cargo vendor > .cargo/config.toml
+          toolchain: "1.62"
+          target: aarch64-unknown-linux-musl
+          override: true
       - name: Create Release
+        if: github.event_name == 'create' && github.event.ref_type == 'tag'
         id: create_release
         uses: actions/create-release@v1
         env:
@@ -66,20 +51,8 @@ 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:
@@ -90,6 +63,7 @@ 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:
@@ -100,6 +74,7 @@ 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:
@@ -110,6 +85,7 @@ 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:
@@ -119,7 +95,18 @@ 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:
@@ -130,6 +117,7 @@ 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:
@@ -139,3 +127,24 @@ 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

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

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "cloud-hypervisor"
-version = "26.0.0"
+version = "29.0.0"
 authors = ["The Cloud Hypervisor Authors"]
 edition = "2021"
 default-run = "cloud-hypervisor"
@@ -10,58 +10,59 @@ 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 = "1.60"
 
 [profile.release]
 lto = true
+codegen-units = 1
+opt-level = "s"
+strip = true
 
 [dependencies]
-anyhow = "1.0.62"
+anyhow = "1.0.68"
 api_client = { path = "api_client" }
-clap = { version = "3.2.17", features = ["wrap_help","cargo"] }
+clap = { version = "4.0.32", features = ["wrap_help","cargo","string"] }
 epoll = "4.3.1"
 event_monitor = { path = "event_monitor" }
 hypervisor = { path = "hypervisor" }
-libc = "0.2.131"
+libc = "0.2.139"
 log = { version = "0.4.17", features = ["std"] }
 option_parser = { path = "option_parser" }
-seccompiler = "0.2.0"
-serde_json = "1.0.83"
+seccompiler = "0.3.0"
+serde_json = "1.0.89"
 signal-hook = "0.3.14"
-thiserror = "1.0.32"
+thiserror = "1.0.38"
+tpm = { path = "tpm"}
+tracer = { path = "tracer" }
 vmm = { path = "vmm" }
-vmm-sys-util = "0.10.0"
-vm-memory = "0.8.0"
-
-[build-dependencies]
-clap = { version = "3.2.17", features = ["cargo"] }
+vmm-sys-util = "0.11.0"
+vm-memory = "0.10.0"
 
 # List of patched crates
 [patch.crates-io]
-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" }
+kvm-bindings = { git = "https://github.com/cloud-hypervisor/kvm-bindings", branch = "ch-v0.6.0-tdx" }
 versionize_derive = { git = "https://github.com/cloud-hypervisor/versionize_derive", branch = "ch" }
 
 [dev-dependencies]
 dirs = "4.0.0"
 net_util = { path = "net_util" }
-once_cell = "1.13.1"
-serde_json = "1.0.83"
+once_cell = "1.16.0"
+serde_json = "1.0.89"
 test_infra = { path = "test_infra" }
 wait-timeout = "0.2.0"
 
 [features]
-default = ["common", "kvm"]
-# Common features for all hypervisors
-common = ["fwdebug"]
-amx = ["vmm/amx"]
-cmos = ["vmm/cmos"]
-fwdebug = ["vmm/fwdebug"]
-gdb = ["vmm/gdb"]
+default = ["kvm"]
 guest_debug = ["vmm/guest_debug"]
 kvm = ["vmm/kvm"]
 mshv = ["vmm/mshv"]
 tdx = ["vmm/tdx"]
+tracing = ["vmm/tracing", "tracer/tracing"]
 
 [workspace]
 members = [
@@ -79,7 +80,9 @@ members = [
     "performance-metrics",
     "qcow",
     "rate_limiter",
+    "serial_buffer",
     "test_infra",
+    "tracer",
     "vfio_user",
     "vhdx",
     "vhost_user_block",

Jenkinsfile

@@ -1,467 +1,449 @@
 def runWorkers = true
-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 fuzzer files only changes') {
-					when {
-						expression {
-							return fuzzFileOnly()
-						}
-					}
-					steps {
-						script {
-							runWorkers = false
-							echo "Fuzzer cargo files 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 '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 ('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 (musl)') {
-					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 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 '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-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 'jammy' } }
-					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')
-				}
-			}
-		}
-	}
+pipeline {
+    agent none
+    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('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-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 - 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')
+                }
+            }
+        }
+    }
 }
 
 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/ ${distro} main\" | sudo tee /etc/apt/sources.list.d/azure-cli.list"
-	sh "sudo apt update"
-	sh "sudo apt install -y azure-cli"
+    sh 'sudo apt install -y ca-certificates curl apt-transport-https lsb-release gnupg'
+    sh 'curl -sL https://packages.microsoft.com/keys/microsoft.asc | gpg --dearmor | sudo tee /etc/apt/trusted.gpg.d/microsoft.gpg > /dev/null'
+    sh "echo \"deb [arch=${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 boolean docsFileOnly() {
+def boolean skipWorkerBuild() {
     if (env.CHANGE_TARGET == null) {
-        return false;
+        return false
     }
 
-    return sh(
+    if (sh(
         returnStatus: true,
         script: "git diff --name-only origin/${env.CHANGE_TARGET}... | grep -v '\\.md'"
-    ) != 0
-}
-
-def boolean fuzzFileOnly() {
-    if (env.CHANGE_TARGET == null) {
-        return false;
+    ) != 0) {
+        return true
     }
 
-    return sh(
+    if (sh(
         returnStatus: true,
         script: "git diff --name-only origin/${env.CHANGE_TARGET}... | grep -v -E 'fuzz/'"
-    ) != 0
+    ) != 0) {
+        return true
+    }
+
+    if (sh(
+        returnStatus: true,
+        script: "git diff --name-only origin/${env.CHANGE_TARGET}... | grep -v -E '.github/'"
+    ) != 0) {
+        return true
+    }
+
+    return false
 }

README.md

@@ -4,42 +4,43 @@
     - [Architectures](#architectures)
     - [Guest OS](#guest-os)
 - [2. Getting Started](#2-getting-started)
-  - [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)
+  - [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)
       - [Disk image](#disk-image)
       - [Booting the guest VM](#booting-the-guest-vm)
 - [3. Status](#3-status)
   - [Hot Plug](#hot-plug)
   - [Device Model](#device-model)
-  - [TODO](#todo)
-- [4. `rust-vmm` project dependency](#4-rust-vmm-project-dependency)
-  - [Firecracker and crosvm](#firecracker-and-crosvm)
+  - [Roadmap](#roadmap)
+- [4. Relationship with _Rust VMM_ Project](#4-relationship-with-rust-vmm-project)
+  - [Differences with Firecracker and crosvm](#differences-with-firecracker-and-crosvm)
 - [5. Community](#5-community)
   - [Contribute](#contribute)
-  - [Join us](#join-us)
+  - [Slack](#slack)
+  - [Mailing list](#mailing-list)
   - [Security issues](#security-issues)
 
 # 1. What is Cloud Hypervisor?
 
 Cloud Hypervisor is an open source Virtual Machine Monitor (VMM) that runs on
-top of [KVM](https://www.kernel.org/doc/Documentation/virtual/kvm/api.txt)
-hypervisor and Microsoft Hypervisor (MSHV).
+top of the [KVM](https://www.kernel.org/doc/Documentation/virtual/kvm/api.txt)
+hypervisor and the Microsoft Hypervisor (MSHV).
 
-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
+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
 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
 
@@ -59,7 +60,7 @@ based on the [rust-vmm](https://github.com/rust-vmm) crates.
 ### Architectures
 
 Cloud Hypervisor supports the `x86-64` and `AArch64` architectures. There are
-some small differences in functionality between the two architectures
+minor differences in functionality between the two architectures
 (see [#1125](https://github.com/cloud-hypervisor/cloud-hypervisor/issues/1125)).
 
 ### Guest OS
@@ -68,187 +69,186 @@ Cloud Hypervisor supports `64-bit Linux` and Windows 10/Windows Server 2019.
 
 # 2. Getting Started
 
-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).
+The following sections describe how to build and run Cloud Hypervisor.
 
-## Preparation
+## Prerequisites for AArch64
 
-We create a folder to build and run `cloud-hypervisor` at `$HOME/cloud-hypervisor`
+- AArch64 servers (recommended) or development boards equipped with the GICv3
+  interrupt controller.
+
+## Host OS
+
+For required KVM functionality the minimum host kernel version is 4.11.  For
+adequate performance the minimum recommended host kernel version is 5.6. 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 Hyperivor
+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.
 
 ```shell
-$ export CLOUDH=$HOME/cloud-hypervisor
-$ mkdir $CLOUDH
-```
-
-## Install prerequisites
-
-You need to install some prerequisite packages in order to build and test Cloud
-Hypervisor. Here, all the steps are based on Ubuntu, for other Linux
-distributions please replace the package manager and package name.
-
-```shell
-# Install build-essential, git, and qemu-utils
-$ sudo apt install git build-essential qemu-utils
-# 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
-```
-
-## 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/0.4.1/hypervisor-fw
-$ popd
+$ wget https://github.com/cloud-hypervisor/rust-hypervisor-firmware/releases/download/0.4.2/hypervisor-fw
 ```
 
+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.
+
 ```shell
-$ pushd $CLOUDH
-$ sudo setcap cap_net_admin+ep ./cloud-hypervisor/target/release/cloud-hypervisor
-$ ./cloud-hypervisor/target/release/cloud-hypervisor \
+$ sudo setcap cap_net_admin+ep ./cloud-hypervisor
+$ ./create-cloud-init.sh
+$ ./cloud-hypervisor \
 	--kernel ./hypervisor-fw \
-	--disk path=focal-server-cloudimg-amd64.raw \
+	--disk path=focal-server-cloudimg-amd64.raw path=/tmp/ubuntu-cloudinit.img \
 	--cpus boot=4 \
 	--memory size=1024M \
 	--net "tap=,mac=,ip=,mask="
-$ popd
 ```
 
-Multiple arguments can be given to the `--disk` parameter.
+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`.
 
-### Custom kernel and disk image
+```shell
+$ ./cloud-hypervisor \
+	--kernel ./hypervisor-fw \
+	--disk path=focal-server-cloudimg-amd64.raw path=/tmp/ubuntu-cloudinit.img \
+	--cpus boot=4 \
+	--memory size=1024M \
+	--net "tap=,mac=,ip=,mask=" \
+	--serial tty \
+	--console off
+```
 
-#### Building your kernel
+### Custom Kernel and Disk Image
 
-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.
+#### 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.
 
 To build the kernel:
 
 ```shell
-
 # Clone the Cloud Hypervisor Linux branch
-$ 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 cloud-hypervisor kernel config to build your kernel
-$ cp $CLOUDH/cloud-hypervisor/resources/linux-config-x86_64 .config
+# Use the x86-64 cloud-hypervisor kernel config to build your kernel for x86-64
+$ 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
 $ KCFLAGS="-Wa,-mx86-used-note=no" make bzImage -j `nproc`
+# Do native build of the AArch64 kernel
+$ make -j `nproc`
 $ popd
 ```
 
-The `vmlinux` kernel image will then be located at
+For x86-64, 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, we will use a Ubuntu cloud image that contains a root
-partition:
+For the disk image the same Ubuntu image as before can be used. This contains
+an `ext4` root filesystem.
 
 ```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
-$ popd
+$ 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
 ```
 
 #### Booting the guest VM
 
-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:
+These sample commands boot the disk image using the custom kernel whilst also
+supplying the desired kernel command line.
+
+- x86-64
 
 ```shell
-$ pushd $CLOUDH
-$ sudo setcap cap_net_admin+ep ./cloud-hypervisor/target/release/cloud-hypervisor
-$ ./cloud-hypervisor/target/release/cloud-hypervisor \
+$ sudo setcap cap_net_admin+ep ./cloud-hypervisor
+$ ./create-cloud-init.sh
+$ ./cloud-hypervisor \
 	--kernel ./linux-cloud-hypervisor/arch/x86/boot/compressed/vmlinux.bin \
-	--disk path=focal-server-cloudimg-amd64.raw \
+	--disk path=focal-server-cloudimg-amd64.raw path=/tmp/ubuntu-cloudinit.img \
 	--cmdline "console=hvc0 root=/dev/vda1 rw" \
 	--cpus boot=4 \
 	--memory size=1024M \
 	--net "tap=,mac=,ip=,mask="
 ```
 
-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:
+- AArch64
 
+```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 path=/tmp/ubuntu-cloudinit.img \
+	--cmdline "console=hvc0 root=/dev/vda1 rw" \
+	--cpus boot=4 \
+	--memory size=1024M \
+	--net "tap=,mac=,ip=,mask="
 ```
-$ ./cloud-hypervisor/target/release/cloud-hypervisor \
+
+If earlier kernel messages are required the serial console should be used instead of `virtio-console`.
+
+- x86-64
+
+```shell
+$ ./cloud-hypervisor \
 	--kernel ./linux-cloud-hypervisor/arch/x86/boot/compressed/vmlinux.bin \
 	--console off \
 	--serial tty \
@@ -259,10 +259,24 @@ $ ./cloud-hypervisor/target/release/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
@@ -280,16 +294,17 @@ Currently the following items are **not** guaranteed across updates:
 * The following features are considered experimental and may change
   substantially between releases: TDX, vfio-user, vDPA.
 
-As of 2022-04-05, the following cloud images are supported:
+Further details can be found in the [release documentation](docs/releases.md).
 
-- [Ubuntu Bionic](https://cloud-images.ubuntu.com/bionic/current/) (cloudimg)
-- [Ubuntu Focal](https://cloud-images.ubuntu.com/focal/current/) (cloudimg)
-- [Ubuntu Jammy](https://cloud-images.ubuntu.com/jammy/current/) (cloudimg)
+As of 2023-01-03, 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/))
 
 Direct kernel boot to userspace should work with a rootfs from most
-distributions.
-
-Further details can be found in the [release documentation](docs/releases.md).
+distributions although you may need to enable exotic filesystem types in the
+reference kernel configuration (e.g. XFS or btrfs.)
 
 ## Hot Plug
 
@@ -302,14 +317,12 @@ Cloud Hypervisor supports hotplug of CPUs, passthrough devices (VFIO),
 Details of the device model can be found in this
 [documentation](docs/device_model.md).
 
-## TODO
+## Roadmap
 
-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.
+The project roadmap is tracked through a [GitHub
+project](https://github.com/orgs/cloud-hypervisor/projects/6).
 
-# 4. `rust-vmm` project dependency
+# 4. Relationship with _Rust VMM_ Project
 
 In order to satisfy the design goal of having a high-performance,
 security-focused hypervisor the decision was made to use the
@@ -318,39 +331,26 @@ 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 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:
+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.
 
-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
+## Differences with Firecracker and crosvm
 
 A large part of the Cloud Hypervisor code is based on either the Firecracker or
-the crosvm projects implementations. Both of these are VMMs written in Rust with
-a focus on safety and security, like Cloud Hypervisor.
+the crosvm project's implementations. Both of these are VMMs written in Rust
+with a focus on safety and security, like Cloud Hypervisor.
 
-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 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.
 
-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.
+The Cloud Hypervisor community thanks the communities of both the Firecracker
+and crosvm projects for their excellent work.
 
 # 5. Community
 
@@ -360,21 +360,28 @@ repository.
 
 ## Contribute
 
-We are working on building a global, diverse and collaborative community around
-the Cloud Hypervisor project. Anyone who is interested in
+The project strongly believes in 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.
 
-We believe that contributing to a open source project like Cloud Hypervisor
-covers a lot more than just sending code. Testing, documentation, pull request
+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.
 
-## Join us
+## Slack
 
 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/).
 
+## 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/).
+
 ## Security issues
 
 Please contact the maintainers listed in the MAINTAINERS.md file with security issues.

acpi_tables/Cargo.toml

@@ -5,4 +5,4 @@ authors = ["The Cloud Hypervisor Authors"]
 edition = "2021"
 
 [dependencies]
-vm-memory = "0.8.0"
+vm-memory = "0.10.0"

acpi_tables/src/sdt.rs

@@ -38,7 +38,6 @@ pub struct Sdt {
     data: Vec<u8>,
 }
 
-#[allow(clippy::len_without_is_empty)]
 impl Sdt {
     pub fn new(
         signature: [u8; 4],
@@ -98,6 +97,7 @@ impl Sdt {
     /// 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());
+        // SAFETY: The assertion above makes sure we don't do out of bounds write.
         unsafe {
             *(((self.data.as_mut_ptr() as usize) + offset) as *mut T) = value;
         }
@@ -123,6 +123,10 @@ impl Sdt {
     pub fn len(&self) -> usize {
         self.data.len()
     }
+
+    pub fn is_empty(&self) -> bool {
+        self.data.is_empty()
+    }
 }
 
 #[cfg(test)]

api_client/Cargo.toml

@@ -5,4 +5,4 @@ authors = ["The Cloud Hypervisor Authors"]
 edition = "2021"
 
 [dependencies]
-vmm-sys-util = "0.10.0"
+vmm-sys-util = "0.11.0"

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()])
 }
@@ -141,18 +141,19 @@ fn parse_http_response(socket: &mut dyn Read) -> Result<Option<String>, Error> {
     }
 }
 
-pub fn simple_api_command_with_fds<T: Read + Write + ScmSocket>(
+/// 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>(
     socket: &mut T,
     method: &str,
-    c: &str,
+    full_command: &str,
     request_body: Option<&str>,
     request_fds: Vec<RawFd>,
-) -> Result<(), Error> {
+) -> Result<Option<String>, Error> {
     socket
         .send_with_fds(
             &[format!(
-                "{} /api/v1/vm.{} HTTP/1.1\r\nHost: localhost\r\nAccept: */*\r\n",
-                method, c
+                "{method} /api/v1/{full_command} HTTP/1.1\r\nHost: localhost\r\nAccept: */*\r\n"
             )
             .as_bytes()],
             &request_fds,
@@ -175,12 +176,69 @@ pub fn simple_api_command_with_fds<T: Read + Write + ScmSocket>(
 
     socket.flush().map_err(Error::Socket)?;
 
-    if let Some(body) = parse_http_response(socket)? {
-        println!("{}", body);
+    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());
     }
+
     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

@@ -9,20 +9,20 @@ default = []
 tdx = []
 
 [dependencies]
-anyhow = "1.0.62"
+anyhow = "1.0.68"
 byteorder = "1.4.3"
 hypervisor = { path = "../hypervisor" }
-libc = "0.2.131"
-linux-loader = { version = "0.4.0", features = ["elf", "bzimage", "pe"] }
+libc = "0.2.139"
+linux-loader = { version = "0.8.1", features = ["elf", "bzimage", "pe"] }
 log = "0.4.17"
-serde = { version = "1.0.143", features = ["rc", "derive"] }
-thiserror = "1.0.32"
-uuid = "1.1.2"
-versionize = "0.1.6"
+serde = { version = "1.0.151", features = ["rc", "derive"] }
+thiserror = "1.0.38"
+uuid = "1.2.2"
+versionize = "0.1.9"
 versionize_derive = "0.1.4"
-vm-memory = { version = "0.8.0", features = ["backend-mmap", "backend-bitmap"] }
+vm-memory = { version = "0.10.0", features = ["backend-mmap", "backend-bitmap"] }
 vm-migration = { path = "../vm-migration" }
-vmm-sys-util = { version = "0.10.0", features = ["with-serde"] }
+vmm-sys-util = { version = "0.11.0", features = ["with-serde"] }
 
 [target.'cfg(target_arch = "aarch64")'.dependencies]
 fdt_parser = { version = "0.1.4", package = "fdt" }

arch/src/aarch64/fdt.rs

@@ -51,8 +51,6 @@ 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;
@@ -121,7 +119,7 @@ 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, vcpu_mpidr.len())?;
+        create_pmu_node(&mut fdt)?;
     }
     create_clock_node(&mut fdt)?;
     create_psci_node(&mut fdt)?;
@@ -162,7 +160,7 @@ fn create_cpu_nodes(
     let num_cpus = vcpu_mpidr.len();
 
     for (cpu_id, mpidr) in vcpu_mpidr.iter().enumerate().take(num_cpus) {
-        let cpu_name = format!("cpu@{:x}", cpu_id);
+        let cpu_name = format!("cpu@{cpu_id:x}");
         let cpu_node = fdt.begin_node(&cpu_name)?;
         fdt.property_string("device_type", "cpu")?;
         fdt.property_string("compatible", "arm,arm-v8")?;
@@ -194,15 +192,15 @@ fn create_cpu_nodes(
 
         // Create device tree nodes with regard of above mapping.
         for cluster_idx in 0..packages {
-            let cluster_name = format!("cluster{:x}", cluster_idx);
+            let cluster_name = format!("cluster{cluster_idx:x}");
             let cluster_node = fdt.begin_node(&cluster_name)?;
 
             for core_idx in 0..cores_per_package {
-                let core_name = format!("core{:x}", core_idx);
+                let core_name = format!("core{core_idx:x}");
                 let core_node = fdt.begin_node(&core_name)?;
 
                 for thread_idx in 0..threads_per_core {
-                    let thread_name = format!("thread{:x}", thread_idx);
+                    let thread_name = format!("thread{thread_idx:x}");
                     let thread_node = fdt.begin_node(&thread_name)?;
                     let cpu_idx = threads_per_core * cores_per_package * cluster_idx
                         + threads_per_core * core_idx
@@ -249,7 +247,7 @@ fn create_memory_node(
                     node_memory_addr = memory_region_start_addr;
                 }
             }
-            let memory_node_name = format!("memory@{:x}", node_memory_addr);
+            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)?;
@@ -261,7 +259,7 @@ fn create_memory_node(
         if last_addr < super::layout::MEM_32BIT_RESERVED_START.raw_value() {
             // Case 1: all RAM is under the hole
             let mem_size = last_addr - super::layout::RAM_START.raw_value() + 1;
-            let mem_reg_prop = [super::layout::RAM_START.raw_value() as u64, mem_size as u64];
+            let mem_reg_prop = [super::layout::RAM_START.raw_value(), mem_size];
             let memory_node = fdt.begin_node("memory")?;
             fdt.property_string("device_type", "memory")?;
             fdt.property_array_u64("reg", &mem_reg_prop)?;
@@ -271,7 +269,7 @@ fn create_memory_node(
             // Region 1: RAM before the hole
             let mem_size = super::layout::MEM_32BIT_RESERVED_START.raw_value()
                 - super::layout::RAM_START.raw_value();
-            let mem_reg_prop = [super::layout::RAM_START.raw_value() as u64, mem_size as u64];
+            let mem_reg_prop = [super::layout::RAM_START.raw_value(), mem_size];
             let memory_node_name = format!("memory@{:x}", super::layout::RAM_START.raw_value());
             let memory_node = fdt.begin_node(&memory_node_name)?;
             fdt.property_string("device_type", "memory")?;
@@ -280,10 +278,7 @@ fn create_memory_node(
 
             // Region 2: RAM after the hole
             let mem_size = last_addr - super::layout::RAM_64BIT_START.raw_value() + 1;
-            let mem_reg_prop = [
-                super::layout::RAM_64BIT_START.raw_value() as u64,
-                mem_size as u64,
-            ];
+            let mem_reg_prop = [super::layout::RAM_64BIT_START.raw_value(), mem_size];
             let memory_node_name =
                 format!("memory@{:x}", super::layout::RAM_64BIT_START.raw_value());
             let memory_node = fdt.begin_node(&memory_node_name)?;
@@ -305,7 +300,7 @@ fn create_chosen_node(
     fdt.property_string("bootargs", cmdline)?;
 
     if let Some(initrd_config) = initrd {
-        let initrd_start = initrd_config.address.raw_value() as u64;
+        let initrd_start = initrd_config.address.raw_value();
         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)?;
@@ -541,16 +536,9 @@ fn create_devices_node<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::Buil
     Ok(())
 }
 
-fn create_pmu_node(fdt: &mut FdtWriter, cpu_nums: usize) -> FdtWriterResult<()> {
-    let num_cpus = cpu_nums as u64 as u32;
+fn create_pmu_node(fdt: &mut FdtWriter) -> FdtWriterResult<()> {
     let compatible = "arm,armv8-pmuv3";
-    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 irq = [GIC_FDT_IRQ_TYPE_PPI, AARCH64_PMU_IRQ, IRQ_TYPE_LEVEL_HI];
 
     let pmu_node = fdt.begin_node("pmu")?;
     fdt.property_string("compatible", compatible)?;
@@ -679,7 +667,7 @@ 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@{:x}", virtio_iommu_bdf);
+                let virtio_iommu_node_name = format!("virtio_iommu@{virtio_iommu_bdf:x}");
                 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/layout.rs

@@ -59,7 +59,7 @@ pub const 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.
-pub const MAPPED_IO_START: GuestAddress = GuestAddress(0x0900_0000);
+const MAPPED_IO_START: GuestAddress = GuestAddress(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
@@ -73,7 +73,7 @@ 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 = GuestAddress(0x0900_0000);
+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);
 
@@ -98,6 +98,11 @@ pub const RAM_START: GuestAddress = GuestAddress(0x4000_0000);
 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);
 

arch/src/aarch64/mod.rs

@@ -19,7 +19,7 @@ use std::collections::HashMap;
 use std::convert::TryInto;
 use std::fmt::Debug;
 use std::sync::{Arc, Mutex};
-use vm_memory::{Address, GuestAddress, GuestMemory, GuestUsize};
+use vm_memory::{Address, GuestAddress, GuestMemory, GuestMemoryAtomic, GuestUsize};
 
 /// Errors thrown while configuring aarch64 system.
 #[derive(Debug)]
@@ -64,9 +64,9 @@ pub struct EntryPoint {
 pub fn configure_vcpu(
     vcpu: &Arc<dyn hypervisor::Vcpu>,
     id: u8,
-    kernel_entry_point: Option<EntryPoint>,
+    boot_setup: Option<(EntryPoint, &GuestMemoryAtomic<GuestMemoryMmap>)>,
 ) -> super::Result<u64> {
-    if let Some(kernel_entry_point) = kernel_entry_point {
+    if let Some((kernel_entry_point, _guest_memory)) = boot_setup {
         vcpu.setup_regs(
             id,
             kernel_entry_point.entry_addr.raw_value(),
@@ -108,7 +108,7 @@ pub fn arch_memory_regions(size: GuestUsize) -> Vec<(GuestAddress, usize, Region
 
     // RAM space
     // Case1: guest memory fits before the gap
-    if size as u64 <= ram_32bit_space_size {
+    if size <= ram_32bit_space_size {
         regions.push((layout::RAM_START, size as usize, RegionType::Ram));
     // Case2: guest memory extends beyond the gap
     } else {
@@ -231,7 +231,7 @@ mod tests {
             layout::MEM_32BIT_RESERVED_START.unchecked_offset_from(layout::RAM_START) as usize;
         assert_eq!(6, regions.len());
         assert_eq!(layout::RAM_START, regions[3].0);
-        assert_eq!(ram_32bit_space_size as usize, regions[3].1);
+        assert_eq!(ram_32bit_space_size, regions[3].1);
         assert_eq!(RegionType::Ram, regions[3].2);
         assert_eq!(RegionType::Reserved, regions[5].2);
         assert_eq!(RegionType::Ram, regions[4].2);

arch/src/aarch64/regs.rs

@@ -39,3 +39,6 @@ macro_rules! arm64_sys_reg {
 }
 
 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);

arch/src/aarch64/uefi.rs

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

arch/src/lib.rs

@@ -99,7 +99,7 @@ pub use x86_64::{
 #[cfg(target_arch = "x86_64")]
 #[inline(always)]
 fn pagesize() -> usize {
-    // Trivially safe
+    // SAFETY: Trivially safe
     unsafe { libc::sysconf(libc::_SC_PAGESIZE) as usize }
 }
 
@@ -145,7 +145,7 @@ 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:?}")
     }
 }
 

arch/src/x86_64/layout.rs

@@ -107,6 +107,11 @@ 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

@@ -125,9 +125,11 @@ struct MemmapTableEntryWrapper(hvm_memmap_table_entry);
 #[derive(Copy, Clone, Default)]
 struct ModlistEntryWrapper(hvm_modlist_entry);
 
-// SAFETY: These data structures only contain a series of integers
+// SAFETY: data structure 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
@@ -231,7 +233,7 @@ impl CpuidPatch {
     ) {
         let mut entry_found = false;
         for entry in cpuid.iter_mut() {
-            if entry.function == function && (index == None || index.unwrap() == entry.index) {
+            if entry.function == function && (index.is_none() || index.unwrap() == entry.index) {
                 entry_found = true;
                 match reg {
                     CpuidReg::EAX => {
@@ -545,13 +547,30 @@ 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}");
     let cpuid_patches = vec![
         // Patch tsc deadline timer bit
         CpuidPatch {
@@ -631,6 +650,20 @@ pub fn generate_common_cpuid(
                     }
                 }
             }
+            // 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);
@@ -661,6 +694,7 @@ 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,
@@ -725,8 +759,7 @@ pub fn generate_common_cpuid(
 pub fn configure_vcpu(
     vcpu: &Arc<dyn hypervisor::Vcpu>,
     id: u8,
-    kernel_entry_point: Option<EntryPoint>,
-    vm_memory: &GuestMemoryAtomic<GuestMemoryMmap>,
+    boot_setup: Option<(EntryPoint, &GuestMemoryAtomic<GuestMemoryMmap>)>,
     cpuid: Vec<CpuIdEntry>,
     kvm_hyperv: bool,
 ) -> super::Result<()> {
@@ -743,12 +776,12 @@ pub fn configure_vcpu(
     }
 
     regs::setup_msrs(vcpu).map_err(Error::MsrsConfiguration)?;
-    if let Some(kernel_entry_point) = kernel_entry_point {
+    if let Some((kernel_entry_point, guest_memory)) = boot_setup {
         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(&vm_memory.memory(), vcpu).map_err(Error::SregsConfiguration)?;
+            regs::setup_sregs(&guest_memory.memory(), vcpu).map_err(Error::SregsConfiguration)?;
         }
     }
     interrupts::set_lint(vcpu).map_err(|e| Error::LocalIntConfiguration(e.into()))?;
@@ -764,11 +797,11 @@ pub fn arch_memory_regions(size: GuestUsize) -> Vec<(GuestAddress, usize, Region
         .checked_add(layout::MEM_32BIT_DEVICES_SIZE)
         .expect("32-bit reserved region is too large");
 
-    let requested_memory_size = GuestAddress(size as u64);
+    let requested_memory_size = GuestAddress(size);
     let mut regions = Vec::new();
 
     // case1: guest memory fits before the gap
-    if size as u64 <= layout::MEM_32BIT_RESERVED_START.raw_value() {
+    if size <= layout::MEM_32BIT_RESERVED_START.raw_value() {
         regions.push((GuestAddress(0), size as usize, RegionType::Ram));
     // case2: guest memory extends beyond the gap
     } else {
@@ -865,7 +898,7 @@ 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() as u64;
+    start_info.0.cmdline_paddr = cmdline_addr.raw_value();
     start_info.0.memmap_paddr = layout::MEMMAP_START.raw_value();
 
     if let Some(rsdp_addr) = rsdp_addr {
@@ -934,7 +967,7 @@ fn configure_pvh(
         add_memmap_entry(
             &mut memmap,
             sgx_epc_region.start().raw_value(),
-            sgx_epc_region.size() as u64,
+            sgx_epc_region.size(),
             E820_RESERVED,
         );
     }
@@ -1011,6 +1044,7 @@ pub fn initramfs_load_addr(
 }
 
 pub fn get_host_cpu_phys_bits() -> u8 {
+    // SAFETY: call cpuid with valid leaves
     unsafe {
         let leaf = x86_64::__cpuid(0x8000_0000);
 
@@ -1121,12 +1155,13 @@ 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() as u64;
+        let size = epc_section.size();
         let eax = (start & 0xffff_f000) as u32 | 0x1;
         let ebx = (start >> 32) as u32;
         let ecx = (size & 0xffff_f000) as u32 | (leaf.ecx & 0xf);
@@ -1302,6 +1337,6 @@ 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

@@ -37,11 +37,17 @@ 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)]
@@ -95,7 +101,7 @@ const CPU_FEATURE_APIC: u32 = 0x200;
 const CPU_FEATURE_FPU: u32 = 0x001;
 
 fn compute_checksum<T: Copy>(v: &T) -> u8 {
-    // Safe because we are only reading the bytes within the size of the `T` reference `v`.
+    // SAFETY: 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() {
@@ -294,7 +300,7 @@ 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

@@ -124,7 +124,7 @@ pub fn configure_segments_and_sregs(
     mem: &GuestMemoryMmap,
     sregs: &mut SpecialRegisters,
 ) -> Result<()> {
-    let gdt_table: [u64; BOOT_GDT_MAX as usize] = {
+    let gdt_table: [u64; BOOT_GDT_MAX] = {
         // Configure GDT entries as specified by PVH boot protocol
         [
             gdt_entry(0, 0, 0),               // NULL

arch/src/x86_64/smbios.rs

@@ -48,10 +48,10 @@ impl Display for Error {
             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),
+            ParseUuid(e) => format!("Failure to parse uuid: {e}"),
         };
 
-        write!(f, "SMBIOS error: {}", description)
+        write!(f, "SMBIOS error: {description}")
     }
 }
 
@@ -67,7 +67,7 @@ const PCI_SUPPORTED: u64 = 1 << 7;
 const IS_VIRTUAL_MACHINE: u8 = 1 << 4;
 
 fn compute_checksum<T: Copy>(v: &T) -> u8 {
-    // Safe because we are only reading the bytes within the size of the `T` reference `v`.
+    // SAFETY: 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() {
@@ -145,11 +145,15 @@ struct SmbiosEndOfTable {
     handle: u16,
 }
 
-// SAFETY: These data structures only contain a series of integers
+// SAFETY: data structure 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>(

arch/src/x86_64/tdx/mod.rs

@@ -4,7 +4,9 @@
 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)]
@@ -13,6 +15,8 @@ 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")]
@@ -21,8 +25,13 @@ 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)]
@@ -64,7 +73,72 @@ impl Default for TdvfSectionType {
     }
 }
 
-pub fn parse_tdvf_sections(file: &mut File) -> Result<Vec<TdvfSection>, TdvfError> {
+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,
+                ));
+            }
+        }
+    }
+
+    // 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.
+
     // 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
@@ -74,13 +148,21 @@ pub fn parse_tdvf_sections(file: &mut File) -> Result<Vec<TdvfSection>, TdvfErro
     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;
 
-    file.seek(SeekFrom::Start(descriptor_offset))
+    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)
         .map_err(TdvfError::ReadDescriptor)?;
 
     let mut descriptor: TdvfDescriptor = Default::default();
-    // Safe as we read exactly the size of the descriptor header
+    // SAFETY: 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,
@@ -107,7 +189,7 @@ pub fn parse_tdvf_sections(file: &mut File) -> Result<Vec<TdvfSection>, TdvfErro
     let mut sections = Vec::new();
     sections.resize_with(descriptor.num_sections as usize, TdvfSection::default);
 
-    // Safe as we read exactly the advertised sections
+    // SAFETY: we read exactly the advertised sections
     file.read_exact(unsafe {
         std::slice::from_raw_parts_mut(
             sections.as_mut_ptr() as *mut u8,
@@ -116,7 +198,7 @@ pub fn parse_tdvf_sections(file: &mut File) -> Result<Vec<TdvfSection>, TdvfErro
     })
     .map_err(TdvfError::ReadDescriptor)?;
 
-    Ok(sections)
+    Ok((sections, guid_found))
 }
 
 #[repr(u16)]
@@ -211,12 +293,17 @@ struct TdPayload {
     payload_info: PayloadInfo,
 }
 
-// SAFETY: These data structures only contain a series of integers
+// SAFETY: data structure 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 {
@@ -312,12 +399,19 @@ 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 */
@@ -445,9 +539,9 @@ 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!("{:x?}", section)
+            eprintln!("{section:x?}")
         }
     }
 }

block_util/Cargo.toml

@@ -8,17 +8,18 @@ edition = "2021"
 default = []
 
 [dependencies]
-io-uring = "0.5.3"
-libc = "0.2.131"
+io-uring = "0.5.11"
+libc = "0.2.139"
 log = "0.4.17"
 qcow = { path = "../qcow" }
-thiserror = "1.0.32"
-versionize = "0.1.6"
+smallvec = "1.10.0"
+thiserror = "1.0.38"
+versionize = "0.1.9"
 versionize_derive = "0.1.4"
 vhdx = { path = "../vhdx" }
 virtio-bindings = { version = "0.1.0", features = ["virtio-v5_0_0"] }
-virtio-queue = "0.5.0"
-vm-memory = { version = "0.8.0", features = ["backend-mmap", "backend-atomic", "backend-bitmap"] }
+virtio-queue = "0.7.0"
+vm-memory = { version = "0.10.0", features = ["backend-mmap", "backend-atomic", "backend-bitmap"] }
 vm-virtio = { path = "../vm-virtio" }
-vmm-sys-util = "0.10.0"
+vmm-sys-util = "0.11.0"
 

block_util/src/async_io.rs

@@ -3,7 +3,6 @@
 // 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;
@@ -54,19 +53,21 @@ 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(),
@@ -75,9 +76,7 @@ impl DiskTopology {
                     BlockSize::PhysicalBlock => BLKPBSZGET(),
                     BlockSize::MinimumIo => BLKIOMIN(),
                     BlockSize::OptimalIo => BLKIOOPT(),
-                }
-                .try_into()
-                .unwrap(),
+                } as _,
                 &mut block_size,
             )
         };
@@ -132,15 +131,15 @@ pub trait AsyncIo: Send {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()>;
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()>;
     fn fsync(&mut self, user_data: Option<u64>) -> AsyncIoResult<()>;
-    fn complete(&mut self) -> Vec<(u64, i32)>;
+    fn next_completed_request(&mut self) -> Option<(u64, i32)>;
 }

block_util/src/fixed_vhd_async.rs

@@ -64,7 +64,7 @@ impl AsyncIo for FixedVhdAsync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         if offset as u64 >= self.size {
@@ -83,7 +83,7 @@ impl AsyncIo for FixedVhdAsync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         if offset as u64 >= self.size {
@@ -104,7 +104,7 @@ impl AsyncIo for FixedVhdAsync {
         self.raw_file_async.fsync(user_data)
     }
 
-    fn complete(&mut self) -> Vec<(u64, i32)> {
-        self.raw_file_async.complete()
+    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
+        self.raw_file_async.next_completed_request()
     }
 }

block_util/src/fixed_vhd_sync.rs

@@ -62,7 +62,7 @@ impl AsyncIo for FixedVhdSync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         if offset as u64 >= self.size {
@@ -81,7 +81,7 @@ impl AsyncIo for FixedVhdSync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         if offset as u64 >= self.size {
@@ -101,7 +101,7 @@ impl AsyncIo for FixedVhdSync {
         self.raw_file_sync.fsync(user_data)
     }
 
-    fn complete(&mut self) -> Vec<(u64, i32)> {
-        self.raw_file_sync.complete()
+    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
+        self.raw_file_sync.next_completed_request()
     }
 }

block_util/src/lib.rs

@@ -22,8 +22,10 @@ pub mod vhdx_sync;
 
 use crate::async_io::{AsyncIo, AsyncIoError, AsyncIoResult};
 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::fs::File;
 use std::io::{self, IoSlice, IoSliceMut, Read, Seek, SeekFrom, Write};
@@ -32,6 +34,7 @@ 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;
@@ -51,23 +54,23 @@ pub const SECTOR_SIZE: u64 = 0x01 << SECTOR_SHIFT;
 
 #[derive(Error, Debug)]
 pub enum Error {
-    #[error("Guest gave us bad memory addresses.")]
+    #[error("Guest gave us bad memory addresses")]
     GuestMemory(GuestMemoryError),
-    #[error("Guest gave us offsets that would have overflowed a usize.")]
+    #[error("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.")]
+    #[error("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.")]
+    #[error("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.")]
+    #[error("Guest gave us too few descriptors in a descriptor chain")]
     DescriptorChainTooShort,
-    #[error("Guest gave us a descriptor that was too short to use.")]
+    #[error("Guest gave us a descriptor that was too short to use")]
     DescriptorLengthTooSmall,
-    #[error("Getting a block's metadata fails for any reason.")]
+    #[error("Getting a block's metadata fails for any reason")]
     GetFileMetadata,
-    #[error("The requested operation would cause a seek beyond disk end.")]
+    #[error("The requested operation would cause a seek beyond disk end")]
     InvalidOffset,
-    #[error("The requested operation does not support multiple descriptors.")]
+    #[error("The requested operation does not support multiple descriptors")]
     TooManyDescriptors,
 }
 
@@ -195,10 +198,11 @@ pub struct AlignedOperation {
 pub struct Request {
     pub request_type: RequestType,
     pub sector: u64,
-    pub data_descriptors: Vec<(GuestAddress, u32)>,
+    pub data_descriptors: SmallVec<[(GuestAddress, u32); 1]>,
     pub status_addr: GuestAddress,
     pub writeback: bool,
     pub aligned_operations: Vec<AlignedOperation>,
+    pub start: Instant,
 }
 
 impl Request {
@@ -226,10 +230,11 @@ impl Request {
         let mut req = Request {
             request_type: request_type(desc_chain.memory(), hdr_desc_addr)?,
             sector: sector(desc_chain.memory(), hdr_desc_addr)?,
-            data_descriptors: Vec::new(),
+            data_descriptors: SmallVec::with_capacity(1),
             status_addr: GuestAddress(0),
             writeback: true,
             aligned_operations: Vec::new(),
+            start: Instant::now(),
         };
 
         let status_desc;
@@ -249,6 +254,7 @@ 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);
@@ -353,7 +359,8 @@ impl Request {
         let request_type = self.request_type;
         let offset = (sector << SECTOR_SHIFT) as libc::off_t;
 
-        let mut iovecs = Vec::new();
+        let mut iovecs: SmallVec<[libc::iovec; 1]> =
+            SmallVec::with_capacity(self.data_descriptors.len());
         for (data_addr, data_len) in &self.data_descriptors {
             if *data_len == 0 {
                 continue;
@@ -381,7 +388,7 @@ 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();
-                // Safe because layout has non-zero size
+                // SAFETY: layout has non-zero size
                 let aligned_ptr = unsafe { alloc_zeroed(layout) };
                 if aligned_ptr.is_null() {
                     return Err(ExecuteError::TemporaryBufferAllocation(
@@ -392,7 +399,7 @@ impl Request {
                 // We need to perform the copy beforehand in case we're writing
                 // data out.
                 if request_type == RequestType::Out {
-                    // Safe because destination buffer has been allocated with
+                    // SAFETY: destination buffer has been allocated with
                     // the proper size.
                     unsafe {
                         std::ptr::copy(origin_ptr as *const u8, aligned_ptr, *data_len as usize)
@@ -430,12 +437,12 @@ 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 => {
@@ -467,7 +474,7 @@ 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 {
-                // Safe because origin buffer has been allocated with the
+                // SAFETY: origin buffer has been allocated with the
                 // proper size.
                 unsafe {
                     std::ptr::copy(
@@ -479,7 +486,7 @@ impl Request {
             }
 
             // Free the temporary aligned buffer.
-            // Safe because aligned_ptr was allocated by alloc_zeroed with the same
+            // SAFETY: aligned_ptr was allocated by alloc_zeroed with the same
             // layout
             unsafe {
                 dealloc(
@@ -528,8 +535,9 @@ pub struct VirtioBlockGeometry {
     pub sectors: u8,
 }
 
-// SAFETY: these data structures only contain a series of integers
+// SAFETY: data structure 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
@@ -566,15 +574,15 @@ pub fn block_io_uring_is_supported() -> bool {
         return false;
     }
 
-    // Check IORING_OP_READ is supported
-    if !probe.is_supported(opcode::Read::CODE) {
-        info!("{} IORING_OP_READ operation not supported", error_msg);
+    // 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_WRITE is supported
-    if !probe.is_supported(opcode::Write::CODE) {
-        info!("{} IORING_OP_WRITE operation not supported", error_msg);
+    // Check IORING_OP_WRITEV is supported
+    if !probe.is_supported(opcode::Writev::CODE) {
+        info!("{} IORING_OP_WRITEV operation not supported", error_msg);
         return false;
     }
 
@@ -588,14 +596,15 @@ where
     fn read_vectored_sync(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
         eventfd: &EventFd,
-        completion_list: &mut Vec<(u64, i32)>,
+        completion_list: &mut VecDeque<(u64, i32)>,
     ) -> AsyncIoResult<()> {
         // Convert libc::iovec into IoSliceMut
         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) }));
         }
 
@@ -611,7 +620,7 @@ where
                 .map_err(AsyncIoError::ReadVectored)?
         };
 
-        completion_list.push((user_data, result as i32));
+        completion_list.push_back((user_data, result as i32));
         eventfd.write(1).unwrap();
 
         Ok(())
@@ -620,14 +629,15 @@ where
     fn write_vectored_sync(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
         eventfd: &EventFd,
-        completion_list: &mut Vec<(u64, i32)>,
+        completion_list: &mut VecDeque<(u64, i32)>,
     ) -> AsyncIoResult<()> {
         // Convert libc::iovec into IoSlice
         let mut slices = Vec::new();
         for iovec in iovecs.iter() {
+            // SAFETY: on Linux IoSliceMut wraps around libc::iovec
             slices.push(IoSlice::new(unsafe { std::mem::transmute(*iovec) }));
         }
 
@@ -643,7 +653,7 @@ where
                 .map_err(AsyncIoError::WriteVectored)?
         };
 
-        completion_list.push((user_data, result as i32));
+        completion_list.push_back((user_data, result as i32));
         eventfd.write(1).unwrap();
 
         Ok(())
@@ -653,7 +663,7 @@ where
         &mut self,
         user_data: Option<u64>,
         eventfd: &EventFd,
-        completion_list: &mut Vec<(u64, i32)>,
+        completion_list: &mut VecDeque<(u64, i32)>,
     ) -> AsyncIoResult<()> {
         let result: i32 = {
             let mut file = self.file();
@@ -665,7 +675,7 @@ where
         };
 
         if let Some(user_data) = user_data {
-            completion_list.push((user_data, result));
+            completion_list.push_back((user_data, result));
             eventfd.write(1).unwrap();
         }
 

block_util/src/qcow_sync.rs

@@ -5,6 +5,7 @@
 use crate::async_io::{AsyncIo, AsyncIoResult, DiskFile, DiskFileError, DiskFileResult};
 use crate::AsyncAdaptor;
 use qcow::{QcowFile, RawFile, Result as QcowResult};
+use std::collections::VecDeque;
 use std::fs::File;
 use std::io::{Seek, SeekFrom};
 use std::sync::{Arc, Mutex, MutexGuard};
@@ -26,7 +27,7 @@ impl DiskFile for QcowDiskSync {
     fn size(&mut self) -> DiskFileResult<u64> {
         let mut file = self.qcow_file.lock().unwrap();
 
-        Ok(file.seek(SeekFrom::End(0)).map_err(DiskFileError::Size)? as u64)
+        file.seek(SeekFrom::End(0)).map_err(DiskFileError::Size)
     }
 
     fn new_async_io(&self, _ring_depth: u32) -> DiskFileResult<Box<dyn AsyncIo>> {
@@ -37,7 +38,7 @@ impl DiskFile for QcowDiskSync {
 pub struct QcowSync {
     qcow_file: Arc<Mutex<QcowFile>>,
     eventfd: EventFd,
-    completion_list: Vec<(u64, i32)>,
+    completion_list: VecDeque<(u64, i32)>,
 }
 
 impl QcowSync {
@@ -46,7 +47,7 @@ impl QcowSync {
             qcow_file,
             eventfd: EventFd::new(libc::EFD_NONBLOCK)
                 .expect("Failed creating EventFd for QcowSync"),
-            completion_list: Vec::new(),
+            completion_list: VecDeque::new(),
         }
     }
 }
@@ -65,7 +66,7 @@ impl AsyncIo for QcowSync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         self.qcow_file.read_vectored_sync(
@@ -80,7 +81,7 @@ impl AsyncIo for QcowSync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         self.qcow_file.write_vectored_sync(
@@ -97,7 +98,7 @@ impl AsyncIo for QcowSync {
             .fsync_sync(user_data, &self.eventfd, &mut self.completion_list)
     }
 
-    fn complete(&mut self) -> Vec<(u64, i32)> {
-        self.completion_list.drain(..).collect()
+    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
+        self.completion_list.pop_front()
     }
 }

block_util/src/raw_async.rs

@@ -23,10 +23,9 @@ impl RawFileDisk {
 
 impl DiskFile for RawFileDisk {
     fn size(&mut self) -> DiskFileResult<u64> {
-        Ok(self
-            .file
+        self.file
             .seek(SeekFrom::End(0))
-            .map_err(DiskFileError::Size)? as u64)
+            .map_err(DiskFileError::Size)
     }
 
     fn new_async_io(&self, ring_depth: u32) -> DiskFileResult<Box<dyn AsyncIo>> {
@@ -77,12 +76,12 @@ impl AsyncIo for RawFileAsync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         let (submitter, mut sq, _) = self.io_uring.split();
 
-        // Safe because we know the file descriptor is valid and we
+        // SAFETY: we know the file descriptor is valid and we
         // relied on vm-memory to provide the buffer address.
         let _ = unsafe {
             sq.push(
@@ -105,12 +104,12 @@ impl AsyncIo for RawFileAsync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         let (submitter, mut sq, _) = self.io_uring.split();
 
-        // Safe because we know the file descriptor is valid and we
+        // SAFETY: we know the file descriptor is valid and we
         // relied on vm-memory to provide the buffer address.
         let _ = unsafe {
             sq.push(
@@ -134,7 +133,7 @@ impl AsyncIo for RawFileAsync {
         if let Some(user_data) = user_data {
             let (submitter, mut sq, _) = self.io_uring.split();
 
-            // Safe because we know the file descriptor is valid.
+            // SAFETY: we know the file descriptor is valid.
             let _ = unsafe {
                 sq.push(
                     &opcode::Fsync::new(types::Fd(self.fd))
@@ -149,20 +148,17 @@ 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 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
+    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
+        self.io_uring
+            .completion()
+            .next()
+            .map(|entry| (entry.user_data(), entry.result()))
     }
 }

block_util/src/raw_sync.rs

@@ -5,6 +5,7 @@
 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};
@@ -22,10 +23,9 @@ impl RawFileDiskSync {
 
 impl DiskFile for RawFileDiskSync {
     fn size(&mut self) -> DiskFileResult<u64> {
-        Ok(self
-            .file
+        self.file
             .seek(SeekFrom::End(0))
-            .map_err(DiskFileError::Size)? as u64)
+            .map_err(DiskFileError::Size)
     }
 
     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: Vec<(u64, i32)>,
+    completion_list: VecDeque<(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: Vec::new(),
+            completion_list: VecDeque::new(),
         }
     }
 }
@@ -66,9 +66,10 @@ impl AsyncIo for RawFileSync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
+        // SAFETY: FFI call with valid arguments
         let result = unsafe {
             libc::preadv(
                 self.fd as libc::c_int,
@@ -81,7 +82,7 @@ impl AsyncIo for RawFileSync {
             return Err(AsyncIoError::ReadVectored(std::io::Error::last_os_error()));
         }
 
-        self.completion_list.push((user_data, result as i32));
+        self.completion_list.push_back((user_data, result as i32));
         self.eventfd.write(1).unwrap();
 
         Ok(())
@@ -90,9 +91,10 @@ impl AsyncIo for RawFileSync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
+        // SAFETY: FFI call with valid arguments
         let result = unsafe {
             libc::pwritev(
                 self.fd as libc::c_int,
@@ -105,27 +107,28 @@ impl AsyncIo for RawFileSync {
             return Err(AsyncIoError::WriteVectored(std::io::Error::last_os_error()));
         }
 
-        self.completion_list.push((user_data, result as i32));
+        self.completion_list.push_back((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((user_data, result as i32));
+            self.completion_list.push_back((user_data, result));
             self.eventfd.write(1).unwrap();
         }
 
         Ok(())
     }
 
-    fn complete(&mut self) -> Vec<(u64, i32)> {
-        self.completion_list.drain(..).collect()
+    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
+        self.completion_list.pop_front()
     }
 }

block_util/src/vhdx_sync.rs

@@ -4,6 +4,7 @@
 
 use crate::async_io::{AsyncIo, AsyncIoResult, DiskFile, DiskFileError, DiskFileResult};
 use crate::AsyncAdaptor;
+use std::collections::VecDeque;
 use std::fs::File;
 use std::sync::{Arc, Mutex, MutexGuard};
 use vhdx::vhdx::{Result as VhdxResult, Vhdx};
@@ -37,7 +38,7 @@ impl DiskFile for VhdxDiskSync {
 pub struct VhdxSync {
     vhdx_file: Arc<Mutex<Vhdx>>,
     eventfd: EventFd,
-    completion_list: Vec<(u64, i32)>,
+    completion_list: VecDeque<(u64, i32)>,
 }
 
 impl VhdxSync {
@@ -45,7 +46,7 @@ impl VhdxSync {
         Ok(VhdxSync {
             vhdx_file,
             eventfd: EventFd::new(libc::EFD_NONBLOCK)?,
-            completion_list: Vec::new(),
+            completion_list: VecDeque::new(),
         })
     }
 }
@@ -64,7 +65,7 @@ impl AsyncIo for VhdxSync {
     fn read_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         self.vhdx_file.read_vectored_sync(
@@ -79,7 +80,7 @@ impl AsyncIo for VhdxSync {
     fn write_vectored(
         &mut self,
         offset: libc::off_t,
-        iovecs: Vec<libc::iovec>,
+        iovecs: &[libc::iovec],
         user_data: u64,
     ) -> AsyncIoResult<()> {
         self.vhdx_file.write_vectored_sync(
@@ -96,7 +97,7 @@ impl AsyncIo for VhdxSync {
             .fsync_sync(user_data, &self.eventfd, &mut self.completion_list)
     }
 
-    fn complete(&mut self) -> Vec<(u64, i32)> {
-        self.completion_list.drain(..).collect()
+    fn next_completed_request(&mut self) -> Option<(u64, i32)> {
+        self.completion_list.pop_front()
     }
 }

build.rs

@@ -3,15 +3,12 @@
 // 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() + crate_version!();
+    let mut version = "v".to_owned() + env!("CARGO_PKG_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;
@@ -23,5 +20,5 @@ fn main() {
     // 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=BUILT_VERSION={}", version);
+    println!("cargo:rustc-env=BUILT_VERSION={version}");
 }

devices/Cargo.toml

@@ -6,24 +6,25 @@ edition = "2021"
 
 [dependencies]
 acpi_tables = { path = "../acpi_tables" }
-anyhow = "1.0.62"
+anyhow = "1.0.68"
 arch = { path = "../arch" }
 bitflags = "1.3.2"
 byteorder = "1.4.3"
 hypervisor = { path = "../hypervisor" }
-libc = "0.2.131"
+libc = "0.2.139"
 log = "0.4.17"
-versionize = "0.1.6"
+phf = { version = "0.11.1", features = ["macros"] }
+thiserror = "1.0.38"
+tpm = { path = "../tpm" }
+versionize = "0.1.9"
 versionize_derive = "0.1.4"
 vm-device = { path = "../vm-device" }
-vm-memory = "0.8.0"
+vm-memory = "0.10.0"
 vm-migration = { path = "../vm-migration" }
-vmm-sys-util = "0.10.0"
+vmm-sys-util = "0.11.0"
 
 [target.'cfg(target_arch = "aarch64")'.dependencies]
 arch = { path = "../arch" }
 
 [features]
 default = []
-cmos = []
-fwdebug = []

devices/src/acpi.rs

@@ -105,18 +105,17 @@ impl BusDevice for AcpiGedDevice {
 impl Aml for AcpiGedDevice {
     fn append_aml_bytes(&self, bytes: &mut Vec<u8>) {
         aml::Device::new(
-            "_SB_.GED_".into(),
+            "_SB_.GEC_".into(),
             vec![
-                &aml::Name::new("_HID".into(), &"ACPI0013"),
-                &aml::Name::new("_UID".into(), &aml::ZERO),
+                &aml::Name::new("_HID".into(), &aml::EisaName::new("PNP0A06")),
+                &aml::Name::new("_UID".into(), &"Generic Event Controller"),
                 &aml::Name::new(
                     "_CRS".into(),
-                    &aml::ResourceTemplate::new(vec![&aml::Interrupt::new(
+                    &aml::ResourceTemplate::new(vec![&aml::AddressSpace::new_memory(
+                        aml::AddressSpaceCachable::NotCacheable,
                         true,
-                        true,
-                        false,
-                        false,
-                        self.ged_irq,
+                        self.address.0,
+                        self.address.0 + GED_DEVICE_ACPI_SIZE as u64 - 1,
                     )]),
                 ),
                 &aml::OpRegion::new(
@@ -132,8 +131,8 @@ impl Aml for AcpiGedDevice {
                     vec![aml::FieldEntry::Named(*b"GDAT", 8)],
                 ),
                 &aml::Method::new(
-                    "_EVT".into(),
-                    1,
+                    "ESCN".into(),
+                    0,
                     true,
                     vec![
                         &aml::Store::new(&aml::Local(0), &aml::Path::new("GDAT")),
@@ -164,6 +163,30 @@ impl Aml for AcpiGedDevice {
                 ),
             ],
         )
+        .append_aml_bytes(bytes);
+        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![])],
+                ),
+            ],
+        )
         .append_aml_bytes(bytes)
     }
 }

devices/src/gic.rs

@@ -6,14 +6,17 @@ use super::interrupt_controller::{Error, InterruptController};
 extern crate arch;
 use anyhow::anyhow;
 use arch::layout;
-use hypervisor::{arch::aarch64::gic::Vgic, CpuState};
+use hypervisor::{
+    arch::aarch64::gic::{Vgic, VgicConfig},
+    CpuState, GicState,
+};
 use std::result;
 use std::sync::{Arc, Mutex};
 use vm_device::interrupt::{
     InterruptIndex, InterruptManager, InterruptSourceConfig, InterruptSourceGroup,
     LegacyIrqSourceConfig, MsiIrqGroupConfig,
 };
-use vm_memory::Address;
+use vm_memory::address::Address;
 use vm_migration::{Migratable, MigratableError, Pausable, Snapshot, Snapshottable, Transportable};
 use vmm_sys_util::eventfd::EventFd;
 
@@ -22,6 +25,7 @@ type Result<T> = result::Result<T, Error>;
 // Reserve 32 IRQs for legacy devices.
 pub const IRQ_LEGACY_BASE: usize = 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
@@ -36,8 +40,9 @@ pub struct Gic {
 
 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 {
@@ -46,38 +51,34 @@ impl Gic {
             })
             .map_err(Error::CreateInterruptSourceGroup)?;
 
-        Ok(Gic {
-            interrupt_source_group,
-            vgic: None,
-        })
-    }
-
-    pub fn create_vgic(
-        &mut self,
-        vm: &Arc<dyn hypervisor::Vm>,
-        vcpu_count: u64,
-    ) -> Result<Arc<Mutex<dyn Vgic>>> {
         let vgic = vm
-            .create_vgic(
-                vcpu_count,
-                layout::GIC_V3_DIST_START.raw_value(),
-                layout::GIC_V3_DIST_SIZE,
-                layout::GIC_V3_REDIST_SIZE,
-                layout::GIC_V3_ITS_SIZE,
-                layout::IRQ_NUM,
-            )
+            .create_vgic(Gic::create_default_config(vcpu_count as u64))
             .map_err(Error::CreateGic)?;
-        self.vgic = Some(vgic.clone());
-        Ok(vgic.clone())
+
+        let gic = Gic {
+            interrupt_source_group,
+            vgic: Some(vgic),
+        };
+        gic.enable()?;
+
+        Ok(gic)
     }
 
-    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);
+    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)
     }
-}
 
-impl InterruptController for Gic {
     fn enable(&self) -> Result<()> {
         // Set irqfd for legacy interrupts
         self.interrupt_source_group
@@ -103,6 +104,33 @@ impl InterruptController for Gic {
         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<()> {
@@ -118,27 +146,15 @@ impl InterruptController for Gic {
     }
 }
 
-pub const GIC_V3_ITS_SNAPSHOT_ID: &str = "gic-v3-its";
 impl Snapshottable for Gic {
     fn id(&self) -> String {
-        GIC_V3_ITS_SNAPSHOT_ID.to_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(&self.id(), &state)
-    }
-
-    fn restore(&mut self, snapshot: Snapshot) -> std::result::Result<(), MigratableError> {
-        let vgic = self.vgic.as_ref().unwrap().clone();
-        vgic.lock()
-            .unwrap()
-            .set_state(&snapshot.to_state(&self.id())?)
-            .map_err(|e| {
-                MigratableError::Restore(anyhow!("Could not restore GICv3ITS state {:?}", e))
-            })?;
-        Ok(())
+        Snapshot::new_from_state(&state)
     }
 }
 

devices/src/interrupt_controller.rs

@@ -24,8 +24,12 @@ 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>;
@@ -55,8 +59,6 @@ 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

@@ -10,7 +10,6 @@
 // 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};
@@ -194,6 +193,7 @@ 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 {
@@ -202,17 +202,47 @@ 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.
-        Ok(Ioapic {
+        let ioapic = Ioapic {
             id,
-            id_reg: 0,
-            reg_sel: 0,
-            reg_entries: [0x10000; NUM_IOAPIC_PINS],
-            used_entries: [false; NUM_IOAPIC_PINS],
+            id_reg,
+            reg_sel,
+            reg_entries,
+            used_entries,
             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)?;
+                }
+            }
+        }
+
+        Ok(ioapic)
     }
 
     fn ioapic_write(&mut self, val: u32) {
@@ -299,21 +329,6 @@ impl Ioapic {
         }
     }
 
-    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];
 
@@ -423,18 +438,7 @@ impl Snapshottable for Ioapic {
     }
 
     fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        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
-                ))
-            })
+        Snapshot::new_from_versioned_state(&self.state())
     }
 }
 

devices/src/legacy/cmos.rs

@@ -97,7 +97,7 @@ impl BusDevice for Cmos {
                 let day;
                 let month;
                 let year;
-                // The clock_gettime and gmtime_r calls are safe as long as the structs they are
+                // SAFETY: 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

@@ -40,11 +40,11 @@ impl DebugIoPortRange {
 impl fmt::Display for DebugIoPortRange {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         match self {
-            DebugIoPortRange::Firmware => write!(f, "{}: Firmware", DEBUG_IOPORT_PREFIX),
-            DebugIoPortRange::Bootloader => write!(f, "{}: Bootloader", DEBUG_IOPORT_PREFIX),
-            DebugIoPortRange::Kernel => write!(f, "{}: Kernel", DEBUG_IOPORT_PREFIX),
-            DebugIoPortRange::Userspace => write!(f, "{}: Userspace", DEBUG_IOPORT_PREFIX),
-            DebugIoPortRange::Custom => write!(f, "{}: Custom", DEBUG_IOPORT_PREFIX),
+            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"),
         }
     }
 }

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}.")
             }
         }
     }
@@ -106,18 +106,39 @@ impl VersionMapped for GpioState {}
 
 impl Gpio {
     /// Constructs an PL061 GPIO device.
-    pub fn new(id: String, interrupt: Arc<dyn InterruptSourceGroup>) -> Self {
+    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)
+            };
+
         Self {
             id,
-            data: 0,
-            old_in_data: 0,
-            dir: 0,
-            isense: 0,
-            ibe: 0,
-            iev: 0,
-            im: 0,
-            istate: 0,
-            afsel: 0,
+            data,
+            old_in_data,
+            dir,
+            isense,
+            ibe,
+            iev,
+            im,
+            istate,
+            afsel,
             interrupt,
         }
     }
@@ -136,24 +157,12 @@ 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) as u32;
+        let changed = (self.old_in_data ^ self.data) & !self.dir;
         if changed > 0 {
             self.old_in_data = self.data;
             for i in 0..N_GPIOS {
@@ -319,12 +328,7 @@ impl Snapshottable for Gpio {
     }
 
     fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        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(())
+        Snapshot::new_from_versioned_state(&self.state())
     }
 }
 
@@ -378,6 +382,7 @@ 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/mod.rs

@@ -8,7 +8,7 @@
 mod cmos;
 #[cfg(target_arch = "x86_64")]
 mod debug_port;
-#[cfg(feature = "fwdebug")]
+#[cfg(target_arch = "x86_64")]
 mod fwdebug;
 #[cfg(target_arch = "aarch64")]
 mod gpio_pl061;
@@ -22,7 +22,7 @@ mod uart_pl011;
 pub use self::cmos::Cmos;
 #[cfg(target_arch = "x86_64")]
 pub use self::debug_port::DebugPort;
-#[cfg(feature = "fwdebug")]
+#[cfg(target_arch = "x86_64")]
 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}"),
         }
     }
 }
@@ -61,12 +61,10 @@ 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,
 }
 
@@ -101,7 +99,7 @@ pub struct LocalTime {
 
 impl LocalTime {
     /// Returns the [LocalTime](struct.LocalTime.html) structure for the calling moment.
-    #[allow(dead_code)]
+    #[cfg(test)]
     pub fn now() -> LocalTime {
         let mut timespec = libc::timespec {
             tv_sec: 0,
@@ -121,7 +119,7 @@ impl LocalTime {
             tm_zone: std::ptr::null(),
         };
 
-        // Safe because the parameters are valid.
+        // SAFETY: the parameters are valid.
         unsafe {
             libc::clock_gettime(libc::CLOCK_REALTIME, &mut timespec);
             libc::localtime_r(&timespec.tv_sec, &mut tm);
@@ -173,22 +171,6 @@ 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
@@ -199,7 +181,7 @@ pub fn get_time(clock_type: ClockType) -> u64 {
         tv_sec: 0,
         tv_nsec: 0,
     };
-    // Safe because the parameters are valid.
+    // SAFETY: 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)
 }
@@ -525,7 +507,6 @@ 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

@@ -63,7 +63,6 @@ pub struct Serial {
     id: String,
     interrupt_enable: u8,
     interrupt_identification: u8,
-    interrupt: Arc<dyn InterruptSourceGroup>,
     line_control: u8,
     line_status: u8,
     modem_control: u8,
@@ -71,6 +70,7 @@ pub struct Serial {
     scratch: u8,
     baud_divisor: u16,
     in_buffer: VecDeque<u8>,
+    interrupt: Arc<dyn InterruptSourceGroup>,
     out: Option<Box<dyn io::Write + Send>>,
 }
 
@@ -93,19 +93,56 @@ 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: 0,
-            interrupt_identification: DEFAULT_INTERRUPT_IDENTIFICATION,
+            interrupt_enable,
+            interrupt_identification,
+            line_control,
+            line_status,
+            modem_control,
+            modem_status,
+            scratch,
+            baud_divisor,
+            in_buffer,
             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,
         }
     }
@@ -115,13 +152,18 @@ impl Serial {
         id: String,
         interrupt: Arc<dyn InterruptSourceGroup>,
         out: Box<dyn io::Write + Send>,
+        state: Option<SerialState>,
     ) -> Serial {
-        Self::new(id, interrupt, Some(out))
+        Self::new(id, interrupt, Some(out), state)
     }
 
     /// Constructs a Serial port with no connected output.
-    pub fn new_sink(id: String, interrupt: Arc<dyn InterruptSourceGroup>) -> Serial {
-        Self::new(id, interrupt, None)
+    pub fn new_sink(
+        id: String,
+        interrupt: Arc<dyn InterruptSourceGroup>,
+        state: Option<SerialState>,
+    ) -> Serial {
+        Self::new(id, interrupt, None, state)
     }
 
     pub fn set_out(&mut self, out: Box<dyn io::Write + Send>) {
@@ -199,7 +241,7 @@ impl Serial {
     }
 
     fn handle_write(&mut self, offset: u8, v: u8) -> Result<()> {
-        match offset as u8 {
+        match offset {
             DLAB_LOW if self.is_dlab_set() => {
                 self.baud_divisor = (self.baud_divisor & 0xff00) | u16::from(v)
             }
@@ -242,18 +284,6 @@ 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 {
@@ -304,12 +334,7 @@ impl Snapshottable for Serial {
     }
 
     fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        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(())
+        Snapshot::new_from_versioned_state(&self.state())
     }
 }
 
@@ -384,6 +409,7 @@ 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']);
@@ -404,6 +430,7 @@ 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
@@ -440,6 +467,7 @@ 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
@@ -462,6 +490,7 @@ 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]);
@@ -483,6 +512,7 @@ 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]);
@@ -509,6 +539,7 @@ 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

@@ -60,11 +60,11 @@ 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}"),
         }
     }
 }
@@ -122,24 +122,79 @@ impl Pl011 {
         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: 0x90u32,
-            lcr: 0u32,
-            rsr: 0u32,
-            cr: 0x300u32,
-            dmacr: 0u32,
-            debug: 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,
+            flags,
+            lcr,
+            rsr,
+            cr,
+            dmacr,
+            debug,
+            int_enabled,
+            int_level,
+            read_fifo,
+            ilpr,
+            ibrd,
+            fbrd,
+            ifl,
+            read_count,
+            read_trigger,
             irq,
             out,
             timestamp,
@@ -170,24 +225,6 @@ 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.debug = state.debug;
-        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);
@@ -306,25 +343,25 @@ impl Pl011 {
         let elapsed = self.timestamp.elapsed();
 
         match self.debug {
-            0x00..=0x1f => info!(
+            0x00..=0x1f => warn!(
                 "[Debug I/O port: Firmware code: 0x{:x}] {}.{:>06} seconds",
                 self.debug,
                 elapsed.as_secs(),
                 elapsed.as_micros()
             ),
-            0x20..=0x3f => info!(
+            0x20..=0x3f => warn!(
                 "[Debug I/O port: Bootloader code: 0x{:x}] {}.{:>06} seconds",
                 self.debug,
                 elapsed.as_secs(),
                 elapsed.as_micros()
             ),
-            0x40..=0x5f => info!(
+            0x40..=0x5f => warn!(
                 "[Debug I/O port: Kernel code: 0x{:x}] {}.{:>06} seconds",
                 self.debug,
                 elapsed.as_secs(),
                 elapsed.as_micros()
             ),
-            0x60..=0x7f => info!(
+            0x60..=0x7f => warn!(
                 "[Debug I/O port: Userspace code: 0x{:x}] {}.{:>06} seconds",
                 self.debug,
                 elapsed.as_secs(),
@@ -417,12 +454,7 @@ impl Snapshottable for Pl011 {
     }
 
     fn snapshot(&mut self) -> std::result::Result<Snapshot, MigratableError> {
-        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(())
+        Snapshot::new_from_versioned_state(&self.state())
     }
 }
 
@@ -498,12 +530,13 @@ mod tests {
             Arc::new(TestInterrupt::new(intr_evt.try_clone().unwrap())),
             Some(Box::new(pl011_out.clone())),
             Instant::now(),
+            None,
         );
 
-        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']);
+        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']);
         assert_eq!(
             pl011_out.buf.lock().unwrap().as_slice(),
             &[b'x', b'a', b'b', b'c']
@@ -519,6 +552,7 @@ mod tests {
             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
@@ -529,11 +563,11 @@ mod tests {
         assert_eq!(intr_evt.read().unwrap(), 2);
 
         let mut data = [0u8];
-        pl011.read(0, UARTDR as u64, &mut data);
+        pl011.read(0, UARTDR, &mut data);
         assert_eq!(data[0], b'a');
-        pl011.read(0, UARTDR as u64, &mut data);
+        pl011.read(0, UARTDR, &mut data);
         assert_eq!(data[0], b'b');
-        pl011.read(0, UARTDR as u64, &mut data);
+        pl011.read(0, UARTDR, &mut data);
         assert_eq!(data[0], b'c');
     }
 }

devices/src/lib.rs

@@ -19,6 +19,7 @@ pub mod interrupt_controller;
 #[cfg(target_arch = "x86_64")]
 pub mod ioapic;
 pub mod legacy;
+pub mod tpm;
 
 pub use self::acpi::{AcpiGedDevice, AcpiPmTimerDevice, AcpiShutdownDevice};
 
@@ -32,11 +33,9 @@ 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];
@@ -48,11 +47,9 @@ 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/tpm.rs

@@ -0,0 +1,467 @@
+// 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 phf::phf_map;
+use std::cmp;
+use std::sync::{Arc, Barrier};
+use thiserror::Error;
+use tpm::emulator::{BackendCmd, Emulator};
+use tpm::TPM_CRB_BUFFER_MAX;
+use tpm::TPM_SUCCESS;
+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),
+    #[error("Failed to deliver tpm Command: {0}")]
+    DeliverRequest(#[source] anyhow::Error),
+}
+type Result<T> = anyhow::Result<T, Error>;
+
+/* crb 32-bit registers */
+const CRB_LOC_STATE: u32 = 0x0;
+//Register Fields
+// Field => (start, length)
+// start: lowest bit in the bit field numbered from 0
+// length: length of the bit field
+const CRB_LOC_STATE_FIELDS: phf::Map<&str, [u32; 2]> = phf_map! {
+    "tpmEstablished" => [0, 1],
+    "locAssigned" => [1,1],
+    "activeLocality"=> [2, 3],
+    "reserved" => [5, 2],
+    "tpmRegValidSts" => [7, 1]
+};
+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 CRB_LOC_STS_FIELDS: phf::Map<&str, [u32; 2]> = phf_map! {
+    "Granted" => [0, 1],
+    "beenSeized" => [1,1]
+};
+const CRB_INTF_ID: u32 = 0x30;
+const CRB_INTF_ID_FIELDS: phf::Map<&str, [u32; 2]> = phf_map! {
+    "InterfaceType" => [0, 4],
+    "InterfaceVersion" => [4, 4],
+    "CapLocality" =>  [8, 1],
+    "CapCRBIdleBypass" => [9, 1],
+    "Reserved1" => [10, 1],
+    "CapDataXferSizeSupport" => [11, 2],
+    "CapFIFO" =>  [13, 1],
+    "CapCRB" => [14, 1],
+    "CapIFRes" => [15, 2],
+    "InterfaceSelector" => [17, 2],
+    "IntfSelLock" =>  [19, 1],
+    "Reserved2" => [20, 4],
+    "RID" => [24, 8]
+};
+const CRB_INTF_ID2: u32 = 0x34;
+const CRB_INTF_ID2_FIELDS: phf::Map<&str, [u32; 2]> = phf_map! {
+    "VID" => [0, 16],
+    "DID" => [16, 16]
+};
+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 CRB_CTRL_STS_FIELDS: phf::Map<&str, [u32; 2]> = phf_map! {
+    "tpmSts" => [0, 1],
+    "tpmIdle" => [1, 1]
+};
+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: u32 = CRB_DATA_BUFFER;
+
+// 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;
+
+fn get_fields_map(reg: u32) -> phf::Map<&'static str, [u32; 2]> {
+    match reg {
+        CRB_LOC_STATE => CRB_LOC_STATE_FIELDS,
+        CRB_LOC_STS => CRB_LOC_STS_FIELDS,
+        CRB_INTF_ID => CRB_INTF_ID_FIELDS,
+        CRB_INTF_ID2 => CRB_INTF_ID2_FIELDS,
+        CRB_CTRL_STS => CRB_CTRL_STS_FIELDS,
+        _ => {
+            panic!("Fields in '{reg:?}' register were accessed which are Invalid");
+        }
+    }
+}
+
+/// Set a particular field in a Register
+fn set_reg_field(regs: &mut [u32; TPM_CRB_R_MAX as usize], reg: u32, field: &str, value: u32) {
+    let reg_fields = get_fields_map(reg);
+    if reg_fields.contains_key(field) {
+        let start = reg_fields.get(field).unwrap()[0];
+        let len = reg_fields.get(field).unwrap()[1];
+        let mask = (!(0_u32) >> (32 - len)) << start;
+        regs[reg as usize] = (regs[reg as usize] & !mask) | ((value << start) & mask);
+    } else {
+        error!(
+            "Failed to tpm Register. {:?} is not a valid field in Reg {:#X}",
+            field, reg
+        )
+    }
+}
+
+/// Get the value of a particular field in a Register
+fn get_reg_field(regs: &[u32; TPM_CRB_R_MAX as usize], reg: u32, field: &str) -> u32 {
+    let reg_fields = get_fields_map(reg);
+    if reg_fields.contains_key(field) {
+        let start = reg_fields.get(field).unwrap()[0];
+        let len = reg_fields.get(field).unwrap()[1];
+        let mask = (!(0_u32) >> (32 - len)) << start;
+        (regs[reg as usize] & mask) >> start
+    } else {
+        // TODO: Sensible return value if fields do not exist
+        0x0
+    }
+}
+
+fn locality_from_addr(addr: u32) -> u8 {
+    (addr >> 12) as u8
+}
+
+pub struct Tpm {
+    emulator: Emulator,
+    cmd: Option<BackendCmd>,
+    regs: [u32; TPM_CRB_R_MAX as usize],
+    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,
+            cmd: None,
+            regs: [0; TPM_CRB_R_MAX as usize],
+            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, CRB_LOC_STATE, "locAssigned") == 0 {
+            return TPM_CRB_NO_LOCALITY;
+        }
+        get_reg_field(&self.regs, CRB_LOC_STATE, "activeLocality")
+    }
+
+    fn request_completed(&mut self, result: isize) {
+        self.regs[CRB_CTRL_START as usize] = !CRB_START_INVOKE;
+        if result != 0 {
+            set_reg_field(&mut self.regs, CRB_CTRL_STS, "tpmSts", 1);
+        }
+    }
+
+    fn reset(&mut self) -> Result<()> {
+        let cur_buff_size = self.emulator.get_buffer_size().unwrap();
+        self.regs = [0; TPM_CRB_R_MAX as usize];
+        set_reg_field(&mut self.regs, CRB_LOC_STATE, "tpmRegValidSts", 1);
+        set_reg_field(&mut self.regs, CRB_CTRL_STS, "tpmIdle", 1);
+        set_reg_field(
+            &mut self.regs,
+            CRB_INTF_ID,
+            "InterfaceType",
+            CRB_INTF_TYPE_CRB_ACTIVE,
+        );
+        set_reg_field(
+            &mut self.regs,
+            CRB_INTF_ID,
+            "InterfaceVersion",
+            CRB_INTF_VERSION_CRB,
+        );
+        set_reg_field(
+            &mut self.regs,
+            CRB_INTF_ID,
+            "CapLocality",
+            CRB_INTF_CAP_LOCALITY_0_ONLY,
+        );
+        set_reg_field(
+            &mut self.regs,
+            CRB_INTF_ID,
+            "CapCRBIdleBypass",
+            CRB_INTF_CAP_IDLE_FAST,
+        );
+        set_reg_field(
+            &mut self.regs,
+            CRB_INTF_ID,
+            "CapDataXferSizeSupport",
+            CRB_INTF_CAP_XFER_SIZE_64,
+        );
+        set_reg_field(
+            &mut self.regs,
+            CRB_INTF_ID,
+            "CapFIFO",
+            CRB_INTF_CAP_FIFO_NOT_SUPPORTED,
+        );
+        set_reg_field(
+            &mut self.regs,
+            CRB_INTF_ID,
+            "CapCRB",
+            CRB_INTF_CAP_CRB_SUPPORTED,
+        );
+        set_reg_field(
+            &mut self.regs,
+            CRB_INTF_ID,
+            "InterfaceSelector",
+            CRB_INTF_IF_SELECTOR_CRB,
+        );
+        set_reg_field(&mut self.regs, CRB_INTF_ID, "RID", 0b0000);
+        set_reg_field(&mut self.regs, CRB_INTF_ID2, "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
+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, CRB_CTRL_STS, "tpmIdle", 0);
+                    }
+                    CRB_CTRL_REQ_GO_IDLE => {
+                        set_reg_field(&mut self.regs, CRB_CTRL_STS, "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;
+
+                        self.cmd = Some(BackendCmd {
+                            locality: locality as u8,
+                            input: self.data_buff[0..self.data_buff_len].to_vec(),
+                            input_len: cmp::min(self.data_buff_len, TPM_CRB_BUFFER_MAX),
+                            output: self.data_buff.to_vec(),
+                            output_len: TPM_CRB_BUFFER_MAX,
+                            selftest_done: false,
+                        });
+
+                        let mut cmd = self.cmd.as_ref().unwrap().clone();
+                        let output = self.emulator.deliver_request(&mut cmd).map_err(|e| {
+                            Error::DeliverRequest(anyhow!(
+                                "Failed to deliver tpm request. Error :{:?}",
+                                e
+                            ))
+                        });
+                        //TODO: drop the copy here
+                        self.data_buff.fill(0);
+                        self.data_buff.clone_from_slice(output.unwrap().as_slice());
+
+                        self.request_completed(TPM_SUCCESS as isize);
+                    }
+                }
+                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, CRB_LOC_STATE, "locAssigned", 0);
+                            set_reg_field(&mut self.regs, CRB_LOC_STS, "Granted", 0);
+                        }
+                        CRB_LOC_CTRL_REQUEST_ACCESS => {
+                            set_reg_field(&mut self.regs, CRB_LOC_STS, "Granted", 1);
+                            set_reg_field(&mut self.regs, CRB_LOC_STS, "beenSeized", 0);
+                            set_reg_field(&mut self.regs, CRB_LOC_STATE, "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 as usize] = [0; TPM_CRB_R_MAX as usize];
+        set_reg_field(&mut regs, CRB_INTF_ID, "RID", 0xAC);
+        assert_eq!(
+            get_reg_field(&regs, CRB_INTF_ID, "RID"),
+            0xAC,
+            concat!("Test: ", stringify!(set_get_reg_field))
+        );
+    }
+}

docs/arm64.md

@@ -1,152 +0,0 @@
-# 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/building.md

@@ -0,0 +1,86 @@
+- [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
+# 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/custom-image.md

@@ -88,7 +88,7 @@ Ubuntu distributions.
 
 ```bash
 apt update
-apt install fio iperf iperf3 socat stress
+apt install fio iperf iperf3 socat stress cpuid
 ```
 
 ### Remove counterproductive packages
@@ -158,3 +158,172 @@ 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-20221118-1.raw
+mv jammy-server-cloudimg-amd64-custom-20221118-1.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](docs/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/gdb.md

@@ -2,10 +2,10 @@
 
 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 `gdb` feature enabled:
+To enable debugging with GDB, build with the `guest_debug` feature enabled:
 
 ```bash
-cargo build --features gdb
+cargo build --features guest_debug
 ```
 
 To use the `--gdb` option, specify the Unix Domain Socket with `--path` that Cloud Hypervisor will use to communicate with the host's GDB:
@@ -44,4 +44,4 @@ Continuing.
 
 Breakpoint 1, 0x00000000001121b7 in ?? ()
 (gdb)
-```
+```

docs/intel_tdx.md

@@ -61,7 +61,8 @@ meaning it will be printing guest kernel logs to the `virtio-console` device.
 
 ```bash
 ./cloud-hypervisor \
-    --tdx firmware=edk2-staging/Build/OvmfCh/RELEASE_GCC5/FV/OVMF.fd \
+    --platform tdx=on
+    --firmware edk2-staging/Build/OvmfCh/RELEASE_GCC5/FV/OVMF.fd \
     --cpus boot=1 \
     --memory size=1G \
     --disk path=tdx_guest_img
@@ -72,7 +73,8 @@ firmware:
 
 ```bash
 ./cloud-hypervisor \
-    --tdx firmware=edk2-staging/Build/OvmfCh/DEBUG_GCC5/FV/OVMF.fd \
+    --platform tdx=on
+    --firmware edk2-staging/Build/OvmfCh/DEBUG_GCC5/FV/OVMF.fd \
     --cpus boot=1 \
     --memory size=1G \
     --disk path=tdx_guest_img \
@@ -95,7 +97,8 @@ option as well.
 
 ```bash
 ./cloud-hypervisor \
-    --tdx firmware=tdshim \
+    --platform tdx=on
+    --firmware tdshim \
     --kernel bzImage \
     --cmdline "root=/dev/vda3 console=hvc0 rw"
     --cpus boot=1 \

docs/memory.md

@@ -20,12 +20,13 @@ 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" [default: size=512M]
+--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]
 ```
 
 ### `size`
@@ -117,6 +118,9 @@ 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_
 
 ```
@@ -140,6 +144,9 @@ 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_
@@ -171,6 +178,25 @@ _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
@@ -271,6 +297,9 @@ 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.
 
@@ -298,6 +327,9 @@ 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_

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 -f ./cloud-hypervisor ...
+strace --decode-pids=comm -f ./cloud-hypervisor ...
 ```

docs/tpm.md

@@ -0,0 +1,122 @@
+# 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

@@ -0,0 +1,42 @@
+# 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

@@ -2,17 +2,19 @@
 
 Cloud Hypervisor supports UEFI boot through the utilization of the EDK II based UEFI firmware. 
 
-## Building UEFI Firmware
+## Building UEFI Firmware for x86-64
 
 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/cloud-hypervisor/edk2 -b ch
+git clone https://github.com/tianocore/edk2
 cd edk2
 . edksetup.sh
 git submodule update --init
@@ -27,11 +29,40 @@ 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` 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` (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.
 
 # Links
 

docs/vsock.md

@@ -0,0 +1,84 @@
+# VSOCK support
+
+VSOCK provides a way for guest and host to communicate through a socket. VSOCK sockets support both stream and datagram types.
+
+The `virtio-vsock` is based on the [Firecracker](https://github.com/firecracker-microvm/firecracker/blob/main/docs/vsock.md) implementation, where additional details can be found.
+
+## What is a CID?
+
+CID is a 32-bit context identifier describing the source or destination. In combination with the port, the complete addressing can be achieved to describe multiple listeners running on the same machine.
+
+The table below depicts the well known CID values:
+
+| CID | Description |
+|-----|-------------|
+| -1  | Random CID | 
+|  0  | Hypervisor | 
+|  1  | Loopback | 
+|  2  | Host | 
+
+## Prerequisites
+
+### Kernel Requirements
+
+Host kernel: CONFIG_VHOST_VSOCK
+
+Guest kernel: CONFIG_VIRTIO_VSOCKETS
+
+### Nested VM support
+
+Linux __v5.5__ or newer is required for the L1 VM.
+
+### Loopback support
+
+Linux __v5.6__ or newer is required.
+
+## Establishing VSOCK Connection
+
+VSOCK device becomes available with `--vsock` option passed by the VM start. Cloud Hypervisor can be invoked for instance as below:
+
+```bash
+cloud-hypervisor \
+	--cpus boot=1 \
+	--memory size=4G \
+	--firmware CLOUDHV.fd \
+	--disk path=jammy-server-cloudimg.raw \
+	--vsock cid=3,socket=/tmp/ch.vsock
+```
+
+The examples use __socat__ `>=1.7.4` to illustrate the VSOCK functionality. However, there are other tools supporting VSOCK, like [ncat](https://stefano-garzarella.github.io/posts/2019-11-08-kvmforum-2019-vsock/).
+
+### Connecting from Host to Guest
+
+The host starts to listen on the defined port:
+
+`$ socat - VSOCK-LISTEN:1234`
+
+Once the host is listening, the guest can send data:
+
+`echo -e "CONNECT 1234\\nHello from host!" | socat - UNIX-CONNECT:/tmp/ch.vsock
+ 
+Note the string `CONNECT <port>` prepended to the actual data. It is possible for the guest to start listening on different ports, thus the specific command is needed to instruct VSOCK to which listener the host wants to connect. It needs to be sent once per connection. Once the connection established, data transfers can take place directly.
+
+### Connecting from Guest to Host
+
+This first requires a listening UNIX socket on the host side. The UNIX socket path has to be constructed by using the socket path used at the VM launch time with appended `_` and the port number to be used on the guest side. As in the example above, if we'd intended to connect from the guest to the port `1234`, the Unix socket path on the host side would be `/tmp/ch.vsock_1234`.
+
+Also note that the CID used on the guest side is the well known CID value `2`.
+
+Listening on the host side:
+
+`$ socat - UNIX-LISTEN:/tmp/ch.vsock_1234`
+
+From the guest:
+
+`$ echo -e "Hello from guest!" | socat - VSOCK-CONNECT:2:1234`
+
+## Links
+
+- [virtio-vsock in QEMU, Firecracker and Linux: Status, Performance and Challenges](https://kvmforum2019.sched.com/event/TmwK)
+- [Leveraging virtio-vsock in the cloud and containers](https://archive.fosdem.org/2021/schedule/event/vai_virtio_vsock/)
+- [VSOCK man page](https://manpages.ubuntu.com/manpages/focal/man7/vsock.7.html)
+- [https://stefano-garzarella.github.io/posts/2020-02-20-vsock-nested-vms-loopback/](https://stefano-garzarella.github.io/posts/2020-02-20-vsock-nested-vms-loopback/)
+- [https://github.com/firecracker-microvm/firecracker/blob/main/docs/vsock.md](https://github.com/firecracker-microvm/firecracker/blob/main/docs/vsock.md)
+

docs/windows.md

@@ -131,7 +131,7 @@ See also the [links](#Links) section for a more extended SAC documentation.
 
 ## Network
 
-This section illustrates the Windows specific corner points for the VM network configuration. For the extended networking guide, including bridging for multiple VMs, follow [networking.md](networking.md).
+This section illustrates the Windows specific aspects of the VM network configuration.
 
 ### Basic Networking
 

event_monitor/Cargo.toml

@@ -5,6 +5,6 @@ authors = ["The Cloud Hypervisor Authors"]
 edition = "2021"
 
 [dependencies]
-libc = "0.2.131"
-serde = { version = "1.0.143", features = ["rc", "derive"] }
-serde_json = "1.0.83"
+libc = "0.2.139"
+serde = { version = "1.0.151", features = ["rc", "derive"] }
+serde_json = "1.0.89"

event_monitor/src/lib.rs

@@ -16,8 +16,10 @@ static mut MONITOR: Option<(File, Instant)> = None;
 /// This function must only be called once from the main process before any threads
 /// are created to avoid race conditions
 pub fn set_monitor(file: File) -> Result<(), std::io::Error> {
+    // SAFETY: there is only one caller of this function, so MONITOR is written to only once
     assert!(unsafe { MONITOR.is_none() });
     let fd = file.as_raw_fd();
+    // SAFETY: FFI call to configure the fd
     let ret = unsafe {
         let mut flags = libc::fcntl(fd, libc::F_GETFL);
         flags |= libc::O_NONBLOCK;
@@ -26,6 +28,7 @@ pub fn set_monitor(file: File) -> Result<(), std::io::Error> {
     if ret < 0 {
         return Err(std::io::Error::last_os_error());
     }
+    // SAFETY: MONITOR is None. Nobody else can hold a reference to it.
     unsafe {
         MONITOR = Some((file, Instant::now()));
     };
@@ -41,6 +44,7 @@ struct Event<'a> {
 }
 
 pub fn event_log(source: &str, event: &str, properties: Option<&HashMap<Cow<str>, Cow<str>>>) {
+    // SAFETY: MONITOR is always in a valid state (None or Some).
     if let Some((file, start)) = unsafe { MONITOR.as_ref() } {
         let e = Event {
             timestamp: start.elapsed(),

fuzz/Cargo.lock

@@ -11,9 +11,9 @@ dependencies = [
 
 [[package]]
 name = "anyhow"
-version = "1.0.62"
+version = "1.0.68"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1485d4d2cc45e7b201ee3767015c96faa5904387c9d87c6efdd0fb511f12d305"
+checksum = "2cb2f989d18dd141ab8ae82f64d1a8cdd37e0840f73a406896cf5e99502fab61"
 
 [[package]]
 name = "api_client"
@@ -24,9 +24,9 @@ dependencies = [
 
 [[package]]
 name = "arbitrary"
-version = "1.1.3"
+version = "1.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5a7924531f38b1970ff630f03eb20a2fde69db5c590c93b0f3482e95dcc5fd60"
+checksum = "29d47fbf90d5149a107494b15a7dc8d69b351be2db3bb9691740e88ec17fd880"
 
 [[package]]
 name = "arc-swap"
@@ -56,23 +56,6 @@ dependencies = [
  "vmm-sys-util",
 ]
 
-[[package]]
-name = "atty"
-version = "0.2.14"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d9b39be18770d11421cdb1b9947a45dd3f37e93092cbf377614828a319d5fee8"
-dependencies = [
- "hermit-abi",
- "libc",
- "winapi",
-]
-
-[[package]]
-name = "autocfg"
-version = "1.1.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d468802bab17cbc0cc575e9b053f41e72aa36bfa6b7f55e3529ffa43161b97fa"
-
 [[package]]
 name = "bincode"
 version = "1.3.3"
@@ -96,6 +79,7 @@ dependencies = [
  "libc",
  "log",
  "qcow",
+ "smallvec",
  "thiserror",
  "versionize",
  "versionize_derive",
@@ -115,9 +99,12 @@ checksum = "14c189c53d098945499cdfa7ecc63567cf3886b3332b312a5b4585d8d3a6a610"
 
 [[package]]
 name = "cc"
-version = "1.0.73"
+version = "1.0.78"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2fff2a6927b3bb87f9595d67196a70493f627687a71d87a0d692242c33f58c11"
+checksum = "a20104e2335ce8a659d6dd92a51a767a0c062599c73b343fd152cb401e828c3d"
+dependencies = [
+ "jobserver",
+]
 
 [[package]]
 name = "cfg-if"
@@ -127,33 +114,31 @@ checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
 name = "clap"
-version = "3.2.17"
+version = "4.0.32"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "29e724a68d9319343bb3328c9cc2dfde263f4b3142ee1059a9980580171c954b"
+checksum = "a7db700bc935f9e43e88d00b0850dae18a63773cfbec6d8e070fccf7fef89a39"
 dependencies = [
- "atty",
  "bitflags",
  "clap_lex",
- "indexmap",
+ "is-terminal",
  "once_cell",
  "strsim",
  "termcolor",
  "terminal_size",
- "textwrap",
 ]
 
 [[package]]
 name = "clap_lex"
-version = "0.2.4"
+version = "0.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2850f2f5a82cbf437dd5af4d49848fbdfc27c157c3d010345776f952765261c5"
+checksum = "0d4198f73e42b4936b35b5bb248d81d2b595ecb170da0bac7655c54eedfa8da8"
 dependencies = [
  "os_str_bytes",
 ]
 
 [[package]]
 name = "cloud-hypervisor"
-version = "25.0.0"
+version = "28.0.0"
 dependencies = [
  "anyhow",
  "api_client",
@@ -168,6 +153,8 @@ dependencies = [
  "serde_json",
  "signal-hook",
  "thiserror",
+ "tpm",
+ "tracer",
  "vm-memory",
  "vmm",
  "vmm-sys-util",
@@ -183,7 +170,9 @@ dependencies = [
  "epoll",
  "libc",
  "libfuzzer-sys",
+ "linux-loader",
  "micro_http",
+ "net_util",
  "once_cell",
  "qcow",
  "seccompiler",
@@ -214,9 +203,9 @@ checksum = "55626594feae15d266d52440b26ff77de0e22230cf0c113abe619084c1ddc910"
 
 [[package]]
 name = "darling"
-version = "0.14.1"
+version = "0.14.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4529658bdda7fd6769b8614be250cdcfc3aeb0ee72fe66f9e41e5e5eb73eac02"
+checksum = "b0dd3cd20dc6b5a876612a6e5accfe7f3dd883db6d07acfbf14c128f61550dfa"
 dependencies = [
  "darling_core",
  "darling_macro",
@@ -224,9 +213,9 @@ dependencies = [
 
 [[package]]
 name = "darling_core"
-version = "0.14.1"
+version = "0.14.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "649c91bc01e8b1eac09fb91e8dbc7d517684ca6be8ebc75bb9cafc894f9fdb6f"
+checksum = "a784d2ccaf7c98501746bf0be29b2022ba41fd62a2e622af997a03e9f972859f"
 dependencies = [
  "fnv",
  "ident_case",
@@ -238,9 +227,9 @@ dependencies = [
 
 [[package]]
 name = "darling_macro"
-version = "0.14.1"
+version = "0.14.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ddfc69c5bfcbd2fc09a0f38451d2daf0e372e367986a83906d1b0dbc88134fb5"
+checksum = "7618812407e9402654622dd402b0a89dff9ba93badd6540781526117b92aab7e"
 dependencies = [
  "darling_core",
  "quote",
@@ -259,6 +248,9 @@ dependencies = [
  "hypervisor",
  "libc",
  "log",
+ "phf",
+ "thiserror",
+ "tpm",
  "versionize",
  "versionize_derive",
  "vm-device",
@@ -277,6 +269,27 @@ dependencies = [
  "libc",
 ]
 
+[[package]]
+name = "errno"
+version = "0.2.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f639046355ee4f37944e44f60642c6f3a7efa3cf6b78c78a0d989a8ce6c396a1"
+dependencies = [
+ "errno-dragonfly",
+ "libc",
+ "winapi",
+]
+
+[[package]]
+name = "errno-dragonfly"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "aa68f1b12764fab894d2755d2518754e71b4fd80ecfb822714a1206c2aab39bf"
+dependencies = [
+ "cc",
+ "libc",
+]
+
 [[package]]
 name = "event_monitor"
 version = "0.1.0"
@@ -300,26 +313,20 @@ checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
 
 [[package]]
 name = "getrandom"
-version = "0.2.7"
+version = "0.2.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4eb1a864a501629691edf6c15a593b7a51eebaa1e8468e9ddc623de7c9b58ec6"
+checksum = "c05aeb6a22b8f62540c194aac980f2115af067bfe15a0734d7277a768d396b31"
 dependencies = [
  "cfg-if",
  "libc",
  "wasi",
 ]
 
-[[package]]
-name = "hashbrown"
-version = "0.12.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8a9ee70c43aaf417c914396645a0fa852624801b24ebb7ae78fe8272889ac888"
-
 [[package]]
 name = "hermit-abi"
-version = "0.1.20"
+version = "0.2.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c7a30908dbce072eca83216eab939d2290080e00ca71611b96a09e5cdce5f3fa"
+checksum = "ee512640fe35acbfb4bb779db6f0d80704c2cacfa2e39b601ef3e3f47d1ae4c7"
 dependencies = [
  "libc",
 ]
@@ -345,12 +352,11 @@ dependencies = [
 
 [[package]]
 name = "iced-x86"
-version = "1.17.0"
+version = "1.18.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "158f5204401d08f91d19176112146d75e99b3cf745092e268fa7be33e09adcec"
+checksum = "1dd04b950d75b3498320253b17fb92745b2cc79ead8814aede2f7c1bab858bec"
 dependencies = [
  "lazy_static",
- "static_assertions",
 ]
 
 [[package]]
@@ -360,35 +366,56 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b9e0384b61958566e926dc50660321d12159025e767c18e043daf26b70104c39"
 
 [[package]]
-name = "indexmap"
-version = "1.9.1"
+name = "io-lifetimes"
+version = "1.0.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "10a35a97730320ffe8e2d410b5d3b69279b98d2c14bdb8b70ea89ecf7888d41e"
+checksum = "46112a93252b123d31a119a8d1a1ac19deac4fac6e0e8b0df58f0d4e5870e63c"
 dependencies = [
- "autocfg",
- "hashbrown",
+ "libc",
+ "windows-sys",
 ]
 
 [[package]]
 name = "io-uring"
-version = "0.5.3"
+version = "0.5.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "427496d698429d86ff947537dff68facb83fe753b7abea4fab4ac1c4b7a4b94b"
+checksum = "c9ddf18110cedc09617f5a965bdb51adb0c0fb02c791e4ab22f7a21bb25269b0"
 dependencies = [
  "bitflags",
  "libc",
 ]
 
 [[package]]
-name = "itoa"
-version = "1.0.3"
+name = "is-terminal"
+version = "0.4.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6c8af84674fe1f223a982c933a0ee1086ac4d4052aa0fb8060c12c6ad838e754"
+checksum = "28dfb6c8100ccc63462345b67d1bbc3679177c75ee4bf59bf29c8b1d110b8189"
+dependencies = [
+ "hermit-abi",
+ "io-lifetimes",
+ "rustix",
+ "windows-sys",
+]
+
+[[package]]
+name = "itoa"
+version = "1.0.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fad582f4b9e86b6caa621cabeb0963332d92eea04729ab12892c2533951e6440"
+
+[[package]]
+name = "jobserver"
+version = "0.1.25"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "068b1ee6743e4d11fb9c6a1e6064b3693a1b600e7f5f5988047d98b3dc9fb90b"
+dependencies = [
+ "libc",
+]
 
 [[package]]
 name = "kvm-bindings"
-version = "0.5.0"
-source = "git+https://github.com/cloud-hypervisor/kvm-bindings?branch=ch-v0.5.0-tdx#52e56d0e8ef0f6ea32fc0492e6a175b73617a49f"
+version = "0.6.0"
+source = "git+https://github.com/cloud-hypervisor/kvm-bindings?branch=ch-v0.6.0-tdx#7d9ffb47e5b9b1989577258800a0f57c93f1445f"
 dependencies = [
  "serde",
  "serde_derive",
@@ -397,8 +424,9 @@ dependencies = [
 
 [[package]]
 name = "kvm-ioctls"
-version = "0.11.0"
-source = "git+https://github.com/rust-vmm/kvm-ioctls?branch=main#1e03e29cdfbb0cb108a98de7a78045a5a517f18e"
+version = "0.12.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c3a321cabd827642499c77e27314f388dd83a717a5ca716b86476fb947f73ae4"
 dependencies = [
  "kvm-bindings",
  "libc",
@@ -413,15 +441,15 @@ checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646"
 
 [[package]]
 name = "libc"
-version = "0.2.131"
+version = "0.2.139"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "04c3b4822ccebfa39c02fc03d1534441b22ead323fa0f48bb7ddd8e6ba076a40"
+checksum = "201de327520df007757c1f0adce6e827fe8562fbc28bfd9c15571c66ca1f5f79"
 
 [[package]]
 name = "libfuzzer-sys"
-version = "0.4.3"
+version = "0.4.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "336244aaeab6a12df46480dc585802aa743a72d66b11937844c61bbca84c991d"
+checksum = "c8fff891139ee62800da71b7fd5b508d570b9ad95e614a53c6f453ca08366038"
 dependencies = [
  "arbitrary",
  "cc",
@@ -430,13 +458,19 @@ dependencies = [
 
 [[package]]
 name = "linux-loader"
-version = "0.4.0"
+version = "0.8.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8a5e77493808403a6bd56a301a64ea6b9342e36ea845044bf0dfdf56fe52fa08"
+checksum = "b9259ddbfbb52cc918f6bbc60390004ddd0228cf1d85f402009ff2b3d95de83f"
 dependencies = [
  "vm-memory",
 ]
 
+[[package]]
+name = "linux-raw-sys"
+version = "0.1.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f051f77a7c8e6957c0696eac88f26b0117e54f52d3fc682ab19397a8812846a4"
+
 [[package]]
 name = "log"
 version = "0.4.17"
@@ -449,7 +483,7 @@ dependencies = [
 [[package]]
 name = "micro_http"
 version = "0.1.0"
-source = "git+https://github.com/firecracker-microvm/micro-http?branch=main#863b0370ba7e57f7df5b908ada9e5b44809ccae9"
+source = "git+https://github.com/firecracker-microvm/micro-http?branch=main#fbef706e28c7ca7cce8dfec7ca073f3b359879d3"
 dependencies = [
  "libc",
  "vmm-sys-util",
@@ -485,9 +519,9 @@ dependencies = [
 
 [[package]]
 name = "once_cell"
-version = "1.13.1"
+version = "1.17.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "074864da206b4973b84eb91683020dbefd6a8c3f0f38e054d93954e891935e4e"
+checksum = "6f61fba1741ea2b3d6a1e3178721804bb716a68a6aeba1149b5d52e3d464ea66"
 
 [[package]]
 name = "option_parser"
@@ -495,9 +529,9 @@ version = "0.1.0"
 
 [[package]]
 name = "os_str_bytes"
-version = "6.3.0"
+version = "6.4.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9ff7415e9ae3fff1225851df9e0d9e4e5479f947619774677a63572e55e80eff"
+checksum = "9b7820b9daea5457c9f21c69448905d723fbd21136ccf521748f23fd49e723ee"
 
 [[package]]
 name = "pci"
@@ -523,10 +557,52 @@ dependencies = [
 ]
 
 [[package]]
-name = "proc-macro2"
-version = "1.0.43"
+name = "phf"
+version = "0.11.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0a2ca2c61bc9f3d74d2886294ab7b9853abd9c1ad903a3ac7815c58989bb7bab"
+checksum = "928c6535de93548188ef63bb7c4036bd415cd8f36ad25af44b9789b2ee72a48c"
+dependencies = [
+ "phf_macros",
+ "phf_shared",
+]
+
+[[package]]
+name = "phf_generator"
+version = "0.11.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b1181c94580fa345f50f19d738aaa39c0ed30a600d95cb2d3e23f94266f14fbf"
+dependencies = [
+ "phf_shared",
+ "rand",
+]
+
+[[package]]
+name = "phf_macros"
+version = "0.11.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "92aacdc5f16768709a569e913f7451034034178b05bdc8acda226659a3dccc66"
+dependencies = [
+ "phf_generator",
+ "phf_shared",
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
+[[package]]
+name = "phf_shared"
+version = "0.11.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e1fb5f6f826b772a8d4c0394209441e7d37cbbb967ae9c7e0e8134365c9ee676"
+dependencies = [
+ "siphasher",
+]
+
+[[package]]
+name = "proc-macro2"
+version = "1.0.49"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "57a8eca9f9c4ffde41714334dee777596264c7825420f521abc92b5b5deb63a5"
 dependencies = [
  "unicode-ident",
 ]
@@ -551,6 +627,21 @@ dependencies = [
  "proc-macro2",
 ]
 
+[[package]]
+name = "rand"
+version = "0.8.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "34af8d1a0e25924bc5b7c43c079c942339d8f0a8b57c39049bef581b46327404"
+dependencies = [
+ "rand_core",
+]
+
+[[package]]
+name = "rand_core"
+version = "0.6.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ec0be4795e2f6a28069bec0b5ff3e2ac9bafc99e6a9a7dc3547996c5c816922c"
+
 [[package]]
 name = "rate_limiter"
 version = "0.1.0"
@@ -562,9 +653,9 @@ dependencies = [
 
 [[package]]
 name = "remain"
-version = "0.2.4"
+version = "0.2.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "06df529c0d271b27ac4a2c260f5ce2914b60bd44702cecec7b9f271adbdde23b"
+checksum = "5704e2cda92fd54202f05430725317ba0ea7d0c96b246ca0a92e45177127ba3b"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -581,40 +672,54 @@ dependencies = [
 ]
 
 [[package]]
-name = "ryu"
-version = "1.0.11"
+name = "rustix"
+version = "0.36.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4501abdff3ae82a1c1b477a17252eb69cee9e66eb915c1abaa4f44d873df9f09"
+checksum = "a3807b5d10909833d3e9acd1eb5fb988f79376ff10fce42937de71a449c4c588"
+dependencies = [
+ "bitflags",
+ "errno",
+ "io-lifetimes",
+ "libc",
+ "linux-raw-sys",
+ "windows-sys",
+]
+
+[[package]]
+name = "ryu"
+version = "1.0.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7b4b9743ed687d4b4bcedf9ff5eaa7398495ae14e61cba0a295704edbc7decde"
 
 [[package]]
 name = "seccompiler"
-version = "0.2.0"
+version = "0.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e01d1292a1131b22ccea49f30bd106f1238b5ddeec1a98d39268dcc31d540e68"
+checksum = "6f6575e3c2b3a0fe2ef3e53855b6a8dead7c29f783da5e123d378c8c6a89017e"
 dependencies = [
  "libc",
 ]
 
 [[package]]
 name = "semver"
-version = "1.0.13"
+version = "1.0.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "93f6841e709003d68bb2deee8c343572bf446003ec20a583e76f7b15cebf3711"
+checksum = "58bc9567378fc7690d6b2addae4e60ac2eeea07becb2c64b9f218b53865cba2a"
 
 [[package]]
 name = "serde"
-version = "1.0.143"
+version = "1.0.152"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "53e8e5d5b70924f74ff5c6d64d9a5acd91422117c60f48c4e07855238a254553"
+checksum = "bb7d1f0d3021d347a83e556fc4683dea2ea09d87bccdf88ff5c12545d89d5efb"
 dependencies = [
  "serde_derive",
 ]
 
 [[package]]
 name = "serde_derive"
-version = "1.0.143"
+version = "1.0.152"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d3d8e8de557aee63c26b85b947f5e59b690d0454c753f3adeb5cd7835ab88391"
+checksum = "af487d118eecd09402d70a5d72551860e788df87b464af30e5ea6a38c75c541e"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -623,9 +728,9 @@ dependencies = [
 
 [[package]]
 name = "serde_json"
-version = "1.0.83"
+version = "1.0.91"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "38dd04e3c8279e75b31ef29dbdceebfe5ad89f4d0937213c53f7d49d01b3d5a7"
+checksum = "877c235533714907a8c2464236f5c4b2a17262ef1bd71f38f35ea592c8da6883"
 dependencies = [
  "itoa",
  "ryu",
@@ -634,9 +739,9 @@ dependencies = [
 
 [[package]]
 name = "serde_with"
-version = "2.0.0"
+version = "2.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "89df7a26519371a3cce44fbb914c2819c84d9b897890987fa3ab096491cc0ea8"
+checksum = "25bf4a5a814902cd1014dbccfa4d4560fb8432c779471e96e035602519f82eef"
 dependencies = [
  "serde",
  "serde_with_macros",
@@ -644,9 +749,9 @@ dependencies = [
 
 [[package]]
 name = "serde_with_macros"
-version = "2.0.0"
+version = "2.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "de337f322382fcdfbb21a014f7c224ee041a23785651db67b9827403178f698f"
+checksum = "a1966009f3c05f095697c537312f5415d1e3ed31ce0a56942bac4c771c5c335e"
 dependencies = [
  "darling",
  "proc-macro2",
@@ -654,6 +759,10 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "serial_buffer"
+version = "0.1.0"
+
 [[package]]
 name = "signal-hook"
 version = "0.3.14"
@@ -674,10 +783,16 @@ dependencies = [
 ]
 
 [[package]]
-name = "static_assertions"
-version = "1.1.0"
+name = "siphasher"
+version = "0.3.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a2eb9349b6444b326872e140eb1cf5e7c522154d69e7a0ffb0fb81c06b37543f"
+checksum = "7bd3e3206899af3f8b12af284fafc038cc1dc2b41d1b89dd17297221c5d225de"
+
+[[package]]
+name = "smallvec"
+version = "1.10.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a507befe795404456341dfab10cef66ead4c041f62b8b11bbb92bffe5d0953e0"
 
 [[package]]
 name = "strsim"
@@ -687,9 +802,9 @@ checksum = "73473c0e59e6d5812c5dfe2a064a6444949f089e20eec9a2e5506596494e4623"
 
 [[package]]
 name = "syn"
-version = "1.0.99"
+version = "1.0.107"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "58dbef6ec655055e20b86b15a8cc6d439cca19b667537ac6a1369572d151ab13"
+checksum = "1f4064b5b16e03ae50984a5a8ed5d4f8803e6bc1fd170a3cda91a1be4b18e3f5"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -707,63 +822,77 @@ dependencies = [
 
 [[package]]
 name = "terminal_size"
-version = "0.1.17"
+version = "0.2.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "633c1a546cee861a1a6d0dc69ebeca693bf4296661ba7852b9d21d159e0506df"
+checksum = "cb20089a8ba2b69debd491f8d2d023761cbf196e999218c591fa1e7e15a21907"
 dependencies = [
- "libc",
- "winapi",
-]
-
-[[package]]
-name = "textwrap"
-version = "0.15.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b1141d4d61095b28419e22cb0bbf02755f5e54e0526f97f1e3d1d160e60885fb"
-dependencies = [
- "terminal_size",
+ "rustix",
+ "windows-sys",
 ]
 
 [[package]]
 name = "thiserror"
-version = "1.0.32"
+version = "1.0.37"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f5f6586b7f764adc0231f4c79be7b920e766bb2f3e51b3661cdb263828f19994"
+checksum = "10deb33631e3c9018b9baf9dcbbc4f737320d2b576bac10f6aefa048fa407e3e"
 dependencies = [
  "thiserror-impl",
 ]
 
 [[package]]
 name = "thiserror-impl"
-version = "1.0.32"
+version = "1.0.37"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "12bafc5b54507e0149cdf1b145a5d80ab80a90bcd9275df43d4fff68460f6c21"
+checksum = "982d17546b47146b28f7c22e3d08465f6b8903d0ea13c1660d9d84a6e7adcdbb"
 dependencies = [
  "proc-macro2",
  "quote",
  "syn",
 ]
 
+[[package]]
+name = "tpm"
+version = "0.1.0"
+dependencies = [
+ "anyhow",
+ "byteorder",
+ "libc",
+ "log",
+ "thiserror",
+ "vmm-sys-util",
+]
+
+[[package]]
+name = "tracer"
+version = "0.1.0"
+dependencies = [
+ "libc",
+ "log",
+ "once_cell",
+ "serde",
+ "serde_json",
+]
+
 [[package]]
 name = "unicode-ident"
-version = "1.0.3"
+version = "1.0.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c4f5b37a154999a8f3f98cc23a628d850e154479cd94decf3414696e12e31aaf"
+checksum = "84a22b9f218b40614adcb3f4ff08b703773ad44fa9423e4e0d346d5db86e4ebc"
 
 [[package]]
 name = "uuid"
-version = "1.1.2"
+version = "1.2.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dd6469f4314d5f1ffec476e05f17cc9a78bc7a27a6a857842170bdf8d6f98d2f"
+checksum = "422ee0de9031b5b948b97a8fc04e3aa35230001a722ddd27943e0be31564ce4c"
 dependencies = [
  "getrandom",
 ]
 
 [[package]]
 name = "versionize"
-version = "0.1.6"
+version = "0.1.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7429cf68de8f091b667d27323ed323afd39584a56d533995b12ddd748e5e6ca9"
+checksum = "d6e2495726cf917e7ba7ec8bf0f0fceab543dd38d0a4195ed6bef331e38a290f"
 dependencies = [
  "bincode",
  "crc64",
@@ -788,17 +917,16 @@ dependencies = [
 
 [[package]]
 name = "vfio-bindings"
-version = "0.3.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "43449b404c488f70507dca193debd4bea361fe8089869b947adc19720e464bce"
+version = "0.4.0"
+source = "git+https://github.com/rust-vmm/vfio?branch=main#7ef33f6759ae6d37a54cd66f749aa47c1c81509c"
 dependencies = [
  "vmm-sys-util",
 ]
 
 [[package]]
 name = "vfio-ioctls"
-version = "0.1.0"
-source = "git+https://github.com/rust-vmm/vfio?branch=main#1ff2324e5cb0fccab825f9c96bb9387a1a01863f"
+version = "0.2.0"
+source = "git+https://github.com/rust-vmm/vfio?branch=main#7ef33f6759ae6d37a54cd66f749aa47c1c81509c"
 dependencies = [
  "byteorder",
  "kvm-bindings",
@@ -841,9 +969,9 @@ dependencies = [
 
 [[package]]
 name = "vhost"
-version = "0.4.0"
+version = "0.6.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "53567fd9ab820e4f3cc156f24146882fee3c365194c3e1dea74723265f27fc88"
+checksum = "c9b791c5b0717a0558888a4cf7240cea836f39a99cb342e12ce633dcaa078072"
 dependencies = [
  "bitflags",
  "libc",
@@ -877,6 +1005,7 @@ dependencies = [
  "seccompiler",
  "serde",
  "serde_json",
+ "serial_buffer",
  "thiserror",
  "versionize",
  "versionize_derive",
@@ -893,9 +1022,9 @@ dependencies = [
 
 [[package]]
 name = "virtio-queue"
-version = "0.5.0"
+version = "0.7.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b4f59652909f276e6edd8bf36e9f106480b2202f5f046717b3de14f1b4072a28"
+checksum = "19e927d93d54c365034fd7f31a5f458a1f540de4a37c52e892670dad9692173c"
 dependencies = [
  "log",
  "virtio-bindings",
@@ -927,14 +1056,14 @@ dependencies = [
 
 [[package]]
 name = "vm-fdt"
-version = "0.1.0"
-source = "git+https://github.com/rust-vmm/vm-fdt?branch=main#720e48e435b791ec6cbe8d2b229448b71dcb1ab9"
+version = "0.2.0"
+source = "git+https://github.com/rust-vmm/vm-fdt?branch=main#c5a99ab71b130435927d19b50c85fcd5ce904a8c"
 
 [[package]]
 name = "vm-memory"
-version = "0.8.0"
+version = "0.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "767ed8aaebbff902e02e6d3749dc2baef55e46565f8a6414a065e5baee4b4a81"
+checksum = "688a70366615b45575a424d9c665561c1b5ab2224d494f706b6a6812911a827c"
 dependencies = [
  "arc-swap",
  "libc",
@@ -990,8 +1119,10 @@ dependencies = [
  "seccompiler",
  "serde",
  "serde_json",
+ "serial_buffer",
  "signal-hook",
  "thiserror",
+ "tracer",
  "uuid",
  "versionize",
  "versionize_derive",
@@ -1010,9 +1141,9 @@ dependencies = [
 
 [[package]]
 name = "vmm-sys-util"
-version = "0.10.0"
+version = "0.11.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "08604d7be03eb26e33b3cee3ed4aef2bf550b305d1cca60e84da5d28d3790b62"
+checksum = "cc06a16ee8ebf0d9269aed304030b0d20a866b8b3dd3d4ce532596ac567a0d24"
 dependencies = [
  "bitflags",
  "libc",
@@ -1056,3 +1187,60 @@ name = "winapi-x86_64-pc-windows-gnu"
 version = "0.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
+
+[[package]]
+name = "windows-sys"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5a3e1820f08b8513f676f7ab6c1f99ff312fb97b553d30ff4dd86f9f15728aa7"
+dependencies = [
+ "windows_aarch64_gnullvm",
+ "windows_aarch64_msvc",
+ "windows_i686_gnu",
+ "windows_i686_msvc",
+ "windows_x86_64_gnu",
+ "windows_x86_64_gnullvm",
+ "windows_x86_64_msvc",
+]
+
+[[package]]
+name = "windows_aarch64_gnullvm"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "41d2aa71f6f0cbe00ae5167d90ef3cfe66527d6f613ca78ac8024c3ccab9a19e"
+
+[[package]]
+name = "windows_aarch64_msvc"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dd0f252f5a35cac83d6311b2e795981f5ee6e67eb1f9a7f64eb4500fbc4dcdb4"
+
+[[package]]
+name = "windows_i686_gnu"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fbeae19f6716841636c28d695375df17562ca208b2b7d0dc47635a50ae6c5de7"
+
+[[package]]
+name = "windows_i686_msvc"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "84c12f65daa39dd2babe6e442988fc329d6243fdce47d7d2d155b8d874862246"
+
+[[package]]
+name = "windows_x86_64_gnu"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bf7b1b21b5362cbc318f686150e5bcea75ecedc74dd157d874d754a2ca44b0ed"
+
+[[package]]
+name = "windows_x86_64_gnullvm"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "09d525d2ba30eeb3297665bd434a54297e4170c7f1a44cad4ef58095b4cd2028"
+
+[[package]]
+name = "windows_x86_64_msvc"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f40009d85759725a34da6d89a94e63d7bdc50a862acf0dbc7c8e488f1edcb6f5"

fuzz/Cargo.toml

@@ -12,18 +12,20 @@ cargo-fuzz = true
 block_util = { path = "../block_util" }
 devices = { path = "../devices" }
 epoll = "4.3.1"
-libc = "0.2.131"
-libfuzzer-sys = "0.4.3"
+libc = "0.2.138"
+libfuzzer-sys = "0.4.5"
+linux-loader = { version = "0.8.1", features = ["elf", "bzimage", "pe"] }
 micro_http = { git = "https://github.com/firecracker-microvm/micro-http", branch = "main" }
-once_cell = "1.13.1"
+net_util = { path = "../net_util" }
+once_cell = "1.17.0"
 qcow = { path = "../qcow" }
-seccompiler = "0.2.0"
+seccompiler = "0.3.0"
 vhdx = { path = "../vhdx" }
 virtio-devices = { path = "../virtio-devices" }
-virtio-queue = "0.5.0"
+virtio-queue = "0.7.0"
 vmm = { path = "../vmm" }
-vmm-sys-util = "0.10.0"
-vm-memory = "0.8.0"
+vmm-sys-util = "0.11.0"
+vm-memory = "0.10.0"
 vm-device = { path = "../vm-device" }
 vm-virtio = { path = "../vm-virtio" }
 
@@ -31,14 +33,19 @@ vm-virtio = { path = "../vm-virtio" }
 path = ".."
 
 [patch.crates-io]
-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" }
+kvm-bindings = { git = "https://github.com/cloud-hypervisor/kvm-bindings", branch = "ch-v0.6.0-tdx" }
 versionize_derive = { git = "https://github.com/cloud-hypervisor/versionize_derive", branch = "ch" }
 
 # Prevent this from interfering with workspaces
 [workspace]
 members = ["."]
 
+[[bin]]
+name = "balloon"
+path = "fuzz_targets/balloon.rs"
+test = false
+doc = false
+
 [[bin]]
 name = "block"
 path = "fuzz_targets/block.rs"
@@ -51,18 +58,66 @@ path = "fuzz_targets/cmos.rs"
 test = false
 doc = false
 
+[[bin]]
+name = "console"
+path = "fuzz_targets/console.rs"
+test = false
+doc = false
+
 [[bin]]
 name = "http_api"
 path = "fuzz_targets/http_api.rs"
 test = false
 doc = false
 
+[[bin]]
+name = "iommu"
+path = "fuzz_targets/iommu.rs"
+test = false
+doc = false
+
+[[bin]]
+name = "linux_loader"
+path = "fuzz_targets/linux_loader.rs"
+test = false
+doc = false
+
+[[bin]]
+name = "linux_loader_cmdline"
+path = "fuzz_targets/linux_loader_cmdline.rs"
+test = false
+doc = false
+
+[[bin]]
+name = "mem"
+path = "fuzz_targets/mem.rs"
+test = false
+doc = false
+
+[[bin]]
+name = "net"
+path = "fuzz_targets/net.rs"
+test = false
+doc = false
+
+[[bin]]
+name = "pmem"
+path = "fuzz_targets/pmem.rs"
+test = false
+doc = false
+
 [[bin]]
 name = "qcow"
 path = "fuzz_targets/qcow.rs"
 test = false
 doc = false
 
+[[bin]]
+name = "rng"
+path = "fuzz_targets/rng.rs"
+test = false
+doc = false
+
 [[bin]]
 name = "serial"
 path = "fuzz_targets/serial.rs"
@@ -74,3 +129,9 @@ name = "vhdx"
 path = "fuzz_targets/vhdx.rs"
 test = false
 doc = false
+
+[[bin]]
+name = "watchdog"
+path = "fuzz_targets/watchdog.rs"
+test = false
+doc = false

fuzz/fuzz_targets/balloon.rs

@@ -0,0 +1,153 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use seccompiler::SeccompAction;
+use std::os::unix::io::{AsRawFd, FromRawFd};
+use std::sync::Arc;
+use virtio_devices::{VirtioDevice, VirtioInterrupt, VirtioInterruptType};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic};
+use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+const QUEUE_DATA_SIZE: usize = 4;
+const MEM_SIZE: usize = 512 * 1024;
+const BALLOON_SIZE: u64 = 512 * 1024;
+// Number of queues
+const QUEUE_NUM: usize = 3;
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 64;
+// Descriptor table alignment
+const DESC_TABLE_ALIGN_SIZE: u64 = 16;
+// Avalable ring alignment
+const AVAIL_RING_ALIGN_SIZE: u64 = 2;
+// Used ring alignment
+const USED_RING_ALIGN_SIZE: u64 = 4;
+// Descriptor table size
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Available ring size
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Used ring size
+const USED_RING_SIZE: u64 = 6_u64 + 8 * QUEUE_SIZE as u64;
+
+fuzz_target!(|bytes| {
+    if bytes.len() < QUEUE_DATA_SIZE * QUEUE_NUM
+        || bytes.len() > (QUEUE_DATA_SIZE * QUEUE_NUM + MEM_SIZE)
+    {
+        return;
+    }
+
+    let mut balloon = virtio_devices::Balloon::new(
+        "fuzzer_balloon".to_owned(),
+        BALLOON_SIZE,
+        true,
+        true,
+        SeccompAction::Allow,
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+        None,
+    )
+    .unwrap();
+
+    let queue_data = &bytes[..QUEUE_DATA_SIZE * QUEUE_NUM];
+    let mem_bytes = &bytes[QUEUE_DATA_SIZE * QUEUE_NUM..];
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[(GuestAddress(0), MEM_SIZE)]).unwrap();
+    if mem.write_slice(mem_bytes, GuestAddress(0 as u64)).is_err() {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    // Setup the virt queues with the input bytes
+    let mut queues = setup_virt_queues(
+        &[
+            &queue_data[..QUEUE_DATA_SIZE].try_into().unwrap(),
+            &queue_data[QUEUE_DATA_SIZE..QUEUE_DATA_SIZE * 2]
+                .try_into()
+                .unwrap(),
+            &queue_data[QUEUE_DATA_SIZE * 2..QUEUE_DATA_SIZE * 3]
+                .try_into()
+                .unwrap(),
+        ],
+        0,
+    );
+
+    let inflate_q = queues.remove(0);
+    let inflate_evt = EventFd::new(0).unwrap();
+    let inflate_queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(inflate_evt.as_raw_fd())) };
+    let deflate_q = queues.remove(0);
+    let deflate_evt = EventFd::new(0).unwrap();
+    let deflate_queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(deflate_evt.as_raw_fd())) };
+    let reporting_q = queues.remove(0);
+    let reporting_evt = EventFd::new(0).unwrap();
+    let reporting_queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(reporting_evt.as_raw_fd())) };
+
+    // Kick the 'queue' events before activate the balloon device
+    inflate_queue_evt.write(1).unwrap();
+    deflate_queue_evt.write(1).unwrap();
+    reporting_queue_evt.write(1).unwrap();
+
+    balloon
+        .activate(
+            guest_memory,
+            Arc::new(NoopVirtioInterrupt {}),
+            vec![
+                (0, inflate_q, inflate_evt),
+                (1, deflate_q, deflate_evt),
+                (2, reporting_q, reporting_evt),
+            ],
+        )
+        .ok();
+
+    // Wait for the events to finish and balloon device worker thread to return
+    balloon.wait_for_epoll_threads();
+});
+
+pub struct NoopVirtioInterrupt {}
+
+impl VirtioInterrupt for NoopVirtioInterrupt {
+    fn trigger(&self, _int_type: VirtioInterruptType) -> std::result::Result<(), std::io::Error> {
+        Ok(())
+    }
+}
+
+macro_rules! align {
+    ($n:expr, $align:expr) => {{
+        (($n + $align - 1) / $align) * $align
+    }};
+}
+
+fn setup_virt_queues(bytes: &[&[u8; QUEUE_DATA_SIZE]], base_addr: u64) -> Vec<Queue> {
+    let mut queues = Vec::new();
+    let mut base_addr = base_addr;
+    for b in bytes {
+        let mut q = Queue::new(QUEUE_SIZE).unwrap();
+
+        let desc_table_addr = align!(base_addr, DESC_TABLE_ALIGN_SIZE);
+        let avail_ring_addr = align!(desc_table_addr + DESC_TABLE_SIZE, AVAIL_RING_ALIGN_SIZE);
+        let used_ring_addr = align!(avail_ring_addr + AVAIL_RING_SIZE, USED_RING_ALIGN_SIZE);
+        q.try_set_desc_table_address(GuestAddress(desc_table_addr))
+            .unwrap();
+        q.try_set_avail_ring_address(GuestAddress(avail_ring_addr))
+            .unwrap();
+        q.try_set_used_ring_address(GuestAddress(used_ring_addr))
+            .unwrap();
+
+        q.set_next_avail(b[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+        q.set_next_used(b[1] as u16);
+        q.set_event_idx(b[2] % 2 != 0);
+        q.set_size(b[3] as u16 % QUEUE_SIZE);
+
+        q.set_ready(true);
+        queues.push(q);
+
+        base_addr = used_ring_addr + USED_RING_SIZE;
+    }
+
+    queues
+}

fuzz/fuzz_targets/block.rs

@@ -1,6 +1,10 @@
 // Copyright 2018 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 LICENSE file.
+//
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0 AND BSD-3-Clause
 
 #![no_main]
 
@@ -16,37 +20,35 @@ use std::sync::Arc;
 use virtio_devices::{Block, VirtioDevice, VirtioInterrupt, VirtioInterruptType};
 use virtio_queue::{Queue, QueueT};
 use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic};
-use vmm::{EpollContext, EpollDispatch};
 use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
 
 type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
 
-const MEM_SIZE: u64 = 256 * 1024 * 1024;
-const QUEUE_SIZE: u16 = 16; // Max entries in the queue.
+const QUEUE_DATA_SIZE: usize = 4;
+const MEM_SIZE: usize = 256 * 1024 * 1024;
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 256;
+// Guest physical address for descriptor table.
+const DESC_TABLE_ADDR: u64 = 0;
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Guest physical address for available ring
+const AVAIL_RING_ADDR: u64 = DESC_TABLE_ADDR + DESC_TABLE_SIZE;
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Guest physical address for used ring (requires to 4-bytes aligned)
+const USED_RING_ADDR: u64 = (AVAIL_RING_ADDR + AVAIL_RING_SIZE + 3) & !3_u64;
 
 fuzz_target!(|bytes| {
-    if bytes.len() as u64 > MEM_SIZE {
+    if bytes.len() < QUEUE_DATA_SIZE || bytes.len() > (QUEUE_DATA_SIZE + MEM_SIZE) {
         return;
     }
 
-    let mem = GuestMemoryMmap::from_ranges(&[(GuestAddress(0), MEM_SIZE as usize)]).unwrap();
-    if mem.write_slice(bytes, GuestAddress(0 as u64)).is_err() {
-        return;
-    }
-
-    let guest_memory = GuestMemoryAtomic::new(mem);
-
-    let mut q = Queue::new(QUEUE_SIZE).unwrap();
-    q.set_ready(true);
-    q.set_size(QUEUE_SIZE / 2);
-
-    let evt = EventFd::new(0).unwrap();
-    let queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(evt.as_raw_fd())) };
+    let queue_data = &bytes[..QUEUE_DATA_SIZE];
+    let mem_bytes = &bytes[QUEUE_DATA_SIZE..];
 
+    // Create a virtio-block device backed by a synchronous raw file
     let shm = memfd_create(&ffi::CString::new("fuzz").unwrap(), 0).unwrap();
     let disk_file: File = unsafe { File::from_raw_fd(shm) };
     let qcow_disk = Box::new(RawFileDiskSync::new(disk_file)) as Box<dyn DiskFile>;
-
     let mut block = Block::new(
         "tmp".to_owned(),
         qcow_disk,
@@ -58,9 +60,26 @@ fuzz_target!(|bytes| {
         SeccompAction::Allow,
         None,
         EventFd::new(EFD_NONBLOCK).unwrap(),
+        None,
     )
     .unwrap();
 
+    // Setup the virt queue with the input bytes
+    let q = setup_virt_queue(queue_data.try_into().unwrap());
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[(GuestAddress(0), MEM_SIZE)]).unwrap();
+    if mem.write_slice(mem_bytes, GuestAddress(0 as u64)).is_err() {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    let evt = EventFd::new(0).unwrap();
+    let queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(evt.as_raw_fd())) };
+
+    // Kick the 'queue' event before activate the block device
+    queue_evt.write(1).unwrap();
+
     block
         .activate(
             guest_memory,
@@ -69,11 +88,8 @@ fuzz_target!(|bytes| {
         )
         .ok();
 
-    queue_evt.write(77).unwrap(); // Rings the doorbell, any byte will do.
-
-    wait_queue_event_processed(queue_evt);
-
-    block.reset(); // Ensure the virtio-block device thread is killed and joined
+    // Wait for the events to finish and block device worker thread to return
+    block.wait_for_epoll_threads();
 });
 
 fn memfd_create(name: &ffi::CStr, flags: u32) -> Result<RawFd, io::Error> {
@@ -94,28 +110,20 @@ impl VirtioInterrupt for NoopVirtioInterrupt {
     }
 }
 
-fn wait_queue_event_processed(queue_evt: EventFd) {
-    let mut epoll = EpollContext::new().unwrap();
-    epoll.add_event(&queue_evt, EpollDispatch::Api).unwrap();
+fn setup_virt_queue(bytes: &[u8; QUEUE_DATA_SIZE]) -> Queue {
+    let mut q = Queue::new(QUEUE_SIZE).unwrap();
+    q.set_next_avail(bytes[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+    q.set_next_used(bytes[1] as u16);
+    q.set_event_idx(bytes[2] % 2 != 0);
+    q.set_size(bytes[3] as u16 % QUEUE_SIZE);
 
-    let epoll_fd = epoll.as_raw_fd();
-    let mut events = vec![epoll::Event::new(epoll::Events::empty(), 0); 1];
+    q.try_set_desc_table_address(GuestAddress(DESC_TABLE_ADDR))
+        .unwrap();
+    q.try_set_avail_ring_address(GuestAddress(AVAIL_RING_ADDR))
+        .unwrap();
+    q.try_set_used_ring_address(GuestAddress(USED_RING_ADDR))
+        .unwrap();
+    q.set_ready(true);
 
-    loop {
-        let num_events = match epoll::wait(epoll_fd, 0, &mut events[..]) {
-            Ok(res) => res,
-            Err(e) => {
-                if e.kind() == io::ErrorKind::Interrupted {
-                    continue;
-                }
-
-                println!("Warning: Unexpected epoll::wait() error!");
-                break;
-            }
-        };
-
-        if num_events == 0 {
-            break;
-        }
-    }
+    q
 }

fuzz/fuzz_targets/console.rs

@@ -0,0 +1,194 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use seccompiler::SeccompAction;
+use std::fs::File;
+use std::io::Write;
+use std::os::unix::io::{AsRawFd, FromRawFd, RawFd};
+use std::sync::Arc;
+use virtio_devices::{VirtioDevice, VirtioInterrupt, VirtioInterruptType};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic};
+use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+macro_rules! align {
+    ($n:expr, $align:expr) => {{
+        (($n + $align - 1) / $align) * $align
+    }};
+}
+
+const CONSOLE_INPUT_SIZE: usize = 128;
+const QUEUE_DATA_SIZE: usize = 4;
+const MEM_SIZE: usize = 32 * 1024 * 1024;
+// Guest memory gap
+const GUEST_MEM_GAP: u64 = 1 * 1024 * 1024;
+// Guest physical address for the first virt queue
+const BASE_VIRT_QUEUE_ADDR: u64 = MEM_SIZE as u64 + GUEST_MEM_GAP;
+// Number of queues
+const QUEUE_NUM: usize = 2;
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 256;
+// Descriptor table alignment
+const DESC_TABLE_ALIGN_SIZE: u64 = 16;
+// Used ring alignment
+const USED_RING_ALIGN_SIZE: u64 = 4;
+// Descriptor table size
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Available ring size
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Padding size before used ring
+const PADDING_SIZE: u64 = align!(AVAIL_RING_SIZE, USED_RING_ALIGN_SIZE) - AVAIL_RING_SIZE;
+// Used ring size
+const USED_RING_SIZE: u64 = 6_u64 + 8 * QUEUE_SIZE as u64;
+// Virtio-queue size in bytes
+const QUEUE_BYTES_SIZE: usize = align!(
+    DESC_TABLE_SIZE + AVAIL_RING_SIZE + PADDING_SIZE + USED_RING_SIZE,
+    DESC_TABLE_ALIGN_SIZE
+) as usize;
+
+fuzz_target!(|bytes| {
+    if bytes.len() < (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE) * QUEUE_NUM + CONSOLE_INPUT_SIZE
+        || bytes.len()
+            > (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE) * QUEUE_NUM + CONSOLE_INPUT_SIZE + MEM_SIZE
+    {
+        return;
+    }
+
+    let (pipe_rx, mut pipe_tx) = create_pipe().unwrap();
+    let output = unsafe {
+        File::from_raw_fd(
+            memfd_create(&std::ffi::CString::new("fuzz_console_output").unwrap()).unwrap(),
+        )
+    };
+    let endpoint = virtio_devices::Endpoint::FilePair(output, pipe_rx);
+
+    let (mut console, _) = virtio_devices::Console::new(
+        "fuzzer_console".to_owned(),
+        endpoint,
+        None,  // resize_pipe
+        false, // iommu
+        SeccompAction::Allow,
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+        None,
+    )
+    .unwrap();
+
+    let console_input_bytes = &bytes[..CONSOLE_INPUT_SIZE];
+    let queue_data = &bytes[CONSOLE_INPUT_SIZE..CONSOLE_INPUT_SIZE + QUEUE_DATA_SIZE * QUEUE_NUM];
+    let queue_bytes = &bytes[CONSOLE_INPUT_SIZE + QUEUE_DATA_SIZE * QUEUE_NUM
+        ..CONSOLE_INPUT_SIZE + (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE) * QUEUE_NUM];
+    let mem_bytes = &bytes[CONSOLE_INPUT_SIZE + (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE) * QUEUE_NUM..];
+
+    // Setup the virt queues with the input bytes
+    let mut queues = setup_virt_queues(
+        &[
+            &queue_data[..QUEUE_DATA_SIZE].try_into().unwrap(),
+            &queue_data[QUEUE_DATA_SIZE..QUEUE_DATA_SIZE * 2]
+                .try_into()
+                .unwrap(),
+        ],
+        BASE_VIRT_QUEUE_ADDR,
+    );
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[
+        (GuestAddress(0), MEM_SIZE),
+        (GuestAddress(BASE_VIRT_QUEUE_ADDR), queue_bytes.len()),
+    ])
+    .unwrap();
+    if mem
+        .write_slice(queue_bytes, GuestAddress(BASE_VIRT_QUEUE_ADDR))
+        .is_err()
+    {
+        return;
+    }
+    if mem.write_slice(mem_bytes, GuestAddress(0 as u64)).is_err() {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    let input_queue = queues.remove(0);
+    let input_evt = EventFd::new(0).unwrap();
+    let input_queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(input_evt.as_raw_fd())) };
+    let output_queue = queues.remove(0);
+    let output_evt = EventFd::new(0).unwrap();
+    let output_queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(output_evt.as_raw_fd())) };
+
+    // Kick the 'queue' events and endpoint event before activate the console device
+    input_queue_evt.write(1).unwrap();
+    output_queue_evt.write(1).unwrap();
+    pipe_tx.write_all(console_input_bytes).unwrap(); // To use fuzzed data;
+
+    console
+        .activate(
+            guest_memory,
+            Arc::new(NoopVirtioInterrupt {}),
+            vec![(0, input_queue, input_evt), (1, output_queue, output_evt)],
+        )
+        .unwrap();
+
+    // Wait for the events to finish and console device worker thread to return
+    console.wait_for_epoll_threads();
+});
+
+pub struct NoopVirtioInterrupt {}
+
+impl VirtioInterrupt for NoopVirtioInterrupt {
+    fn trigger(&self, _int_type: VirtioInterruptType) -> std::result::Result<(), std::io::Error> {
+        Ok(())
+    }
+}
+
+fn setup_virt_queues(bytes: &[&[u8; QUEUE_DATA_SIZE]], base_addr: u64) -> Vec<Queue> {
+    let mut queues = Vec::new();
+    for (i, b) in bytes.iter().enumerate() {
+        let mut q = Queue::new(QUEUE_SIZE).unwrap();
+
+        let desc_table_addr = base_addr + (QUEUE_BYTES_SIZE * i) as u64;
+        let avail_ring_addr = desc_table_addr + DESC_TABLE_SIZE;
+        let used_ring_addr = avail_ring_addr + PADDING_SIZE + AVAIL_RING_SIZE;
+        q.try_set_desc_table_address(GuestAddress(desc_table_addr))
+            .unwrap();
+        q.try_set_avail_ring_address(GuestAddress(avail_ring_addr))
+            .unwrap();
+        q.try_set_used_ring_address(GuestAddress(used_ring_addr))
+            .unwrap();
+
+        q.set_next_avail(b[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+        q.set_next_used(b[1] as u16);
+        q.set_event_idx(b[2] % 2 != 0);
+        q.set_size(b[3] as u16 % QUEUE_SIZE);
+
+        q.set_ready(true);
+        queues.push(q);
+    }
+
+    queues
+}
+
+fn memfd_create(name: &std::ffi::CStr) -> Result<RawFd, std::io::Error> {
+    let res = unsafe { libc::syscall(libc::SYS_memfd_create, name.as_ptr(), 0) };
+
+    if res < 0 {
+        Err(std::io::Error::last_os_error())
+    } else {
+        Ok(res as RawFd)
+    }
+}
+
+fn create_pipe() -> Result<(File, File), std::io::Error> {
+    let mut pipe = [-1; 2];
+    if unsafe { libc::pipe2(pipe.as_mut_ptr(), libc::O_CLOEXEC) } == -1 {
+        return Err(std::io::Error::last_os_error());
+    }
+    let rx = unsafe { File::from_raw_fd(pipe[0]) };
+    let tx = unsafe { File::from_raw_fd(pipe[1]) };
+
+    Ok((rx, tx))
+}

fuzz/fuzz_targets/iommu.rs

@@ -0,0 +1,145 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use seccompiler::SeccompAction;
+use std::os::unix::io::{AsRawFd, FromRawFd};
+use std::sync::Arc;
+use virtio_devices::{VirtioDevice, VirtioInterrupt, VirtioInterruptType};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic};
+use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+macro_rules! align {
+    ($n:expr, $align:expr) => {{
+        (($n + $align - 1) / $align) * $align
+    }};
+}
+
+const QUEUE_DATA_SIZE: usize = 4;
+const MEM_SIZE: usize = 32 * 1024 * 1024;
+// Reuse what's being done from DeviceManager::get_msi_iova_space()
+const IOVA_SPACE_SIZE: usize = (0xfeef_ffff - 0xfee0_0000) + 1;
+
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 256;
+// Descriptor table alignment
+const DESC_TABLE_ALIGN_SIZE: u64 = 16;
+// Avalable ring alignment
+const AVAIL_RING_ALIGN_SIZE: u64 = 2;
+// Used ring alignment
+const USED_RING_ALIGN_SIZE: u64 = 4;
+// Descriptor table size
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Available ring size
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Used ring size
+const USED_RING_SIZE: u64 = 6_u64 + 8 * QUEUE_SIZE as u64;
+
+// Guest memory gap
+const GUEST_MEM_GAP: u64 = 1 * 1024 * 1024;
+// Guest physical address for descriptor table.
+const DESC_TABLE_ADDR: u64 = align!(MEM_SIZE as u64 + GUEST_MEM_GAP, DESC_TABLE_ALIGN_SIZE);
+// Guest physical address for available ring
+const AVAIL_RING_ADDR: u64 = align!(DESC_TABLE_ADDR + DESC_TABLE_SIZE, AVAIL_RING_ALIGN_SIZE);
+// Guest physical address for used ring
+const USED_RING_ADDR: u64 = align!(AVAIL_RING_ADDR + AVAIL_RING_SIZE, USED_RING_ALIGN_SIZE);
+// Virtio-queue size in bytes
+const QUEUE_BYTES_SIZE: usize = (USED_RING_ADDR + USED_RING_SIZE - DESC_TABLE_ADDR) as usize;
+
+fuzz_target!(|bytes| {
+    if bytes.len() < (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE)
+        || bytes.len() > (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE + MEM_SIZE)
+    {
+        return;
+    }
+
+    let (mut iommu, _) = virtio_devices::Iommu::new(
+        "fuzzer_iommu".to_owned(),
+        SeccompAction::Allow,
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+        ((MEM_SIZE - IOVA_SPACE_SIZE) as u64, (MEM_SIZE - 1) as u64),
+        None,
+    )
+    .unwrap();
+
+    let queue_data = &bytes[..QUEUE_DATA_SIZE];
+    let queue_bytes = &bytes[QUEUE_DATA_SIZE..QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE];
+    let mem_bytes = &bytes[QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE..];
+
+    // Setup the request queue with the input bytes
+    let request_queue = setup_virt_queue(queue_data.try_into().unwrap());
+    // Given the "event queue" events are not handled from the current
+    // implementation of virtio-iommu, we simply setup the 'event_queue'
+    // with exactly the same content as the 'request_queue'.
+    let _event_queue = setup_virt_queue(queue_data.try_into().unwrap());
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[
+        (GuestAddress(0), MEM_SIZE),
+        (GuestAddress(DESC_TABLE_ADDR), QUEUE_BYTES_SIZE),
+    ])
+    .unwrap();
+    if mem
+        .write_slice(queue_bytes, GuestAddress(DESC_TABLE_ADDR))
+        .is_err()
+    {
+        return;
+    }
+    if mem.write_slice(mem_bytes, GuestAddress(0 as u64)).is_err() {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    let request_evt = EventFd::new(0).unwrap();
+    let request_queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(request_evt.as_raw_fd())) };
+    let _event_evt = EventFd::new(0).unwrap();
+
+    // Kick the 'queue' event before activate the vIOMMU device
+    request_queue_evt.write(1).unwrap();
+
+    iommu
+        .activate(
+            guest_memory,
+            Arc::new(NoopVirtioInterrupt {}),
+            vec![
+                (0, request_queue, request_evt),
+                (0, _event_queue, _event_evt),
+            ],
+        )
+        .ok();
+
+    // Wait for the events to finish and vIOMMU device worker thread to return
+    iommu.wait_for_epoll_threads();
+});
+
+pub struct NoopVirtioInterrupt {}
+
+impl VirtioInterrupt for NoopVirtioInterrupt {
+    fn trigger(&self, _int_type: VirtioInterruptType) -> std::result::Result<(), std::io::Error> {
+        Ok(())
+    }
+}
+
+fn setup_virt_queue(bytes: &[u8; QUEUE_DATA_SIZE]) -> Queue {
+    let mut q = Queue::new(QUEUE_SIZE).unwrap();
+    q.set_next_avail(bytes[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+    q.set_next_used(bytes[1] as u16);
+    q.set_event_idx(bytes[2] % 2 != 0);
+    q.set_size(bytes[3] as u16 % QUEUE_SIZE);
+
+    q.try_set_desc_table_address(GuestAddress(DESC_TABLE_ADDR))
+        .unwrap();
+    q.try_set_avail_ring_address(GuestAddress(AVAIL_RING_ADDR))
+        .unwrap();
+    q.try_set_used_ring_address(GuestAddress(USED_RING_ADDR))
+        .unwrap();
+    q.set_ready(true);
+
+    q
+}

fuzz/fuzz_targets/linux_loader.rs

@@ -0,0 +1,50 @@
+// Copyright 2018 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 LICENSE file.
+//
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0 AND BSD-3-Clause
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use linux_loader::loader::KernelLoader;
+use std::ffi;
+use std::fs::File;
+use std::io;
+use std::io::{Seek, SeekFrom, Write};
+use std::os::unix::io::{FromRawFd, RawFd};
+use vm_memory::{bitmap::AtomicBitmap, GuestAddress};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+const MEM_SIZE: usize = 256 * 1024 * 1024;
+// From 'arch::x86_64::layout::HIGH_RAM_START'
+const HIGH_RAM_START: GuestAddress = GuestAddress(0x100000);
+
+fuzz_target!(|bytes| {
+    let shm = memfd_create(&ffi::CString::new("fuzz_load_kernel").unwrap(), 0).unwrap();
+    let mut kernel_file: File = unsafe { File::from_raw_fd(shm) };
+    kernel_file.write_all(&bytes).unwrap();
+    kernel_file.seek(SeekFrom::Start(0)).unwrap();
+
+    let guest_memory = GuestMemoryMmap::from_ranges(&[(GuestAddress(0), MEM_SIZE)]).unwrap();
+    linux_loader::loader::elf::Elf::load(
+        &guest_memory,
+        None,
+        &mut kernel_file,
+        Some(HIGH_RAM_START),
+    )
+    .ok();
+});
+
+fn memfd_create(name: &ffi::CStr, flags: u32) -> Result<RawFd, io::Error> {
+    let res = unsafe { libc::syscall(libc::SYS_memfd_create, name.as_ptr(), flags) };
+
+    if res < 0 {
+        Err(io::Error::last_os_error())
+    } else {
+        Ok(res as RawFd)
+    }
+}

fuzz/fuzz_targets/linux_loader_cmdline.rs

@@ -0,0 +1,34 @@
+// Copyright 2018 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 LICENSE file.
+//
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0 AND BSD-3-Clause
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use vm_memory::{bitmap::AtomicBitmap, GuestAddress};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+const MEM_SIZE: usize = 256 * 1024 * 1024;
+// From 'arch::x86_64::layout::CMDLINE_START'
+const CMDLINE_START: GuestAddress = GuestAddress(0x20000);
+
+fuzz_target!(|bytes| {
+    let payload_config = vmm::config::PayloadConfig {
+        firmware: None,
+        kernel: None,
+        cmdline: Some(String::from_utf8_lossy(&bytes).to_string()),
+        initramfs: None,
+    };
+    let kernel_cmdline = match vmm::vm::Vm::generate_cmdline(&payload_config) {
+        Ok(cmdline) => cmdline,
+        _ => return,
+    };
+    let guest_memory = GuestMemoryMmap::from_ranges(&[(GuestAddress(0), MEM_SIZE)]).unwrap();
+
+    linux_loader::loader::load_cmdline(&guest_memory, CMDLINE_START, &kernel_cmdline).ok();
+});

fuzz/fuzz_targets/mem.rs

@@ -0,0 +1,179 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use seccompiler::SeccompAction;
+use std::os::unix::io::{AsRawFd, FromRawFd};
+use std::sync::{Arc, Mutex};
+use virtio_devices::{BlocksState, Mem, VirtioDevice, VirtioInterrupt, VirtioInterruptType};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic};
+use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+type GuestRegionMmap = vm_memory::GuestRegionMmap<AtomicBitmap>;
+
+macro_rules! align {
+    ($n:expr, $align:expr) => {{
+        (($n + $align - 1) / $align) * $align
+    }};
+}
+
+const VIRTIO_MEM_DATA_SIZE: usize = 1;
+const QUEUE_DATA_SIZE: usize = 4;
+// The size of the guest memory for the virtio-mem region
+const MEM_SIZE: usize = 128 * 1024 * 1024;
+// The start address of the virtio-mem region in the guest memory
+const VIRTIO_MEM_REGION_ADDRESS: u64 = 0;
+
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 64;
+// Descriptor table alignment
+const DESC_TABLE_ALIGN_SIZE: u64 = 16;
+// Avalable ring alignment
+const AVAIL_RING_ALIGN_SIZE: u64 = 2;
+// Used ring alignment
+const USED_RING_ALIGN_SIZE: u64 = 4;
+// Descriptor table size
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Available ring size
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Used ring size
+const USED_RING_SIZE: u64 = 6_u64 + 8 * QUEUE_SIZE as u64;
+
+// Guest memory gap
+const GUEST_MEM_GAP: u64 = 1 * 1024 * 1024;
+// Guest physical address for descriptor table.
+const DESC_TABLE_ADDR: u64 = align!(MEM_SIZE as u64 + GUEST_MEM_GAP, DESC_TABLE_ALIGN_SIZE);
+// Guest physical address for available ring
+const AVAIL_RING_ADDR: u64 = align!(DESC_TABLE_ADDR + DESC_TABLE_SIZE, AVAIL_RING_ALIGN_SIZE);
+// Guest physical address for used ring
+const USED_RING_ADDR: u64 = align!(AVAIL_RING_ADDR + AVAIL_RING_SIZE, USED_RING_ALIGN_SIZE);
+// Virtio-queue size in bytes
+const QUEUE_BYTES_SIZE: usize = (USED_RING_ADDR + USED_RING_SIZE - DESC_TABLE_ADDR) as usize;
+
+fuzz_target!(|bytes| {
+    if bytes.len() < VIRTIO_MEM_DATA_SIZE + QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE
+        || bytes.len() > (VIRTIO_MEM_DATA_SIZE + QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE + MEM_SIZE)
+    {
+        return;
+    }
+
+    let virtio_mem_data = &bytes[..VIRTIO_MEM_DATA_SIZE];
+    let queue_data = &bytes[VIRTIO_MEM_DATA_SIZE..VIRTIO_MEM_DATA_SIZE + QUEUE_DATA_SIZE];
+    let queue_bytes = &bytes[VIRTIO_MEM_DATA_SIZE + QUEUE_DATA_SIZE
+        ..VIRTIO_MEM_DATA_SIZE + QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE];
+    let mem_bytes = &bytes[VIRTIO_MEM_DATA_SIZE + QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE..];
+
+    // Create a virtio-mem device based on the input bytes;
+    let (mut virtio_mem, virtio_mem_region) =
+        create_dummy_virtio_mem(virtio_mem_data.try_into().unwrap());
+
+    // Setup the virt queue with the input bytes
+    let q = setup_virt_queue(queue_data.try_into().unwrap());
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[
+        (GuestAddress(DESC_TABLE_ADDR), QUEUE_BYTES_SIZE), // guest region for the virt queue
+    ])
+    .unwrap();
+    if mem
+        .write_slice(queue_bytes, GuestAddress(DESC_TABLE_ADDR))
+        .is_err()
+    {
+        return;
+    }
+    // Add the memory region for the virtio-mem device
+    let mem = mem.insert_region(virtio_mem_region).unwrap();
+    if mem
+        .write_slice(mem_bytes, GuestAddress(VIRTIO_MEM_REGION_ADDRESS))
+        .is_err()
+    {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    let evt = EventFd::new(0).unwrap();
+    let queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(evt.as_raw_fd())) };
+
+    // Kick the 'queue' event before activate the virtio-mem device
+    queue_evt.write(1).unwrap();
+
+    virtio_mem
+        .activate(
+            guest_memory,
+            Arc::new(NoopVirtioInterrupt {}),
+            vec![(0, q, evt)],
+        )
+        .ok();
+
+    // Wait for the events to finish and virtio-mem device worker thread to return
+    virtio_mem.wait_for_epoll_threads();
+});
+
+pub struct NoopVirtioInterrupt {}
+
+impl VirtioInterrupt for NoopVirtioInterrupt {
+    fn trigger(&self, _int_type: VirtioInterruptType) -> std::result::Result<(), std::io::Error> {
+        Ok(())
+    }
+}
+
+// Create a dummy virtio-mem device for fuzzing purpose only
+fn create_dummy_virtio_mem(bytes: &[u8; VIRTIO_MEM_DATA_SIZE]) -> (Mem, Arc<GuestRegionMmap>) {
+    let numa_id = if bytes[0] % 2 != 0 { Some(0) } else { None };
+
+    let region = vmm::memory_manager::MemoryManager::create_ram_region(
+        &None,
+        0,
+        GuestAddress(VIRTIO_MEM_REGION_ADDRESS),
+        MEM_SIZE,
+        false,
+        false,
+        false,
+        None,
+        numa_id,
+        None,
+        false,
+    )
+    .unwrap();
+
+    let blocks_state = Arc::new(Mutex::new(BlocksState::new(region.size() as u64)));
+
+    (
+        Mem::new(
+            "fuzzer_mem".to_owned(),
+            &region,
+            SeccompAction::Allow,
+            numa_id.map(|i| i as u16),
+            0,
+            false,
+            EventFd::new(EFD_NONBLOCK).unwrap(),
+            blocks_state.clone(),
+            None,
+        )
+        .unwrap(),
+        region,
+    )
+}
+
+fn setup_virt_queue(bytes: &[u8; QUEUE_DATA_SIZE]) -> Queue {
+    let mut q = Queue::new(QUEUE_SIZE).unwrap();
+    q.set_next_avail(bytes[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+    q.set_next_used(bytes[1] as u16);
+    q.set_event_idx(bytes[2] % 2 != 0);
+    q.set_size(bytes[3] as u16 % QUEUE_SIZE);
+
+    q.try_set_desc_table_address(GuestAddress(DESC_TABLE_ADDR))
+        .unwrap();
+    q.try_set_avail_ring_address(GuestAddress(AVAIL_RING_ADDR))
+        .unwrap();
+    q.try_set_used_ring_address(GuestAddress(USED_RING_ADDR))
+        .unwrap();
+    q.set_ready(true);
+
+    q
+}

fuzz/fuzz_targets/net.rs

@@ -0,0 +1,285 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use seccompiler::SeccompAction;
+use std::fs::File;
+use std::io::{Read, Write};
+use std::os::unix::io::{AsRawFd, FromRawFd};
+use std::sync::Arc;
+use virtio_devices::{VirtioDevice, VirtioInterrupt, VirtioInterruptType};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic};
+use vmm::EpollContext;
+use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+macro_rules! align {
+    ($n:expr, $align:expr) => {{
+        (($n + $align - 1) / $align) * $align
+    }};
+}
+
+const TAP_INPUT_SIZE: usize = 128;
+const QUEUE_DATA_SIZE: usize = 4;
+const MEM_SIZE: usize = 32 * 1024 * 1024;
+// Guest memory gap
+const GUEST_MEM_GAP: u64 = 1 * 1024 * 1024;
+// Guest physical address for the first virt queue
+const BASE_VIRT_QUEUE_ADDR: u64 = MEM_SIZE as u64 + GUEST_MEM_GAP;
+// Number of queues
+const QUEUE_NUM: usize = 2;
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 256;
+// Descriptor table alignment
+const DESC_TABLE_ALIGN_SIZE: u64 = 16;
+// Used ring alignment
+const USED_RING_ALIGN_SIZE: u64 = 4;
+// Descriptor table size
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Available ring size
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Padding size before used ring
+const PADDING_SIZE: u64 = align!(AVAIL_RING_SIZE, USED_RING_ALIGN_SIZE) - AVAIL_RING_SIZE;
+// Used ring size
+const USED_RING_SIZE: u64 = 6_u64 + 8 * QUEUE_SIZE as u64;
+// Virtio-queue size in bytes
+const QUEUE_BYTES_SIZE: usize = align!(
+    DESC_TABLE_SIZE + AVAIL_RING_SIZE + PADDING_SIZE + USED_RING_SIZE,
+    DESC_TABLE_ALIGN_SIZE
+) as usize;
+
+fuzz_target!(|bytes| {
+    if bytes.len() < TAP_INPUT_SIZE + (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE) * QUEUE_NUM
+        || bytes.len()
+            > TAP_INPUT_SIZE + (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE) * QUEUE_NUM + MEM_SIZE
+    {
+        return;
+    }
+
+    let (dummy_tap_frontend, dummy_tap_backend) = create_socketpair().unwrap();
+    let if_name = "fuzzer_tap_name".as_bytes().to_vec();
+    let tap = net_util::Tap::new_for_fuzzing(dummy_tap_frontend, if_name);
+
+    let mut net = virtio_devices::Net::new_with_tap(
+        "fuzzer_net".to_owned(),
+        vec![tap],
+        None,  // guest_mac
+        false, // iommu
+        QUEUE_NUM,
+        QUEUE_SIZE,
+        SeccompAction::Allow,
+        None,
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+        None,
+        true,
+        true,
+        true,
+    )
+    .unwrap();
+
+    let tap_input_bytes = &bytes[..TAP_INPUT_SIZE];
+    let queue_data = &bytes[TAP_INPUT_SIZE..TAP_INPUT_SIZE + QUEUE_DATA_SIZE * QUEUE_NUM];
+    let queue_bytes = &bytes[TAP_INPUT_SIZE + QUEUE_DATA_SIZE * QUEUE_NUM
+        ..TAP_INPUT_SIZE + (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE) * QUEUE_NUM];
+    let mem_bytes = &bytes[TAP_INPUT_SIZE + (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE) * QUEUE_NUM..];
+
+    // Setup the virt queues with the input bytes
+    let mut queues = setup_virt_queues(
+        &[
+            &queue_data[..QUEUE_DATA_SIZE].try_into().unwrap(),
+            &queue_data[QUEUE_DATA_SIZE..QUEUE_DATA_SIZE * 2]
+                .try_into()
+                .unwrap(),
+        ],
+        BASE_VIRT_QUEUE_ADDR,
+    );
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[
+        (GuestAddress(0), MEM_SIZE),
+        (GuestAddress(BASE_VIRT_QUEUE_ADDR), queue_bytes.len()),
+    ])
+    .unwrap();
+    if mem
+        .write_slice(queue_bytes, GuestAddress(BASE_VIRT_QUEUE_ADDR))
+        .is_err()
+    {
+        return;
+    }
+    if mem.write_slice(mem_bytes, GuestAddress(0 as u64)).is_err() {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    let input_queue = queues.remove(0);
+    let input_evt = EventFd::new(0).unwrap();
+    let input_queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(input_evt.as_raw_fd())) };
+    let output_queue = queues.remove(0);
+    let output_evt = EventFd::new(0).unwrap();
+    let output_queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(output_evt.as_raw_fd())) };
+
+    // Start the thread of dummy tap backend to handle the rx and tx from the virtio-net
+    let exit_evt = EventFd::new(libc::EFD_NONBLOCK).unwrap();
+    let tap_backend_thread = {
+        let dummy_tap_backend = dummy_tap_backend.try_clone().unwrap();
+        let tap_input_bytes: [u8; TAP_INPUT_SIZE] = tap_input_bytes[..].try_into().unwrap();
+        let exit_evt = exit_evt.try_clone().unwrap();
+        std::thread::Builder::new()
+            .name("dummy_tap_backend".to_string())
+            .spawn(move || {
+                tap_backend_stub(dummy_tap_backend, &tap_input_bytes, exit_evt);
+            })
+            .unwrap()
+    };
+
+    // Kick the 'queue' events and endpoint event before activate the net device
+    input_queue_evt.write(1).unwrap();
+    output_queue_evt.write(1).unwrap();
+
+    net.activate(
+        guest_memory,
+        Arc::new(NoopVirtioInterrupt {}),
+        vec![(0, input_queue, input_evt), (1, output_queue, output_evt)],
+    )
+    .unwrap();
+
+    // Wait for the events to finish and net device worker thread to return
+    net.wait_for_epoll_threads();
+    // Terminate the thread for the dummy tap backend
+    exit_evt.write(1).ok();
+    tap_backend_thread.join().unwrap();
+});
+
+pub struct NoopVirtioInterrupt {}
+
+impl VirtioInterrupt for NoopVirtioInterrupt {
+    fn trigger(&self, _int_type: VirtioInterruptType) -> std::result::Result<(), std::io::Error> {
+        Ok(())
+    }
+}
+
+fn setup_virt_queues(bytes: &[&[u8; QUEUE_DATA_SIZE]], base_addr: u64) -> Vec<Queue> {
+    let mut queues = Vec::new();
+    for (i, b) in bytes.iter().enumerate() {
+        let mut q = Queue::new(QUEUE_SIZE).unwrap();
+
+        let desc_table_addr = base_addr + (QUEUE_BYTES_SIZE * i) as u64;
+        let avail_ring_addr = desc_table_addr + DESC_TABLE_SIZE;
+        let used_ring_addr = avail_ring_addr + PADDING_SIZE + AVAIL_RING_SIZE;
+        q.try_set_desc_table_address(GuestAddress(desc_table_addr))
+            .unwrap();
+        q.try_set_avail_ring_address(GuestAddress(avail_ring_addr))
+            .unwrap();
+        q.try_set_used_ring_address(GuestAddress(used_ring_addr))
+            .unwrap();
+
+        q.set_next_avail(b[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+        q.set_next_used(b[1] as u16);
+        q.set_event_idx(b[2] % 2 != 0);
+        q.set_size(b[3] as u16 % QUEUE_SIZE);
+
+        q.set_ready(true);
+        queues.push(q);
+    }
+
+    queues
+}
+
+fn create_socketpair() -> Result<(File, File), std::io::Error> {
+    let mut fds = [-1, -1];
+    unsafe {
+        let ret = libc::socketpair(
+            libc::AF_UNIX,
+            libc::SOCK_STREAM | libc::SOCK_NONBLOCK,
+            0,
+            fds.as_mut_ptr(),
+        );
+        if ret == -1 {
+            return Err(std::io::Error::last_os_error());
+        }
+    }
+
+    let socket1 = unsafe { File::from_raw_fd(fds[0]) };
+    let socket2 = unsafe { File::from_raw_fd(fds[1]) };
+    Ok((socket1, socket2))
+}
+
+enum EpollEvent {
+    Exit = 0,
+    Rx = 1,
+    Tx = 2,
+    Unknown,
+}
+
+impl From<u64> for EpollEvent {
+    fn from(v: u64) -> Self {
+        use EpollEvent::*;
+        match v {
+            0 => Exit,
+            1 => Rx,
+            2 => Tx,
+            _ => Unknown,
+        }
+    }
+}
+
+// Handle the rx and tx requests from the virtio-net device
+fn tap_backend_stub(
+    mut dummy_tap: File,
+    tap_input_bytes: &[u8; TAP_INPUT_SIZE],
+    exit_evt: EventFd,
+) {
+    let mut epoll = EpollContext::new().unwrap();
+    epoll
+        .add_event_custom(&exit_evt, EpollEvent::Exit as u64, epoll::Events::EPOLLIN)
+        .unwrap();
+    let dummy_tap_write = dummy_tap.try_clone().unwrap();
+    epoll
+        .add_event_custom(
+            &dummy_tap_write,
+            EpollEvent::Rx as u64,
+            epoll::Events::EPOLLOUT,
+        )
+        .unwrap();
+    epoll
+        .add_event_custom(&dummy_tap, EpollEvent::Tx as u64, epoll::Events::EPOLLIN)
+        .unwrap();
+
+    let epoll_fd = epoll.as_raw_fd();
+    let mut events = vec![epoll::Event::new(epoll::Events::empty(), 0); 3];
+    loop {
+        let num_events = match epoll::wait(epoll_fd, -1, &mut events[..]) {
+            Ok(num_events) => num_events,
+            Err(e) => match e.raw_os_error() {
+                Some(libc::EAGAIN) | Some(libc::EINTR) => continue,
+                _ => panic!("Unexpected epoll::wait error!"),
+            },
+        };
+
+        for event in events.iter().take(num_events) {
+            let dispatch_event: EpollEvent = event.data.into();
+            match dispatch_event {
+                EpollEvent::Exit => {
+                    return;
+                }
+                EpollEvent::Rx => {
+                    dummy_tap.write_all(tap_input_bytes).unwrap();
+                    break;
+                }
+                EpollEvent::Tx => {
+                    let mut buffer = Vec::new();
+                    dummy_tap.read_to_end(&mut buffer).ok();
+                    break;
+                }
+                _ => {
+                    panic!("Unexpected Epoll event");
+                }
+            }
+        }
+    }
+}

fuzz/fuzz_targets/pmem.rs

@@ -0,0 +1,151 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+
+use libc::{MAP_NORESERVE, MAP_PRIVATE, PROT_READ, PROT_WRITE};
+use libfuzzer_sys::fuzz_target;
+use seccompiler::SeccompAction;
+use std::ffi;
+use std::fs::File;
+use std::io;
+use std::os::unix::io::{AsRawFd, FromRawFd, RawFd};
+use std::sync::Arc;
+use virtio_devices::{Pmem, UserspaceMapping, VirtioDevice, VirtioInterrupt, VirtioInterruptType};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::guest_memory::FileOffset;
+use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic, MmapRegion};
+use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+const QUEUE_DATA_SIZE: usize = 4;
+const MEM_SIZE: usize = 256 * 1024 * 1024;
+const PMEM_FILE_SIZE: usize = 128 * 1024 * 1024;
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 256;
+// Guest physical address for descriptor table.
+const DESC_TABLE_ADDR: u64 = 0;
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Guest physical address for available ring
+const AVAIL_RING_ADDR: u64 = DESC_TABLE_ADDR + DESC_TABLE_SIZE;
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Guest physical address for used ring (requires to 4-bytes aligned)
+const USED_RING_ADDR: u64 = (AVAIL_RING_ADDR + AVAIL_RING_SIZE + 3) & !3_u64;
+
+fuzz_target!(|bytes| {
+    if bytes.len() < QUEUE_DATA_SIZE || bytes.len() > (QUEUE_DATA_SIZE + MEM_SIZE) {
+        return;
+    }
+
+    let mut pmem = create_dummy_pmem();
+    let queue_data = &bytes[..QUEUE_DATA_SIZE];
+    let mem_bytes = &bytes[QUEUE_DATA_SIZE..];
+
+    // Setup the virt queue with the input bytes
+    let q = setup_virt_queue(queue_data.try_into().unwrap());
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[(GuestAddress(0), MEM_SIZE)]).unwrap();
+    if mem.write_slice(mem_bytes, GuestAddress(0 as u64)).is_err() {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    let evt = EventFd::new(0).unwrap();
+    let queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(evt.as_raw_fd())) };
+
+    // Kick the 'queue' event before activate the pmem device
+    queue_evt.write(1).unwrap();
+
+    pmem.activate(
+        guest_memory,
+        Arc::new(NoopVirtioInterrupt {}),
+        vec![(0, q, evt)],
+    )
+    .ok();
+
+    // Wait for the events to finish and pmem device worker thread to return
+    pmem.wait_for_epoll_threads();
+});
+
+fn memfd_create_with_size(name: &ffi::CStr, flags: u32, size: usize) -> Result<RawFd, io::Error> {
+    let fd = unsafe { libc::syscall(libc::SYS_memfd_create, name.as_ptr(), flags) };
+    if fd < 0 {
+        return Err(io::Error::last_os_error());
+    }
+
+    let res = unsafe { libc::syscall(libc::SYS_ftruncate, fd, size) };
+    if res < 0 {
+        return Err(io::Error::last_os_error());
+    }
+
+    Ok(fd as RawFd)
+}
+
+pub struct NoopVirtioInterrupt {}
+
+impl VirtioInterrupt for NoopVirtioInterrupt {
+    fn trigger(&self, _int_type: VirtioInterruptType) -> std::result::Result<(), std::io::Error> {
+        Ok(())
+    }
+}
+
+// Create a dummy virtio-pmem device for fuzzing purpose only
+fn create_dummy_pmem() -> Pmem {
+    let shm =
+        memfd_create_with_size(&ffi::CString::new("fuzz").unwrap(), 0, PMEM_FILE_SIZE).unwrap();
+    let file: File = unsafe { File::from_raw_fd(shm) };
+
+    // The fuzzer is focusing on the virtio-pmem code that processes guest inputs (e.g. virt queues),
+    // so dummy mappings (both mmap and user mapping) does the job and is faster (using smaller amount of memory).
+    let dummy_mapping_size = 1;
+    let cloned_file = file.try_clone().unwrap();
+    let dummy_mmap_region = MmapRegion::build(
+        Some(FileOffset::new(cloned_file, 0)),
+        dummy_mapping_size,
+        PROT_READ | PROT_WRITE,
+        MAP_NORESERVE | MAP_PRIVATE,
+    )
+    .unwrap();
+    let guest_addr = GuestAddress(0);
+    let dummy_user_mapping = UserspaceMapping {
+        host_addr: dummy_mmap_region.as_ptr() as u64,
+        mem_slot: 0,
+        addr: guest_addr,
+        len: dummy_mapping_size as u64,
+        mergeable: false,
+    };
+
+    Pmem::new(
+        "tmp".to_owned(),
+        file,
+        guest_addr,
+        dummy_user_mapping,
+        dummy_mmap_region,
+        false,
+        SeccompAction::Allow,
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+        None,
+    )
+    .unwrap()
+}
+
+fn setup_virt_queue(bytes: &[u8; QUEUE_DATA_SIZE]) -> Queue {
+    let mut q = Queue::new(QUEUE_SIZE).unwrap();
+    q.set_next_avail(bytes[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+    q.set_next_used(bytes[1] as u16);
+    q.set_event_idx(bytes[2] % 2 != 0);
+    q.set_size(bytes[3] as u16 % QUEUE_SIZE);
+
+    q.try_set_desc_table_address(GuestAddress(DESC_TABLE_ADDR))
+        .unwrap();
+    q.try_set_avail_ring_address(GuestAddress(AVAIL_RING_ADDR))
+        .unwrap();
+    q.try_set_used_ring_address(GuestAddress(USED_RING_ADDR))
+        .unwrap();
+    q.set_ready(true);
+
+    q
+}

fuzz/fuzz_targets/rng.rs

@@ -0,0 +1,135 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use seccompiler::SeccompAction;
+use std::os::unix::io::{AsRawFd, FromRawFd};
+use std::sync::Arc;
+use virtio_devices::{VirtioDevice, VirtioInterrupt, VirtioInterruptType};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic};
+use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+macro_rules! align {
+    ($n:expr, $align:expr) => {{
+        (($n + $align - 1) / $align) * $align
+    }};
+}
+
+const QUEUE_DATA_SIZE: usize = 4;
+const MEM_SIZE: usize = 1 * 1024 * 1024;
+
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 256;
+// Descriptor table alignment
+const DESC_TABLE_ALIGN_SIZE: u64 = 16;
+// Avalable ring alignment
+const AVAIL_RING_ALIGN_SIZE: u64 = 2;
+// Used ring alignment
+const USED_RING_ALIGN_SIZE: u64 = 4;
+// Descriptor table size
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Available ring size
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Used ring size
+const USED_RING_SIZE: u64 = 6_u64 + 8 * QUEUE_SIZE as u64;
+
+// Guest memory gap
+const GUEST_MEM_GAP: u64 = 1 * 1024 * 1024;
+// Guest physical address for descriptor table.
+const DESC_TABLE_ADDR: u64 = align!(MEM_SIZE as u64 + GUEST_MEM_GAP, DESC_TABLE_ALIGN_SIZE);
+// Guest physical address for available ring
+const AVAIL_RING_ADDR: u64 = align!(DESC_TABLE_ADDR + DESC_TABLE_SIZE, AVAIL_RING_ALIGN_SIZE);
+// Guest physical address for used ring
+const USED_RING_ADDR: u64 = align!(AVAIL_RING_ADDR + AVAIL_RING_SIZE, USED_RING_ALIGN_SIZE);
+// Virtio-queue size in bytes
+const QUEUE_BYTES_SIZE: usize = (USED_RING_ADDR + USED_RING_SIZE - DESC_TABLE_ADDR) as usize;
+
+fuzz_target!(|bytes| {
+    if bytes.len() < (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE)
+        || bytes.len() > (QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE + MEM_SIZE)
+    {
+        return;
+    }
+
+    let mut rng = virtio_devices::Rng::new(
+        "fuzzer_rng".to_owned(),
+        "/dev/urandom",
+        false,
+        SeccompAction::Allow,
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+        None,
+    )
+    .unwrap();
+
+    let queue_data = &bytes[..QUEUE_DATA_SIZE];
+    let queue_bytes = &bytes[QUEUE_DATA_SIZE..QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE];
+    let mem_bytes = &bytes[QUEUE_DATA_SIZE + QUEUE_BYTES_SIZE..];
+
+    // Setup the virt queue with the input bytes
+    let q = setup_virt_queue(queue_data.try_into().unwrap());
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[
+        (GuestAddress(0), MEM_SIZE),
+        (GuestAddress(DESC_TABLE_ADDR), QUEUE_BYTES_SIZE),
+    ])
+    .unwrap();
+    if mem
+        .write_slice(queue_bytes, GuestAddress(DESC_TABLE_ADDR))
+        .is_err()
+    {
+        return;
+    }
+    if mem.write_slice(mem_bytes, GuestAddress(0 as u64)).is_err() {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    let evt = EventFd::new(0).unwrap();
+    let queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(evt.as_raw_fd())) };
+
+    // Kick the 'queue' event before activate the rng device
+    queue_evt.write(1).unwrap();
+
+    rng.activate(
+        guest_memory,
+        Arc::new(NoopVirtioInterrupt {}),
+        vec![(0, q, evt)],
+    )
+    .ok();
+
+    // Wait for the events to finish and rng device worker thread to return
+    rng.wait_for_epoll_threads();
+});
+
+pub struct NoopVirtioInterrupt {}
+
+impl VirtioInterrupt for NoopVirtioInterrupt {
+    fn trigger(&self, _int_type: VirtioInterruptType) -> std::result::Result<(), std::io::Error> {
+        Ok(())
+    }
+}
+
+fn setup_virt_queue(bytes: &[u8; QUEUE_DATA_SIZE]) -> Queue {
+    let mut q = Queue::new(QUEUE_SIZE).unwrap();
+    q.set_next_avail(bytes[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+    q.set_next_used(bytes[1] as u16);
+    q.set_event_idx(bytes[2] % 2 != 0);
+    q.set_size(bytes[3] as u16 % QUEUE_SIZE);
+
+    q.try_set_desc_table_address(GuestAddress(DESC_TABLE_ADDR))
+        .unwrap();
+    q.try_set_avail_ring_address(GuestAddress(AVAIL_RING_ADDR))
+        .unwrap();
+    q.try_set_used_ring_address(GuestAddress(USED_RING_ADDR))
+        .unwrap();
+    q.set_ready(true);
+
+    q
+}

fuzz/fuzz_targets/serial.rs

@@ -15,6 +15,7 @@ fuzz_target!(|bytes| {
     let mut serial = Serial::new_sink(
         "serial".into(),
         Arc::new(TestInterrupt::new(EventFd::new(EFD_NONBLOCK).unwrap())),
+        None,
     );
 
     let mut i = 0;

fuzz/fuzz_targets/watchdog.rs

@@ -0,0 +1,100 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use seccompiler::SeccompAction;
+use std::os::unix::io::{AsRawFd, FromRawFd};
+use std::sync::Arc;
+use virtio_devices::{VirtioDevice, VirtioInterrupt, VirtioInterruptType};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::{bitmap::AtomicBitmap, Bytes, GuestAddress, GuestMemoryAtomic};
+use vmm_sys_util::eventfd::{EventFd, EFD_NONBLOCK};
+
+type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
+
+const QUEUE_DATA_SIZE: usize = 4;
+const MEM_SIZE: usize = 256 * 1024 * 1024;
+// Max entries in the queue.
+const QUEUE_SIZE: u16 = 256;
+// Guest physical address for descriptor table.
+const DESC_TABLE_ADDR: u64 = 0;
+const DESC_TABLE_SIZE: u64 = 16_u64 * QUEUE_SIZE as u64;
+// Guest physical address for available ring
+const AVAIL_RING_ADDR: u64 = DESC_TABLE_ADDR + DESC_TABLE_SIZE;
+const AVAIL_RING_SIZE: u64 = 6_u64 + 2 * QUEUE_SIZE as u64;
+// Guest physical address for used ring (requires to 4-bytes aligned)
+const USED_RING_ADDR: u64 = (AVAIL_RING_ADDR + AVAIL_RING_SIZE + 3) & !3_u64;
+
+fuzz_target!(|bytes| {
+    if bytes.len() < QUEUE_DATA_SIZE || bytes.len() > (QUEUE_DATA_SIZE + MEM_SIZE) {
+        return;
+    }
+
+    let mut watchdog = virtio_devices::Watchdog::new(
+        "fuzzer_watchdog".to_owned(),
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+        SeccompAction::Allow,
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+        None,
+    )
+    .unwrap();
+
+    let queue_data = &bytes[..QUEUE_DATA_SIZE];
+    let mem_bytes = &bytes[QUEUE_DATA_SIZE..];
+
+    // Setup the virt queue with the input bytes
+    let q = setup_virt_queue(queue_data.try_into().unwrap());
+
+    // Setup the guest memory with the input bytes
+    let mem = GuestMemoryMmap::from_ranges(&[(GuestAddress(0), MEM_SIZE)]).unwrap();
+    if mem.write_slice(mem_bytes, GuestAddress(0 as u64)).is_err() {
+        return;
+    }
+    let guest_memory = GuestMemoryAtomic::new(mem);
+
+    let evt = EventFd::new(0).unwrap();
+    let queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(evt.as_raw_fd())) };
+
+    // Kick the 'queue' event before activate the watchdog device
+    queue_evt.write(1).unwrap();
+
+    watchdog
+        .activate(
+            guest_memory,
+            Arc::new(NoopVirtioInterrupt {}),
+            vec![(0, q, evt)],
+        )
+        .ok();
+
+    // Wait for the events to finish and watchdog device worker thread to return
+    watchdog.wait_for_epoll_threads();
+});
+
+pub struct NoopVirtioInterrupt {}
+
+impl VirtioInterrupt for NoopVirtioInterrupt {
+    fn trigger(&self, _int_type: VirtioInterruptType) -> std::result::Result<(), std::io::Error> {
+        Ok(())
+    }
+}
+
+fn setup_virt_queue(bytes: &[u8; QUEUE_DATA_SIZE]) -> Queue {
+    let mut q = Queue::new(QUEUE_SIZE).unwrap();
+    q.set_next_avail(bytes[0] as u16); // 'u8' is enough given the 'QUEUE_SIZE' is small
+    q.set_next_used(bytes[1] as u16);
+    q.set_event_idx(bytes[2] % 2 != 0);
+    q.set_size(bytes[3] as u16 % QUEUE_SIZE);
+
+    q.try_set_desc_table_address(GuestAddress(DESC_TABLE_ADDR))
+        .unwrap();
+    q.try_set_avail_ring_address(GuestAddress(AVAIL_RING_ADDR))
+        .unwrap();
+    q.try_set_used_ring_address(GuestAddress(USED_RING_ADDR))
+        .unwrap();
+    q.set_ready(true);
+
+    q
+}

hypervisor/Cargo.toml

@@ -11,25 +11,25 @@ mshv = ["mshv-ioctls", "mshv-bindings"]
 tdx = []
 
 [dependencies]
-anyhow = "1.0.62"
+anyhow = "1.0.68"
 byteorder = "1.4.3"
-thiserror = "1.0.32"
-libc = "0.2.131"
+thiserror = "1.0.38"
+libc = "0.2.139"
 log = "0.4.17"
-kvm-ioctls = { version = "0.11.0", optional = true }
-kvm-bindings = { git = "https://github.com/cloud-hypervisor/kvm-bindings", branch = "ch-v0.5.0-tdx", features = ["with-serde", "fam-wrappers"], optional  = true }
+kvm-ioctls = { version = "0.12.0", optional = true }
+kvm-bindings = { git = "https://github.com/cloud-hypervisor/kvm-bindings", branch = "ch-v0.6.0-tdx", features = ["with-serde", "fam-wrappers"], optional  = true }
 mshv-bindings = { git = "https://github.com/rust-vmm/mshv", branch = "main", features = ["with-serde", "fam-wrappers"], optional  = true }
 mshv-ioctls = { git = "https://github.com/rust-vmm/mshv", branch = "main", optional  = true}
-serde = { version = "1.0.143", features = ["rc", "derive"] }
-serde_with = { version = "2.0.0", default-features = false, features = ["macros"] }
+serde = { version = "1.0.151", features = ["rc", "derive"] }
+serde_with = { version = "2.1.0", default-features = false, features = ["macros"] }
 vfio-ioctls = { git = "https://github.com/rust-vmm/vfio", branch = "main", default-features = false }
-vm-memory = { version = "0.8.0", features = ["backend-mmap", "backend-atomic"] }
-vmm-sys-util = { version = "0.10.0", features = ["with-serde"] }
+vm-memory = { version = "0.10.0", features = ["backend-mmap", "backend-atomic"] }
+vmm-sys-util = { version = "0.11.0", features = ["with-serde"] }
 
 [target.'cfg(target_arch = "x86_64")'.dependencies.iced-x86]
-version = "1.17.0"
+version = "1.18.0"
 default-features = false
 features = ["std", "decoder", "op_code_info", "instr_info", "fast_fmt"]
 
 [dev-dependencies]
-env_logger = "0.9.0"
+env_logger = "0.10.0"

hypervisor/src/arch/aarch64/gic.rs

@@ -16,6 +16,18 @@ pub enum Error {
 }
 pub type Result<T> = result::Result<T, Error>;
 
+#[derive(Debug)]
+pub struct VgicConfig {
+    pub vcpu_count: u64,
+    pub dist_addr: u64,
+    pub dist_size: u64,
+    pub redists_addr: u64,
+    pub redists_size: u64,
+    pub msi_addr: u64,
+    pub msi_size: u64,
+    pub nr_irqs: u32,
+}
+
 /// Hypervisor agnostic interface for a virtualized GIC
 pub trait Vgic: Send + Sync {
     /// Returns the fdt compatibility property of the device

hypervisor/src/arch/emulator/mod.rs

@@ -24,10 +24,10 @@ impl<T: Debug> Display for Exception<T> {
             self.vector,
             self.ip,
             self.error
-                .map(|e| format!(": error {:x}", e))
+                .map(|e| format!(": error {e:x}"))
                 .unwrap_or_else(|| "".to_owned()),
             self.payload
-                .map(|payload| format!(": payload {:x}", payload))
+                .map(|payload| format!(": payload {payload:x}"))
                 .unwrap_or_else(|| "".to_owned())
         )
     }

hypervisor/src/arch/x86/emulator/instructions/cmp.rs

@@ -38,11 +38,7 @@ fn calc_rflags_cpazso(op0: u64, op1: u64, op_size: usize) -> u64 {
     // AF cares about the lowest 4 bits (nibble). msb_shift is 3 in this case.
     let af = ((cout >> 3) & 0x1) << AF_SHIFT;
 
-    let zf = if result & (!0u64 >> (63 - msb_shift)) == 0 {
-        1
-    } else {
-        0
-    } << ZF_SHIFT;
+    let zf = u64::from(result & (!0u64 >> (63 - msb_shift)) == 0) << ZF_SHIFT;
 
     let sf = ((result >> msb_shift) & 0x1) << SF_SHIFT;
 

hypervisor/src/arch/x86/emulator/instructions/mod.rs

@@ -56,7 +56,7 @@ fn get_op<T: CpuStateManager>(
         OpKind::Immediate16 => insn.immediate16() as u64,
         OpKind::Immediate32 => insn.immediate32() as u64,
         OpKind::Immediate32to64 => insn.immediate32to64() as u64,
-        OpKind::Immediate64 => insn.immediate64() as u64,
+        OpKind::Immediate64 => insn.immediate64(),
         k => return Err(PlatformError::InvalidOperand(anyhow!("{:?}", k))),
     };
 

hypervisor/src/arch/x86/mod.rs

@@ -272,6 +272,7 @@ impl LapicState {
         use std::io::Cursor;
         use std::mem;
 
+        // SAFETY: plain old data type
         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).
@@ -290,6 +291,7 @@ impl LapicState {
         use std::io::Cursor;
         use std::mem;
 
+        // SAFETY: plain old data type
         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).

hypervisor/src/cpu.rs

@@ -19,7 +19,6 @@ use crate::kvm::{TdxExitDetails, TdxExitStatus};
 use crate::CpuState;
 use crate::MpState;
 use thiserror::Error;
-#[cfg(target_arch = "x86_64")]
 use vm_memory::GuestAddress;
 
 #[derive(Error, Debug)]
@@ -49,12 +48,12 @@ pub enum HypervisorCpuError {
     ///
     /// Setting floating point registers error
     ///
-    #[error("Failed to set special register: {0}")]
+    #[error("Failed to set floating point registers: {0}")]
     SetFloatingPointRegs(#[source] anyhow::Error),
     ///
     /// Getting floating point register error
     ///
-    #[error("Failed to get special register: {0}")]
+    #[error("Failed to get floating points registers: {0}")]
     GetFloatingPointRegs(#[source] anyhow::Error),
     ///
     /// Setting Cpuid error
@@ -347,7 +346,6 @@ pub trait Vcpu: Send + Sync {
     fn notify_guest_clock_paused(&self) -> Result<()> {
         Ok(())
     }
-    #[cfg(target_arch = "x86_64")]
     ///
     /// Sets debug registers to set hardware breakpoints and/or enable single step.
     ///
@@ -408,7 +406,9 @@ pub trait Vcpu: Send + Sync {
     /// Initialize TDX support on the vCPU
     ///
     #[cfg(feature = "tdx")]
-    fn tdx_init(&self, hob_address: u64) -> Result<()>;
+    fn tdx_init(&self, _hob_address: u64) -> Result<()> {
+        unimplemented!()
+    }
     ///
     /// Set the "immediate_exit" state
     ///
@@ -417,12 +417,16 @@ pub trait Vcpu: Send + Sync {
     ///
     /// Returns the details about TDX exit reason
     ///
-    fn get_tdx_exit_details(&mut self) -> Result<TdxExitDetails>;
+    fn get_tdx_exit_details(&mut self) -> Result<TdxExitDetails> {
+        unimplemented!()
+    }
     #[cfg(feature = "tdx")]
     ///
     /// Set the status code for TDX exit
     ///
-    fn set_tdx_status(&mut self, status: TdxExitStatus);
+    fn set_tdx_status(&mut self, _status: TdxExitStatus) {
+        unimplemented!()
+    }
     #[cfg(target_arch = "x86_64")]
     ///
     /// Return the list of initial MSR entries for a VCPU

hypervisor/src/hypervisor.rs

@@ -117,12 +117,20 @@ pub trait Hypervisor: Send + Sync {
     }
     #[cfg(target_arch = "aarch64")]
     ///
-    /// Retrieve AArch64 host maximum IPA size supported by KVM.
+    /// Retrieve AArch64 host maximum IPA size supported by KVM
     ///
     fn get_host_ipa_limit(&self) -> i32;
     ///
     /// Retrieve TDX capabilities
     ///
     #[cfg(feature = "tdx")]
-    fn tdx_capabilities(&self) -> Result<TdxCapabilities>;
+    fn tdx_capabilities(&self) -> Result<TdxCapabilities> {
+        unimplemented!()
+    }
+    ///
+    /// Get the number of supported hardware breakpoints
+    ///
+    fn get_guest_debug_hw_bps(&self) -> usize {
+        unimplemented!()
+    }
 }

hypervisor/src/kvm/aarch64/gic/mod.rs

@@ -4,7 +4,7 @@ mod dist_regs;
 mod icc_regs;
 mod redist_regs;
 
-use crate::arch::aarch64::gic::{Error, Result, Vgic};
+use crate::arch::aarch64::gic::{Error, Result, Vgic, VgicConfig};
 use crate::device::HypervisorDeviceError;
 use crate::kvm::{kvm_bindings, KvmVm};
 use crate::{CpuState, Vm};
@@ -133,9 +133,7 @@ impl KvmGicV3Its {
     /// Setup the device-specific attributes
     fn init_device_attributes(&mut self, vm: &KvmVm, nr_irqs: u32) -> Result<()> {
         // GicV3 part attributes
-        /* Setting up the distributor attribute.
-         We are placing the GIC below 1GB so we need to substract the size of the distributor.
-        */
+        /* Setting up the distributor attribute. */
         Self::set_device_attribute(
             &self.device,
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ADDR,
@@ -144,9 +142,7 @@ impl KvmGicV3Its {
             0,
         )?;
 
-        /* Setting up the redistributors' attribute.
-        We are calculating here the start of the redistributors address. We have one per CPU.
-        */
+        /* Setting up the redistributors' attribute. */
         Self::set_device_attribute(
             &self.device,
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ADDR,
@@ -248,37 +244,26 @@ impl KvmGicV3Its {
 
     /// Method to initialize the GIC device
     #[allow(clippy::new_ret_no_self)]
-    pub fn new(
-        vm: &dyn Vm,
-        vcpu_count: u64,
-        dist_addr: u64,
-        dist_size: u64,
-        redist_size: u64,
-        msi_size: u64,
-        nr_irqs: u32,
-    ) -> Result<KvmGicV3Its> {
+    pub fn new(vm: &dyn Vm, config: VgicConfig) -> Result<KvmGicV3Its> {
         // This is inside KVM module
         let vm = vm.as_any().downcast_ref::<KvmVm>().expect("Wrong VM type?");
 
         let vgic = Self::create_device(vm)?;
-        let redists_size: u64 = redist_size * vcpu_count;
-        let redists_addr: u64 = dist_addr - redists_size;
-        let msi_addr: u64 = redists_addr - msi_size;
 
         let mut gic_device = KvmGicV3Its {
             device: vgic,
             its_device: None,
-            gicr_typers: vec![0; vcpu_count.try_into().unwrap()],
-            dist_addr,
-            dist_size,
-            redists_addr,
-            redists_size,
-            msi_addr,
-            msi_size,
-            vcpu_count,
+            gicr_typers: vec![0; config.vcpu_count.try_into().unwrap()],
+            dist_addr: config.dist_addr,
+            dist_size: config.dist_size,
+            redists_addr: config.redists_addr,
+            redists_size: config.redists_size,
+            msi_addr: config.msi_addr,
+            msi_size: config.msi_size,
+            vcpu_count: config.vcpu_count,
         };
 
-        gic_device.init_device_attributes(vm, nr_irqs)?;
+        gic_device.init_device_attributes(vm, config.nr_irqs)?;
 
         Ok(gic_device)
     }
@@ -500,23 +485,30 @@ mod tests {
     use crate::aarch64::gic::{
         get_dist_regs, get_icc_regs, get_redist_regs, set_dist_regs, set_icc_regs, set_redist_regs,
     };
+    use crate::arch::aarch64::gic::VgicConfig;
     use crate::kvm::KvmGicV3Its;
 
+    fn create_test_vgic_config() -> VgicConfig {
+        VgicConfig {
+            vcpu_count: 1,
+            dist_addr: 0x0900_0000 - 0x01_0000,
+            dist_size: 0x01_0000,
+            // dist_addr - redists_size
+            redists_addr: 0x0900_0000 - 0x01_0000 - 0x02_0000,
+            redists_size: 0x02_0000,
+            // redists_addr - msi_size
+            msi_addr: 0x0900_0000 - 0x01_0000 - 0x02_0000 - 0x02_0000,
+            msi_size: 0x02_0000,
+            nr_irqs: 256,
+        }
+    }
+
     #[test]
     fn test_create_gic() {
         let hv = crate::new().unwrap();
         let vm = hv.create_vm().unwrap();
 
-        assert!(KvmGicV3Its::new(
-            &*vm,
-            1,
-            0x0900_0000 - 0x01_0000,
-            0x01_0000,
-            0x02_0000,
-            0x02_0000,
-            256
-        )
-        .is_ok());
+        assert!(KvmGicV3Its::new(&*vm, create_test_vgic_config()).is_ok());
     }
 
     #[test]
@@ -524,16 +516,7 @@ mod tests {
         let hv = crate::new().unwrap();
         let vm = hv.create_vm().unwrap();
         let _ = vm.create_vcpu(0, None).unwrap();
-        let gic = KvmGicV3Its::new(
-            &*vm,
-            1,
-            0x0900_0000 - 0x01_0000,
-            0x01_0000,
-            0x02_0000,
-            0x02_0000,
-            256,
-        )
-        .expect("Cannot create gic");
+        let gic = KvmGicV3Its::new(&*vm, create_test_vgic_config()).expect("Cannot create gic");
 
         let res = get_dist_regs(&gic.device);
         assert!(res.is_ok());
@@ -549,16 +532,7 @@ mod tests {
         let hv = crate::new().unwrap();
         let vm = hv.create_vm().unwrap();
         let _ = vm.create_vcpu(0, None).unwrap();
-        let gic = KvmGicV3Its::new(
-            &*vm,
-            1,
-            0x0900_0000 - 0x01_0000,
-            0x01_0000,
-            0x02_0000,
-            0x02_0000,
-            256,
-        )
-        .expect("Cannot create gic");
+        let gic = KvmGicV3Its::new(&*vm, create_test_vgic_config()).expect("Cannot create gic");
 
         let gicr_typer = vec![123];
         let res = get_redist_regs(&gic.device, &gicr_typer);
@@ -575,16 +549,7 @@ mod tests {
         let hv = crate::new().unwrap();
         let vm = hv.create_vm().unwrap();
         let _ = vm.create_vcpu(0, None).unwrap();
-        let gic = KvmGicV3Its::new(
-            &*vm,
-            1,
-            0x0900_0000 - 0x01_0000,
-            0x01_0000,
-            0x02_0000,
-            0x02_0000,
-            256,
-        )
-        .expect("Cannot create gic");
+        let gic = KvmGicV3Its::new(&*vm, create_test_vgic_config()).expect("Cannot create gic");
 
         let gicr_typer = vec![123];
         let res = get_icc_regs(&gic.device, &gicr_typer);
@@ -602,14 +567,7 @@ mod tests {
         let vm = hv.create_vm().unwrap();
         let _ = vm.create_vcpu(0, None).unwrap();
         let gic = vm
-            .create_vgic(
-                1,
-                0x0900_0000 - 0x01_0000,
-                0x01_0000,
-                0x02_0000,
-                0x02_0000,
-                256,
-            )
+            .create_vgic(create_test_vgic_config())
             .expect("Cannot create gic");
 
         assert!(gic.lock().unwrap().save_data_tables().is_ok());

hypervisor/src/kvm/aarch64/gic/redist_regs.rs

@@ -39,8 +39,8 @@ 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 as u64
-    | KVM_REG_SIZE_U64 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);
@@ -200,13 +200,7 @@ 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 = {
-            if index == vcpu_states.len() - 1 {
-                1
-            } else {
-                0
-            }
-        };
+        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

hypervisor/src/kvm/aarch64/mod.rs

@@ -24,8 +24,8 @@ pub use {kvm_ioctls::Cap, kvm_ioctls::Kvm};
 // This macro gets the offset of a structure (i.e `str`) member (i.e `field`) without having
 // an instance of that structure.
 #[macro_export]
-macro_rules! offset__of {
-    ($str:ty, $($field:ident)+) => ({
+macro_rules! offset_of {
+    ($str:ty, $field:ident) => {{
         let tmp: std::mem::MaybeUninit<$str> = std::mem::MaybeUninit::uninit();
         let base = tmp.as_ptr();
 
@@ -34,14 +34,16 @@ macro_rules! offset__of {
         // SAFETY: The pointer is valid and aligned, just not initialised. Using `addr_of` ensures
         // that we don't actually read from `base` (which would be UB) nor create an intermediate
         // reference.
-        let member = unsafe { core::ptr::addr_of!((*base).$($field)*) } as *const u8;
+        let member = unsafe { core::ptr::addr_of!((*base).$field) } as *const u8;
 
         // Avoid warnings when nesting `unsafe` blocks.
         #[allow(unused_unsafe)]
         // SAFETY: The two pointers are within the same allocated object `tmp`. All requirements
         // from offset_from are upheld.
-        unsafe { member.offset_from(base as *const u8) as usize }
-    });
+        unsafe {
+            member.offset_from(base as *const u8) as usize
+        }
+    }};
 }
 
 // Following are macros that help with getting the ID of a aarch64 core register.
@@ -97,8 +99,7 @@ pub fn is_system_register(regid: u64) -> bool {
 
     assert!(
         !(size != KVM_REG_SIZE_U32 && size != KVM_REG_SIZE_U64),
-        "Unexpected register size for system register {}",
-        size
+        "Unexpected register size for system register {size}"
     );
 
     true

hypervisor/src/kvm/mod.rs

@@ -16,14 +16,14 @@ pub use crate::aarch64::{
     VcpuKvmState,
 };
 #[cfg(target_arch = "aarch64")]
-use crate::arch::aarch64::gic::Vgic;
+use crate::arch::aarch64::gic::{Vgic, VgicConfig};
 use crate::cpu;
 use crate::hypervisor;
 use crate::vec_with_array_field;
 use crate::vm::{self, InterruptSourceConfig, VmOps};
 use crate::HypervisorType;
 #[cfg(target_arch = "aarch64")]
-use crate::{arm64_core_reg_id, offset__of};
+use crate::{arm64_core_reg_id, offset_of};
 use kvm_ioctls::{NoDatamatch, VcpuFd, VmFd};
 use std::any::Any;
 use std::collections::HashMap;
@@ -60,8 +60,8 @@ use crate::{
 use aarch64::{RegList, Register, StandardRegisters};
 #[cfg(target_arch = "x86_64")]
 use kvm_bindings::{
-    kvm_enable_cap, kvm_guest_debug, kvm_msr_entry, MsrList, KVM_CAP_HYPERV_SYNIC,
-    KVM_CAP_SPLIT_IRQCHIP, KVM_GUESTDBG_ENABLE, KVM_GUESTDBG_SINGLESTEP, KVM_GUESTDBG_USE_HW_BP,
+    kvm_enable_cap, kvm_msr_entry, MsrList, KVM_CAP_HYPERV_SYNIC, KVM_CAP_SPLIT_IRQCHIP,
+    KVM_GUESTDBG_USE_HW_BP,
 };
 #[cfg(target_arch = "x86_64")]
 use x86_64::check_required_kvm_extensions;
@@ -74,16 +74,17 @@ pub use kvm_bindings;
 #[cfg(feature = "tdx")]
 use kvm_bindings::KVMIO;
 pub use kvm_bindings::{
-    kvm_clock_data, kvm_create_device, kvm_device_type_KVM_DEV_TYPE_VFIO, kvm_irq_routing,
-    kvm_irq_routing_entry, kvm_mp_state, kvm_userspace_memory_region, KVM_IRQ_ROUTING_IRQCHIP,
-    KVM_IRQ_ROUTING_MSI, KVM_MEM_LOG_DIRTY_PAGES, KVM_MEM_READONLY, KVM_MSI_VALID_DEVID,
+    kvm_clock_data, kvm_create_device, kvm_device_type_KVM_DEV_TYPE_VFIO, kvm_guest_debug,
+    kvm_irq_routing, kvm_irq_routing_entry, kvm_mp_state, kvm_userspace_memory_region,
+    KVM_GUESTDBG_ENABLE, KVM_GUESTDBG_SINGLESTEP, KVM_IRQ_ROUTING_IRQCHIP, KVM_IRQ_ROUTING_MSI,
+    KVM_MEM_LOG_DIRTY_PAGES, KVM_MEM_READONLY, KVM_MSI_VALID_DEVID,
 };
 #[cfg(target_arch = "aarch64")]
 use kvm_bindings::{
-    kvm_regs, user_fpsimd_state, user_pt_regs, KVM_NR_SPSR, KVM_REG_ARM64, KVM_REG_ARM64_SYSREG,
-    KVM_REG_ARM64_SYSREG_CRM_MASK, KVM_REG_ARM64_SYSREG_CRN_MASK, KVM_REG_ARM64_SYSREG_OP0_MASK,
-    KVM_REG_ARM64_SYSREG_OP1_MASK, KVM_REG_ARM64_SYSREG_OP2_MASK, KVM_REG_ARM_CORE,
-    KVM_REG_SIZE_U128, KVM_REG_SIZE_U32, KVM_REG_SIZE_U64,
+    kvm_regs, user_fpsimd_state, user_pt_regs, KVM_GUESTDBG_USE_HW, KVM_NR_SPSR, KVM_REG_ARM64,
+    KVM_REG_ARM64_SYSREG, KVM_REG_ARM64_SYSREG_CRM_MASK, KVM_REG_ARM64_SYSREG_CRN_MASK,
+    KVM_REG_ARM64_SYSREG_OP0_MASK, KVM_REG_ARM64_SYSREG_OP1_MASK, KVM_REG_ARM64_SYSREG_OP2_MASK,
+    KVM_REG_ARM_CORE, KVM_REG_SIZE_U128, KVM_REG_SIZE_U32, KVM_REG_SIZE_U64,
 };
 pub use kvm_ioctls;
 pub use kvm_ioctls::{Cap, Kvm};
@@ -413,25 +414,9 @@ impl vm::Vm for KvmVm {
     ///
     /// Creates a virtual GIC device.
     ///
-    fn create_vgic(
-        &self,
-        vcpu_count: u64,
-        dist_addr: u64,
-        dist_size: u64,
-        redist_size: u64,
-        msi_size: u64,
-        nr_irqs: u32,
-    ) -> vm::Result<Arc<Mutex<dyn Vgic>>> {
-        let gic_device = KvmGicV3Its::new(
-            self,
-            vcpu_count,
-            dist_addr,
-            dist_size,
-            redist_size,
-            msi_size,
-            nr_irqs,
-        )
-        .map_err(|e| vm::HypervisorVmError::CreateVgic(anyhow!("Vgic error {:?}", e)))?;
+    fn create_vgic(&self, config: VgicConfig) -> vm::Result<Arc<Mutex<dyn Vgic>>> {
+        let gic_device = KvmGicV3Its::new(self, config)
+            .map_err(|e| vm::HypervisorVmError::CreateVgic(anyhow!("Vgic error {:?}", e)))?;
         Ok(Arc::new(Mutex::new(gic_device)))
     }
     ///
@@ -837,7 +822,7 @@ impl vm::Vm for KvmVm {
         tdx_command(
             &self.fd.as_raw_fd(),
             TdxCommand::InitMemRegion,
-            if measure { 1 } else { 0 },
+            u32::from(measure),
             &data as *const _ as u64,
         )
         .map_err(vm::HypervisorVmError::InitMemRegionTdx)
@@ -1075,6 +1060,20 @@ impl hypervisor::Hypervisor for KvmHypervisor {
 
         Ok(data)
     }
+
+    ///
+    /// Get the number of supported hardware breakpoints
+    ///
+    fn get_guest_debug_hw_bps(&self) -> usize {
+        #[cfg(target_arch = "x86_64")]
+        {
+            4
+        }
+        #[cfg(target_arch = "aarch64")]
+        {
+            self.kvm.get_guest_debug_hw_bps() as usize
+        }
+    }
 }
 /// Vcpu struct for KVM
 pub struct KvmVcpu {
@@ -1115,7 +1114,7 @@ impl cpu::Vcpu for KvmVcpu {
     #[cfg(target_arch = "aarch64")]
     fn get_regs(&self) -> cpu::Result<StandardRegisters> {
         let mut state: StandardRegisters = kvm_regs::default();
-        let mut off = offset__of!(user_pt_regs, regs);
+        let mut off = offset_of!(user_pt_regs, regs);
         // There are 31 user_pt_regs:
         // https://elixir.free-electrons.com/linux/v4.14.174/source/arch/arm64/include/uapi/asm/ptrace.h#L72
         // These actually are the general-purpose registers of the Armv8-a
@@ -1124,84 +1123,99 @@ impl cpu::Vcpu for KvmVcpu {
             state.regs.regs[i] = self
                 .fd
                 .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off))
-                .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?;
+                .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
+                .try_into()
+                .unwrap();
             off += std::mem::size_of::<u64>();
         }
 
         // We are now entering the "Other register" section of the ARMv8-a architecture.
         // First one, stack pointer.
-        let off = offset__of!(user_pt_regs, sp);
+        let off = offset_of!(user_pt_regs, sp);
         state.regs.sp = self
             .fd
             .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off))
-            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?;
+            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
+            .try_into()
+            .unwrap();
 
         // Second one, the program counter.
-        let off = offset__of!(user_pt_regs, pc);
+        let off = offset_of!(user_pt_regs, pc);
         state.regs.pc = self
             .fd
             .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off))
-            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?;
+            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
+            .try_into()
+            .unwrap();
 
         // Next is the processor state.
-        let off = offset__of!(user_pt_regs, pstate);
+        let off = offset_of!(user_pt_regs, pstate);
         state.regs.pstate = self
             .fd
             .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off))
-            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?;
+            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
+            .try_into()
+            .unwrap();
 
         // The stack pointer associated with EL1
-        let off = offset__of!(kvm_regs, sp_el1);
+        let off = offset_of!(kvm_regs, sp_el1);
         state.sp_el1 = self
             .fd
             .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off))
-            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?;
+            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
+            .try_into()
+            .unwrap();
 
         // Exception Link Register for EL1, when taking an exception to EL1, this register
         // holds the address to which to return afterwards.
-        let off = offset__of!(kvm_regs, elr_el1);
+        let off = offset_of!(kvm_regs, elr_el1);
         state.elr_el1 = self
             .fd
             .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off))
-            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?;
+            .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
+            .try_into()
+            .unwrap();
 
         // Saved Program Status Registers, there are 5 of them used in the kernel.
-        let mut off = offset__of!(kvm_regs, spsr);
+        let mut off = offset_of!(kvm_regs, spsr);
         for i in 0..KVM_NR_SPSR as usize {
             state.spsr[i] = self
                 .fd
                 .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off))
-                .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?;
+                .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
+                .try_into()
+                .unwrap();
             off += std::mem::size_of::<u64>();
         }
 
         // Now moving on to floting point registers which are stored in the user_fpsimd_state in the kernel:
         // https://elixir.free-electrons.com/linux/v4.9.62/source/arch/arm64/include/uapi/asm/kvm.h#L53
-        let mut off = offset__of!(kvm_regs, fp_regs) + offset__of!(user_fpsimd_state, vregs);
+        let mut off = offset_of!(kvm_regs, fp_regs) + offset_of!(user_fpsimd_state, vregs);
         for i in 0..32 {
             state.fp_regs.vregs[i] = self
                 .fd
                 .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U128, off))
-                .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
-                .into();
+                .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?;
             off += mem::size_of::<u128>();
         }
 
         // Floating-point Status Register
-        let off = offset__of!(kvm_regs, fp_regs) + offset__of!(user_fpsimd_state, fpsr);
+        let off = offset_of!(kvm_regs, fp_regs) + offset_of!(user_fpsimd_state, fpsr);
         state.fp_regs.fpsr = self
             .fd
             .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U32, off))
             .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
-            as u32;
+            .try_into()
+            .unwrap();
 
         // Floating-point Control Register
-        let off = offset__of!(kvm_regs, fp_regs) + offset__of!(user_fpsimd_state, fpcr);
+        let off = offset_of!(kvm_regs, fp_regs) + offset_of!(user_fpsimd_state, fpcr);
         state.fp_regs.fpcr = self
             .fd
             .get_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U32, off))
             .map_err(|e| cpu::HypervisorCpuError::GetCoreRegister(e.into()))?
-            as u32;
+            .try_into()
+            .unwrap();
         Ok(state)
     }
     #[cfg(target_arch = "x86_64")]
@@ -1224,74 +1238,92 @@ impl cpu::Vcpu for KvmVcpu {
     fn set_regs(&self, state: &StandardRegisters) -> cpu::Result<()> {
         // The function follows the exact identical order from `state`. Look there
         // for some additional info on registers.
-        let mut off = offset__of!(user_pt_regs, regs);
+        let mut off = offset_of!(user_pt_regs, regs);
         for i in 0..31 {
             self.fd
                 .set_one_reg(
                     arm64_core_reg_id!(KVM_REG_SIZE_U64, off),
-                    state.regs.regs[i],
+                    state.regs.regs[i].into(),
                 )
                 .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
             off += std::mem::size_of::<u64>();
         }
 
-        let off = offset__of!(user_pt_regs, sp);
+        let off = offset_of!(user_pt_regs, sp);
         self.fd
-            .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off), state.regs.sp)
+            .set_one_reg(
+                arm64_core_reg_id!(KVM_REG_SIZE_U64, off),
+                state.regs.sp.into(),
+            )
             .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
 
-        let off = offset__of!(user_pt_regs, pc);
+        let off = offset_of!(user_pt_regs, pc);
         self.fd
-            .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off), state.regs.pc)
+            .set_one_reg(
+                arm64_core_reg_id!(KVM_REG_SIZE_U64, off),
+                state.regs.pc.into(),
+            )
             .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
 
-        let off = offset__of!(user_pt_regs, pstate);
+        let off = offset_of!(user_pt_regs, pstate);
         self.fd
-            .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off), state.regs.pstate)
+            .set_one_reg(
+                arm64_core_reg_id!(KVM_REG_SIZE_U64, off),
+                state.regs.pstate.into(),
+            )
             .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
 
-        let off = offset__of!(kvm_regs, sp_el1);
+        let off = offset_of!(kvm_regs, sp_el1);
         self.fd
-            .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off), state.sp_el1)
+            .set_one_reg(
+                arm64_core_reg_id!(KVM_REG_SIZE_U64, off),
+                state.sp_el1.into(),
+            )
             .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
 
-        let off = offset__of!(kvm_regs, elr_el1);
+        let off = offset_of!(kvm_regs, elr_el1);
         self.fd
-            .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off), state.elr_el1)
+            .set_one_reg(
+                arm64_core_reg_id!(KVM_REG_SIZE_U64, off),
+                state.elr_el1.into(),
+            )
             .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
 
-        let mut off = offset__of!(kvm_regs, spsr);
+        let mut off = offset_of!(kvm_regs, spsr);
         for i in 0..KVM_NR_SPSR as usize {
             self.fd
-                .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, off), state.spsr[i])
+                .set_one_reg(
+                    arm64_core_reg_id!(KVM_REG_SIZE_U64, off),
+                    state.spsr[i].into(),
+                )
                 .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
             off += std::mem::size_of::<u64>();
         }
 
-        let mut off = offset__of!(kvm_regs, fp_regs) + offset__of!(user_fpsimd_state, vregs);
+        let mut off = offset_of!(kvm_regs, fp_regs) + offset_of!(user_fpsimd_state, vregs);
         for i in 0..32 {
             self.fd
                 .set_one_reg(
                     arm64_core_reg_id!(KVM_REG_SIZE_U128, off),
-                    state.fp_regs.vregs[i] as u64,
+                    state.fp_regs.vregs[i],
                 )
                 .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
             off += mem::size_of::<u128>();
         }
 
-        let off = offset__of!(kvm_regs, fp_regs) + offset__of!(user_fpsimd_state, fpsr);
+        let off = offset_of!(kvm_regs, fp_regs) + offset_of!(user_fpsimd_state, fpsr);
         self.fd
             .set_one_reg(
                 arm64_core_reg_id!(KVM_REG_SIZE_U32, off),
-                state.fp_regs.fpsr as u64,
+                state.fp_regs.fpsr.into(),
             )
             .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
 
-        let off = offset__of!(kvm_regs, fp_regs) + offset__of!(user_fpsimd_state, fpcr);
+        let off = offset_of!(kvm_regs, fp_regs) + offset_of!(user_fpsimd_state, fpcr);
         self.fd
             .set_one_reg(
                 arm64_core_reg_id!(KVM_REG_SIZE_U32, off),
-                state.fp_regs.fpcr as u64,
+                state.fp_regs.fpcr.into(),
             )
             .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
         Ok(())
@@ -1582,7 +1614,6 @@ impl cpu::Vcpu for KvmVcpu {
 
         Ok(())
     }
-    #[cfg(target_arch = "x86_64")]
     ///
     /// Sets debug registers to set hardware breakpoints and/or enable single step.
     ///
@@ -1591,32 +1622,47 @@ impl cpu::Vcpu for KvmVcpu {
         addrs: &[vm_memory::GuestAddress],
         singlestep: bool,
     ) -> cpu::Result<()> {
-        if addrs.len() > 4 {
-            return Err(cpu::HypervisorCpuError::SetDebugRegs(anyhow!(
-                "Support 4 breakpoints at most but {} addresses are passed",
-                addrs.len()
-            )));
-        }
-
         let mut dbg = kvm_guest_debug {
+            #[cfg(target_arch = "x86_64")]
             control: KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW_BP,
+            #[cfg(target_arch = "aarch64")]
+            control: KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW,
             ..Default::default()
         };
         if singlestep {
             dbg.control |= KVM_GUESTDBG_SINGLESTEP;
         }
 
-        // Set bits 9 and 10.
-        // bit 9: GE (global exact breakpoint enable) flag.
-        // bit 10: always 1.
-        dbg.arch.debugreg[7] = 0x0600;
+        // Set the debug registers.
+        // Here we assume that the number of addresses do not exceed what
+        // `Hypervisor::get_guest_debug_hw_bps()` specifies.
+        #[cfg(target_arch = "x86_64")]
+        {
+            // Set bits 9 and 10.
+            // bit 9: GE (global exact breakpoint enable) flag.
+            // bit 10: always 1.
+            dbg.arch.debugreg[7] = 0x0600;
 
-        for (i, addr) in addrs.iter().enumerate() {
-            dbg.arch.debugreg[i] = addr.0;
-            // Set global breakpoint enable flag
-            dbg.arch.debugreg[7] |= 2 << (i * 2);
+            for (i, addr) in addrs.iter().enumerate() {
+                dbg.arch.debugreg[i] = addr.0;
+                // Set global breakpoint enable flag
+                dbg.arch.debugreg[7] |= 2 << (i * 2);
+            }
+        }
+        #[cfg(target_arch = "aarch64")]
+        {
+            for (i, addr) in addrs.iter().enumerate() {
+                // DBGBCR_EL1 (Debug Breakpoint Control Registers, D13.3.2):
+                // bit 0: 1 (Enabled)
+                // bit 1~2: 0b11 (PMC = EL1/EL0)
+                // bit 5~8: 0b1111 (BAS = AArch64)
+                // others: 0
+                dbg.arch.dbg_bcr[i] = 0b1u64 | 0b110u64 | 0b1_1110_0000u64;
+                // DBGBVR_EL1 (Debug Breakpoint Value Registers, D13.3.3):
+                // bit 2~52: VA[2:52]
+                dbg.arch.dbg_bvr[i] = (!0u64 >> 11) & addr.0;
+            }
         }
-
         self.fd
             .set_guest_debug(&dbg)
             .map_err(|e| cpu::HypervisorCpuError::SetDebugRegs(e.into()))
@@ -1654,8 +1700,8 @@ impl cpu::Vcpu for KvmVcpu {
         // it to the corresponding KVM ID, and call `KVM_GET_ONE_REG` API to
         // get the value of the system parameter.
         //
-        let id: u64 = KVM_REG_ARM64 as u64
-            | KVM_REG_SIZE_U64 as u64
+        let id: u64 = KVM_REG_ARM64
+            | KVM_REG_SIZE_U64
             | KVM_REG_ARM64_SYSREG as u64
             | ((((sys_reg) >> 5)
                 & (KVM_REG_ARM64_SYSREG_OP0_MASK
@@ -1663,9 +1709,12 @@ impl cpu::Vcpu for KvmVcpu {
                     | KVM_REG_ARM64_SYSREG_CRN_MASK
                     | KVM_REG_ARM64_SYSREG_CRM_MASK
                     | KVM_REG_ARM64_SYSREG_OP2_MASK)) as u64);
-        self.fd
+        Ok(self
+            .fd
             .get_one_reg(id)
-            .map_err(|e| cpu::HypervisorCpuError::GetSysRegister(e.into()))
+            .map_err(|e| cpu::HypervisorCpuError::GetSysRegister(e.into()))?
+            .try_into()
+            .unwrap())
     }
     ///
     /// Configure core registers for a given CPU.
@@ -1684,32 +1733,35 @@ impl cpu::Vcpu for KvmVcpu {
         const PSTATE_FAULT_BITS_64: u64 =
             PSR_MODE_EL1h | PSR_A_BIT | PSR_F_BIT | PSR_I_BIT | PSR_D_BIT;
 
-        let kreg_off = offset__of!(kvm_regs, regs);
+        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;
+        let pstate = offset_of!(user_pt_regs, pstate) + kreg_off;
         self.fd
             .set_one_reg(
                 arm64_core_reg_id!(KVM_REG_SIZE_U64, pstate),
-                PSTATE_FAULT_BITS_64,
+                PSTATE_FAULT_BITS_64.into(),
             )
             .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
 
         // 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;
+            let pc = offset_of!(user_pt_regs, pc) + kreg_off;
             self.fd
-                .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, pc), boot_ip as u64)
+                .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, pc), boot_ip.into())
                 .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
 
             // 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;
+            let regs0 = offset_of!(user_pt_regs, regs) + kreg_off;
             self.fd
-                .set_one_reg(arm64_core_reg_id!(KVM_REG_SIZE_U64, regs0), fdt_start)
+                .set_one_reg(
+                    arm64_core_reg_id!(KVM_REG_SIZE_U64, regs0),
+                    fdt_start.into(),
+                )
                 .map_err(|e| cpu::HypervisorCpuError::SetCoreRegister(e.into()))?;
         }
         Ok(())
@@ -1876,7 +1928,9 @@ impl cpu::Vcpu for KvmVcpu {
                 addr: self
                     .fd
                     .get_one_reg(*index)
-                    .map_err(|e| cpu::HypervisorCpuError::GetSysRegister(e.into()))?,
+                    .map_err(|e| cpu::HypervisorCpuError::GetSysRegister(e.into()))?
+                    .try_into()
+                    .unwrap(),
             });
         }
 
@@ -1981,7 +2035,7 @@ impl cpu::Vcpu for KvmVcpu {
         // Set system registers
         for reg in &state.sys_regs {
             self.fd
-                .set_one_reg(reg.id, reg.addr)
+                .set_one_reg(reg.id, reg.addr.into())
                 .map_err(|e| cpu::HypervisorCpuError::SetSysRegister(e.into()))?;
         }
 
@@ -2012,6 +2066,7 @@ impl cpu::Vcpu for KvmVcpu {
     #[cfg(feature = "tdx")]
     fn get_tdx_exit_details(&mut self) -> cpu::Result<TdxExitDetails> {
         let kvm_run = self.fd.get_kvm_run();
+        // SAFETY: accessing a union field in a valid structure
         let tdx_vmcall = unsafe { &mut kvm_run.__bindgen_anon_1.tdx.u.vmcall };
 
         tdx_vmcall.status_code = TDG_VP_VMCALL_INVALID_OPERAND;
@@ -2035,6 +2090,7 @@ impl cpu::Vcpu for KvmVcpu {
     #[cfg(feature = "tdx")]
     fn set_tdx_status(&mut self, status: TdxExitStatus) {
         let kvm_run = self.fd.get_kvm_run();
+        // SAFETY: accessing a union field in a valid structure
         let tdx_vmcall = unsafe { &mut kvm_run.__bindgen_anon_1.tdx.u.vmcall };
 
         tdx_vmcall.status_code = match status {

hypervisor/src/lib.rs

@@ -18,8 +18,6 @@
 //! - arm64
 //!
 
-#![allow(clippy::significant_drop_in_scrutinee)]
-
 #[macro_use]
 extern crate anyhow;
 #[cfg(target_arch = "x86_64")]
@@ -63,7 +61,9 @@ pub use vm::{
 
 #[derive(Debug, Copy, Clone)]
 pub enum HypervisorType {
+    #[cfg(feature = "kvm")]
     Kvm,
+    #[cfg(feature = "mshv")]
     Mshv,
 }
 

hypervisor/src/mshv/mod.rs

@@ -446,10 +446,10 @@ impl cpu::Vcpu for MshvVcpu {
                             /* Advance RIP and update RAX */
                             let arr_reg_name_value = [
                                 (
-                                    hv_register_name::HV_X64_REGISTER_RIP,
+                                    hv_x64_register_name_HV_X64_REGISTER_RIP,
                                     info.header.rip + insn_len,
                                 ),
-                                (hv_register_name::HV_X64_REGISTER_RAX, ret_rax),
+                                (hv_x64_register_name_HV_X64_REGISTER_RAX, ret_rax),
                             ];
                             set_registers_64!(self.fd, arr_reg_name_value)
                                 .map_err(|e| cpu::HypervisorCpuError::SetRegister(e.into()))?;
@@ -458,12 +458,13 @@ impl cpu::Vcpu for MshvVcpu {
                         _ => {}
                     }
 
-                    // SAFETY: access_info is valid, otherwise we won't be here
                     assert!(
+                        // SAFETY: access_info is valid, otherwise we won't be here
                         (unsafe { access_info.__bindgen_anon_1.string_op() } != 1),
                         "String IN/OUT not supported"
                     );
                     assert!(
+                        // SAFETY: access_info is valid, otherwise we won't be here
                         (unsafe { access_info.__bindgen_anon_1.rep_prefix() } != 1),
                         "Rep IN/OUT not supported"
                     );
@@ -494,10 +495,10 @@ impl cpu::Vcpu for MshvVcpu {
                     /* Advance RIP and update RAX */
                     let arr_reg_name_value = [
                         (
-                            hv_register_name::HV_X64_REGISTER_RIP,
+                            hv_x64_register_name_HV_X64_REGISTER_RIP,
                             info.header.rip + insn_len,
                         ),
-                        (hv_register_name::HV_X64_REGISTER_RAX, ret_rax),
+                        (hv_x64_register_name_HV_X64_REGISTER_RAX, ret_rax),
                     ];
                     set_registers_64!(self.fd, arr_reg_name_value)
                         .map_err(|e| cpu::HypervisorCpuError::SetRegister(e.into()))?;
@@ -703,7 +704,6 @@ impl cpu::Vcpu for MshvVcpu {
             msr!(msr_index::MSR_LSTAR),
             msr!(msr_index::MSR_KERNEL_GS_BASE),
             msr!(msr_index::MSR_SYSCALL_MASK),
-            msr!(msr_index::MSR_IA32_TSC),
             msr_data!(msr_index::MSR_MTRRdefType, MTRR_ENABLE | MTRR_MEM_TYPE_WB),
         ]
         .to_vec()
@@ -1093,7 +1093,7 @@ impl vm::Vm for MshvVm {
             flags,
             guest_pfn: guest_phys_addr >> PAGE_SHIFT,
             size: memory_size,
-            userspace_addr: userspace_addr as u64,
+            userspace_addr,
         }
         .into()
     }

hypervisor/src/vm.rs

@@ -11,7 +11,7 @@
 #[cfg(target_arch = "aarch64")]
 use crate::aarch64::VcpuInit;
 #[cfg(target_arch = "aarch64")]
-use crate::arch::aarch64::gic::Vgic;
+use crate::arch::aarch64::gic::{Vgic, VgicConfig};
 #[cfg(feature = "tdx")]
 use crate::arch::x86::CpuIdEntry;
 use crate::cpu::Vcpu;
@@ -273,15 +273,7 @@ pub trait Vm: Send + Sync + Any {
     /// Creates a new KVM vCPU file descriptor and maps the memory corresponding
     fn create_vcpu(&self, id: u8, vm_ops: Option<Arc<dyn VmOps>>) -> Result<Arc<dyn Vcpu>>;
     #[cfg(target_arch = "aarch64")]
-    fn create_vgic(
-        &self,
-        vcpu_count: u64,
-        dist_addr: u64,
-        dist_size: u64,
-        redist_size: u64,
-        msi_size: u64,
-        nr_irqs: u32,
-    ) -> Result<Arc<Mutex<dyn Vgic>>>;
+    fn create_vgic(&self, config: VgicConfig) -> Result<Arc<Mutex<dyn Vgic>>>;
 
     /// Registers an event to be signaled whenever a certain address is written to.
     fn register_ioevent(
@@ -334,19 +326,25 @@ pub trait Vm: Send + Sync + Any {
     fn get_dirty_log(&self, slot: u32, base_gpa: u64, memory_size: u64) -> Result<Vec<u64>>;
     #[cfg(feature = "tdx")]
     /// Initalize TDX on this VM
-    fn tdx_init(&self, cpuid: &[CpuIdEntry], max_vcpus: u32) -> Result<()>;
+    fn tdx_init(&self, _cpuid: &[CpuIdEntry], _max_vcpus: u32) -> Result<()> {
+        unimplemented!()
+    }
     #[cfg(feature = "tdx")]
     /// Finalize the configuration of TDX on this VM
-    fn tdx_finalize(&self) -> Result<()>;
+    fn tdx_finalize(&self) -> Result<()> {
+        unimplemented!()
+    }
     #[cfg(feature = "tdx")]
     /// Initalize a TDX memory region for this VM
     fn tdx_init_memory_region(
         &self,
-        host_address: u64,
-        guest_address: u64,
-        size: u64,
-        measure: bool,
-    ) -> Result<()>;
+        _host_address: u64,
+        _guest_address: u64,
+        _size: u64,
+        _measure: bool,
+    ) -> Result<()> {
+        unimplemented!()
+    }
     /// Downcast to the underlying hypervisor VM type
     fn as_any(&self) -> &dyn Any;
 }

net_gen/Cargo.toml

@@ -5,4 +5,4 @@ authors = ["The Chromium OS Authors"]
 edition = "2021"
 
 [dependencies]
-vmm-sys-util = "0.10.0"
+vmm-sys-util = "0.11.0"

net_gen/src/iff.rs

@@ -1187,6 +1187,7 @@ pub struct ifreq {
 
 impl Default for ifreq {
     fn default() -> Self {
+        // SAFETY: all zeros is a valid pattern for this data type
         unsafe { std::mem::zeroed() }
     }
 }

net_util/Cargo.toml

@@ -6,23 +6,23 @@ edition = "2021"
 
 [dependencies]
 epoll = "4.3.1"
-getrandom = "0.2.7"
-libc = "0.2.131"
+getrandom = "0.2.8"
+libc = "0.2.139"
 log = "0.4.17"
 net_gen = { path = "../net_gen" }
 rate_limiter = { path = "../rate_limiter" }
-serde = "1.0.143"
-thiserror = "1.0.32"
-versionize = "0.1.6"
+serde = "1.0.151"
+thiserror = "1.0.38"
+versionize = "0.1.9"
 versionize_derive = "0.1.4"
 virtio-bindings = "0.1.0"
-virtio-queue = "0.5.0"
-vm-memory = { version = "0.8.0", features = ["backend-mmap", "backend-atomic", "backend-bitmap"] }
+virtio-queue = "0.7.0"
+vm-memory = { version = "0.10.0", features = ["backend-mmap", "backend-atomic", "backend-bitmap"] }
 vm-virtio = { path = "../vm-virtio" }
-vmm-sys-util = "0.10.0"
+vmm-sys-util = "0.11.0"
 
 [dev-dependencies]
-once_cell = "1.13.1"
+once_cell = "1.16.0"
 pnet = "0.31.0"
 pnet_datalink = "0.31.0"
-serde_json = "1.0.83"
+serde_json = "1.0.89"

net_util/src/lib.rs

@@ -66,32 +66,34 @@ fn create_sockaddr(ip_addr: net::Ipv4Addr) -> net_gen::sockaddr {
     let addr_in = net_gen::sockaddr_in {
         sin_family: net_gen::AF_INET as u16,
         sin_port: 0,
+        // SAFETY: ip_addr can be safely transmute to in_addr
         sin_addr: unsafe { mem::transmute(ip_addr.octets()) },
         __pad: [0; 8usize],
     };
 
+    // SAFETY: addr_in can be safely transmute to sockaddr
     unsafe { mem::transmute(addr_in) }
 }
 
 fn create_inet_socket() -> Result<net::UdpSocket> {
-    // This is safe since we check the return value.
+    // SAFETY: we check the return value.
     let sock = unsafe { libc::socket(libc::AF_INET, libc::SOCK_DGRAM, 0) };
     if sock < 0 {
         return Err(Error::CreateSocket(IoError::last_os_error()));
     }
 
-    // This is safe; nothing else will use or hold onto the raw sock fd.
+    // SAFETY: nothing else will use or hold onto the raw sock fd.
     Ok(unsafe { net::UdpSocket::from_raw_fd(sock) })
 }
 
 fn create_unix_socket() -> Result<net::UdpSocket> {
-    // This is safe since we check the return value.
+    // SAFETY: we check the return value.
     let sock = unsafe { libc::socket(libc::AF_UNIX, libc::SOCK_DGRAM, 0) };
     if sock < 0 {
         return Err(Error::CreateSocket(IoError::last_os_error()));
     }
 
-    // This is safe; nothing else will use or hold onto the raw sock fd.
+    // SAFETY: nothing else will use or hold onto the raw sock fd.
     Ok(unsafe { net::UdpSocket::from_raw_fd(sock) })
 }
 
@@ -131,17 +133,19 @@ pub fn build_net_config_space(
     config: &mut VirtioNetConfig,
     mac: MacAddr,
     num_queues: usize,
+    mtu: Option<u16>,
     avail_features: &mut u64,
 ) {
     config.mac.copy_from_slice(mac.get_bytes());
     *avail_features |= 1 << VIRTIO_NET_F_MAC;
 
-    build_net_config_space_with_mq(config, num_queues, avail_features);
+    build_net_config_space_with_mq(config, num_queues, mtu, avail_features);
 }
 
 pub fn build_net_config_space_with_mq(
     config: &mut VirtioNetConfig,
     num_queues: usize,
+    mtu: Option<u16>,
     avail_features: &mut u64,
 ) {
     let num_queue_pairs = (num_queues / 2) as u16;
@@ -151,6 +155,9 @@ pub fn build_net_config_space_with_mq(
         config.max_virtqueue_pairs = num_queue_pairs;
         *avail_features |= 1 << VIRTIO_NET_F_MQ;
     }
+    if let Some(mtu) = mtu {
+        config.mtu = mtu;
+    }
 }
 
 pub fn virtio_features_to_tap_offload(features: u64) -> c_uint {

net_util/src/mac.rs

@@ -125,7 +125,7 @@ impl<'de> Deserialize<'de> for MacAddr {
     {
         let s = String::deserialize(deserializer)?;
         MacAddr::parse_str(&s)
-            .map_err(|e| D::Error::custom(format!("The provided MAC address is invalid: {}", e)))
+            .map_err(|e| D::Error::custom(format!("The provided MAC address is invalid: {e}")))
     }
 }
 
@@ -161,7 +161,7 @@ mod tests {
 
         let mac = MacAddr::parse_str("12:34:56:78:9a:BC").unwrap();
 
-        println!("parsed MAC address: {}", mac);
+        println!("parsed MAC address: {mac}");
 
         let bytes = mac.get_bytes();
         assert_eq!(bytes, [0x12u8, 0x34, 0x56, 0x78, 0x9a, 0xbc]);

net_util/src/open_tap.rs

@@ -12,9 +12,9 @@ use thiserror::Error;
 pub enum Error {
     #[error("Failed to convert an hexadecimal string into an integer: {0}")]
     ConvertHexStringToInt(std::num::ParseIntError),
-    #[error("Error related to the multiqueue support (no support TAP side).")]
+    #[error("Error related to the multiqueue support (no support TAP side)")]
     MultiQueueNoTapSupport,
-    #[error("Error related to the multiqueue support (no support device side).")]
+    #[error("Error related to the multiqueue support (no support device side)")]
     MultiQueueNoDeviceSupport,
     #[error("Failed to read the TAP flags from sysfs: {0}")]
     ReadSysfsTunFlags(io::Error),
@@ -30,6 +30,8 @@ pub enum Error {
     TapGetMac(TapError),
     #[error("Setting vnet header size failed: {0}")]
     TapSetVnetHdrSize(TapError),
+    #[error("Setting MTU failed: {0}")]
+    TapSetMtu(TapError),
     #[error("Enabling tap interface failed: {0}")]
     TapEnable(TapError),
 }
@@ -39,7 +41,7 @@ type Result<T> = std::result::Result<T, Error>;
 fn check_mq_support(if_name: &Option<&str>, queue_pairs: usize) -> Result<()> {
     if let Some(tap_name) = if_name {
         let mq = queue_pairs > 1;
-        let path = format!("/sys/class/net/{}/tun_flags", tap_name);
+        let path = format!("/sys/class/net/{tap_name}/tun_flags");
         // interface does not exist, check is not required
         if !Path::new(&path).exists() {
             return Ok(());
@@ -63,6 +65,7 @@ pub fn open_tap(
     ip_addr: Option<Ipv4Addr>,
     netmask: Option<Ipv4Addr>,
     host_mac: &mut Option<MacAddr>,
+    mtu: Option<u16>,
     num_rx_q: usize,
     flags: Option<i32>,
 ) -> Result<Vec<Tap>> {
@@ -95,6 +98,9 @@ pub fn open_tap(
             } else {
                 *host_mac = Some(tap.get_mac_addr().map_err(Error::TapGetMac)?)
             }
+            if let Some(mtu) = mtu {
+                tap.set_mtu(mtu as i32).map_err(Error::TapSetMtu)?;
+            }
             tap.enable().map_err(Error::TapEnable)?;
 
             tap.set_vnet_hdr_size(vnet_hdr_size)

net_util/src/queue_pair.rs

@@ -83,6 +83,7 @@ impl TxVirtio {
             }
 
             let len = if !iovecs.is_empty() {
+                // SAFETY: FFI call with correct arguments
                 let result = unsafe {
                     libc::writev(
                         tap.as_raw_fd() as libc::c_int,
@@ -104,6 +105,10 @@ impl TxVirtio {
                     return Err(NetQueuePairError::WriteTap(e));
                 }
 
+                if (result as usize) < vnet_hdr_len() {
+                    return Err(NetQueuePairError::InvalidVirtioNetHeader);
+                }
+
                 self.counter_bytes += Wrapping(result as u64 - vnet_hdr_len() as u64);
                 self.counter_frames += Wrapping(1);
 
@@ -185,7 +190,7 @@ impl RxVirtio {
                         .translate_gva(access_platform, desc.len() as usize),
                     10,
                 )
-                .unwrap();
+                .ok_or(NetQueuePairError::DescriptorInvalidHeader)?;
             let mut next_desc = Some(desc);
 
             let mut iovecs = Vec::new();
@@ -217,6 +222,7 @@ impl RxVirtio {
             }
 
             let len = if !iovecs.is_empty() {
+                // SAFETY: FFI call with correct arguments
                 let result = unsafe {
                     libc::readv(
                         tap.as_raw_fd() as libc::c_int,
@@ -238,6 +244,10 @@ impl RxVirtio {
                     return Err(NetQueuePairError::ReadTap(e));
                 }
 
+                if (result as usize) < vnet_hdr_len() {
+                    return Err(NetQueuePairError::InvalidVirtioNetHeader);
+                }
+
                 // Write num_buffers to guest memory. We simply write 1 as we
                 // never spread the frame over more than one descriptor chain.
                 desc_chain
@@ -288,7 +298,7 @@ pub struct NetCounters {
 
 #[derive(Error, Debug)]
 pub enum NetQueuePairError {
-    #[error("No memory configured.")]
+    #[error("No memory configured")]
     NoMemoryConfigured,
     #[error("Error registering listener: {0}")]
     RegisterListener(io::Error),
@@ -302,9 +312,9 @@ pub enum NetQueuePairError {
     GuestMemory(vm_memory::GuestMemoryError),
     #[error("Returned an error while iterating through the queue: {0}")]
     QueueIteratorFailed(virtio_queue::Error),
-    #[error("Descriptor chain is too short.")]
+    #[error("Descriptor chain is too short")]
     DescriptorChainTooShort,
-    #[error("Descriptor chain does not contain valid descriptors.")]
+    #[error("Descriptor chain does not contain valid descriptors")]
     DescriptorChainInvalid,
     #[error("Failed to determine if queue needed notification: {0}")]
     QueueNeedsNotification(virtio_queue::Error),
@@ -312,6 +322,10 @@ pub enum NetQueuePairError {
     QueueEnableNotification(virtio_queue::Error),
     #[error("Failed to add used index to the queue: {0}")]
     QueueAddUsed(virtio_queue::Error),
+    #[error("Descriptor with invalid virtio-net header")]
+    DescriptorInvalidHeader,
+    #[error("Invalid virtio-net header")]
+    InvalidVirtioNetHeader,
 }
 
 pub struct NetQueuePair {