build: Release v28.2 (bug fix release)

Signed-off-by: Bo Chen <chen.bo@intel.com>

.github/workflows/build.yaml

@@ -29,35 +29,27 @@ jobs:
       - name: Install Rust toolchain (${{ matrix.rust }})
         uses: actions-rs/toolchain@v1
         with:
-            toolchain: ${{ matrix.rust }}
-            target: ${{ matrix.target }}
-            override: true
-
-      - name: Debug Check (default features)
-        run: |
-          git rev-list origin/main..$GITHUB_SHA | xargs -t -I % sh -c 'git checkout %; cargo check --tests --all --target=${{ matrix.target }}'
-          git checkout $GITHUB_SHA
+          toolchain: ${{ matrix.rust }}
+          target: ${{ matrix.target }}
+          override: true
 
       - name: Build (default features)
         run: cargo rustc --locked --bin cloud-hypervisor -- -D warnings
 
-      - 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
 
       - 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
-
       - name: Build (default features + guest_debug)
         run: cargo rustc --locked --bin cloud-hypervisor --features "guest_debug" -- -D warnings
 
-      - 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
+
+      - name: Build (mshv + kvm)
+        run: cargo rustc --locked --bin cloud-hypervisor --no-default-features --features "mshv,kvm"  -- -D warnings
 
       - name: Release Build (default features)
         run: cargo build --locked --all --release --target=${{ matrix.target }}

.github/workflows/quality-aarch64.yaml

@@ -1,47 +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: ${{ matrix.target }}
-            override: true
-            components: rustfmt, clippy
-      - name: Formatting (rustfmt)
-        run: cargo fmt -- --check
-
-      - name: Clippy (common + kvm)
-        uses: actions-rs/cargo@v1
-        with:
-          use-cross: true
-          command: clippy
-          args: --target=${{ 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

@@ -14,14 +14,21 @@ jobs:
           - stable
         target:
           - x86_64-unknown-linux-gnu
+          - aarch64-unknown-linux-gnu
         experimental: [false]
         include:
           - rust: beta
             target: x86_64-unknown-linux-gnu
             experimental: true
+          - rust: beta
+            target: aarch64-unknown-linux-gnu
+            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:
@@ -30,26 +37,68 @@ jobs:
           override: true
           components: rustfmt, clippy
 
+      - name: Debug Check (default features)
+        if: ${{ matrix.target == 'x86_64-unknown-linux-gnu' }}
+        run: |
+          set -e
+          commits=$(git rev-list origin/${{ github.base_ref }}..${{ github.sha }})
+          for commit in $commits; do git checkout $commit; cargo check --tests --all --target=${{ 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: --locked --all --all-targets --no-default-features --tests --features "kvm" -- -D warnings
 
       - 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: --locked --all --all-targets --tests -- -D warnings
 
       - 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: --locked --all --all-targets --tests --features "guest_debug" -- -D warnings
 
-      - 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: --locked --all --all-targets --tests --features "tracing" -- -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: --locked --all --all-targets --no-default-features --tests --features "mshv" -- -D warnings
+
+      - name: Clippy (mshv + kvm)
+        if: ${{ matrix.target == 'x86_64-unknown-linux-gnu' }}
+        uses: actions-rs/cargo@v1
+        with:
+          use-cross: ${{ matrix.target != 'x86_64-unknown-linux-gnu' }}
+          command: clippy
+          args: --locked --all --all-targets --no-default-features --tests --features "mshv,kvm" -- -D warnings
+
+      - 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: --locked --all --all-targets --no-default-features --tests --features "tdx,kvm" -- -D warnings
 
       - name: Check build did not modify any files
         run: test -z "$(git status --porcelain)"

.github/workflows/release.yaml

@@ -1,28 +1,28 @@
 name: Cloud Hypervisor Release
-on: [create]
+on: [pull_request, create]
 
 jobs:
   release:
-    if: github.event_name == 'create' && github.event.ref_type == 'tag'
+    if: (github.event_name == 'create' && github.event.ref_type == 'tag') || github.event_name == 'pull_request'
     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-${{ github.event.ref }}
       - name: Install Rust toolchain (x86_64-unknown-linux-gnu)
         uses: actions-rs/toolchain@v1
         with:
-            toolchain: "1.62"
-            target: x86_64-unknown-linux-gnu
+          toolchain: "1.62"
+          target: x86_64-unknown-linux-gnu
       - name: Install Rust toolchain (x86_64-unknown-linux-musl)
         uses: actions-rs/toolchain@v1
         with:
-            toolchain: "1.62"
-            target: x86_64-unknown-linux-musl
+          toolchain: "1.62"
+          target: x86_64-unknown-linux-musl
       - name: Build
         uses: actions-rs/cargo@v1
         with:
@@ -35,28 +35,14 @@ jobs:
             toolchain: "1.62"
             command: build
             args: --all --release --target=x86_64-unknown-linux-musl
-      - name: Strip cloud-hypervisor binaries
-        run: strip target/*/release/cloud-hypervisor
       - name: Install Rust toolchain (aarch64-unknown-linux-musl)
         uses: actions-rs/toolchain@v1
         with:
-            toolchain: "1.62"
-            target: aarch64-unknown-linux-musl
-            override: true
-      - name: Static Build (AArch64)
-        uses: actions-rs/cargo@v1
-        with:
-            use-cross: true
-            command: build
-            args: --all --release --target=aarch64-unknown-linux-musl
-      - name: Vendor
-        working-directory: ../cloud-hypervisor-${{ 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 +52,8 @@ 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 +64,7 @@ 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 +75,7 @@ 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 +86,7 @@ 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 +96,18 @@ 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 +118,7 @@ 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 +128,24 @@ 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

CONTRIBUTING.md

@@ -17,6 +17,23 @@ We follow the [Rust Style](https://github.com/rust-dev-tools/fmt-rfcs/blob/maste
 convention and enforce it through the Continuous Integration (CI) process calling into `rustfmt`
 for each submitted Pull Request (PR).
 
+## Basic Checks
+
+Please consider creating the following hook as `.git/hooks/pre-commit` in order
+to ensure basic correctness of your code. You can extend this further if you
+have specific features that you regularly develop against.
+
+```sh
+#!/bin/sh
+
+cargo fmt -- --check || exit 1
+cargo check --locked --all --all-targets --tests || exit 1
+cargo clippy --locked --all --all-targets --tests -- -D warnings || exit 1
+```
+
+You will need to `chmod +x .git/hooks/pre-commit` to have it run on every
+commit you make.
+
 ## Certificate of Origin
 
 In order to get a clear contribution chain of trust we use the [signed-off-by language](https://01.org/community/signed-process)

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "cloud-hypervisor"
-version = "25.0.0"
+version = "28.2.0"
 authors = ["The Cloud Hypervisor Authors"]
 edition = "2021"
 default-run = "cloud-hypervisor"
@@ -14,54 +14,53 @@ rust-version = "1.60"
 
 [profile.release]
 lto = true
+codegen-units = 1
+opt-level = "s"
+strip = true
 
 [dependencies]
-anyhow = "1.0.58"
+anyhow = "1.0.66"
 api_client = { path = "api_client" }
-clap = { version = "3.2.8", features = ["wrap_help","cargo"] }
+clap = { version = "4.0.29", features = ["wrap_help","cargo","string"] }
 epoll = "4.3.1"
 event_monitor = { path = "event_monitor" }
 hypervisor = { path = "hypervisor" }
-libc = "0.2.126"
+libc = "0.2.138"
 log = { version = "0.4.17", features = ["std"] }
 option_parser = { path = "option_parser" }
-seccompiler = "0.2.0"
-serde_json = "1.0.82"
+seccompiler = "0.3.0"
+serde_json = "1.0.89"
 signal-hook = "0.3.14"
-thiserror = "1.0.31"
+thiserror = "1.0.37"
+tpm = { path = "tpm"}
+tracer = { path = "tracer" }
 vmm = { path = "vmm" }
-vmm-sys-util = "0.9.0"
-vm-memory = "0.8.0"
+vmm-sys-util = "0.11.0"
+vm-memory = "0.10.0"
 
 [build-dependencies]
-clap = { version = "3.2.8", features = ["cargo"] }
+clap = { version = "4.0.29", features = ["cargo"] }
 
 # 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 = "1.13.0"
-serde_json = "1.0.82"
+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 +78,9 @@ members = [
     "performance-metrics",
     "qcow",
     "rate_limiter",
+    "serial_buffer",
     "test_infra",
+    "tracer",
     "vfio_user",
     "vhdx",
     "vhost_user_block",

Jenkinsfile

@@ -1,413 +1,333 @@
 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 cargo files only changes') {
-					when {
-						expression {
-							return fuzzCargoFileOnly()
-						}
-					}
-					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 'focal' } }
-					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
-						}
-					}
-					stages {
-						stage ('Checkout') {
-							steps {
-								checkout scm
-							}
-						}
-						stage ('Run unit tests') {
-							steps {
-								sh "scripts/dev_cli.sh tests --unit --libc musl"
-							}
-						}
-						stage ('Run integration tests') {
-							options {
-								timeout(time: 1, unit: 'HOURS')
-							}
-							steps {
-								sh "sudo modprobe openvswitch"
-								sh "scripts/dev_cli.sh tests --integration --libc musl"
-							}
-						}
-					}
-					post {
-						always {
-							sh "sudo chown -R jenkins.jenkins ${WORKSPACE}"
-							deleteDir()
-						}
-					}
-				}
-				stage ('Worker build (musl)') {
-					agent { node { label 'focal' } }
-					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 'focal' } }
-					when {
-						beforeAgent true
-						expression {
-							return runWorkers
-						}
-					}
-					environment {
-        					AZURE_CONNECTION_STRING = credentials('46b4e7d6-315f-4cc1-8333-b58780863b9b')
-					}
-					stages {
-						stage ('Checkout') {
-							steps {
-								checkout scm
-							}
-						}
-						stage ('Install azure-cli') {
-							steps {
-								installAzureCli()
-							}
-						}
-						stage ('Download assets') {
-							steps {
-								sh "mkdir ${env.HOME}/workloads"
-								sh 'az storage blob download --container-name private-images --file "$HOME/workloads/windows-server-2019.raw" --name windows-server-2019.raw --connection-string "$AZURE_CONNECTION_STRING"'
-							}
-						}
-						stage ('Run Windows guest integration tests') {
-							options {
-								timeout(time: 1, unit: 'HOURS')
-							}
-							steps {
-								sh "scripts/dev_cli.sh tests --integration-windows"
-							}
-						}
-						stage ('Run Windows guest integration tests for musl') {
-							options {
-								timeout(time: 1, unit: 'HOURS')
-							}
-							steps {
-								sh "scripts/dev_cli.sh tests --integration-windows --libc musl"
-							}
-						}
-					}
-				}
-				stage ('Worker build - Live Migration') {
-					agent { node { label 'focal-small' } }
-					when {
-						beforeAgent true
-						expression {
-							return runWorkers
-						}
-					}
-					stages {
-						stage ('Checkout') {
-							steps {
-								checkout scm
-							}
-						}
-						stage ('Run live-migration integration tests') {
-							options {
-								timeout(time: 1, unit: 'HOURS')
-							}
-							steps {
-								sh "sudo modprobe openvswitch"
-								sh "scripts/dev_cli.sh tests --integration-live-migration"
-							}
-						}
-						stage ('Run live-migration integration tests for musl') {
-							options {
-								timeout(time: 1, unit: 'HOURS')
-							}
-							steps {
-								sh "sudo modprobe openvswitch"
-								sh "scripts/dev_cli.sh tests --integration-live-migration --libc musl"
-							}
-						}
-					}
-				}
-				stage ('Worker build - Metrics') {
-					agent { node { label 'focal-metrics' } }
-					when {
-						branch 'main'
-						beforeAgent true
-						expression {
-							return runWorkers
-						}
-					}
-					environment {
-						METRICS_PUBLISH_KEY = credentials('52e0945f-ce7a-43d1-87af-67d1d87cc40f')
-					}
-					stages {
-						stage ('Checkout') {
-							steps {
-								checkout scm
-							}
-						}
-						stage ('Run metrics tests') {
-							options {
-								timeout(time: 1, unit: 'HOURS')
-							}
-							steps {
-								sh 'scripts/dev_cli.sh tests --metrics -- -- --report-file /root/workloads/metrics.json'
-							}
-						}
-						stage ('Upload metrics report') {
-							steps {
-								sh 'curl -X PUT https://cloud-hypervisor-metrics.azurewebsites.net/api/publishmetrics -H "x-functions-key: $METRICS_PUBLISH_KEY" -T ~/workloads/metrics.json'
-							}
-						}
-					}
-				}
-			}
-		}
-	}
-	post {
-		regression {
-			script {
-				if (env.BRANCH_NAME == 'main') {
-					slackSend (color: '#ff0000', message: '"main" branch build is now failing')
-				}
-			}
-		}
-		fixed {
-			script {
-				if (env.BRANCH_NAME == 'main') {
-					slackSend (color: '#00ff00', message: '"main" branch build is now fixed')
-				}
-			}
-		}
-	}
+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('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'
+                            }
+                        }
+                    }
+                }
+            }
+        }
+    }
+    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() {
-	sh "sudo apt install -y ca-certificates curl apt-transport-https lsb-release gnupg"
-	sh "curl -sL https://packages.microsoft.com/keys/microsoft.asc | gpg --dearmor | sudo tee /etc/apt/trusted.gpg.d/microsoft.gpg > /dev/null"
-	sh "echo \"deb [arch=amd64] https://packages.microsoft.com/repos/azure-cli/ focal main\" | sudo tee /etc/apt/sources.list.d/azure-cli.list"
-	sh "sudo apt update"
-	sh "sudo apt install -y azure-cli"
+def installAzureCli(distro, arch) {
+    sh 'sudo apt install -y ca-certificates curl apt-transport-https lsb-release gnupg'
+    sh 'curl -sL https://packages.microsoft.com/keys/microsoft.asc | gpg --dearmor | sudo tee /etc/apt/trusted.gpg.d/microsoft.gpg > /dev/null'
+    sh "echo \"deb [arch=${arch}] https://packages.microsoft.com/repos/azure-cli/ ${distro} main\" | sudo tee /etc/apt/sources.list.d/azure-cli.list"
+    sh 'sudo apt update'
+    sh 'sudo apt install -y azure-cli'
 }
 
 def boolean docsFileOnly() {
     if (env.CHANGE_TARGET == null) {
-        return false;
+        return false
     }
 
     return sh(
@@ -416,13 +336,13 @@ def boolean docsFileOnly() {
     ) != 0
 }
 
-def boolean fuzzCargoFileOnly() {
+def boolean fuzzFileOnly() {
     if (env.CHANGE_TARGET == null) {
-        return false;
+        return false
     }
 
     return sh(
         returnStatus: true,
-        script: "git diff --name-only origin/${env.CHANGE_TARGET}... | grep -v -E 'fuzz\\/Cargo.(toml|lock)'"
+        script: "git diff --name-only origin/${env.CHANGE_TARGET}... | grep -v -E 'fuzz/'"
     ) != 0
 }

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
@@ -68,142 +69,121 @@ 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 on the
+`x86-64` platform. For getting started on the `AArch64` platform, please refer
+to the
+[AArch64 documentation](docs/arm64.md).
 
-## Preparation
+## Host OS
 
-We create a folder to build and run `cloud-hypervisor` at `$HOME/cloud-hypervisor`
+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
+
+The instructions below are for the `x86-64` platform. For `AArch64` please see
+the [AArch64 specific documentation](docs/arm64.md).
+
+Cloud Hypervisor supports direct kernel boot (if the kernel is 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`.)
+
+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 $CLOUDH
-```
-
-## Install prerequisites
-
-You need to install some prerequisite packages in order to build and test Cloud
-Hypervisor. Here, all the steps are based on Ubuntu, for other Linux
-distributions please replace the package manager and package name.
-
-```shell
-# Install build-essential, git, and qemu-img
-$ sudo apt install git build-essential qemu-img
-# 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/0.4.0/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 into a `vmlinux` ELF kernel (compiled with PVH support). 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 $CLOUDH/cloud-hypervisor/resources/linux-config-x86_64 .config
+$ wget https://raw.githubusercontent.com/cloud-hypervisor/cloud-hypervisor/main/resources/linux-config-x86_64
+$ cp linux-config-x86_64 .config
 $ KCFLAGS="-Wa,-mx86-used-note=no" make bzImage -j `nproc`
 $ popd
 ```
@@ -213,42 +193,34 @@ The `vmlinux` kernel image will then be located at
 
 #### 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
 ```
 
 #### 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.
 
 ```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.
+If earlier kernel messages are required the serial console should be used instead of `virtio-console`.
 
-When in need for earlier debug messages, using the legacy serial device based
-console is preferred:
-
-```
-$ ./cloud-hypervisor/target/release/cloud-hypervisor \
+```./cloud-hypervisor \
 	--kernel ./linux-cloud-hypervisor/arch/x86/boot/compressed/vmlinux.bin \
 	--console off \
 	--serial tty \
@@ -261,8 +233,8 @@ $ ./cloud-hypervisor/target/release/cloud-hypervisor \
 
 # 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 +252,18 @@ 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 2022-10-13, the following cloud images are supported:
+
+- [Ubuntu Bionic](https://cloud-images.ubuntu.com/bionic/current/) (bionic-server-cloudimg-amd64.img)
+- [Ubuntu Focal](https://cloud-images.ubuntu.com/focal/current/) (focal-server-cloudimg-amd64.img)
+- [Ubuntu Jammy](https://cloud-images.ubuntu.com/jammy/current/) (jammy-server-cloudimg-amd64.img )
+- [Fedora 36](https://fedora.mirrorservice.org/fedora/linux/releases/36/Cloud/x86_64/images/) (Fedora-Cloud-Base-36-1.5.x86_64.raw.xz)
 
 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 +276,12 @@ 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 +290,26 @@ 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 +319,28 @@ 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

@@ -4,6 +4,7 @@
 //
 
 #[repr(packed)]
+#[derive(Clone, Copy)]
 pub struct GenericAddress {
     pub address_space_id: u8,
     pub register_bit_width: u8,

api_client/Cargo.toml

@@ -5,4 +5,4 @@ authors = ["The Cloud Hypervisor Authors"]
 edition = "2021"
 
 [dependencies]
-vmm-sys-util = "0.9.0"
+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()])
 }
@@ -101,6 +101,10 @@ fn parse_http_response(socket: &mut dyn Read) -> Result<Option<String>, Error> {
     loop {
         let mut bytes = vec![0; 256];
         let count = socket.read(&mut bytes).map_err(Error::Socket)?;
+        // If the return value is 0, the peer has performed an orderly shutdown.
+        if count == 0 {
+            break;
+        }
         res.push_str(std::str::from_utf8(&bytes[0..count]).unwrap());
 
         // End of headers
@@ -127,7 +131,7 @@ fn parse_http_response(socket: &mut dyn Read) -> Result<Option<String>, Error> {
             }
         }
     }
-    let body_string = content_length.and(Some(String::from(&res[body_offset.unwrap()..])));
+    let body_string = content_length.and(body_offset.map(|o| String::from(&res[o..])));
     let status_code = get_status_code(&res)?;
 
     if status_code.is_server_error() {
@@ -137,18 +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,
@@ -171,12 +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

@@ -9,21 +9,21 @@ default = []
 tdx = []
 
 [dependencies]
-acpi_tables = { path = "../acpi_tables" }
-anyhow = "1.0.58"
+anyhow = "1.0.66"
 byteorder = "1.4.3"
 hypervisor = { path = "../hypervisor" }
-libc = "0.2.126"
-linux-loader = { version = "0.4.0", features = ["elf", "bzimage", "pe"] }
+libc = "0.2.138"
+linux-loader = { version = "0.8.1", features = ["elf", "bzimage", "pe"] }
 log = "0.4.17"
-serde = { version = "1.0.138", features = ["rc", "derive"] }
-thiserror = "1.0.31"
-versionize = "0.1.6"
+serde = { version = "1.0.150", features = ["rc", "derive"] }
+thiserror = "1.0.37"
+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 = "0.9.0", features = ["with-serde"] }
+vmm-sys-util = { version = "0.11.0", features = ["with-serde"] }
 
 [target.'cfg(target_arch = "aarch64")'.dependencies]
-fdt_parser = { version = "0.1.3", package = 'fdt'}
+fdt_parser = { version = "0.1.4", package = "fdt" }
 vm-fdt = { git = "https://github.com/rust-vmm/vm-fdt", branch = "main" }

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)?;
@@ -541,16 +539,9 @@ 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 +670,7 @@ 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

@@ -6,6 +6,8 @@
 pub mod fdt;
 /// Layout for this aarch64 system.
 pub mod layout;
+/// Module for system registers definition
+pub mod regs;
 /// Module for loading UEFI binary.
 pub mod uefi;
 
@@ -73,7 +75,9 @@ pub fn configure_vcpu(
         .map_err(Error::RegsConfiguration)?;
     }
 
-    let mpidr = vcpu.read_mpidr().map_err(Error::VcpuRegMpidr)?;
+    let mpidr = vcpu
+        .get_sys_reg(regs::MPIDR_EL1)
+        .map_err(Error::VcpuRegMpidr)?;
     Ok(mpidr)
 }
 

arch/src/aarch64/regs.rs

@@ -0,0 +1,44 @@
+// Copyright 2022 Arm Limited (or its affiliates). All rights reserved.
+// SPDX-License-Identifier: Apache-2.0
+
+///
+/// AArch64 system register encoding:
+/// See https://developer.arm.com/documentation/ddi0487 (chapter D12)
+///
+///   31      22  21 20 19 18 16 15 12 11  8 7   5 4  0
+///  +----------+---+-----+-----+-----+-----+-----+----+
+///  |1101010100| L | op0 | op1 | CRn | CRm | op2 | Rt |
+///  +----------+---+-----+-----+-----+-----+-----+----+
+///
+/// Notes:
+/// - L and Rt are reserved as implementation defined fields, ignored.
+///
+const SYSREG_HEAD: u32 = 0b1101010100u32 << 22;
+const SYSREG_OP0_SHIFT: u32 = 19;
+const SYSREG_OP0_MASK: u32 = 0b11u32 << 19;
+const SYSREG_OP1_SHIFT: u32 = 16;
+const SYSREG_OP1_MASK: u32 = 0b111u32 << 16;
+const SYSREG_CRN_SHIFT: u32 = 12;
+const SYSREG_CRN_MASK: u32 = 0b1111u32 << 12;
+const SYSREG_CRM_SHIFT: u32 = 8;
+const SYSREG_CRM_MASK: u32 = 0b1111u32 << 8;
+const SYSREG_OP2_SHIFT: u32 = 5;
+const SYSREG_OP2_MASK: u32 = 0b111u32 << 5;
+
+/// Define the ID of system registers
+#[macro_export]
+macro_rules! arm64_sys_reg {
+    ($name: tt, $op0: tt, $op1: tt, $crn: tt, $crm: tt, $op2: tt) => {
+        pub const $name: u32 = SYSREG_HEAD
+            | ((($op0 as u32) << SYSREG_OP0_SHIFT) & SYSREG_OP0_MASK as u32)
+            | ((($op1 as u32) << SYSREG_OP1_SHIFT) & SYSREG_OP1_MASK as u32)
+            | ((($crn as u32) << SYSREG_CRN_SHIFT) & SYSREG_CRN_MASK as u32)
+            | ((($crm as u32) << SYSREG_CRM_SHIFT) & SYSREG_CRM_MASK as u32)
+            | ((($op2 as u32) << SYSREG_OP2_SHIFT) & SYSREG_OP2_MASK as u32);
+    };
+}
+
+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

@@ -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/interrupts.rs

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

arch/src/x86_64/layout.rs

@@ -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

@@ -15,7 +15,7 @@ pub mod regs;
 use crate::GuestMemoryMmap;
 use crate::InitramfsConfig;
 use crate::RegionType;
-use hypervisor::x86_64::{CpuId, CpuIdEntry, CPUID_FLAG_VALID_INDEX};
+use hypervisor::arch::x86::{CpuIdEntry, CPUID_FLAG_VALID_INDEX};
 use hypervisor::HypervisorError;
 use linux_loader::loader::bootparam::boot_params;
 use linux_loader::loader::elf::start_info::{
@@ -202,7 +202,7 @@ impl From<Error> for super::Error {
     }
 }
 
-#[allow(dead_code, clippy::upper_case_acronyms)]
+#[allow(clippy::upper_case_acronyms)]
 #[derive(Copy, Clone, Debug)]
 pub enum CpuidReg {
     EAX,
@@ -223,17 +223,15 @@ pub struct CpuidPatch {
 
 impl CpuidPatch {
     pub fn set_cpuid_reg(
-        cpuid: &mut CpuId,
+        cpuid: &mut Vec<CpuIdEntry>,
         function: u32,
         index: Option<u32>,
         reg: CpuidReg,
         value: u32,
     ) {
-        let entries = cpuid.as_mut_slice();
-
         let mut entry_found = false;
-        for entry in entries.iter_mut() {
-            if entry.function == function && (index == None || index.unwrap() == entry.index) {
+        for entry in cpuid.iter_mut() {
+            if entry.function == function && (index.is_none() || index.unwrap() == entry.index) {
                 entry_found = true;
                 match reg {
                     CpuidReg::EAX => {
@@ -279,16 +277,12 @@ impl CpuidPatch {
                 }
             }
 
-            if let Err(e) = cpuid.push(entry) {
-                error!("Failed adding new CPUID entry: {:?}", e);
-            }
+            cpuid.push(entry);
         }
     }
 
-    pub fn patch_cpuid(cpuid: &mut CpuId, patches: Vec<CpuidPatch>) {
-        let entries = cpuid.as_mut_slice();
-
-        for entry in entries.iter_mut() {
+    pub fn patch_cpuid(cpuid: &mut [CpuIdEntry], patches: Vec<CpuidPatch>) {
+        for entry in cpuid {
             for patch in patches.iter() {
                 if entry.function == patch.function && entry.index == patch.index {
                     if let Some(flags_bit) = patch.flags_bit {
@@ -312,16 +306,15 @@ impl CpuidPatch {
     }
 
     pub fn is_feature_enabled(
-        cpuid: &CpuId,
+        cpuid: &[CpuIdEntry],
         function: u32,
         index: u32,
         reg: CpuidReg,
         feature_bit: usize,
     ) -> bool {
-        let entries = cpuid.as_slice();
         let mask = 1 << feature_bit;
 
-        for entry in entries.iter() {
+        for entry in cpuid {
             if entry.function == function && entry.index == index {
                 let reg_val = match reg {
                     CpuidReg::EAX => entry.eax,
@@ -470,12 +463,12 @@ impl CpuidFeatureEntry {
     }
 
     fn get_features_from_cpuid(
-        cpuid: &CpuId,
+        cpuid: &[CpuIdEntry],
         feature_entry_list: &[CpuidFeatureEntry],
     ) -> Vec<u32> {
         let mut features = vec![0; feature_entry_list.len()];
         for (i, feature_entry) in feature_entry_list.iter().enumerate() {
-            for cpuid_entry in cpuid.as_slice().iter() {
+            for cpuid_entry in cpuid {
                 if cpuid_entry.function == feature_entry.function
                     && cpuid_entry.index == feature_entry.index
                 {
@@ -505,8 +498,8 @@ impl CpuidFeatureEntry {
     // The function returns `Error` (a.k.a. "incompatible"), when the CPUID features from `src_vm_cpuid`
     // is not a subset of those of the `dest_vm_cpuid`.
     pub fn check_cpuid_compatibility(
-        src_vm_cpuid: &CpuId,
-        dest_vm_cpuid: &CpuId,
+        src_vm_cpuid: &[CpuIdEntry],
+        dest_vm_cpuid: &[CpuIdEntry],
     ) -> Result<(), Error> {
         let feature_entry_list = &Self::checked_feature_entry_list();
         let src_vm_features = Self::get_features_from_cpuid(src_vm_cpuid, feature_entry_list);
@@ -558,7 +551,7 @@ pub fn generate_common_cpuid(
     phys_bits: u8,
     kvm_hyperv: bool,
     #[cfg(feature = "tdx")] tdx_enabled: bool,
-) -> super::Result<CpuId> {
+) -> super::Result<Vec<CpuIdEntry>> {
     let cpuid_patches = vec![
         // Patch tsc deadline timer bit
         CpuidPatch {
@@ -638,6 +631,20 @@ 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);
@@ -669,16 +676,14 @@ pub fn generate_common_cpuid(
     for i in 0x8000_0002..=0x8000_0004 {
         cpuid.retain(|c| c.function != i);
         let leaf = unsafe { std::arch::x86_64::__cpuid(i) };
-        cpuid
-            .push(CpuIdEntry {
-                function: i,
-                eax: leaf.eax,
-                ebx: leaf.ebx,
-                ecx: leaf.ecx,
-                edx: leaf.edx,
-                ..Default::default()
-            })
-            .map_err(Error::CpuidIdentification)?;
+        cpuid.push(CpuIdEntry {
+            function: i,
+            eax: leaf.eax,
+            ebx: leaf.ebx,
+            ecx: leaf.ecx,
+            edx: leaf.edx,
+            ..Default::default()
+        });
     }
 
     if kvm_hyperv {
@@ -687,56 +692,44 @@ pub fn generate_common_cpuid(
         cpuid.retain(|c| c.function != 0x4000_0001);
         // See "Hypervisor Top Level Functional Specification" for details
         // Compliance with "Hv#1" requires leaves up to 0x4000_000a
-        cpuid
-            .push(CpuIdEntry {
-                function: 0x40000000,
-                eax: 0x4000000a, // Maximum cpuid leaf
-                ebx: 0x756e694c, // "Linu"
-                ecx: 0x564b2078, // "x KV"
-                edx: 0x7648204d, // "M Hv"
-                ..Default::default()
-            })
-            .map_err(Error::CpuidKvmHyperV)?;
-        cpuid
-            .push(CpuIdEntry {
-                function: 0x40000001,
-                eax: 0x31237648, // "Hv#1"
-                ..Default::default()
-            })
-            .map_err(Error::CpuidKvmHyperV)?;
-        cpuid
-            .push(CpuIdEntry {
-                function: 0x40000002,
-                eax: 0x3839,  // "Build number"
-                ebx: 0xa0000, // "Version"
-                ..Default::default()
-            })
-            .map_err(Error::CpuidKvmHyperV)?;
-        cpuid
-            .push(CpuIdEntry {
-                function: 0x4000_0003,
-                eax: 1 << 1 // AccessPartitionReferenceCounter
+        cpuid.push(CpuIdEntry {
+            function: 0x40000000,
+            eax: 0x4000000a, // Maximum cpuid leaf
+            ebx: 0x756e694c, // "Linu"
+            ecx: 0x564b2078, // "x KV"
+            edx: 0x7648204d, // "M Hv"
+            ..Default::default()
+        });
+        cpuid.push(CpuIdEntry {
+            function: 0x40000001,
+            eax: 0x31237648, // "Hv#1"
+            ..Default::default()
+        });
+        cpuid.push(CpuIdEntry {
+            function: 0x40000002,
+            eax: 0x3839,  // "Build number"
+            ebx: 0xa0000, // "Version"
+            ..Default::default()
+        });
+        cpuid.push(CpuIdEntry {
+            function: 0x4000_0003,
+            eax: 1 << 1 // AccessPartitionReferenceCounter
                    | 1 << 2 // AccessSynicRegs
                    | 1 << 3 // AccessSyntheticTimerRegs
                    | 1 << 9, // AccessPartitionReferenceTsc
-                edx: 1 << 3, // CPU dynamic partitioning
-                ..Default::default()
-            })
-            .map_err(Error::CpuidKvmHyperV)?;
-        cpuid
-            .push(CpuIdEntry {
-                function: 0x4000_0004,
-                eax: 1 << 5, // Recommend relaxed timing
-                ..Default::default()
-            })
-            .map_err(Error::CpuidKvmHyperV)?;
+            edx: 1 << 3, // CPU dynamic partitioning
+            ..Default::default()
+        });
+        cpuid.push(CpuIdEntry {
+            function: 0x4000_0004,
+            eax: 1 << 5, // Recommend relaxed timing
+            ..Default::default()
+        });
         for i in 0x4000_0005..=0x4000_000a {
-            cpuid
-                .push(CpuIdEntry {
-                    function: i,
-                    ..Default::default()
-                })
-                .map_err(Error::CpuidKvmHyperV)?;
+            cpuid.push(CpuIdEntry {
+                function: i,
+                ..Default::default()
+            });
         }
     }
 
@@ -748,7 +741,7 @@ pub fn configure_vcpu(
     id: u8,
     kernel_entry_point: Option<EntryPoint>,
     vm_memory: &GuestMemoryAtomic<GuestMemoryMmap>,
-    cpuid: CpuId,
+    cpuid: Vec<CpuIdEntry>,
     kvm_hyperv: bool,
 ) -> super::Result<()> {
     // Per vCPU CPUID changes; common are handled via generate_common_cpuid()
@@ -785,11 +778,11 @@ 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 {
@@ -840,13 +833,16 @@ pub fn configure_system(
     rsdp_addr: Option<GuestAddress>,
     sgx_epc_region: Option<SgxEpcRegion>,
     serial_number: Option<&str>,
+    uuid: Option<&str>,
+    oem_strings: Option<&[&str]>,
 ) -> super::Result<()> {
     // Write EBDA address to location where ACPICA expects to find it
     guest_mem
         .write_obj((layout::EBDA_START.0 >> 4) as u16, layout::EBDA_POINTER)
         .map_err(Error::EbdaSetup)?;
 
-    let size = smbios::setup_smbios(guest_mem, serial_number).map_err(Error::SmbiosSetup)?;
+    let size = smbios::setup_smbios(guest_mem, serial_number, uuid, oem_strings)
+        .map_err(Error::SmbiosSetup)?;
 
     // Place the MP table after the SMIOS table aligned to 16 bytes
     let offset = GuestAddress(layout::SMBIOS_START).unchecked_add(size);
@@ -883,7 +879,7 @@ 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 {
@@ -952,7 +948,7 @@ 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,
         );
     }
@@ -1056,7 +1052,7 @@ pub fn get_host_cpu_phys_bits() -> u8 {
 }
 
 fn update_cpuid_topology(
-    cpuid: &mut CpuId,
+    cpuid: &mut Vec<CpuIdEntry>,
     threads_per_core: u8,
     cores_per_die: u8,
     dies_per_package: u8,
@@ -1120,7 +1116,10 @@ fn update_cpuid_topology(
 
 // The goal is to update the CPUID sub-leaves to reflect the number of EPC
 // sections exposed to the guest.
-fn update_cpuid_sgx(cpuid: &mut CpuId, epc_sections: Vec<SgxEpcSection>) -> Result<(), Error> {
+fn update_cpuid_sgx(
+    cpuid: &mut Vec<CpuIdEntry>,
+    epc_sections: Vec<SgxEpcSection>,
+) -> Result<(), Error> {
     // Something's wrong if there's no EPC section.
     if epc_sections.is_empty() {
         return Err(Error::NoSgxEpcSection);
@@ -1141,7 +1140,7 @@ fn update_cpuid_sgx(cpuid: &mut CpuId, epc_sections: Vec<SgxEpcSection>) -> Resu
     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);
@@ -1196,6 +1195,8 @@ mod tests {
             Some(layout::RSDP_POINTER),
             None,
             None,
+            None,
+            None,
         );
         assert!(config_err.is_err());
 
@@ -1209,7 +1210,18 @@ mod tests {
             .collect();
         let gm = GuestMemoryMmap::from_ranges(&ram_regions).unwrap();
 
-        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None, None).unwrap();
+        configure_system(
+            &gm,
+            GuestAddress(0),
+            &None,
+            no_vcpus,
+            None,
+            None,
+            None,
+            None,
+            None,
+        )
+        .unwrap();
 
         // Now assigning some memory that is equal to the start of the 32bit memory hole.
         let mem_size = 3328 << 20;
@@ -1220,9 +1232,31 @@ mod tests {
             .map(|r| (r.0, r.1))
             .collect();
         let gm = GuestMemoryMmap::from_ranges(&ram_regions).unwrap();
-        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None, None).unwrap();
+        configure_system(
+            &gm,
+            GuestAddress(0),
+            &None,
+            no_vcpus,
+            None,
+            None,
+            None,
+            None,
+            None,
+        )
+        .unwrap();
 
-        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None, None).unwrap();
+        configure_system(
+            &gm,
+            GuestAddress(0),
+            &None,
+            no_vcpus,
+            None,
+            None,
+            None,
+            None,
+            None,
+        )
+        .unwrap();
 
         // Now assigning some memory that falls after the 32bit memory hole.
         let mem_size = 3330 << 20;
@@ -1233,9 +1267,31 @@ mod tests {
             .map(|r| (r.0, r.1))
             .collect();
         let gm = GuestMemoryMmap::from_ranges(&ram_regions).unwrap();
-        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None, None).unwrap();
+        configure_system(
+            &gm,
+            GuestAddress(0),
+            &None,
+            no_vcpus,
+            None,
+            None,
+            None,
+            None,
+            None,
+        )
+        .unwrap();
 
-        configure_system(&gm, GuestAddress(0), &None, no_vcpus, None, None, None).unwrap();
+        configure_system(
+            &gm,
+            GuestAddress(0),
+            &None,
+            no_vcpus,
+            None,
+            None,
+            None,
+            None,
+            None,
+        )
+        .unwrap();
     }
 
     #[test]
@@ -1260,6 +1316,6 @@ 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

@@ -294,7 +294,7 @@ mod tests {
             mpspec::MP_IOAPIC => mem::size_of::<MpcIoapicWrapper>(),
             mpspec::MP_INTSRC => mem::size_of::<MpcIntsrcWrapper>(),
             mpspec::MP_LINTSRC => mem::size_of::<MpcLintsrcWrapper>(),
-            _ => panic!("unrecognized mpc table entry type: {}", type_),
+            _ => panic!("unrecognized mpc table entry type: {type_}"),
         }
     }
 

arch/src/x86_64/regs.rs

@@ -10,7 +10,7 @@ use crate::layout::{BOOT_GDT_START, BOOT_IDT_START, PVH_INFO_START};
 use crate::GuestMemoryMmap;
 use hypervisor::arch::x86::gdt::{gdt_entry, segment_from_gdt};
 use hypervisor::arch::x86::regs::CR0_PE;
-use hypervisor::x86_64::{FpuState, SpecialRegisters, StandardRegisters};
+use hypervisor::arch::x86::{FpuState, SpecialRegisters, StandardRegisters};
 use std::sync::Arc;
 use std::{mem, result};
 use vm_memory::{Address, Bytes, GuestMemory, GuestMemoryError};
@@ -66,7 +66,7 @@ pub fn setup_fpu(vcpu: &Arc<dyn hypervisor::Vcpu>) -> Result<()> {
 ///
 /// * `vcpu` - Structure for the VCPU that holds the VCPU's fd.
 pub fn setup_msrs(vcpu: &Arc<dyn hypervisor::Vcpu>) -> Result<()> {
-    vcpu.set_msrs(&hypervisor::x86_64::boot_msr_entries())
+    vcpu.set_msrs(&vcpu.boot_msr_entries())
         .map_err(Error::SetModelSpecificRegisters)?;
 
     Ok(())
@@ -124,7 +124,7 @@ pub fn configure_segments_and_sregs(
     mem: &GuestMemoryMmap,
     sregs: &mut SpecialRegisters,
 ) -> Result<()> {
-    let gdt_table: [u64; BOOT_GDT_MAX 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

@@ -12,10 +12,10 @@ use std::fmt::{self, Display};
 use std::mem;
 use std::result;
 use std::slice;
+use uuid::Uuid;
 use vm_memory::ByteValued;
 use vm_memory::{Address, Bytes, GuestAddress};
 
-#[allow(unused_variables)]
 #[derive(Debug)]
 pub enum Error {
     /// There was too little guest memory to store the entire SMBIOS table.
@@ -28,6 +28,8 @@ pub enum Error {
     WriteSmbiosEp,
     /// Failure to write additional data to memory
     WriteData,
+    /// Failure to parse uuid, uuid format may be error
+    ParseUuid(uuid::Error),
 }
 
 impl std::error::Error for Error {}
@@ -37,14 +39,19 @@ impl Display for Error {
         use self::Error::*;
 
         let description = match self {
-            NotEnoughMemory => "There was too little guest memory to store the SMBIOS table",
-            AddressOverflow => "The SMBIOS table has too little address space to be stored",
-            Clear => "Failure while zeroing out the memory for the SMBIOS table",
-            WriteSmbiosEp => "Failure to write SMBIOS entrypoint structure",
-            WriteData => "Failure to write additional data to memory",
+            NotEnoughMemory => {
+                "There was too little guest memory to store the SMBIOS table".to_string()
+            }
+            AddressOverflow => {
+                "The SMBIOS table has too little address space to be stored".to_string()
+            }
+            Clear => "Failure while zeroing out the memory for the SMBIOS table".to_string(),
+            WriteSmbiosEp => "Failure to write SMBIOS entrypoint structure".to_string(),
+            WriteData => "Failure to write additional data to memory".to_string(),
+            ParseUuid(e) => format!("Failure to parse uuid: {e}"),
         };
 
-        write!(f, "SMBIOS error: {}", description)
+        write!(f, "SMBIOS error: {description}")
     }
 }
 
@@ -54,6 +61,7 @@ pub type Result<T> = result::Result<T, Error>;
 const SM3_MAGIC_IDENT: &[u8; 5usize] = b"_SM3_";
 const BIOS_INFORMATION: u8 = 0;
 const SYSTEM_INFORMATION: u8 = 1;
+const OEM_STRINGS: u8 = 11;
 const END_OF_TABLE: u8 = 127;
 const PCI_SUPPORTED: u64 = 1 << 7;
 const IS_VIRTUAL_MACHINE: u8 = 1 << 4;
@@ -68,75 +76,81 @@ fn compute_checksum<T: Copy>(v: &T) -> u8 {
     (!checksum).wrapping_add(1)
 }
 
+#[repr(C)]
 #[repr(packed)]
-#[derive(Default, Copy)]
-pub struct Smbios30Entrypoint {
-    pub signature: [u8; 5usize],
-    pub checksum: u8,
-    pub length: u8,
-    pub majorver: u8,
-    pub minorver: u8,
-    pub docrev: u8,
-    pub revision: u8,
-    pub reserved: u8,
-    pub max_size: u32,
-    pub physptr: u64,
-}
-
-impl Clone for Smbios30Entrypoint {
-    fn clone(&self) -> Self {
-        *self
-    }
+#[derive(Default, Copy, Clone)]
+struct Smbios30Entrypoint {
+    signature: [u8; 5usize],
+    checksum: u8,
+    length: u8,
+    majorver: u8,
+    minorver: u8,
+    docrev: u8,
+    revision: u8,
+    reserved: u8,
+    max_size: u32,
+    physptr: u64,
 }
 
+#[repr(C)]
 #[repr(packed)]
-#[derive(Default, Copy)]
-pub struct SmbiosBiosInfo {
-    pub typ: u8,
-    pub length: u8,
-    pub handle: u16,
-    pub vendor: u8,
-    pub version: u8,
-    pub start_addr: u16,
-    pub release_date: u8,
-    pub rom_size: u8,
-    pub characteristics: u64,
-    pub characteristics_ext1: u8,
-    pub characteristics_ext2: u8,
-}
-
-impl Clone for SmbiosBiosInfo {
-    fn clone(&self) -> Self {
-        *self
-    }
+#[derive(Default, Copy, Clone)]
+struct SmbiosBiosInfo {
+    r#type: u8,
+    length: u8,
+    handle: u16,
+    vendor: u8,
+    version: u8,
+    start_addr: u16,
+    release_date: u8,
+    rom_size: u8,
+    characteristics: u64,
+    characteristics_ext1: u8,
+    characteristics_ext2: u8,
 }
 
+#[repr(C)]
 #[repr(packed)]
-#[derive(Default, Copy)]
-pub struct SmbiosSysInfo {
-    pub typ: u8,
-    pub length: u8,
-    pub handle: u16,
-    pub manufacturer: u8,
-    pub product_name: u8,
-    pub version: u8,
-    pub serial_number: u8,
-    pub uuid: [u8; 16usize],
-    pub wake_up_type: u8,
-    pub sku: u8,
-    pub family: u8,
+#[derive(Default, Copy, Clone)]
+struct SmbiosSysInfo {
+    r#type: u8,
+    length: u8,
+    handle: u16,
+    manufacturer: u8,
+    product_name: u8,
+    version: u8,
+    serial_number: u8,
+    uuid: [u8; 16usize],
+    wake_up_type: u8,
+    sku: u8,
+    family: u8,
 }
 
-impl Clone for SmbiosSysInfo {
-    fn clone(&self) -> Self {
-        *self
-    }
+#[repr(C)]
+#[repr(packed)]
+#[derive(Default, Copy, Clone)]
+struct SmbiosOemStrings {
+    r#type: u8,
+    length: u8,
+    handle: u16,
+    count: u8,
+}
+
+#[repr(C)]
+#[repr(packed)]
+#[derive(Default, Copy, Clone)]
+struct SmbiosEndOfTable {
+    r#type: u8,
+    length: u8,
+    handle: u16,
 }
 
 // SAFETY: These data structures only contain a series of integers
 unsafe impl ByteValued for Smbios30Entrypoint {}
 unsafe impl ByteValued for SmbiosBiosInfo {}
 unsafe impl ByteValued for SmbiosSysInfo {}
+unsafe impl ByteValued for SmbiosOemStrings {}
+unsafe impl ByteValued for SmbiosEndOfTable {}
 
 fn write_and_incr<T: ByteValued>(
     mem: &GuestMemoryMmap,
@@ -162,7 +176,12 @@ fn write_string(
     Ok(curptr)
 }
 
-pub fn setup_smbios(mem: &GuestMemoryMmap, serial_number: Option<&str>) -> Result<u64> {
+pub fn setup_smbios(
+    mem: &GuestMemoryMmap,
+    serial_number: Option<&str>,
+    uuid: Option<&str>,
+    oem_strings: Option<&[&str]>,
+) -> Result<u64> {
     let physptr = GuestAddress(SMBIOS_START)
         .checked_add(mem::size_of::<Smbios30Entrypoint>() as u64)
         .ok_or(Error::NotEnoughMemory)?;
@@ -172,7 +191,7 @@ pub fn setup_smbios(mem: &GuestMemoryMmap, serial_number: Option<&str>) -> Resul
     {
         handle += 1;
         let smbios_biosinfo = SmbiosBiosInfo {
-            typ: BIOS_INFORMATION,
+            r#type: BIOS_INFORMATION,
             length: mem::size_of::<SmbiosBiosInfo>() as u8,
             handle,
             vendor: 1,  // First string written in this section
@@ -189,13 +208,20 @@ pub fn setup_smbios(mem: &GuestMemoryMmap, serial_number: Option<&str>) -> Resul
 
     {
         handle += 1;
+
+        let uuid_number = uuid
+            .map(Uuid::parse_str)
+            .transpose()
+            .map_err(Error::ParseUuid)?
+            .unwrap_or(Uuid::nil());
         let smbios_sysinfo = SmbiosSysInfo {
-            typ: SYSTEM_INFORMATION,
+            r#type: SYSTEM_INFORMATION,
             length: mem::size_of::<SmbiosSysInfo>() as u8,
             handle,
             manufacturer: 1, // First string written in this section
             product_name: 2, // Second string written in this section
             serial_number: serial_number.map(|_| 3).unwrap_or_default(), // 3rd string
+            uuid: uuid_number.to_bytes_le(), // set uuid
             ..Default::default()
         };
         curptr = write_and_incr(mem, smbios_sysinfo, curptr)?;
@@ -207,15 +233,34 @@ pub fn setup_smbios(mem: &GuestMemoryMmap, serial_number: Option<&str>) -> Resul
         curptr = write_and_incr(mem, 0u8, curptr)?;
     }
 
+    if let Some(oem_strings) = oem_strings {
+        handle += 1;
+
+        let smbios_oemstrings = SmbiosOemStrings {
+            r#type: OEM_STRINGS,
+            length: mem::size_of::<SmbiosOemStrings>() as u8,
+            handle,
+            count: oem_strings.len() as u8,
+        };
+
+        curptr = write_and_incr(mem, smbios_oemstrings, curptr)?;
+
+        for s in oem_strings {
+            curptr = write_string(mem, s, curptr)?;
+        }
+
+        curptr = write_and_incr(mem, 0u8, curptr)?;
+    }
+
     {
         handle += 1;
-        let smbios_sysinfo = SmbiosSysInfo {
-            typ: END_OF_TABLE,
-            length: mem::size_of::<SmbiosSysInfo>() as u8,
+        let smbios_end = SmbiosEndOfTable {
+            r#type: END_OF_TABLE,
+            length: mem::size_of::<SmbiosEndOfTable>() as u8,
             handle,
-            ..Default::default()
         };
-        curptr = write_and_incr(mem, smbios_sysinfo, curptr)?;
+        curptr = write_and_incr(mem, smbios_end, curptr)?;
+        curptr = write_and_incr(mem, 0u8, curptr)?;
         curptr = write_and_incr(mem, 0u8, curptr)?;
     }
 
@@ -237,7 +282,7 @@ pub fn setup_smbios(mem: &GuestMemoryMmap, serial_number: Option<&str>) -> Resul
             .map_err(|_| Error::WriteSmbiosEp)?;
     }
 
-    Ok(curptr.unchecked_offset_from(physptr))
+    Ok(curptr.unchecked_offset_from(physptr) + std::mem::size_of::<Smbios30Entrypoint>() as u64)
 }
 
 #[cfg(test)]
@@ -267,7 +312,7 @@ mod tests {
     fn entrypoint_checksum() {
         let mem = GuestMemoryMmap::from_ranges(&[(GuestAddress(SMBIOS_START), 4096)]).unwrap();
 
-        setup_smbios(&mem, None).unwrap();
+        setup_smbios(&mem, None, None, None).unwrap();
 
         let smbios_ep: Smbios30Entrypoint = mem.read_obj(GuestAddress(SMBIOS_START)).unwrap();
 

arch/src/x86_64/tdx/mod.rs

@@ -447,7 +447,7 @@ mod tests {
         let mut f = std::fs::File::open("tdvf.fd").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 +8,17 @@ edition = "2021"
 default = []
 
 [dependencies]
-io-uring = "0.5.2"
-libc = "0.2.126"
+io-uring = "0.5.9"
+libc = "0.2.138"
 log = "0.4.17"
 qcow = { path = "../qcow" }
-thiserror = "1.0.31"
-versionize = "0.1.6"
+thiserror = "1.0.37"
+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 = "0.4.0"
-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 = "0.9.0"
+vmm-sys-util = "0.11.0"
 

block_util/src/async_io.rs

@@ -8,7 +8,7 @@ use std::fs::File;
 use std::os::unix::io::AsRawFd;
 use thiserror::Error;
 use vmm_sys_util::eventfd::EventFd;
-use vmm_sys_util::{ioctl_expr, ioctl_io_nr, ioctl_ioc_nr};
+use vmm_sys_util::{ioctl_io_nr, ioctl_ioc_nr};
 
 #[derive(Error, Debug)]
 pub enum DiskFileError {

block_util/src/lib.rs

@@ -32,6 +32,7 @@ use std::path::Path;
 use std::result;
 use std::sync::Arc;
 use std::sync::MutexGuard;
+use thiserror::Error;
 use versionize::{VersionMap, Versionize, VersionizeResult};
 use versionize_derive::Versionize;
 use virtio_bindings::bindings::virtio_blk::*;
@@ -48,25 +49,25 @@ type GuestMemoryMmap = vm_memory::GuestMemoryMmap<AtomicBitmap>;
 const SECTOR_SHIFT: u8 = 9;
 pub const SECTOR_SIZE: u64 = 0x01 << SECTOR_SHIFT;
 
-#[derive(Debug)]
+#[derive(Error, Debug)]
 pub enum Error {
-    /// Guest gave us bad memory addresses.
+    #[error("Guest gave us bad memory addresses")]
     GuestMemory(GuestMemoryError),
-    /// Guest gave us offsets that would have overflowed a usize.
+    #[error("Guest gave us offsets that would have overflowed a usize")]
     CheckedOffset(GuestAddress, usize),
-    /// 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,
-    /// 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,
-    /// Guest gave us too few descriptors in a descriptor chain.
+    #[error("Guest gave us too few descriptors in a descriptor chain")]
     DescriptorChainTooShort,
-    /// Guest gave us a descriptor that was too short to use.
+    #[error("Guest gave us a descriptor that was too short to use")]
     DescriptorLengthTooSmall,
-    /// Getting a block's metadata fails for any reason.
+    #[error("Getting a block's metadata fails for any reason")]
     GetFileMetadata,
-    /// The requested operation would cause a seek beyond disk end.
+    #[error("The requested operation would cause a seek beyond disk end")]
     InvalidOffset,
-    /// The requested operation does not support multiple descriptors.
+    #[error("The requested operation does not support multiple descriptors")]
     TooManyDescriptors,
 }
 
@@ -102,20 +103,31 @@ pub fn build_disk_image_id(disk_path: &Path) -> Vec<u8> {
     default_disk_image_id
 }
 
-#[derive(Debug)]
+#[derive(Error, Debug)]
 pub enum ExecuteError {
+    #[error("Bad request: {0}")]
     BadRequest(Error),
+    #[error("Falied to flush: {0}")]
     Flush(io::Error),
+    #[error("Failed to read: {0}")]
     Read(GuestMemoryError),
+    #[error("Failed to seek: {0}")]
     Seek(io::Error),
+    #[error("Failed to write: {0}")]
     Write(GuestMemoryError),
+    #[error("Unsupported request: {0}")]
     Unsupported(u32),
+    #[error("Failed to submit io uring: {0}")]
     SubmitIoUring(io::Error),
+    #[error("Failed to get guest address: {0}")]
     GetHostAddress(GuestMemoryError),
+    #[error("Failed to async read: {0}")]
     AsyncRead(AsyncIoError),
+    #[error("Failed to async write: {0}")]
     AsyncWrite(AsyncIoError),
+    #[error("failed to async flush: {0}")]
     AsyncFlush(AsyncIoError),
-    /// Failed allocating a temporary buffer.
+    #[error("Failed allocating a temporary buffer: {0}")]
     TemporaryBufferAllocation(io::Error),
 }
 
@@ -554,15 +566,15 @@ 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;
     }
 

block_util/src/qcow_sync.rs

@@ -26,7 +26,7 @@ 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>> {

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>> {

block_util/src/raw_sync.rs

@@ -22,10 +22,9 @@ 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>> {
@@ -118,7 +117,7 @@ impl AsyncIo for RawFileSync {
         }
 
         if let Some(user_data) = user_data {
-            self.completion_list.push((user_data, result as i32));
+            self.completion_list.push((user_data, result));
             self.eventfd.write(1).unwrap();
         }
 

build.rs

@@ -11,7 +11,7 @@ use std::process::Command;
 fn main() {
     let mut version = "v".to_owned() + crate_version!();
 
-    if let Ok(git_out) = Command::new("git").args(&["describe", "--dirty"]).output() {
+    if let Ok(git_out) = Command::new("git").args(["describe", "--dirty"]).output() {
         if git_out.status.success() {
             if let Ok(git_out_str) = String::from_utf8(git_out.stdout) {
                 version = git_out_str;
@@ -23,5 +23,5 @@ 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

@@ -6,22 +6,25 @@ edition = "2021"
 
 [dependencies]
 acpi_tables = { path = "../acpi_tables" }
-anyhow = "1.0.58"
+anyhow = "1.0.66"
 arch = { path = "../arch" }
 bitflags = "1.3.2"
 byteorder = "1.4.3"
-epoll = "4.3.1"
 hypervisor = { path = "../hypervisor" }
-libc = "0.2.126"
+libc = "0.2.138"
 log = "0.4.17"
-versionize = "0.1.6"
+phf = { version = "0.11.1", features = ["macros"] }
+thiserror = "1.0.37"
+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 = "0.9.0"
+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,
+};
 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;
 
@@ -52,21 +55,28 @@ impl Gic {
         })
     }
 
+    /// 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 create_vgic(
         &mut self,
         vm: &Arc<dyn hypervisor::Vm>,
-        vcpu_count: u64,
+        config: VgicConfig,
     ) -> 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,
-            )
-            .map_err(Error::CreateGic)?;
+        let vgic = vm.create_vgic(config).map_err(Error::CreateGic)?;
         self.vgic = Some(vgic.clone());
         Ok(vgic.clone())
     }

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}.")
             }
         }
     }

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}"),
         }
     }
 }

devices/src/legacy/serial.rs

@@ -199,7 +199,7 @@ 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)
             }

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}"),
         }
     }
 }
@@ -306,25 +306,25 @@ 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(),
@@ -372,7 +372,7 @@ impl BusDevice for Pl011 {
                 UARTIFLS => self.ifl,
                 UARTIMSC => self.int_enabled,
                 UARTRIS => self.int_level,
-                UARTMIS => (self.int_level & self.int_enabled),
+                UARTMIS => self.int_level & self.int_enabled,
                 UARTDMACR => self.dmacr,
                 UARTDEBUG => self.debug,
                 _ => {

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};
 

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/api.md

@@ -1,21 +1,21 @@
 - [Cloud Hypervisor API](#cloud-hypervisor-api)
-  * [External API](#external-api)
-    + [REST API](#rest-api)
-      - [Location and availability](#location-and-availability)
-      - [Endpoints](#endpoints)
-		* [Virtual Machine Manager (VMM) Actions](#virtual-machine-manager-vmm-actions)
-		* [Virtual Machine (VM) Actions](#virtual-machine-vm-actions)
-      - [REST API Examples](#rest-api-examples)
-        * [Create a Virtual Machine](#create-a-virtual-machine)
-	    * [Boot a Virtual Machine](#boot-a-virtual-machine)
-        * [Dump a Virtual Machine Information](#dump-a-virtual-machine-information)
-        * [Reboot a Virtual Machine](#reboot-a-virtual-machine)
-        * [Shut a Virtual Machine Down](#shut-a-virtual-machine-down)
-    + [Command Line Interface](#command-line-interface)
-    + [REST API and CLI Architectural Relationship](#rest-api-and-cli-architectural-relationship)
-  * [Internal API](#internal-api)
-    + [Goals and Design](#goals-and-design)
-  * [End to End Example](#end-to-end-example)
+  - [External API](#external-api)
+    - [REST API](#rest-api)
+    - [Location and availability](#location-and-availability)
+    - [Endpoints](#endpoints)
+      - [Virtual Machine Manager (VMM) Actions](#virtual-machine-manager-vmm-actions)
+      - [Virtual Machine (VM) Actions](#virtual-machine-vm-actions)
+    - [REST API Examples](#rest-api-examples)
+      - [Create a Virtual Machine](#create-a-virtual-machine)
+      - [Boot a Virtual Machine](#boot-a-virtual-machine)
+      - [Dump a Virtual Machine Information](#dump-a-virtual-machine-information)
+      - [Reboot a Virtual Machine](#reboot-a-virtual-machine)
+      - [Shut a Virtual Machine Down](#shut-a-virtual-machine-down)
+    - [Command Line Interface](#command-line-interface)
+    - [REST API and CLI Architectural Relationship](#rest-api-and-cli-architectural-relationship)
+  - [Internal API](#internal-api)
+    - [Goals and Design](#goals-and-design)
+  - [End to End Example](#end-to-end-example)
 
 # Cloud Hypervisor API
 
@@ -71,40 +71,40 @@ The Cloud Hypervisor API exposes the following actions through its endpoints:
 
 #### Virtual Machine Manager (VMM) Actions
 
-Action                              | Endpoint        | Request Body | Response Body              | Prerequisites
-------------------------------------|-----------------|--------------|----------------------------|---------------------------
-Check for the REST API availability | `/vmm.ping`     | N/A          | `/schemas/VmmPingResponse` | N/A
-Shut the VMM down                   | `/vmm.shutdown` | N/A          | N/A                        | The VMM is running
+| Action                              | Endpoint        | Request Body | Response Body              | Prerequisites      |
+| ----------------------------------- | --------------- | ------------ | -------------------------- | ------------------ |
+| Check for the REST API availability | `/vmm.ping`     | N/A          | `/schemas/VmmPingResponse` | N/A                |
+| Shut the VMM down                   | `/vmm.shutdown` | N/A          | N/A                        | The VMM is running |
 
 #### Virtual Machine (VM) Actions
 
-Action                             | Endpoint             | Request Body              | Response Body            | Prerequisites
------------------------------------|----------------------|---------------------------|--------------------------|---------------------------
-Create the VM                      | `/vm.create`         | `/schemas/VmConfig`       | N/A                      | The VM is not created yet
-Delete the VM                      | `/vm.delete`         | N/A                       | N/A                      | N/A
-Boot the VM                        | `/vm.boot`           | N/A                       | N/A                      | The VM is created but not booted
-Shut the VM down                   | `/vm.shutdown`       | N/A                       | N/A                      | The VM is booted
-Reboot the VM                      | `/vm.reboot`         | N/A                       | N/A                      | The VM is booted
-Trigger power button of the VM     | `/vm.power-button`   | N/A                       | N/A                      | The VM is booted
-Pause the VM                       | `/vm.pause`          | N/A                       | N/A                      | The VM is booted
-Resume the VM                      | `/vm.resume`         | N/A                       | N/A                      | The VM is paused
-Task a snapshot of the VM          | `/vm.snapshot`       | `/schemas/VmSnapshotConfig`| N/A                     | The VM is paused
-Perform a coredump of the VM       | `/vm.coredump`       | `/schemas/VmCoredumpData` | N/A                      | The VM is paused
-Restore the VM from a snapshot     | `/vm.restore`        | `/schemas/RestoreConfig`  | N/A                      | The VM is created but not booted
-Add/remove CPUs to/from the VM     | `/vm.resize`         | `/schemas/VmResize`       | N/A                      | The VM is booted
-Add/remove memory from the VM      | `/vm.resize`         | `/schemas/VmResize`       | N/A                      | The VM is booted
-Add/remove memory from a zone      | `/vm.resize-zone`    | `/schemas/VmResizeZone`   | N/A                      | The VM is booted
-Dump the VM information            | `/vm.info`           | N/A                       | `/schemas/VmInfo`        | The VM is created
-Add VFIO PCI device to the VM      | `/vm.add-device`     | `/schemas/VmAddDevice`    | `/schemas/PciDeviceInfo` | The VM is booted
-Add disk device to the VM          | `/vm.add-disk`       | `/schemas/DiskConfig`     | `/schemas/PciDeviceInfo` | The VM is booted
-Add fs device to the VM            | `/vm.add-fs`         | `/schemas/FsConfig`       | `/schemas/PciDeviceInfo` | The VM is booted
-Add pmem device to the VM          | `/vm.add-pmem`       | `/schemas/PmemConfig`     | `/schemas/PciDeviceInfo` | The VM is booted
-Add network device to the VM       | `/vm.add-net`        | `/schemas/NetConfig`      | `/schemas/PciDeviceInfo` | The VM is booted
-Add userspace PCI device to the VM | `/vm.add-user-device`| `/schemas/VmAddUserDevice`| `/schemas/PciDeviceInfo` | The VM is booted
-Add vdpa device to the VM          | `/vm.add-vdpa`       | `/schemas/VdpaConfig`     | `/schemas/PciDeviceInfo` | The VM is booted
-Add vsock device to the VM         | `/vm.add-vsock`      | `/schemas/VsockConfig`    | `/schemas/PciDeviceInfo` | The VM is booted
-Remove device from the VM          | `/vm.remove-device`  | `/schemas/VmRemoveDevice` | N/A                      | The VM is booted
-Dump the VM counters               | `/vm.counters`       | N/A                       | `/schemas/VmCounters`    | The VM is booted
+| Action                             | Endpoint              | Request Body                | Response Body            | Prerequisites                    |
+| ---------------------------------- | --------------------- | --------------------------- | ------------------------ | -------------------------------- |
+| Create the VM                      | `/vm.create`          | `/schemas/VmConfig`         | N/A                      | The VM is not created yet        |
+| Delete the VM                      | `/vm.delete`          | N/A                         | N/A                      | N/A                              |
+| Boot the VM                        | `/vm.boot`            | N/A                         | N/A                      | The VM is created but not booted |
+| Shut the VM down                   | `/vm.shutdown`        | N/A                         | N/A                      | The VM is booted                 |
+| Reboot the VM                      | `/vm.reboot`          | N/A                         | N/A                      | The VM is booted                 |
+| Trigger power button of the VM     | `/vm.power-button`    | N/A                         | N/A                      | The VM is booted                 |
+| Pause the VM                       | `/vm.pause`           | N/A                         | N/A                      | The VM is booted                 |
+| Resume the VM                      | `/vm.resume`          | N/A                         | N/A                      | The VM is paused                 |
+| Task a snapshot of the VM          | `/vm.snapshot`        | `/schemas/VmSnapshotConfig` | N/A                      | The VM is paused                 |
+| Perform a coredump of the VM       | `/vm.coredump`        | `/schemas/VmCoredumpData`   | N/A                      | The VM is paused                 |
+| Restore the VM from a snapshot     | `/vm.restore`         | `/schemas/RestoreConfig`    | N/A                      | The VM is created but not booted |
+| Add/remove CPUs to/from the VM     | `/vm.resize`          | `/schemas/VmResize`         | N/A                      | The VM is booted                 |
+| Add/remove memory from the VM      | `/vm.resize`          | `/schemas/VmResize`         | N/A                      | The VM is booted                 |
+| Add/remove memory from a zone      | `/vm.resize-zone`     | `/schemas/VmResizeZone`     | N/A                      | The VM is booted                 |
+| Dump the VM information            | `/vm.info`            | N/A                         | `/schemas/VmInfo`        | The VM is created                |
+| Add VFIO PCI device to the VM      | `/vm.add-device`      | `/schemas/VmAddDevice`      | `/schemas/PciDeviceInfo` | The VM is booted                 |
+| Add disk device to the VM          | `/vm.add-disk`        | `/schemas/DiskConfig`       | `/schemas/PciDeviceInfo` | The VM is booted                 |
+| Add fs device to the VM            | `/vm.add-fs`          | `/schemas/FsConfig`         | `/schemas/PciDeviceInfo` | The VM is booted                 |
+| Add pmem device to the VM          | `/vm.add-pmem`        | `/schemas/PmemConfig`       | `/schemas/PciDeviceInfo` | The VM is booted                 |
+| Add network device to the VM       | `/vm.add-net`         | `/schemas/NetConfig`        | `/schemas/PciDeviceInfo` | The VM is booted                 |
+| Add userspace PCI device to the VM | `/vm.add-user-device` | `/schemas/VmAddUserDevice`  | `/schemas/PciDeviceInfo` | The VM is booted                 |
+| Add vdpa device to the VM          | `/vm.add-vdpa`        | `/schemas/VdpaConfig`       | `/schemas/PciDeviceInfo` | The VM is booted                 |
+| Add vsock device to the VM         | `/vm.add-vsock`       | `/schemas/VsockConfig`      | `/schemas/PciDeviceInfo` | The VM is booted                 |
+| Remove device from the VM          | `/vm.remove-device`   | `/schemas/VmRemoveDevice`   | N/A                      | The VM is booted                 |
+| Dump the VM counters               | `/vm.counters`        | N/A                         | `/schemas/VmCounters`    | The VM is booted                 |
 
 ### REST API Examples
 
@@ -143,8 +143,7 @@ curl --unix-socket /tmp/cloud-hypervisor.sock -i \
      -H 'Content-Type: application/json'         \
      -d '{
          "cpus":{"boot_vcpus": 4, "max_vcpus": 4},
-         "kernel":{"path":"/opt/clh/kernel/vmlinux-virtio-fs-virtio-iommu"},
-         "cmdline":{"args":"console=ttyS0 console=hvc0 root=/dev/vda1 rw"},
+         "payload":{"kernel":"/opt/clh/kernel/vmlinux-virtio-fs-virtio-iommu", "cmdline":"console=ttyS0 console=hvc0 root=/dev/vda1 rw"},
          "disks":[{"path":"/opt/clh/images/focal-server-cloudimg-amd64.raw"}],
          "rng":{"src":"/dev/urandom"},
          "net":[{"ip":"192.168.10.10", "mask":"255.255.255.0", "mac":"12:34:56:78:90:01"}]
@@ -311,8 +310,7 @@ APIs work together, let's look at a complete VM creation flow, from the
 		-H 'Content-Type: application/json'         \
 		-d '{
 			"cpus":{"boot_vcpus": 4, "max_vcpus": 4},
-			"kernel":{"path":"/opt/clh/kernel/vmlinux-virtio-fs-virtio-iommu"},
-			"cmdline":{"args":"console=ttyS0 console=hvc0 root=/dev/vda1 rw"},
+			"payload":{"kernel":"/opt/clh/kernel/vmlinux-virtio-fs-virtio-iommu", "cmdline":"console=ttyS0 console=hvc0 root=/dev/vda1 rw"},
 			"disks":[{"path":"/opt/clh/images/focal-server-cloudimg-amd64.raw"}],
 			"rng":{"src":"/dev/urandom"},
 			"net":[{"ip":"192.168.10.10", "mask":"255.255.255.0", "mac":"12:34:56:78:90:01"}]

docs/building.md

@@ -0,0 +1,82 @@
+- [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 $CLOUDH
+```
+
+## Install prerequisites
+
+You need to install some prerequisite packages in order to build and test Cloud
+Hypervisor. Here, all the steps are based on Ubuntu, for other Linux
+distributions please replace the package manager and package name.
+
+```shell
+# Install build-essential, git, and qemu-utils
+$ 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 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
+$ 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

@@ -59,6 +59,16 @@ mkdir -p /mnt
 sudo mount -o loop,offset=$((227328 * 512)) focal-server-cloudimg-amd64.raw /mnt
 ```
 
+### Set up DNS
+
+The next step describes changing the root directory to the rootfs contained by 
+the cloud image. For DNS to work in the root directory, you will need to first bind-mount 
+the host `/etc/resolv.conf` onto the mounted linux partition of the cloud image.
+
+```bash
+sudo mount -o bind /etc/resolv.conf /mnt/etc/resolv.conf
+```
+
 ### Change root directory
 
 Changing the root directory will allow us to install new packages to the rootfs
@@ -108,6 +118,7 @@ umount /dev/pts
 umount /proc
 history -c
 exit
+umount /mnt/etc/resolv.conf
 umount /mnt
 ```
 

docs/fs.md

@@ -29,10 +29,11 @@ mkdir /tmp/shared_dir
 _Run virtiofsd_
 ```bash
 ./virtiofsd \
-    -d \
+    --log-level debug \
     --socket-path=/tmp/virtiofs \
     --shared-dir=/tmp/shared_dir \
-    --cache=never
+    --cache=never \
+    --thread-pool-size=$N
 ```
 
 The `cache=never` option is the default when using `virtiofsd` with
@@ -44,6 +45,10 @@ The `cache=always` option will allow the host page cache to be used, which can
 result in better performance for the guest's workload at the cost of increasing
 the footprint on host memory.
 
+The `thread-pool-size` option controls how many IO threads are spawned. For
+very fast storage like NVMe spawning enough worker threads is critical to
+getting an acceptable performance compared to native.
+
 ### Kernel support
 
 Modern Linux kernels (at least v5.10) have support for virtio-fs. Use of older
@@ -90,4 +95,4 @@ Cloud Hypervisor command line, which happens to be `myfs` in this example.
 ## DAX feature
 
 Given the DAX feature is not stable yet from a daemon standpoint, it is not
-available in Cloud Hypervisor.
+available in Cloud Hypervisor.

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 \
@@ -103,8 +106,18 @@ option as well.
     --disk path=tdx_guest_img
 ```
 
-### Guest kernel disables serial ports
+### Guest kernel limitations
+
+#### Serial ports disabled
 
 The latest guest kernel that can be found in the latest image
 `td-guest-rhel8.5.raw` disabled the support for serial ports. This means adding
-`console=ttyS0` will have no effect and will not print any log from the guest.
+`console=ttyS0` will have no effect and will not print any log from the guest.
+
+#### PCI hotplug through ACPI
+
+Unless you run the guest kernel with the parameter `tdx_disable_filter`, ACPI
+devices responsible for handling PCI hotplug (PCI hotplug controller, PCI
+Express Bus and Generic Event Device) will not be allowed, therefore the
+corresponding drivers will not be loaded and the PCI hotplug feature will not
+be supported.

docs/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/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 = "0.2.126"
-serde = { version = "1.0.138", features = ["rc", "derive"] }
-serde_json = "1.0.82"
+libc = "0.2.138"
+serde = { version = "1.0.150", features = ["rc", "derive"] }
+serde_json = "1.0.89"

fuzz/Cargo.lock

@@ -11,9 +11,9 @@ dependencies = [
 
 [[package]]
 name = "anyhow"
-version = "1.0.58"
+version = "1.0.66"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bb07d2053ccdbe10e2af2995a2f116c1330396493dc1269f6a91d0ae82e19704"
+checksum = "216261ddc8289130e551ddcd5ce8a064710c0d064a4d2895c67151c92b5443f6"
 
 [[package]]
 name = "api_client"
@@ -24,21 +24,20 @@ 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"
-version = "1.5.0"
+version = "1.5.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c5d78ce20460b82d3fa150275ed9d55e21064fc7951177baacf86a145c4a4b1f"
+checksum = "983cd8b9d4b02a6dc6ffa557262eb5858a27a0038ffffe21a0f133eaa819a164"
 
 [[package]]
 name = "arch"
 version = "0.1.0"
 dependencies = [
- "acpi_tables",
  "anyhow",
  "byteorder",
  "fdt",
@@ -48,6 +47,7 @@ dependencies = [
  "log",
  "serde",
  "thiserror",
+ "uuid",
  "versionize",
  "versionize_derive",
  "vm-fdt",
@@ -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"
@@ -115,9 +98,12 @@ checksum = "14c189c53d098945499cdfa7ecc63567cf3886b3332b312a5b4585d8d3a6a610"
 
 [[package]]
 name = "cc"
-version = "1.0.73"
+version = "1.0.77"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2fff2a6927b3bb87f9595d67196a70493f627687a71d87a0d692242c33f58c11"
+checksum = "e9f73505338f7d905b19d18738976aae232eb46b8efc15554ffc56deb5d9ebe4"
+dependencies = [
+ "jobserver",
+]
 
 [[package]]
 name = "cfg-if"
@@ -127,33 +113,31 @@ checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
 name = "clap"
-version = "3.2.8"
+version = "4.0.29"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "190814073e85d238f31ff738fcb0bf6910cedeb73376c87cd69291028966fd83"
+checksum = "4d63b9e9c07271b9957ad22c173bae2a4d9a81127680962039296abcd2f8251d"
 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 = "24.0.0"
+version = "28.0.0"
 dependencies = [
  "anyhow",
  "api_client",
@@ -168,6 +152,8 @@ dependencies = [
  "serde_json",
  "signal-hook",
  "thiserror",
+ "tpm",
+ "tracer",
  "vm-memory",
  "vmm",
  "vmm-sys-util",
@@ -179,15 +165,21 @@ version = "0.0.0"
 dependencies = [
  "block_util",
  "cloud-hypervisor",
+ "devices",
+ "epoll",
  "libc",
  "libfuzzer-sys",
+ "micro_http",
+ "once_cell",
  "qcow",
  "seccompiler",
  "vhdx",
  "virtio-devices",
  "virtio-queue",
+ "vm-device",
  "vm-memory",
  "vm-virtio",
+ "vmm",
  "vmm-sys-util",
 ]
 
@@ -206,6 +198,41 @@ version = "1.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "55626594feae15d266d52440b26ff77de0e22230cf0c113abe619084c1ddc910"
 
+[[package]]
+name = "darling"
+version = "0.14.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b0dd3cd20dc6b5a876612a6e5accfe7f3dd883db6d07acfbf14c128f61550dfa"
+dependencies = [
+ "darling_core",
+ "darling_macro",
+]
+
+[[package]]
+name = "darling_core"
+version = "0.14.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a784d2ccaf7c98501746bf0be29b2022ba41fd62a2e622af997a03e9f972859f"
+dependencies = [
+ "fnv",
+ "ident_case",
+ "proc-macro2",
+ "quote",
+ "strsim",
+ "syn",
+]
+
+[[package]]
+name = "darling_macro"
+version = "0.14.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7618812407e9402654622dd402b0a89dff9ba93badd6540781526117b92aab7e"
+dependencies = [
+ "darling_core",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "devices"
 version = "0.1.0"
@@ -215,10 +242,12 @@ dependencies = [
  "arch",
  "bitflags",
  "byteorder",
- "epoll",
  "hypervisor",
  "libc",
  "log",
+ "phf",
+ "thiserror",
+ "tpm",
  "versionize",
  "versionize_derive",
  "vm-device",
@@ -237,6 +266,27 @@ 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"
@@ -248,32 +298,32 @@ dependencies = [
 
 [[package]]
 name = "fdt"
-version = "0.1.3"
+version = "0.1.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b643857cf70949306b81d7e92cb9d47add673868edac9863c4a49c42feaf3f1e"
+checksum = "964f5becd44d069dca0beea2b4bc05639ae7bf3b3f5369c295aff360bb57cca2"
+
+[[package]]
+name = "fnv"
+version = "1.0.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+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 = "0.12.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "db0d4cf898abf0081f964436dc980e96670a0f36863e4b83aaacdb65c9d7ccc3"
-
 [[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",
 ]
@@ -283,15 +333,16 @@ name = "hypervisor"
 version = "0.1.0"
 dependencies = [
  "anyhow",
- "epoll",
+ "byteorder",
  "iced-x86",
  "kvm-bindings",
  "kvm-ioctls",
  "libc",
  "log",
  "serde",
- "serde_json",
+ "serde_with",
  "thiserror",
+ "vfio-ioctls",
  "vm-memory",
  "vmm-sys-util",
 ]
@@ -307,35 +358,62 @@ dependencies = [
 ]
 
 [[package]]
-name = "indexmap"
-version = "1.9.1"
+name = "ident_case"
+version = "1.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "10a35a97730320ffe8e2d410b5d3b69279b98d2c14bdb8b70ea89ecf7888d41e"
+checksum = "b9e0384b61958566e926dc50660321d12159025e767c18e043daf26b70104c39"
+
+[[package]]
+name = "io-lifetimes"
+version = "1.0.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "46112a93252b123d31a119a8d1a1ac19deac4fac6e0e8b0df58f0d4e5870e63c"
 dependencies = [
- "autocfg",
- "hashbrown",
+ "libc",
+ "windows-sys",
 ]
 
 [[package]]
 name = "io-uring"
-version = "0.5.2"
+version = "0.5.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8d75829ed9377bab6c90039fe47b9d84caceb4b5063266142e21bcce6550cda8"
+checksum = "7ba34abb5175052fc1a2227a10d2275b7386c9990167de9786c0b88d8b062330"
 dependencies = [
  "bitflags",
  "libc",
 ]
 
 [[package]]
-name = "itoa"
-version = "1.0.2"
+name = "is-terminal"
+version = "0.4.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "112c678d4050afce233f4f2852bb2eb519230b3cf12f33585275537d7e41578d"
+checksum = "927609f78c2913a6f6ac3c27a4fe87f43e2a35367c0c4b0f8265e8f49a104330"
+dependencies = [
+ "hermit-abi",
+ "io-lifetimes",
+ "rustix",
+ "windows-sys",
+]
+
+[[package]]
+name = "itoa"
+version = "1.0.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4217ad341ebadf8d8e724e264f13e593e0648f5b3e94b3896a5df283be015ecc"
+
+[[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",
@@ -344,8 +422,9 @@ 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",
@@ -360,15 +439,15 @@ checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646"
 
 [[package]]
 name = "libc"
-version = "0.2.126"
+version = "0.2.138"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "349d5a591cd28b49e1d1037471617a32ddcda5731b99419008085f72d5a53836"
+checksum = "db6d7e329c562c5dfab7a46a2afabc8b987ab9a4834c9d1ca04dc54c1546cef8"
 
 [[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",
@@ -377,13 +456,19 @@ 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 = "0.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8f9f08d8963a6c613f4b1a78f4f4a4dbfadf8e6545b2d72861731e4858b8b47f"
+
 [[package]]
 name = "log"
 version = "0.4.17"
@@ -396,7 +481,7 @@ dependencies = [
 [[package]]
 name = "micro_http"
 version = "0.1.0"
-source = "git+https://github.com/firecracker-microvm/micro-http?branch=main#9517a300370a158a7af0996b7eebf040d171e1a4"
+source = "git+https://github.com/firecracker-microvm/micro-http?branch=main#4b18a043e997da5b5f679e3defc279fec908753e"
 dependencies = [
  "libc",
  "vmm-sys-util",
@@ -420,6 +505,7 @@ dependencies = [
  "net_gen",
  "rate_limiter",
  "serde",
+ "thiserror",
  "versionize",
  "versionize_derive",
  "virtio-bindings",
@@ -431,9 +517,9 @@ dependencies = [
 
 [[package]]
 name = "once_cell"
-version = "1.13.0"
+version = "1.16.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "18a6dbe30758c9f83eb00cbea4ac95966305f5a7772f3f42ebfc7fc7eddbd8e1"
+checksum = "86f0b0d4bf799edbc74508c1e8bf170ff5f41238e5f8225603ca7caaae2b7860"
 
 [[package]]
 name = "option_parser"
@@ -441,9 +527,9 @@ version = "0.1.0"
 
 [[package]]
 name = "os_str_bytes"
-version = "6.1.0"
+version = "6.4.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "21326818e99cfe6ce1e524c2a805c189a99b5ae555a35d19f9a284b427d86afa"
+checksum = "9b7820b9daea5457c9f21c69448905d723fbd21136ccf521748f23fd49e723ee"
 
 [[package]]
 name = "pci"
@@ -469,10 +555,52 @@ dependencies = [
 ]
 
 [[package]]
-name = "proc-macro2"
-version = "1.0.40"
+name = "phf"
+version = "0.11.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dd96a1e8ed2596c337f8eae5f24924ec83f5ad5ab21ea8e455d3566c69fbcaf7"
+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 = "1.0.47"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5ea3d908b0e36316caf9e9e2c4625cdde190a7e6f440d794667ed17a1855e725"
 dependencies = [
  "unicode-ident",
 ]
@@ -490,13 +618,28 @@ dependencies = [
 
 [[package]]
 name = "quote"
-version = "1.0.20"
+version = "1.0.21"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3bcdf212e9776fbcb2d23ab029360416bb1706b1aea2d1a5ba002727cbcab804"
+checksum = "bbe448f377a7d6961e30f5955f9b8d106c3f5e449d493ee1b125c1d43c2b5179"
 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"
@@ -508,9 +651,9 @@ dependencies = [
 
 [[package]]
 name = "remain"
-version = "0.2.3"
+version = "0.2.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0c35270ea384ac1762895831cc8acb96f171468e52cec82ed9186f9416209fa4"
+checksum = "1a81bc6a3aaeb180f767bd2bda8c03bac5b93b3b69a783c66f35d16a4df276cf"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -527,40 +670,54 @@ dependencies = [
 ]
 
 [[package]]
-name = "ryu"
-version = "1.0.10"
+name = "rustix"
+version = "0.36.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f3f6f92acf49d1b98f7a81226834412ada05458b7364277387724a237f062695"
+checksum = "a3807b5d10909833d3e9acd1eb5fb988f79376ff10fce42937de71a449c4c588"
+dependencies = [
+ "bitflags",
+ "errno",
+ "io-lifetimes",
+ "libc",
+ "linux-raw-sys",
+ "windows-sys",
+]
+
+[[package]]
+name = "ryu"
+version = "1.0.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4501abdff3ae82a1c1b477a17252eb69cee9e66eb915c1abaa4f44d873df9f09"
 
 [[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 = "1.0.12"
+version = "1.0.14"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a2333e6df6d6598f2b1974829f853c2b4c5f4a6e503c10af918081aa6f8564e1"
+checksum = "e25dfac463d778e353db5be2449d1cce89bd6fd23c9f1ea21310ce6e5a1b29c4"
 
 [[package]]
 name = "serde"
-version = "1.0.138"
+version = "1.0.150"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1578c6245786b9d168c5447eeacfb96856573ca56c9d68fdcf394be134882a47"
+checksum = "e326c9ec8042f1b5da33252c8a37e9ffbd2c9bef0155215b6e6c80c790e05f91"
 dependencies = [
  "serde_derive",
 ]
 
 [[package]]
 name = "serde_derive"
-version = "1.0.138"
+version = "1.0.150"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "023e9b1467aef8a10fb88f25611870ada9800ef7e22afce356bb0d2387b6f27c"
+checksum = "42a3df25b0713732468deadad63ab9da1f1fd75a48a15024b50363f128db627e"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -569,15 +726,41 @@ dependencies = [
 
 [[package]]
 name = "serde_json"
-version = "1.0.82"
+version = "1.0.89"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "82c2c1fdcd807d1098552c5b9a36e425e42e9fbd7c6a37a8425f390f781f7fa7"
+checksum = "020ff22c755c2ed3f8cf162dbb41a7268d934702f3ed3631656ea597e08fc3db"
 dependencies = [
  "itoa",
  "ryu",
  "serde",
 ]
 
+[[package]]
+name = "serde_with"
+version = "2.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "25bf4a5a814902cd1014dbccfa4d4560fb8432c779471e96e035602519f82eef"
+dependencies = [
+ "serde",
+ "serde_with_macros",
+]
+
+[[package]]
+name = "serde_with_macros"
+version = "2.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e3452b4c0f6c1e357f73fdb87cd1efabaa12acf328c7a528e252893baeb3f4aa"
+dependencies = [
+ "darling",
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
+[[package]]
+name = "serial_buffer"
+version = "0.1.0"
+
 [[package]]
 name = "signal-hook"
 version = "0.3.14"
@@ -597,6 +780,12 @@ dependencies = [
  "libc",
 ]
 
+[[package]]
+name = "siphasher"
+version = "0.3.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7bd3e3206899af3f8b12af284fafc038cc1dc2b41d1b89dd17297221c5d225de"
+
 [[package]]
 name = "static_assertions"
 version = "1.1.0"
@@ -611,9 +800,9 @@ checksum = "73473c0e59e6d5812c5dfe2a064a6444949f089e20eec9a2e5506596494e4623"
 
 [[package]]
 name = "syn"
-version = "1.0.98"
+version = "1.0.105"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c50aef8a904de4c23c788f104b7dddc7d6f79c647c7c8ce4cc8f73eb0ca773dd"
+checksum = "60b9b43d45702de4c839cb9b51d9f529c5dd26a4aff255b42b1ebc03e88ee908"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -631,63 +820,77 @@ 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 = "1.0.31"
+version = "1.0.37"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bd829fe32373d27f76265620b5309d0340cb8550f523c1dda251d6298069069a"
+checksum = "10deb33631e3c9018b9baf9dcbbc4f737320d2b576bac10f6aefa048fa407e3e"
 dependencies = [
  "thiserror-impl",
 ]
 
 [[package]]
 name = "thiserror-impl"
-version = "1.0.31"
+version = "1.0.37"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0396bc89e626244658bef819e22d0cc459e795a5ebe878e6ec336d1674a8d79a"
+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 = "1.0.1"
+version = "1.0.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5bd2fe26506023ed7b5e1e315add59d6f584c621d037f9368fea9cfb988f368c"
+checksum = "6ceab39d59e4c9499d4e5a8ee0e2735b891bb7308ac83dfb4e80cad195c9f6f3"
 
 [[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",
@@ -712,17 +915,16 @@ 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#4630612f2ff300e78b6d55be324fb5481aa84661"
+version = "0.2.0"
+source = "git+https://github.com/rust-vmm/vfio?branch=main#7ef33f6759ae6d37a54cd66f749aa47c1c81509c"
 dependencies = [
  "byteorder",
  "kvm-bindings",
@@ -739,7 +941,6 @@ dependencies = [
 name = "vfio_user"
 version = "0.1.0"
 dependencies = [
- "anyhow",
  "libc",
  "log",
  "serde",
@@ -762,14 +963,13 @@ dependencies = [
  "remain",
  "thiserror",
  "uuid",
- "vmm-sys-util",
 ]
 
 [[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",
@@ -803,6 +1003,7 @@ dependencies = [
  "seccompiler",
  "serde",
  "serde_json",
+ "serial_buffer",
  "thiserror",
  "versionize",
  "versionize_derive",
@@ -819,11 +1020,12 @@ dependencies = [
 
 [[package]]
 name = "virtio-queue"
-version = "0.3.0"
+version = "0.7.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "88f2d73c184c18f8acc32dab77fcb6e3af92d53262538d3a68aa474810d6863c"
+checksum = "19e927d93d54c365034fd7f31a5f458a1f540de4a37c52e892670dad9692173c"
 dependencies = [
  "log",
+ "virtio-bindings",
  "vm-memory",
  "vmm-sys-util",
 ]
@@ -844,7 +1046,6 @@ dependencies = [
  "anyhow",
  "hypervisor",
  "serde",
- "serde_json",
  "thiserror",
  "vfio-ioctls",
  "vm-memory",
@@ -853,14 +1054,14 @@ 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",
@@ -885,7 +1086,6 @@ name = "vm-virtio"
 version = "0.1.0"
 dependencies = [
  "log",
- "virtio-bindings",
  "virtio-queue",
  "vm-memory",
 ]
@@ -917,8 +1117,10 @@ dependencies = [
  "seccompiler",
  "serde",
  "serde_json",
+ "serial_buffer",
  "signal-hook",
  "thiserror",
+ "tracer",
  "uuid",
  "versionize",
  "versionize_derive",
@@ -937,9 +1139,9 @@ dependencies = [
 
 [[package]]
 name = "vmm-sys-util"
-version = "0.9.0"
+version = "0.11.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "733537bded03aaa93543f785ae997727b30d1d9f4a03b7861d23290474242e11"
+checksum = "cc06a16ee8ebf0d9269aed304030b0d20a866b8b3dd3d4ce532596ac567a0d24"
 dependencies = [
  "bitflags",
  "libc",
@@ -983,3 +1185,60 @@ 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

@@ -10,23 +10,28 @@ cargo-fuzz = true
 
 [dependencies]
 block_util = { path = "../block_util" }
-libc = "0.2.126"
-libfuzzer-sys = "0.4.3"
+devices = { path = "../devices" }
+epoll = "4.3.1"
+libc = "0.2.135"
+libfuzzer-sys = "0.4.5"
+micro_http = { git = "https://github.com/firecracker-microvm/micro-http", branch = "main" }
+once_cell = "1.16.0"
 qcow = { path = "../qcow" }
-seccompiler = "0.2.0"
+seccompiler = "0.3.0"
 vhdx = { path = "../vhdx" }
 virtio-devices = { path = "../virtio-devices" }
-virtio-queue = "0.4.0"
-vmm-sys-util = "0.9.0"
+virtio-queue = "0.7.0"
+vmm = { path = "../vmm" }
+vmm-sys-util = "0.11.0"
+vm-memory = "0.10.0"
+vm-device = { path = "../vm-device" }
 vm-virtio = { path = "../vm-virtio" }
-vm-memory = "0.8.0"
 
 [dependencies.cloud-hypervisor]
 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
@@ -34,8 +39,8 @@ versionize_derive = { git = "https://github.com/cloud-hypervisor/versionize_deri
 members = ["."]
 
 [[bin]]
-name = "qcow"
-path = "fuzz_targets/qcow.rs"
+name = "balloon"
+path = "fuzz_targets/balloon.rs"
 test = false
 doc = false
 
@@ -45,8 +50,68 @@ path = "fuzz_targets/block.rs"
 test = false
 doc = false
 
+[[bin]]
+name = "cmos"
+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 = "mem"
+path = "fuzz_targets/mem.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"
+test = false
+doc = false
+
 [[bin]]
 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]
 
@@ -9,89 +13,42 @@ use libfuzzer_sys::fuzz_target;
 use seccompiler::SeccompAction;
 use std::ffi;
 use std::fs::File;
-use std::io::{self, Cursor, Read, Seek, SeekFrom};
-use std::mem::size_of;
+use std::io;
 use std::os::unix::io::{AsRawFd, FromRawFd, RawFd};
 use std::path::PathBuf;
 use std::sync::Arc;
 use virtio_devices::{Block, VirtioDevice, VirtioInterrupt, VirtioInterruptType};
-use virtio_queue::{Queue, QueueState};
+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 MEM_SIZE: u64 = 256 * 1024 * 1024;
-const DESC_SIZE: u64 = 16; // Bytes in one virtio descriptor.
-const QUEUE_SIZE: u16 = 16; // Max entries in the queue.
-const CMD_SIZE: usize = 16; // Bytes in the command.
+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| {
-    let size_u64 = size_of::<u64>();
-    let mem = GuestMemoryMmap::from_ranges(&[(GuestAddress(0), MEM_SIZE as usize)]).unwrap();
-
-    // The fuzz data is interpreted as:
-    // starting index 8 bytes
-    // command location 8 bytes
-    // command 16 bytes
-    // descriptors circular buffer 16 bytes * 3
-    if bytes.len() < 4 * size_u64 {
-        // Need an index to start.
+    if bytes.len() < QUEUE_DATA_SIZE || bytes.len() > (QUEUE_DATA_SIZE + MEM_SIZE) {
         return;
     }
 
-    let mut data_image = Cursor::new(bytes);
-
-    let first_index = read_u64(&mut data_image);
-    if first_index > MEM_SIZE / DESC_SIZE {
-        return;
-    }
-    let first_offset = first_index * DESC_SIZE;
-    if first_offset as usize + size_u64 > bytes.len() {
-        return;
-    }
-
-    let command_addr = read_u64(&mut data_image);
-    if command_addr > MEM_SIZE - CMD_SIZE as u64 {
-        return;
-    }
-    if mem
-        .write_slice(
-            &bytes[2 * size_u64..(2 * size_u64) + CMD_SIZE],
-            GuestAddress(command_addr as u64),
-        )
-        .is_err()
-    {
-        return;
-    }
-
-    data_image.seek(SeekFrom::Start(first_offset)).unwrap();
-    let desc_table = read_u64(&mut data_image);
-
-    if mem
-        .write_slice(&bytes[32..], GuestAddress(desc_table as u64))
-        .is_err()
-    {
-        return;
-    }
-
-    let guest_memory = GuestMemoryAtomic::new(mem);
-
-    let mut q = Queue::<
-        GuestMemoryAtomic<GuestMemoryMmap>,
-        QueueState,
-    >::new(guest_memory.clone(), QUEUE_SIZE);
-    q.state.ready = true;
-    q.state.size = QUEUE_SIZE / 2;
-
-    let queue_evts: Vec<EventFd> = vec![EventFd::new(0).unwrap()];
-    let queue_fd = queue_evts[0].as_raw_fd();
-    let queue_evt = unsafe { EventFd::from_raw_fd(libc::dup(queue_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,
@@ -103,27 +60,38 @@ 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,
             Arc::new(NoopVirtioInterrupt {}),
-            vec![q],
-            queue_evts,
+            vec![(0, q, evt)],
         )
         .ok();
 
-    queue_evt.write(77).unwrap(); // Rings the doorbell, any byte will do.
+    // Wait for the events to finish and block device worker thread to return
+    block.wait_for_epoll_threads();
 });
 
-fn read_u64<T: Read>(readable: &mut T) -> u64 {
-    let mut buf = [0u8; size_of::<u64>()];
-    readable.read_exact(&mut buf[..]).unwrap();
-    u64::from_le_bytes(buf)
-}
-
 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) };
 
@@ -137,10 +105,25 @@ fn memfd_create(name: &ffi::CStr, flags: u32) -> Result<RawFd, io::Error> {
 pub struct NoopVirtioInterrupt {}
 
 impl VirtioInterrupt for NoopVirtioInterrupt {
-    fn trigger(
-        &self,
-        _int_type: VirtioInterruptType,
-    ) -> std::result::Result<(), std::io::Error> {
+    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/cmos.rs

@@ -0,0 +1,48 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+use devices::legacy::Cmos;
+use libc::EFD_NONBLOCK;
+use libfuzzer_sys::fuzz_target;
+use vm_device::BusDevice;
+use vmm_sys_util::eventfd::EventFd;
+
+fuzz_target!(|bytes| {
+    // Need at least 16 bytes for the test
+    if bytes.len() < 16 {
+        return;
+    }
+
+    let mut below_4g = [0u8; 8];
+    let mut above_4g = [0u8; 8];
+
+    below_4g.copy_from_slice(&bytes[0..8]);
+    above_4g.copy_from_slice(&bytes[8..16]);
+
+    let mut cmos = Cmos::new(
+        u64::from_le_bytes(below_4g),
+        u64::from_le_bytes(above_4g),
+        EventFd::new(EFD_NONBLOCK).unwrap(),
+    );
+
+    let mut i = 16;
+    while i < bytes.len() {
+        let read = bytes.get(i).unwrap_or(&0) % 2 == 0;
+        i += 1;
+
+        if read {
+            let offset = (bytes.get(i).unwrap_or(&0) % 2) as u64;
+            i += 1;
+            let mut out_bytes = vec![0];
+            cmos.read(0, offset, &mut out_bytes);
+        } else {
+            let offset = (bytes.get(i).unwrap_or(&0) % 2) as u64;
+            i += 1;
+            let data = vec![*bytes.get(i).unwrap_or(&0)];
+            i += 1;
+            cmos.write(0, offset, &data);
+        }
+    }
+});

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/http_api.rs

@@ -0,0 +1,191 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+use libfuzzer_sys::fuzz_target;
+use micro_http::Request;
+use once_cell::sync::Lazy;
+use std::os::unix::io::AsRawFd;
+use std::sync::mpsc::{channel, Receiver};
+use std::thread;
+use vmm::api::{http::*, ApiRequest, ApiResponsePayload};
+use vmm::{EpollContext, EpollDispatch};
+use vmm_sys_util::eventfd::EventFd;
+
+// Need to be ordered for test case reproducibility
+static ROUTES: Lazy<Vec<&Box<dyn EndpointHandler + Sync + Send>>> =
+    Lazy::new(|| HTTP_ROUTES.routes.values().collect());
+
+fuzz_target!(|bytes| {
+    if bytes.len() < 2 {
+        return;
+    }
+
+    let route = ROUTES[bytes[0] as usize % ROUTES.len()];
+    if let Some(request) = generate_request(&bytes[1..]) {
+        let exit_evt = EventFd::new(libc::EFD_NONBLOCK).unwrap();
+        let api_evt = EventFd::new(libc::EFD_NONBLOCK).unwrap();
+        let (api_sender, api_receiver) = channel();
+
+        let http_receiver_thread = {
+            let exit_evt = exit_evt.try_clone().unwrap();
+            let api_evt = api_evt.try_clone().unwrap();
+            thread::Builder::new()
+                .name("http_receiver".to_string())
+                .spawn(move || {
+                    http_receiver_stub(exit_evt, api_evt, api_receiver);
+                })
+                .unwrap()
+        };
+
+        route.handle_request(&request, api_evt, api_sender);
+        exit_evt.write(1).ok();
+        http_receiver_thread.join().unwrap();
+    };
+});
+
+fn generate_request(bytes: &[u8]) -> Option<Request> {
+    let req_method = match bytes[0] % 5 {
+        0 => "GET",
+        1 => "PUT",
+        2 => "PATCH",
+        3 => "POST",
+        _ => "INVALID",
+    };
+    let request_line = format!("{} http://localhost/home HTTP/1.1\r\n", req_method);
+
+    let req_body = &bytes[1..];
+    let request = if req_body.len() > 0 {
+        [
+            format!("{}Content-Length: {}\r\n", request_line, req_body.len()).as_bytes(),
+            req_body,
+        ]
+        .concat()
+    } else {
+        format!("{}\r\n", request_line).as_bytes().to_vec()
+    };
+
+    Request::try_from(&request, None).ok()
+}
+
+fn http_receiver_stub(exit_evt: EventFd, api_evt: EventFd, api_receiver: Receiver<ApiRequest>) {
+    let mut epoll = EpollContext::new().unwrap();
+    epoll.add_event(&exit_evt, EpollDispatch::Exit).unwrap();
+    epoll.add_event(&api_evt, EpollDispatch::Api).unwrap();
+
+    let epoll_fd = epoll.as_raw_fd();
+    let mut events = vec![epoll::Event::new(epoll::Events::empty(), 0); 2];
+    let num_events;
+    loop {
+        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!"),
+            },
+        };
+
+        break;
+    }
+
+    for event in events.iter().take(num_events) {
+        let dispatch_event: EpollDispatch = event.data.into();
+        match dispatch_event {
+            EpollDispatch::Exit => {
+                break;
+            }
+            EpollDispatch::Api => {
+                for _ in 0..api_evt.read().unwrap() {
+                    let api_request = api_receiver.recv().unwrap();
+                    match api_request {
+                        ApiRequest::VmCreate(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmDelete(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmBoot(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmShutdown(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmReboot(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmInfo(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmmPing(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmPause(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmResume(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmSnapshot(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmRestore(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmmShutdown(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmResize(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmResizeZone(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmAddDevice(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmAddUserDevice(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmRemoveDevice(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmAddDisk(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmAddFs(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmAddPmem(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmAddNet(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmAddVdpa(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmAddVsock(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmCounters(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmReceiveMigration(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmSendMigration(_, sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                        ApiRequest::VmPowerButton(sender) => {
+                            sender.send(Ok(ApiResponsePayload::Empty)).unwrap();
+                        }
+                    }
+                }
+            }
+            _ => {
+                panic!("Unexpected Epoll event");
+            }
+        }
+    }
+}

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/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/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

@@ -0,0 +1,73 @@
+// Copyright © 2022 Intel Corporation
+//
+// SPDX-License-Identifier: Apache-2.0
+
+#![no_main]
+use devices::legacy::Serial;
+use libc::EFD_NONBLOCK;
+use libfuzzer_sys::fuzz_target;
+use std::sync::Arc;
+use vm_device::interrupt::{InterruptIndex, InterruptSourceConfig, InterruptSourceGroup};
+use vm_device::BusDevice;
+use vmm_sys_util::eventfd::EventFd;
+
+fuzz_target!(|bytes| {
+    let mut serial = Serial::new_sink(
+        "serial".into(),
+        Arc::new(TestInterrupt::new(EventFd::new(EFD_NONBLOCK).unwrap())),
+    );
+
+    let mut i = 0;
+    while i < bytes.len() {
+        let choice = bytes.get(i).unwrap_or(&0) % 3;
+        i += 1;
+
+        match choice {
+            0 => {
+                let offset = (bytes.get(i).unwrap_or(&0) % 8) as u64;
+                i += 1;
+                let mut out_bytes = vec![0];
+                serial.read(0, offset, &mut out_bytes);
+            }
+            1 => {
+                let offset = (bytes.get(i).unwrap_or(&0) % 8) as u64;
+                i += 1;
+                let data = vec![*bytes.get(i).unwrap_or(&0)];
+                i += 1;
+                serial.write(0, offset, &data);
+            }
+            _ => {
+                let data = vec![*bytes.get(i).unwrap_or(&0)];
+                i += 1;
+                serial.queue_input_bytes(&data).ok();
+            }
+        }
+    }
+});
+
+struct TestInterrupt {
+    event_fd: EventFd,
+}
+
+impl InterruptSourceGroup for TestInterrupt {
+    fn trigger(&self, _index: InterruptIndex) -> Result<(), std::io::Error> {
+        self.event_fd.write(1)
+    }
+    fn update(
+        &self,
+        _index: InterruptIndex,
+        _config: InterruptSourceConfig,
+        _masked: bool,
+    ) -> Result<(), std::io::Error> {
+        Ok(())
+    }
+    fn notifier(&self, _index: InterruptIndex) -> Option<EventFd> {
+        Some(self.event_fd.try_clone().unwrap())
+    }
+}
+
+impl TestInterrupt {
+    fn new(event_fd: EventFd) -> Self {
+        TestInterrupt { event_fd }
+    }
+}

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

@@ -11,19 +11,20 @@ mshv = ["mshv-ioctls", "mshv-bindings"]
 tdx = []
 
 [dependencies]
-anyhow = "1.0.58"
-epoll = "4.3.1"
-thiserror = "1.0.31"
-libc = "0.2.126"
+anyhow = "1.0.66"
+byteorder = "1.4.3"
+thiserror = "1.0.37"
+libc = "0.2.138"
 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 = "1.0.138", features = ["rc", "derive"] }
-serde_json = "1.0.82"
-vm-memory = { version = "0.8.0", features = ["backend-mmap", "backend-atomic"] }
-vmm-sys-util = { version = "0.9.0", features = ["with-serde"] }
+serde = { version = "1.0.150", 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 = "0.10.0", 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"
@@ -31,4 +32,4 @@ 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

@@ -1,9 +1,8 @@
 // Copyright 2022 Arm Limited (or its affiliates). All rights reserved.
 
-use crate::{CpuState, Device, GicState, HypervisorDeviceError, HypervisorVmError};
+use crate::{CpuState, GicState, HypervisorDeviceError, HypervisorVmError};
 use std::any::Any;
 use std::result;
-use std::sync::Arc;
 
 /// Errors thrown while setting up the VGIC.
 #[derive(Debug)]
@@ -17,6 +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
@@ -40,8 +51,6 @@ pub trait Vgic: Send + Sync {
     /// Returns the MSI reg property of the device
     fn msi_properties(&self) -> [u64; 2];
 
-    fn set_its_device(&mut self, its_device: Option<Arc<dyn Device>>);
-
     /// Get the values of GICR_TYPER for each vCPU.
     fn set_gicr_typers(&mut self, vcpu_states: &[CpuState]);
 

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/emulator/mod.rs

@@ -7,8 +7,10 @@
 use crate::arch::emulator::{EmulationError, EmulationResult, PlatformEmulator, PlatformError};
 use crate::arch::x86::emulator::instructions::*;
 use crate::arch::x86::regs::{CR0_PE, EFER_LMA};
-use crate::arch::x86::{segment_type_expand_down, segment_type_ro, Exception, SegmentRegisterOps};
-use crate::x86_64::{SegmentRegister, SpecialRegisters, StandardRegisters};
+use crate::arch::x86::{
+    segment_type_expand_down, segment_type_ro, Exception, SegmentRegister, SpecialRegisters,
+    StandardRegisters,
+};
 use anyhow::Context;
 use iced_x86::*;
 

hypervisor/src/arch/x86/gdt.rs

@@ -8,7 +8,7 @@
 // found in the LICENSE-BSD-3-Clause file.
 
 // For GDT details see arch/x86/include/asm/segment.h
-use crate::x86_64::SegmentRegister;
+use crate::arch::x86::SegmentRegister;
 
 /// Constructor for a conventional segment GDT (or LDT) entry. Derived from the kernel's segment.h.
 pub fn gdt_entry(flags: u16, base: u32, limit: u32) -> u64 {
@@ -51,7 +51,7 @@ fn get_limit(entry: u64) -> u32 {
     // Perform manual limit scaling if G flag is set
     match get_g(entry) {
         0 => limit,
-        _ => ((limit << 12) | 0xFFF), // G flag is either 0 or 1
+        _ => (limit << 12) | 0xFFF, // G flag is either 0 or 1
     }
 }
 
@@ -106,7 +106,6 @@ pub fn segment_from_gdt(entry: u64, table_index: u8) -> SegmentRegister {
         l: get_l(entry),
         g: get_g(entry),
         avl: get_avl(entry),
-        padding: 0,
         unusable: match get_p(entry) {
             0 => 1,
             _ => 0,

hypervisor/src/arch/x86/mod.rs

@@ -13,7 +13,6 @@
 
 pub mod emulator;
 pub mod gdt;
-#[allow(non_upper_case_globals)]
 #[allow(non_camel_case_types)]
 #[allow(non_snake_case)]
 #[allow(non_upper_case_globals)]
@@ -27,7 +26,7 @@ pub const MTRR_MEM_TYPE_WB: u64 = 0x6;
 pub const NUM_IOAPIC_PINS: usize = 24;
 
 // X86 Exceptions
-#[allow(dead_code, clippy::upper_case_acronyms)]
+#[allow(clippy::upper_case_acronyms)]
 #[derive(Clone, Debug)]
 pub enum Exception {
     DE = 0,  // Divide Error
@@ -53,40 +52,86 @@ pub enum Exception {
 
 pub mod regs;
 
-// Abstracted segment register ops.
-// Each x86 hypervisor should implement those.
-pub trait SegmentRegisterOps {
-    // Segment type
-    fn segment_type(&self) -> u8;
-    fn set_segment_type(&mut self, val: u8);
+#[derive(Debug, Default, Copy, Clone, PartialEq, Eq)]
+#[cfg_attr(feature = "with-serde", derive(Deserialize, Serialize))]
+pub struct SegmentRegister {
+    pub base: u64,
+    pub limit: u32,
+    pub selector: u16,
+    pub type_: u8,
+    pub present: u8,
+    pub dpl: u8,
+    pub db: u8,
+    pub s: u8,
+    pub l: u8,
+    pub g: u8,
+    pub avl: u8,
+    pub unusable: u8,
+}
 
-    // Descriptor Privilege Level (DPL)
-    fn dpl(&self) -> u8;
-    fn set_dpl(&mut self, val: u8);
+impl SegmentRegister {
+    pub fn segment_type(&self) -> u8 {
+        self.type_
+    }
+    pub fn set_segment_type(&mut self, val: u8) {
+        self.type_ = val;
+    }
 
-    // Granularity
-    fn granularity(&self) -> u8;
-    fn set_granularity(&mut self, val: u8);
+    pub fn dpl(&self) -> u8 {
+        self.dpl
+    }
 
-    // Memory Presence
-    fn present(&self) -> u8;
-    fn set_present(&mut self, val: u8);
+    pub fn set_dpl(&mut self, val: u8) {
+        self.dpl = val;
+    }
 
-    // Long mode
-    fn long(&self) -> u8;
-    fn set_long(&mut self, val: u8);
+    pub fn present(&self) -> u8 {
+        self.present
+    }
 
-    // Available for system use (AVL)
-    fn avl(&self) -> u8;
-    fn set_avl(&mut self, val: u8);
+    pub fn set_present(&mut self, val: u8) {
+        self.present = val;
+    }
 
-    // Descriptor type (System or code/data)
-    fn desc_type(&self) -> u8;
-    fn set_desc_type(&mut self, val: u8);
+    pub fn long(&self) -> u8 {
+        self.l
+    }
 
-    // D/B
-    fn db(&self) -> u8;
-    fn set_db(&mut self, val: u8);
+    pub fn set_long(&mut self, val: u8) {
+        self.l = val;
+    }
+
+    pub fn avl(&self) -> u8 {
+        self.avl
+    }
+
+    pub fn set_avl(&mut self, val: u8) {
+        self.avl = val;
+    }
+
+    pub fn desc_type(&self) -> u8 {
+        self.s
+    }
+
+    pub fn set_desc_type(&mut self, val: u8) {
+        self.s = val;
+    }
+
+    pub fn granularity(&self) -> u8 {
+        self.g
+    }
+
+    pub fn set_granularity(&mut self, val: u8) {
+        self.g = val;
+    }
+
+    pub fn db(&self) -> u8 {
+        self.db
+    }
+
+    pub fn set_db(&mut self, val: u8) {
+        self.db = val;
+    }
 }
 
 // Code segment
@@ -115,7 +160,6 @@ macro_rules! msr {
         MsrEntry {
             index: $msr,
             data: 0x0,
-            ..Default::default()
         }
     };
 }
@@ -125,7 +169,143 @@ macro_rules! msr_data {
         MsrEntry {
             index: $msr,
             data: $data,
-            ..Default::default()
         }
     };
 }
+
+#[derive(Debug, Default, Copy, Clone, PartialEq, Eq)]
+#[cfg_attr(feature = "with-serde", derive(Deserialize, Serialize))]
+pub struct StandardRegisters {
+    pub rax: u64,
+    pub rbx: u64,
+    pub rcx: u64,
+    pub rdx: u64,
+    pub rsi: u64,
+    pub rdi: u64,
+    pub rsp: u64,
+    pub rbp: u64,
+    pub r8: u64,
+    pub r9: u64,
+    pub r10: u64,
+    pub r11: u64,
+    pub r12: u64,
+    pub r13: u64,
+    pub r14: u64,
+    pub r15: u64,
+    pub rip: u64,
+    pub rflags: u64,
+}
+
+#[derive(Debug, Default, Copy, Clone, PartialEq, Eq)]
+#[cfg_attr(feature = "with-serde", derive(Deserialize, Serialize))]
+pub struct DescriptorTable {
+    pub base: u64,
+    pub limit: u16,
+}
+
+#[derive(Debug, Default, Copy, Clone, PartialEq, Eq)]
+#[cfg_attr(feature = "with-serde", derive(Deserialize, Serialize))]
+pub struct SpecialRegisters {
+    pub cs: SegmentRegister,
+    pub ds: SegmentRegister,
+    pub es: SegmentRegister,
+    pub fs: SegmentRegister,
+    pub gs: SegmentRegister,
+    pub ss: SegmentRegister,
+    pub tr: SegmentRegister,
+    pub ldt: SegmentRegister,
+    pub gdt: DescriptorTable,
+    pub idt: DescriptorTable,
+    pub cr0: u64,
+    pub cr2: u64,
+    pub cr3: u64,
+    pub cr4: u64,
+    pub cr8: u64,
+    pub efer: u64,
+    pub apic_base: u64,
+    pub interrupt_bitmap: [u64; 4usize],
+}
+
+#[derive(Debug, Default, Copy, Clone, PartialEq, Eq, serde::Deserialize, serde::Serialize)]
+pub struct CpuIdEntry {
+    pub function: u32,
+    pub index: u32,
+    pub flags: u32,
+    pub eax: u32,
+    pub ebx: u32,
+    pub ecx: u32,
+    pub edx: u32,
+}
+
+pub const CPUID_FLAG_VALID_INDEX: u32 = 1;
+
+#[derive(Default, Debug, Clone, serde::Deserialize, serde::Serialize)]
+pub struct FpuState {
+    pub fpr: [[u8; 16usize]; 8usize],
+    pub fcw: u16,
+    pub fsw: u16,
+    pub ftwx: u8,
+    pub last_opcode: u16,
+    pub last_ip: u64,
+    pub last_dp: u64,
+    pub xmm: [[u8; 16usize]; 16usize],
+    pub mxcsr: u32,
+}
+
+#[serde_with::serde_as]
+#[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
+pub struct LapicState {
+    #[serde_as(as = "[_; 1024usize]")]
+    pub(crate) regs: [::std::os::raw::c_char; 1024usize],
+}
+
+impl Default for LapicState {
+    fn default() -> Self {
+        // SAFETY: this is plain old data structure
+        unsafe { ::std::mem::zeroed() }
+    }
+}
+
+impl LapicState {
+    pub fn get_klapic_reg(&self, reg_offset: usize) -> u32 {
+        use byteorder::{LittleEndian, ReadBytesExt};
+        use std::io::Cursor;
+        use std::mem;
+
+        let sliceu8 = unsafe {
+            // This array is only accessed as parts of a u32 word, so interpret it as a u8 array.
+            // Cursors are only readable on arrays of u8, not i8(c_char).
+            mem::transmute::<&[i8], &[u8]>(&self.regs[reg_offset..])
+        };
+
+        let mut reader = Cursor::new(sliceu8);
+        // Following call can't fail if the offsets defined above are correct.
+        reader
+            .read_u32::<LittleEndian>()
+            .expect("Failed to read klapic register")
+    }
+
+    pub fn set_klapic_reg(&mut self, reg_offset: usize, value: u32) {
+        use byteorder::{LittleEndian, WriteBytesExt};
+        use std::io::Cursor;
+        use std::mem;
+
+        let sliceu8 = unsafe {
+            // This array is only accessed as parts of a u32 word, so interpret it as a u8 array.
+            // Cursors are only readable on arrays of u8, not i8(c_char).
+            mem::transmute::<&mut [i8], &mut [u8]>(&mut self.regs[reg_offset..])
+        };
+
+        let mut writer = Cursor::new(sliceu8);
+        // Following call can't fail if the offsets defined above are correct.
+        writer
+            .write_u32::<LittleEndian>(value)
+            .expect("Failed to write klapic register")
+    }
+}
+
+#[derive(Debug, Default, Copy, Clone, PartialEq, Eq, serde::Deserialize, serde::Serialize)]
+pub struct MsrEntry {
+    pub index: u32,
+    pub data: u64,
+}

hypervisor/src/cpu.rs

@@ -9,28 +9,16 @@
 //
 
 #[cfg(target_arch = "aarch64")]
-use crate::aarch64::VcpuInit;
-#[cfg(target_arch = "aarch64")]
-use crate::aarch64::{RegList, Register, StandardRegisters};
+use crate::aarch64::{RegList, StandardRegisters, VcpuInit};
+#[cfg(target_arch = "x86_64")]
+use crate::arch::x86::{
+    CpuIdEntry, FpuState, LapicState, MsrEntry, SpecialRegisters, StandardRegisters,
+};
 #[cfg(feature = "tdx")]
 use crate::kvm::{TdxExitDetails, TdxExitStatus};
-#[cfg(all(feature = "mshv", target_arch = "x86_64"))]
-use crate::x86_64::SuspendRegisters;
-#[cfg(target_arch = "x86_64")]
-use crate::x86_64::Xsave;
-#[cfg(target_arch = "x86_64")]
-use crate::x86_64::{CpuId, LapicState};
-#[cfg(target_arch = "x86_64")]
-use crate::x86_64::{
-    ExtendedControlRegisters, FpuState, MsrEntries, SpecialRegisters, StandardRegisters, VcpuEvents,
-};
 use crate::CpuState;
-#[cfg(target_arch = "aarch64")]
-use crate::DeviceAttr;
-#[cfg(feature = "kvm")]
 use crate::MpState;
 use thiserror::Error;
-#[cfg(all(feature = "kvm", target_arch = "x86_64"))]
 use vm_memory::GuestAddress;
 
 #[derive(Error, Debug)]
@@ -48,11 +36,6 @@ pub enum HypervisorCpuError {
     #[error("Failed to get standard registers: {0}")]
     GetStandardRegs(#[source] anyhow::Error),
     ///
-    /// Getting suspend registers error
-    ///
-    #[error("Failed to get suspend registers: {0}")]
-    GetSuspendRegs(#[source] anyhow::Error),
-    ///
     /// Setting special register error
     ///
     #[error("Failed to set special registers: {0}")]
@@ -65,12 +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
@@ -248,6 +231,12 @@ pub enum HypervisorCpuError {
     #[cfg(feature = "tdx")]
     #[error("Unknown TDX VM call")]
     UnknownTdxVmCall,
+    #[cfg(target_arch = "aarch64")]
+    ///
+    /// Failed to intialize PMU
+    ///
+    #[error("Failed to initialize PMU")]
+    InitializePmu,
 }
 
 #[derive(Debug)]
@@ -278,22 +267,10 @@ pub type Result<T> = anyhow::Result<T, HypervisorCpuError>;
 /// Trait to represent a generic Vcpu
 ///
 pub trait Vcpu: Send + Sync {
-    #[cfg(target_arch = "x86_64")]
     ///
     /// Returns the vCPU general purpose registers.
     ///
     fn get_regs(&self) -> Result<StandardRegisters>;
-    #[cfg(target_arch = "aarch64")]
-    ///
-    /// Sets vcpu attribute
-    ///
-    fn set_vcpu_attr(&self, attr: &DeviceAttr) -> Result<()>;
-    #[cfg(target_arch = "aarch64")]
-    ///
-    /// Check if vcpu has attribute.
-    ///
-    fn has_vcpu_attr(&self, attr: &DeviceAttr) -> Result<()>;
-    #[cfg(target_arch = "x86_64")]
     ///
     /// Sets the vCPU general purpose registers.
     ///
@@ -322,7 +299,7 @@ pub trait Vcpu: Send + Sync {
     ///
     /// X86 specific call to setup the CPUID registers.
     ///
-    fn set_cpuid2(&self, cpuid: &CpuId) -> Result<()>;
+    fn set_cpuid2(&self, cpuid: &[CpuIdEntry]) -> Result<()>;
     #[cfg(target_arch = "x86_64")]
     ///
     /// X86 specific call to enable HyperV SynIC
@@ -332,7 +309,7 @@ pub trait Vcpu: Send + Sync {
     ///
     /// X86 specific call to retrieve the CPUID registers.
     ///
-    fn get_cpuid2(&self, num_entries: usize) -> Result<CpuId>;
+    fn get_cpuid2(&self, num_entries: usize) -> Result<Vec<CpuIdEntry>>;
     #[cfg(target_arch = "x86_64")]
     ///
     /// Returns the state of the LAPIC (Local Advanced Programmable Interrupt Controller).
@@ -347,117 +324,66 @@ pub trait Vcpu: Send + Sync {
     ///
     /// Returns the model-specific registers (MSR) for this vCPU.
     ///
-    fn get_msrs(&self, msrs: &mut MsrEntries) -> Result<usize>;
+    fn get_msrs(&self, msrs: &mut Vec<MsrEntry>) -> Result<usize>;
     #[cfg(target_arch = "x86_64")]
     ///
     /// Setup the model-specific registers (MSR) for this vCPU.
     ///
-    fn set_msrs(&self, msrs: &MsrEntries) -> Result<usize>;
-    #[cfg(feature = "kvm")]
+    fn set_msrs(&self, msrs: &[MsrEntry]) -> Result<usize>;
     ///
     /// Returns the vcpu's current "multiprocessing state".
     ///
     fn get_mp_state(&self) -> Result<MpState>;
-    #[cfg(feature = "kvm")]
     ///
     /// Sets the vcpu's current "multiprocessing state".
     ///
     fn set_mp_state(&self, mp_state: MpState) -> Result<()>;
     #[cfg(target_arch = "x86_64")]
     ///
-    /// X86 specific call that returns the vcpu's current "xsave struct".
-    ///
-    fn get_xsave(&self) -> Result<Xsave>;
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// X86 specific call that sets the vcpu's current "xsave struct".
-    ///
-    fn set_xsave(&self, xsave: &Xsave) -> Result<()>;
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// X86 specific call that returns the vcpu's current "xcrs".
-    ///
-    fn get_xcrs(&self) -> Result<ExtendedControlRegisters>;
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// X86 specific call that sets the vcpu's current "xcrs".
-    ///
-    fn set_xcrs(&self, xcrs: &ExtendedControlRegisters) -> Result<()>;
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// Returns currently pending exceptions, interrupts, and NMIs as well as related
-    /// states of the vcpu.
-    ///
-    fn get_vcpu_events(&self) -> Result<VcpuEvents>;
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// Sets pending exceptions, interrupts, and NMIs as well as related states
-    /// of the vcpu.
-    ///
-    fn set_vcpu_events(&self, events: &VcpuEvents) -> Result<()>;
-    #[cfg(all(feature = "kvm", target_arch = "x86_64"))]
-    ///
     /// Let the guest know that it has been paused, which prevents from
     /// potential soft lockups when being resumed.
     ///
-    fn notify_guest_clock_paused(&self) -> Result<()>;
-    #[cfg(all(feature = "kvm", target_arch = "x86_64"))]
+    fn notify_guest_clock_paused(&self) -> Result<()> {
+        Ok(())
+    }
     ///
     /// Sets debug registers to set hardware breakpoints and/or enable single step.
     ///
-    fn set_guest_debug(&self, addrs: &[GuestAddress], singlestep: bool) -> Result<()>;
+    fn set_guest_debug(&self, _addrs: &[GuestAddress], _singlestep: bool) -> Result<()> {
+        Err(HypervisorCpuError::SetDebugRegs(anyhow!("unimplemented")))
+    }
     ///
     /// Sets the type of CPU to be exposed to the guest and optional features.
     ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
+    #[cfg(target_arch = "aarch64")]
     fn vcpu_init(&self, kvi: &VcpuInit) -> Result<()>;
     ///
-    /// Sets the value of one register for this vCPU.
-    ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
-    fn set_reg(&self, reg_id: u64, data: u64) -> Result<()>;
-    ///
-    /// Sets the value of one register for this vCPU.
-    ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
-    fn get_reg(&self, reg_id: u64) -> Result<u64>;
-    ///
     /// Gets a list of the guest registers that are supported for the
     /// KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
     ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
+    #[cfg(target_arch = "aarch64")]
     fn get_reg_list(&self, reg_list: &mut RegList) -> Result<()>;
     ///
-    /// Save the state of the core registers.
+    /// Gets the value of a system register
     ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
-    fn core_registers(&self, state: &mut StandardRegisters) -> Result<()>;
-    ///
-    /// Restore the state of the core registers.
-    ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
-    fn set_core_registers(&self, state: &StandardRegisters) -> Result<()>;
-    ///
-    /// Save the state of the system registers.
-    ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
-    fn system_registers(&self, state: &mut Vec<Register>) -> Result<()>;
-    ///
-    /// Restore the state of the system registers.
-    ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
-    fn set_system_registers(&self, state: &[Register]) -> Result<()>;
-    ///
-    /// Read the MPIDR - Multiprocessor Affinity Register.
-    ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
-    fn read_mpidr(&self) -> Result<u64>;
+    #[cfg(target_arch = "aarch64")]
+    fn get_sys_reg(&self, sys_reg: u32) -> Result<u64>;
     ///
     /// Configure core registers for a given CPU.
     ///
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
+    #[cfg(target_arch = "aarch64")]
     fn setup_regs(&self, cpu_id: u8, boot_ip: u64, fdt_start: u64) -> Result<()>;
     ///
+    /// Check if the CPU supports PMU
+    ///
+    #[cfg(target_arch = "aarch64")]
+    fn has_pmu_support(&self) -> bool;
+    ///
+    /// Initialize PMU
+    ///
+    #[cfg(target_arch = "aarch64")]
+    fn init_pmu(&self, irq: u32) -> Result<()>;
+    ///
     /// Retrieve the vCPU state.
     /// This function is necessary to snapshot the VM
     ///
@@ -480,25 +406,30 @@ pub trait Vcpu: Send + Sync {
     /// Initialize TDX support on the vCPU
     ///
     #[cfg(feature = "tdx")]
-    fn tdx_init(&self, hob_address: u64) -> Result<()>;
-    #[cfg(all(feature = "mshv", target_arch = "x86_64"))]
-    ///
-    /// Return suspend registers(explicit and intercept suspend registers)
-    ///
-    fn get_suspend_regs(&self) -> Result<SuspendRegisters>;
-    #[cfg(feature = "kvm")]
+    fn tdx_init(&self, _hob_address: u64) -> Result<()> {
+        unimplemented!()
+    }
     ///
     /// Set the "immediate_exit" state
     ///
-    fn set_immediate_exit(&self, exit: bool);
+    fn set_immediate_exit(&self, _exit: bool) {}
     #[cfg(feature = "tdx")]
     ///
     /// 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
+    ///
+    fn boot_msr_entries(&self) -> Vec<MsrEntry>;
 }

hypervisor/src/device.rs

@@ -9,8 +9,6 @@
 // Copyright 2020, ARM Limited
 //
 
-use crate::DeviceAttr;
-use std::os::unix::io::AsRawFd;
 use thiserror::Error;
 
 #[derive(Error, Debug)]
@@ -28,20 +26,3 @@ pub enum HypervisorDeviceError {
     #[error("Failed to get device attribute: {0}")]
     GetDeviceAttribute(#[source] anyhow::Error),
 }
-
-///
-/// Result type for returning from a function
-///
-pub type Result<T> = std::result::Result<T, HypervisorDeviceError>;
-
-///
-/// Trait to represent a device
-///
-/// This crate provides a hypervisor-agnostic interfaces for device
-///
-pub trait Device: Send + Sync + AsRawFd {
-    /// Set device attribute.
-    fn set_device_attr(&self, attr: &DeviceAttr) -> Result<()>;
-    /// Get device attribute.
-    fn get_device_attr(&self, attr: &mut DeviceAttr) -> Result<()>;
-}

hypervisor/src/hypervisor.rs

@@ -7,13 +7,12 @@
 // Copyright 2018-2019 CrowdStrike, Inc.
 //
 //
+#[cfg(target_arch = "x86_64")]
+use crate::arch::x86::CpuIdEntry;
 #[cfg(feature = "tdx")]
 use crate::kvm::TdxCapabilities;
 use crate::vm::Vm;
-#[cfg(target_arch = "x86_64")]
-use crate::x86_64::CpuId;
-#[cfg(target_arch = "x86_64")]
-use crate::x86_64::MsrList;
+use crate::HypervisorType;
 use std::sync::Arc;
 use thiserror::Error;
 
@@ -21,6 +20,11 @@ use thiserror::Error;
 ///
 ///
 pub enum HypervisorError {
+    ///
+    /// Hypervisor availability check error
+    ///
+    #[error("Failed to check availability of the hypervisor: {0}")]
+    HypervisorAvailableCheck(#[source] anyhow::Error),
     ///
     /// hypervisor creation error
     ///
@@ -84,6 +88,10 @@ pub type Result<T> = std::result::Result<T, HypervisorError>;
 /// This crate provides a hypervisor-agnostic interfaces
 ///
 pub trait Hypervisor: Send + Sync {
+    ///
+    /// Returns the type of the hypervisor
+    ///
+    fn hypervisor_type(&self) -> HypervisorType;
     ///
     /// Create a Vm using the underlying hypervisor
     /// Return a hypervisor-agnostic Vm trait object
@@ -100,26 +108,29 @@ pub trait Hypervisor: Send + Sync {
     ///
     /// Get the supported CpuID
     ///
-    fn get_cpuid(&self) -> Result<CpuId>;
+    fn get_cpuid(&self) -> Result<Vec<CpuIdEntry>>;
     ///
     /// Check particular extensions if any
     ///
     fn check_required_extensions(&self) -> Result<()> {
         Ok(())
     }
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// Retrieve the list of MSRs supported by the hypervisor.
-    ///
-    fn get_msr_list(&self) -> Result<MsrList>;
     #[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/dist_regs.rs

@@ -2,11 +2,11 @@
 // SPDX-License-Identifier: Apache-2.0
 
 use crate::arch::aarch64::gic::{Error, Result};
+use crate::device::HypervisorDeviceError;
 use crate::kvm::kvm_bindings::{
     kvm_device_attr, KVM_DEV_ARM_VGIC_GRP_DIST_REGS, KVM_DEV_ARM_VGIC_GRP_NR_IRQS,
 };
-use crate::Device;
-use std::sync::Arc;
+use kvm_ioctls::DeviceFd;
 
 /*
  Distributor registers as detailed at page 456 from
@@ -77,7 +77,7 @@ static VGIC_DIST_REGS: &[DistReg] = &[
     VGIC_DIST_REG!(GICD_IPRIORITYR, 8, 0),
 ];
 
-fn dist_attr_access(gic: &Arc<dyn Device>, offset: u32, val: &u32, set: bool) -> Result<()> {
+fn dist_attr_access(gic: &DeviceFd, offset: u32, val: &u32, set: bool) -> Result<()> {
     let mut gic_dist_attr = kvm_device_attr {
         group: KVM_DEV_ARM_VGIC_GRP_DIST_REGS,
         attr: offset as u64,
@@ -85,28 +85,30 @@ fn dist_attr_access(gic: &Arc<dyn Device>, offset: u32, val: &u32, set: bool) ->
         flags: 0,
     };
     if set {
-        gic.set_device_attr(&gic_dist_attr)
-            .map_err(Error::SetDeviceAttribute)?;
+        gic.set_device_attr(&gic_dist_attr).map_err(|e| {
+            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
+        })?;
     } else {
-        gic.get_device_attr(&mut gic_dist_attr)
-            .map_err(Error::GetDeviceAttribute)?;
+        gic.get_device_attr(&mut gic_dist_attr).map_err(|e| {
+            Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
+        })?;
     }
     Ok(())
 }
 
 /// Get the distributor control register.
-pub fn read_ctlr(gic: &Arc<dyn Device>) -> Result<u32> {
+pub fn read_ctlr(gic: &DeviceFd) -> Result<u32> {
     let val: u32 = 0;
     dist_attr_access(gic, GICD_CTLR, &val, false)?;
     Ok(val)
 }
 
 /// Set the distributor control register.
-pub fn write_ctlr(gic: &Arc<dyn Device>, val: u32) -> Result<()> {
+pub fn write_ctlr(gic: &DeviceFd, val: u32) -> Result<()> {
     dist_attr_access(gic, GICD_CTLR, &val, true)
 }
 
-fn get_interrupts_num(gic: &Arc<dyn Device>) -> Result<u32> {
+fn get_interrupts_num(gic: &DeviceFd) -> Result<u32> {
     let num_irq = 0;
 
     let mut nr_irqs_attr = kvm_device_attr {
@@ -115,12 +117,13 @@ fn get_interrupts_num(gic: &Arc<dyn Device>) -> Result<u32> {
         addr: &num_irq as *const u32 as u64,
         flags: 0,
     };
-    gic.get_device_attr(&mut nr_irqs_attr)
-        .map_err(Error::GetDeviceAttribute)?;
+    gic.get_device_attr(&mut nr_irqs_attr).map_err(|e| {
+        Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
+    })?;
     Ok(num_irq)
 }
 
-fn compute_reg_len(gic: &Arc<dyn Device>, reg: &DistReg, base: u32) -> Result<u32> {
+fn compute_reg_len(gic: &DeviceFd, reg: &DistReg, base: u32) -> Result<u32> {
     // FIXME:
     // Redefine some GIC constants to avoid the dependency on `layout` crate.
     // This is temporary solution, will be fixed in future refactoring.
@@ -147,7 +150,7 @@ fn compute_reg_len(gic: &Arc<dyn Device>, reg: &DistReg, base: u32) -> Result<u3
 }
 
 /// Set distributor registers of the GIC.
-pub fn set_dist_regs(gic: &Arc<dyn Device>, state: &[u32]) -> Result<()> {
+pub fn set_dist_regs(gic: &DeviceFd, state: &[u32]) -> Result<()> {
     let mut idx = 0;
 
     for dreg in VGIC_DIST_REGS {
@@ -164,7 +167,7 @@ pub fn set_dist_regs(gic: &Arc<dyn Device>, state: &[u32]) -> Result<()> {
     Ok(())
 }
 /// Get distributor registers of the GIC.
-pub fn get_dist_regs(gic: &Arc<dyn Device>) -> Result<Vec<u32>> {
+pub fn get_dist_regs(gic: &DeviceFd) -> Result<Vec<u32>> {
     let mut state = Vec::new();
 
     for dreg in VGIC_DIST_REGS {

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

@@ -3,14 +3,14 @@
 // SPDX-License-Identifier: Apache-2.0
 
 use crate::arch::aarch64::gic::{Error, Result};
+use crate::device::HypervisorDeviceError;
 use crate::kvm::kvm_bindings::{
     kvm_device_attr, KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS, KVM_REG_ARM64_SYSREG_CRM_MASK,
     KVM_REG_ARM64_SYSREG_CRM_SHIFT, KVM_REG_ARM64_SYSREG_CRN_MASK, KVM_REG_ARM64_SYSREG_CRN_SHIFT,
     KVM_REG_ARM64_SYSREG_OP0_MASK, KVM_REG_ARM64_SYSREG_OP0_SHIFT, KVM_REG_ARM64_SYSREG_OP1_MASK,
     KVM_REG_ARM64_SYSREG_OP1_SHIFT, KVM_REG_ARM64_SYSREG_OP2_MASK, KVM_REG_ARM64_SYSREG_OP2_SHIFT,
 };
-use crate::Device;
-use std::sync::Arc;
+use kvm_ioctls::DeviceFd;
 
 const KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT: u32 = 32;
 const KVM_DEV_ARM_VGIC_V3_MPIDR_MASK: u64 = 0xffffffff << KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT as u64;
@@ -79,13 +79,7 @@ static VGIC_ICC_REGS: &[u64] = &[
     SYS_ICC_AP1R3_EL1,
 ];
 
-fn icc_attr_access(
-    gic: &Arc<dyn Device>,
-    offset: u64,
-    typer: u64,
-    val: &u32,
-    set: bool,
-) -> Result<()> {
+fn icc_attr_access(gic: &DeviceFd, offset: u64, typer: u64, val: &u32, set: bool) -> Result<()> {
     let mut gic_icc_attr = kvm_device_attr {
         group: KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS,
         attr: ((typer & KVM_DEV_ARM_VGIC_V3_MPIDR_MASK) | offset), // this needs the mpidr
@@ -93,17 +87,19 @@ fn icc_attr_access(
         flags: 0,
     };
     if set {
-        gic.set_device_attr(&gic_icc_attr)
-            .map_err(Error::SetDeviceAttribute)?;
+        gic.set_device_attr(&gic_icc_attr).map_err(|e| {
+            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
+        })?;
     } else {
-        gic.get_device_attr(&mut gic_icc_attr)
-            .map_err(Error::GetDeviceAttribute)?;
+        gic.get_device_attr(&mut gic_icc_attr).map_err(|e| {
+            Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
+        })?;
     }
     Ok(())
 }
 
 /// Get ICC registers.
-pub fn get_icc_regs(gic: &Arc<dyn Device>, gicr_typer: &[u64]) -> Result<Vec<u32>> {
+pub fn get_icc_regs(gic: &DeviceFd, gicr_typer: &[u64]) -> Result<Vec<u32>> {
     let mut state: Vec<u32> = Vec::new();
     // We need this for the ICC_AP<m>R<n>_EL1 registers.
     let mut num_priority_bits = 0;
@@ -156,7 +152,7 @@ pub fn get_icc_regs(gic: &Arc<dyn Device>, gicr_typer: &[u64]) -> Result<Vec<u32
 }
 
 /// Set ICC registers.
-pub fn set_icc_regs(gic: &Arc<dyn Device>, gicr_typer: &[u64], state: &[u32]) -> Result<()> {
+pub fn set_icc_regs(gic: &DeviceFd, gicr_typer: &[u64], state: &[u32]) -> Result<()> {
     let mut num_priority_bits = 0;
     let mut idx = 0;
     for ix in gicr_typer {

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

@@ -4,16 +4,17 @@ mod dist_regs;
 mod icc_regs;
 mod redist_regs;
 
-use crate::arch::aarch64::gic::{Error, Result, Vgic};
-use crate::kvm::kvm_bindings;
-use crate::{CpuState, Device, Vm};
+use crate::arch::aarch64::gic::{Error, Result, Vgic, VgicConfig};
+use crate::device::HypervisorDeviceError;
+use crate::kvm::{kvm_bindings, KvmVm};
+use crate::{CpuState, Vm};
 use dist_regs::{get_dist_regs, read_ctlr, set_dist_regs, write_ctlr};
 use icc_regs::{get_icc_regs, set_icc_regs};
+use kvm_ioctls::DeviceFd;
 use redist_regs::{construct_gicr_typers, get_redist_regs, set_redist_regs};
 use serde::{Deserialize, Serialize};
 use std::any::Any;
 use std::convert::TryInto;
-use std::sync::Arc;
 
 const GITS_CTLR: u32 = 0x0000;
 const GITS_IIDR: u32 = 0x0004;
@@ -27,7 +28,7 @@ const GITS_BASER: u32 = 0x0100;
 /// This is a helper function to get/set the ITS device attribute depending
 /// the bool parameter `set` provided.
 pub fn gicv3_its_attr_access(
-    its_device: &Arc<dyn Device>,
+    its_device: &DeviceFd,
     group: u32,
     attr: u32,
     val: &u64,
@@ -40,20 +41,22 @@ pub fn gicv3_its_attr_access(
         flags: 0,
     };
     if set {
-        its_device
-            .set_device_attr(&gicv3_its_attr)
-            .map_err(Error::SetDeviceAttribute)
+        its_device.set_device_attr(&gicv3_its_attr).map_err(|e| {
+            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
+        })
     } else {
         its_device
             .get_device_attr(&mut gicv3_its_attr)
-            .map_err(Error::GetDeviceAttribute)
+            .map_err(|e| {
+                Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
+            })
     }
 }
 
 /// Function that saves/restores ITS tables into guest RAM.
 ///
 /// The tables get flushed to guest RAM whenever the VM gets stopped.
-pub fn gicv3_its_tables_access(its_device: &Arc<dyn Device>, save: bool) -> Result<()> {
+pub fn gicv3_its_tables_access(its_device: &DeviceFd, save: bool) -> Result<()> {
     let attr = if save {
         u64::from(kvm_bindings::KVM_DEV_ARM_ITS_SAVE_TABLES)
     } else {
@@ -68,15 +71,15 @@ pub fn gicv3_its_tables_access(its_device: &Arc<dyn Device>, save: bool) -> Resu
     };
     its_device
         .set_device_attr(&init_gic_attr)
-        .map_err(Error::SetDeviceAttribute)
+        .map_err(|e| Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into())))
 }
 
 pub struct KvmGicV3Its {
-    /// The hypervisor agnostic device for the GicV3
-    device: Arc<dyn Device>,
+    /// The KVM device for the GicV3
+    device: DeviceFd,
 
-    /// The hypervisor agnostic device for the Its device
-    its_device: Option<Arc<dyn Device>>,
+    /// The KVM device for the Its device
+    its_device: Option<DeviceFd>,
 
     /// Vector holding values of GICR_TYPER for each vCPU
     gicr_typers: Vec<u64>,
@@ -127,33 +130,21 @@ impl KvmGicV3Its {
         kvm_bindings::kvm_device_type_KVM_DEV_TYPE_ARM_VGIC_V3
     }
 
-    fn device(&self) -> &Arc<dyn Device> {
-        &self.device
-    }
-
-    fn its_device(&self) -> Option<&Arc<dyn Device>> {
-        self.its_device.as_ref()
-    }
-
     /// Setup the device-specific attributes
-    fn init_device_attributes(&mut self, vm: &dyn Vm, nr_irqs: u32) -> Result<()> {
+    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(),
+            &self.device,
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ADDR,
             u64::from(kvm_bindings::KVM_VGIC_V3_ADDR_TYPE_DIST),
             &self.dist_addr as *const u64 as u64,
             0,
         )?;
 
-        /* Setting up the redistributors' attribute.
-        We are calculating here the start of the redistributors address. We have one per CPU.
-        */
+        /* Setting up the redistributors' attribute. */
         Self::set_device_attribute(
-            self.device(),
+            &self.device,
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ADDR,
             u64::from(kvm_bindings::KVM_VGIC_V3_ADDR_TYPE_REDIST),
             &self.redists_addr as *const u64 as u64,
@@ -171,6 +162,9 @@ impl KvmGicV3Its {
             .create_device(&mut its_device)
             .map_err(Error::CreateGic)?;
 
+        // We know vm is KvmVm
+        let its_fd = its_fd.to_kvm().unwrap();
+
         Self::set_device_attribute(
             &its_fd,
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ADDR,
@@ -187,14 +181,14 @@ impl KvmGicV3Its {
             0,
         )?;
 
-        self.set_its_device(Some(its_fd));
+        self.its_device = Some(its_fd);
 
         /* We need to tell the kernel how many irqs to support with this vgic.
          * See the `layout` module for details.
          */
         let nr_irqs_ptr = &nr_irqs as *const u32;
         Self::set_device_attribute(
-            self.device(),
+            &self.device,
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_NR_IRQS,
             0,
             nr_irqs_ptr as u64,
@@ -205,7 +199,7 @@ impl KvmGicV3Its {
          * See https://code.woboq.org/linux/linux/virt/kvm/arm/vgic/vgic-kvm-device.c.html#211.
          */
         Self::set_device_attribute(
-            self.device(),
+            &self.device,
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_CTRL,
             u64::from(kvm_bindings::KVM_DEV_ARM_VGIC_CTRL_INIT),
             0,
@@ -214,19 +208,24 @@ impl KvmGicV3Its {
     }
 
     /// Create a KVM Vgic device
-    fn create_device(vm: &dyn Vm) -> Result<Arc<dyn Device>> {
+    fn create_device(vm: &KvmVm) -> Result<DeviceFd> {
         let mut gic_device = kvm_bindings::kvm_create_device {
             type_: Self::version(),
             fd: 0,
             flags: 0,
         };
 
-        vm.create_device(&mut gic_device).map_err(Error::CreateGic)
+        let device_fd = vm
+            .create_device(&mut gic_device)
+            .map_err(Error::CreateGic)?;
+
+        // We know for sure this is a KVM fd
+        Ok(device_fd.to_kvm().unwrap())
     }
 
     /// Set a GIC device attribute
     fn set_device_attribute(
-        device: &Arc<dyn Device>,
+        device: &DeviceFd,
         group: u32,
         attr: u64,
         addr: u64,
@@ -238,41 +237,33 @@ impl KvmGicV3Its {
             attr,
             addr,
         };
-        device
-            .set_device_attr(&attr)
-            .map_err(Error::SetDeviceAttribute)
+        device.set_device_attr(&attr).map_err(|e| {
+            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
+        })
     }
 
     /// 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)
     }
@@ -312,10 +303,6 @@ impl Vgic for KvmGicV3Its {
         [self.msi_addr, self.msi_size]
     }
 
-    fn set_its_device(&mut self, its_device: Option<Arc<dyn Device>>) {
-        self.its_device = its_device;
-    }
-
     fn set_gicr_typers(&mut self, vcpu_states: &[CpuState]) {
         let gicr_typers = construct_gicr_typers(vcpu_states);
         self.gicr_typers = gicr_typers;
@@ -329,18 +316,18 @@ impl Vgic for KvmGicV3Its {
     fn state(&self) -> Result<Gicv3ItsState> {
         let gicr_typers = self.gicr_typers.clone();
 
-        let gicd_ctlr = read_ctlr(self.device())?;
+        let gicd_ctlr = read_ctlr(&self.device)?;
 
-        let dist_state = get_dist_regs(self.device())?;
+        let dist_state = get_dist_regs(&self.device)?;
 
-        let rdist_state = get_redist_regs(self.device(), &gicr_typers)?;
+        let rdist_state = get_redist_regs(&self.device, &gicr_typers)?;
 
-        let icc_state = get_icc_regs(self.device(), &gicr_typers)?;
+        let icc_state = get_icc_regs(&self.device, &gicr_typers)?;
 
         let its_baser_state: [u64; 8] = [0; 8];
         for i in 0..8 {
             gicv3_its_attr_access(
-                self.its_device().unwrap(),
+                self.its_device.as_ref().unwrap(),
                 kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
                 GITS_BASER + i * 8,
                 &its_baser_state[i as usize],
@@ -350,7 +337,7 @@ impl Vgic for KvmGicV3Its {
 
         let its_ctlr_state: u64 = 0;
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_CTLR,
             &its_ctlr_state,
@@ -359,7 +346,7 @@ impl Vgic for KvmGicV3Its {
 
         let its_cbaser_state: u64 = 0;
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_CBASER,
             &its_cbaser_state,
@@ -368,7 +355,7 @@ impl Vgic for KvmGicV3Its {
 
         let its_creadr_state: u64 = 0;
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_CREADR,
             &its_creadr_state,
@@ -377,7 +364,7 @@ impl Vgic for KvmGicV3Its {
 
         let its_cwriter_state: u64 = 0;
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_CWRITER,
             &its_cwriter_state,
@@ -386,7 +373,7 @@ impl Vgic for KvmGicV3Its {
 
         let its_iidr_state: u64 = 0;
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_IIDR,
             &its_iidr_state,
@@ -411,17 +398,17 @@ impl Vgic for KvmGicV3Its {
     fn set_state(&mut self, state: &Gicv3ItsState) -> Result<()> {
         let gicr_typers = self.gicr_typers.clone();
 
-        write_ctlr(self.device(), state.gicd_ctlr)?;
+        write_ctlr(&self.device, state.gicd_ctlr)?;
 
-        set_dist_regs(self.device(), &state.dist)?;
+        set_dist_regs(&self.device, &state.dist)?;
 
-        set_redist_regs(self.device(), &gicr_typers, &state.rdist)?;
+        set_redist_regs(&self.device, &gicr_typers, &state.rdist)?;
 
-        set_icc_regs(self.device(), &gicr_typers, &state.icc)?;
+        set_icc_regs(&self.device, &gicr_typers, &state.icc)?;
 
         //Restore GICv3ITS registers
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_IIDR,
             &state.its_iidr,
@@ -429,7 +416,7 @@ impl Vgic for KvmGicV3Its {
         )?;
 
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_CBASER,
             &state.its_cbaser,
@@ -437,7 +424,7 @@ impl Vgic for KvmGicV3Its {
         )?;
 
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_CREADR,
             &state.its_creadr,
@@ -445,7 +432,7 @@ impl Vgic for KvmGicV3Its {
         )?;
 
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_CWRITER,
             &state.its_cwriter,
@@ -454,7 +441,7 @@ impl Vgic for KvmGicV3Its {
 
         for i in 0..8 {
             gicv3_its_attr_access(
-                self.its_device().unwrap(),
+                self.its_device.as_ref().unwrap(),
                 kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
                 GITS_BASER + i * 8,
                 &state.its_baser[i as usize],
@@ -463,10 +450,10 @@ impl Vgic for KvmGicV3Its {
         }
 
         // Restore ITS tables
-        gicv3_its_tables_access(self.its_device().unwrap(), false)?;
+        gicv3_its_tables_access(self.its_device.as_ref().unwrap(), false)?;
 
         gicv3_its_attr_access(
-            self.its_device().unwrap(),
+            self.its_device.as_ref().unwrap(),
             kvm_bindings::KVM_DEV_ARM_VGIC_GRP_ITS_REGS,
             GITS_CTLR,
             &state.its_ctlr,
@@ -485,11 +472,11 @@ impl Vgic for KvmGicV3Its {
             addr: 0,
             flags: 0,
         };
-        self.device()
-            .set_device_attr(&init_gic_attr)
-            .map_err(Error::SetDeviceAttribute)?;
+        self.device.set_device_attr(&init_gic_attr).map_err(|e| {
+            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
+        })?;
         // Flush ITS tables to guest RAM.
-        gicv3_its_tables_access(self.its_device().unwrap(), true)
+        gicv3_its_tables_access(self.its_device.as_ref().unwrap(), true)
     }
 }
 
@@ -498,23 +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]
@@ -522,23 +516,14 @@ 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());
+        let res = get_dist_regs(&gic.device);
         assert!(res.is_ok());
         let state = res.unwrap();
         assert_eq!(state.len(), 568);
 
-        let res = set_dist_regs(gic.device(), &state);
+        let res = set_dist_regs(&gic.device, &state);
         assert!(res.is_ok());
     }
 
@@ -547,25 +532,16 @@ 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);
+        let res = get_redist_regs(&gic.device, &gicr_typer);
         assert!(res.is_ok());
         let state = res.unwrap();
         println!("{}", state.len());
         assert!(state.len() == 24);
 
-        assert!(set_redist_regs(gic.device(), &gicr_typer, &state).is_ok());
+        assert!(set_redist_regs(&gic.device, &gicr_typer, &state).is_ok());
     }
 
     #[test]
@@ -573,25 +549,16 @@ 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);
+        let res = get_icc_regs(&gic.device, &gicr_typer);
         assert!(res.is_ok());
         let state = res.unwrap();
         println!("{}", state.len());
         assert!(state.len() == 9);
 
-        assert!(set_icc_regs(gic.device(), &gicr_typer, &state).is_ok());
+        assert!(set_icc_regs(&gic.device, &gicr_typer, &state).is_ok());
     }
 
     #[test]
@@ -600,14 +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

@@ -3,9 +3,16 @@
 // SPDX-License-Identifier: Apache-2.0
 
 use crate::arch::aarch64::gic::{Error, Result};
-use crate::kvm::kvm_bindings::{kvm_device_attr, KVM_DEV_ARM_VGIC_GRP_REDIST_REGS};
-use crate::{CpuState, Device};
-use std::sync::Arc;
+use crate::device::HypervisorDeviceError;
+use crate::kvm::kvm_bindings::{
+    kvm_device_attr, KVM_DEV_ARM_VGIC_GRP_REDIST_REGS, KVM_REG_ARM64, KVM_REG_ARM64_SYSREG,
+    KVM_REG_ARM64_SYSREG_OP0_MASK, KVM_REG_ARM64_SYSREG_OP0_SHIFT, KVM_REG_ARM64_SYSREG_OP2_MASK,
+    KVM_REG_ARM64_SYSREG_OP2_SHIFT, KVM_REG_SIZE_U64,
+};
+use crate::kvm::Register;
+use crate::kvm::VcpuKvmState;
+use crate::CpuState;
+use kvm_ioctls::DeviceFd;
 
 // Relevant redistributor registers that we want to save/restore.
 const GICR_CTLR: u32 = 0x0000;
@@ -32,6 +39,12 @@ 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
+    | KVM_REG_ARM64_SYSREG as u64
+    | (((3_u64) << KVM_REG_ARM64_SYSREG_OP0_SHIFT) & KVM_REG_ARM64_SYSREG_OP0_MASK as u64)
+    | (((5_u64) << KVM_REG_ARM64_SYSREG_OP2_SHIFT) & KVM_REG_ARM64_SYSREG_OP2_MASK as u64);
+
 /// This is how we represent the registers of a distributor.
 /// It is relrvant their offset from the base address of the
 /// distributor.
@@ -83,31 +96,27 @@ static VGIC_SGI_REGS: &[RdistReg] = &[
     VGIC_RDIST_REG!(GICR_IPRIORITYR0, 32),
 ];
 
-fn redist_attr_access(
-    gic: &Arc<dyn Device>,
-    offset: u32,
-    typer: u64,
-    val: &u32,
-    set: bool,
-) -> Result<()> {
-    let mut gic_dist_attr = kvm_device_attr {
+fn redist_attr_access(gic: &DeviceFd, offset: u32, typer: u64, val: &u32, set: bool) -> Result<()> {
+    let mut gic_redist_attr = kvm_device_attr {
         group: KVM_DEV_ARM_VGIC_GRP_REDIST_REGS,
         attr: (typer & KVM_DEV_ARM_VGIC_V3_MPIDR_MASK) | (offset as u64), // this needs the mpidr
         addr: val as *const u32 as u64,
         flags: 0,
     };
     if set {
-        gic.set_device_attr(&gic_dist_attr)
-            .map_err(Error::SetDeviceAttribute)?;
+        gic.set_device_attr(&gic_redist_attr).map_err(|e| {
+            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
+        })?;
     } else {
-        gic.get_device_attr(&mut gic_dist_attr)
-            .map_err(Error::GetDeviceAttribute)?;
+        gic.get_device_attr(&mut gic_redist_attr).map_err(|e| {
+            Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
+        })?;
     }
     Ok(())
 }
 
 fn access_redists_aux(
-    gic: &Arc<dyn Device>,
+    gic: &DeviceFd,
     gicr_typer: &[u64],
     state: &mut Vec<u32>,
     reg_list: &[RdistReg],
@@ -137,7 +146,7 @@ fn access_redists_aux(
 }
 
 /// Get redistributor registers.
-pub fn get_redist_regs(gic: &Arc<dyn Device>, gicr_typer: &[u64]) -> Result<Vec<u32>> {
+pub fn get_redist_regs(gic: &DeviceFd, gicr_typer: &[u64]) -> Result<Vec<u32>> {
     let mut state = Vec::new();
     let mut idx: usize = 0;
     access_redists_aux(
@@ -154,7 +163,7 @@ pub fn get_redist_regs(gic: &Arc<dyn Device>, gicr_typer: &[u64]) -> Result<Vec<
 }
 
 /// Set redistributor registers.
-pub fn set_redist_regs(gic: &Arc<dyn Device>, gicr_typer: &[u64], state: &[u32]) -> Result<()> {
+pub fn set_redist_regs(gic: &DeviceFd, gicr_typer: &[u64], state: &[u32]) -> Result<()> {
     let mut idx: usize = 0;
     let mut mut_state = state.to_owned();
     access_redists_aux(
@@ -190,15 +199,16 @@ 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 last = {
-            if index == vcpu_states.len() - 1 {
-                1
-            } else {
-                0
-            }
-        };
+        let state: VcpuKvmState = state.clone().into();
+        let last = (index == vcpu_states.len() - 1) as u64;
+        // state.sys_regs is a big collection of system registers, including MIPDR_EL1
+        let mpidr: Vec<Register> = state
+            .sys_regs
+            .into_iter()
+            .filter(|reg| reg.id == KVM_ARM64_SYSREG_MPIDR_EL1)
+            .collect();
         //calculate affinity
-        let mut cpu_affid = state.mpidr & 1095233437695;
+        let mut cpu_affid = mpidr[0].addr & 1095233437695;
         cpu_affid = ((cpu_affid & 0xFF00000000) >> 8) | (cpu_affid & 0xFFFFFF);
         gicr_typers.push((cpu_affid << 32) | (1 << 24) | (index as u64) << 8 | (last << 4));
     }

hypervisor/src/kvm/aarch64/mod.rs

@@ -12,12 +12,8 @@ pub mod gic;
 
 use crate::kvm::{KvmError, KvmResult};
 use kvm_bindings::{
-    kvm_mp_state, kvm_one_reg, kvm_regs, KVM_REG_ARM64, KVM_REG_ARM64_SYSREG,
-    KVM_REG_ARM64_SYSREG_CRM_MASK, KVM_REG_ARM64_SYSREG_CRM_SHIFT, KVM_REG_ARM64_SYSREG_CRN_MASK,
-    KVM_REG_ARM64_SYSREG_CRN_SHIFT, KVM_REG_ARM64_SYSREG_OP0_MASK, KVM_REG_ARM64_SYSREG_OP0_SHIFT,
-    KVM_REG_ARM64_SYSREG_OP1_MASK, KVM_REG_ARM64_SYSREG_OP1_SHIFT, KVM_REG_ARM64_SYSREG_OP2_MASK,
-    KVM_REG_ARM64_SYSREG_OP2_SHIFT, KVM_REG_ARM_COPROC_MASK, KVM_REG_ARM_CORE, KVM_REG_SIZE_MASK,
-    KVM_REG_SIZE_U32, KVM_REG_SIZE_U64,
+    kvm_mp_state, kvm_one_reg, kvm_regs, KVM_REG_ARM_COPROC_MASK, KVM_REG_ARM_CORE,
+    KVM_REG_SIZE_MASK, KVM_REG_SIZE_U32, KVM_REG_SIZE_U64,
 };
 pub use kvm_bindings::{
     kvm_one_reg as Register, kvm_regs as StandardRegisters, kvm_vcpu_init as VcpuInit, RegList,
@@ -85,32 +81,6 @@ macro_rules! arm64_core_reg_id {
     };
 }
 
-// This macro computes the ID of a specific ARM64 system register similar to how
-// the kernel C macro does.
-// https://elixir.bootlin.com/linux/v4.20.17/source/arch/arm64/include/uapi/asm/kvm.h#L203
-#[macro_export]
-macro_rules! arm64_sys_reg {
-    ($name: tt, $op0: tt, $op1: tt, $crn: tt, $crm: tt, $op2: tt) => {
-        pub const $name: u64 = KVM_REG_ARM64 as u64
-            | KVM_REG_SIZE_U64 as u64
-            | KVM_REG_ARM64_SYSREG as u64
-            | ((($op0 as u64) << KVM_REG_ARM64_SYSREG_OP0_SHIFT)
-                & KVM_REG_ARM64_SYSREG_OP0_MASK as u64)
-            | ((($op1 as u64) << KVM_REG_ARM64_SYSREG_OP1_SHIFT)
-                & KVM_REG_ARM64_SYSREG_OP1_MASK as u64)
-            | ((($crn as u64) << KVM_REG_ARM64_SYSREG_CRN_SHIFT)
-                & KVM_REG_ARM64_SYSREG_CRN_MASK as u64)
-            | ((($crm as u64) << KVM_REG_ARM64_SYSREG_CRM_SHIFT)
-                & KVM_REG_ARM64_SYSREG_CRM_MASK as u64)
-            | ((($op2 as u64) << KVM_REG_ARM64_SYSREG_OP2_SHIFT)
-                & KVM_REG_ARM64_SYSREG_OP2_MASK as u64);
-    };
-}
-
-// Constant imported from the Linux kernel:
-// https://elixir.bootlin.com/linux/v4.20.17/source/arch/arm64/include/asm/sysreg.h#L135
-arm64_sys_reg!(MPIDR_EL1, 3, 0, 0, 0, 5);
-
 /// Specifies whether a particular register is a system register or not.
 /// The kernel splits the registers on aarch64 in core registers and system registers.
 /// So, below we get the system registers by checking that they are not core registers.
@@ -127,8 +97,7 @@ 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
@@ -149,8 +118,4 @@ pub struct VcpuKvmState {
     pub mp_state: kvm_mp_state,
     pub core_regs: kvm_regs,
     pub sys_regs: Vec<kvm_one_reg>,
-    // We will be using the mpidr for passing it to the VmState.
-    // The VmState will give this away for saving restoring the icc and redistributor
-    // registers.
-    pub mpidr: u64,
 }

hypervisor/src/kvm/x86_64/mod.rs

@@ -8,7 +8,10 @@
 //
 //
 
-use crate::arch::x86::{msr_index, SegmentRegisterOps, MTRR_ENABLE, MTRR_MEM_TYPE_WB};
+use crate::arch::x86::{
+    CpuIdEntry, DescriptorTable, FpuState, LapicState, MsrEntry, SegmentRegister, SpecialRegisters,
+    StandardRegisters, CPUID_FLAG_VALID_INDEX,
+};
 use crate::kvm::{Cap, Kvm, KvmError, KvmResult};
 use serde::{Deserialize, Serialize};
 
@@ -16,101 +19,15 @@ use serde::{Deserialize, Serialize};
 /// Export generically-named wrappers of kvm-bindings for Unix-based platforms
 ///
 pub use {
-    kvm_bindings::kvm_cpuid_entry2 as CpuIdEntry, kvm_bindings::kvm_dtable as DescriptorTable,
-    kvm_bindings::kvm_fpu as FpuState, kvm_bindings::kvm_lapic_state as LapicState,
-    kvm_bindings::kvm_mp_state as MpState, kvm_bindings::kvm_msr_entry as MsrEntry,
-    kvm_bindings::kvm_regs as StandardRegisters, kvm_bindings::kvm_segment as SegmentRegister,
-    kvm_bindings::kvm_sregs as SpecialRegisters, kvm_bindings::kvm_vcpu_events as VcpuEvents,
+    kvm_bindings::kvm_cpuid_entry2, kvm_bindings::kvm_dtable, kvm_bindings::kvm_fpu,
+    kvm_bindings::kvm_lapic_state, kvm_bindings::kvm_mp_state as MpState,
+    kvm_bindings::kvm_msr_entry, kvm_bindings::kvm_regs, kvm_bindings::kvm_segment,
+    kvm_bindings::kvm_sregs, kvm_bindings::kvm_vcpu_events as VcpuEvents,
     kvm_bindings::kvm_xcrs as ExtendedControlRegisters, kvm_bindings::kvm_xsave as Xsave,
     kvm_bindings::CpuId, kvm_bindings::MsrList, kvm_bindings::Msrs as MsrEntries,
-    kvm_bindings::KVM_CPUID_FLAG_SIGNIFCANT_INDEX as CPUID_FLAG_VALID_INDEX,
+    kvm_bindings::KVM_CPUID_FLAG_SIGNIFCANT_INDEX,
 };
 
-impl SegmentRegisterOps for SegmentRegister {
-    fn segment_type(&self) -> u8 {
-        self.type_
-    }
-    fn set_segment_type(&mut self, val: u8) {
-        self.type_ = val;
-    }
-
-    fn dpl(&self) -> u8 {
-        self.dpl
-    }
-
-    fn set_dpl(&mut self, val: u8) {
-        self.dpl = val;
-    }
-
-    fn present(&self) -> u8 {
-        self.present
-    }
-
-    fn set_present(&mut self, val: u8) {
-        self.present = val;
-    }
-
-    fn long(&self) -> u8 {
-        self.l
-    }
-
-    fn set_long(&mut self, val: u8) {
-        self.l = val;
-    }
-
-    fn avl(&self) -> u8 {
-        self.avl
-    }
-
-    fn set_avl(&mut self, val: u8) {
-        self.avl = val;
-    }
-
-    fn desc_type(&self) -> u8 {
-        self.s
-    }
-
-    fn set_desc_type(&mut self, val: u8) {
-        self.s = val;
-    }
-
-    fn granularity(&self) -> u8 {
-        self.g
-    }
-
-    fn set_granularity(&mut self, val: u8) {
-        self.g = val;
-    }
-
-    fn db(&self) -> u8 {
-        self.db
-    }
-
-    fn set_db(&mut self, val: u8) {
-        self.db = val;
-    }
-}
-
-pub fn boot_msr_entries() -> MsrEntries {
-    MsrEntries::from_entries(&[
-        msr!(msr_index::MSR_IA32_SYSENTER_CS),
-        msr!(msr_index::MSR_IA32_SYSENTER_ESP),
-        msr!(msr_index::MSR_IA32_SYSENTER_EIP),
-        msr!(msr_index::MSR_STAR),
-        msr!(msr_index::MSR_CSTAR),
-        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_IA32_MISC_ENABLE,
-            msr_index::MSR_IA32_MISC_ENABLE_FAST_STRING as u64
-        ),
-        msr_data!(msr_index::MSR_MTRRdefType, MTRR_ENABLE | MTRR_MEM_TYPE_WB),
-    ])
-    .unwrap()
-}
-
 ///
 /// Check KVM extension for Linux
 ///
@@ -137,14 +54,275 @@ pub fn check_required_kvm_extensions(kvm: &Kvm) -> KvmResult<()> {
 }
 #[derive(Clone, Serialize, Deserialize)]
 pub struct VcpuKvmState {
-    pub cpuid: CpuId,
-    pub msrs: MsrEntries,
+    pub cpuid: Vec<CpuIdEntry>,
+    pub msrs: Vec<MsrEntry>,
     pub vcpu_events: VcpuEvents,
-    pub regs: StandardRegisters,
-    pub sregs: SpecialRegisters,
+    pub regs: kvm_regs,
+    pub sregs: kvm_sregs,
     pub fpu: FpuState,
     pub lapic_state: LapicState,
     pub xsave: Xsave,
     pub xcrs: ExtendedControlRegisters,
     pub mp_state: MpState,
 }
+
+impl From<StandardRegisters> for kvm_regs {
+    fn from(regs: StandardRegisters) -> Self {
+        Self {
+            rax: regs.rax,
+            rbx: regs.rbx,
+            rcx: regs.rcx,
+            rdx: regs.rdx,
+            rsi: regs.rsi,
+            rdi: regs.rdi,
+            rsp: regs.rsp,
+            rbp: regs.rbp,
+            r8: regs.r8,
+            r9: regs.r9,
+            r10: regs.r10,
+            r11: regs.r11,
+            r12: regs.r12,
+            r13: regs.r13,
+            r14: regs.r14,
+            r15: regs.r15,
+            rip: regs.rip,
+            rflags: regs.rflags,
+        }
+    }
+}
+
+impl From<kvm_regs> for StandardRegisters {
+    fn from(regs: kvm_regs) -> Self {
+        Self {
+            rax: regs.rax,
+            rbx: regs.rbx,
+            rcx: regs.rcx,
+            rdx: regs.rdx,
+            rsi: regs.rsi,
+            rdi: regs.rdi,
+            rsp: regs.rsp,
+            rbp: regs.rbp,
+            r8: regs.r8,
+            r9: regs.r9,
+            r10: regs.r10,
+            r11: regs.r11,
+            r12: regs.r12,
+            r13: regs.r13,
+            r14: regs.r14,
+            r15: regs.r15,
+            rip: regs.rip,
+            rflags: regs.rflags,
+        }
+    }
+}
+
+impl From<SegmentRegister> for kvm_segment {
+    fn from(s: SegmentRegister) -> Self {
+        Self {
+            base: s.base,
+            limit: s.limit,
+            selector: s.selector,
+            type_: s.type_,
+            present: s.present,
+            dpl: s.dpl,
+            db: s.db,
+            s: s.s,
+            l: s.l,
+            g: s.g,
+            avl: s.avl,
+            unusable: s.unusable,
+            ..Default::default()
+        }
+    }
+}
+
+impl From<kvm_segment> for SegmentRegister {
+    fn from(s: kvm_segment) -> Self {
+        Self {
+            base: s.base,
+            limit: s.limit,
+            selector: s.selector,
+            type_: s.type_,
+            present: s.present,
+            dpl: s.dpl,
+            db: s.db,
+            s: s.s,
+            l: s.l,
+            g: s.g,
+            avl: s.avl,
+            unusable: s.unusable,
+        }
+    }
+}
+
+impl From<DescriptorTable> for kvm_dtable {
+    fn from(dt: DescriptorTable) -> Self {
+        Self {
+            base: dt.base,
+            limit: dt.limit,
+            ..Default::default()
+        }
+    }
+}
+
+impl From<kvm_dtable> for DescriptorTable {
+    fn from(dt: kvm_dtable) -> Self {
+        Self {
+            base: dt.base,
+            limit: dt.limit,
+        }
+    }
+}
+
+impl From<SpecialRegisters> for kvm_sregs {
+    fn from(s: SpecialRegisters) -> Self {
+        Self {
+            cs: s.cs.into(),
+            ds: s.ds.into(),
+            es: s.es.into(),
+            fs: s.fs.into(),
+            gs: s.gs.into(),
+            ss: s.ss.into(),
+            tr: s.tr.into(),
+            ldt: s.ldt.into(),
+            gdt: s.gdt.into(),
+            idt: s.idt.into(),
+            cr0: s.cr0,
+            cr2: s.cr2,
+            cr3: s.cr3,
+            cr4: s.cr4,
+            cr8: s.cr8,
+            efer: s.efer,
+            apic_base: s.apic_base,
+            interrupt_bitmap: s.interrupt_bitmap,
+        }
+    }
+}
+
+impl From<kvm_sregs> for SpecialRegisters {
+    fn from(s: kvm_sregs) -> Self {
+        Self {
+            cs: s.cs.into(),
+            ds: s.ds.into(),
+            es: s.es.into(),
+            fs: s.fs.into(),
+            gs: s.gs.into(),
+            ss: s.ss.into(),
+            tr: s.tr.into(),
+            ldt: s.ldt.into(),
+            gdt: s.gdt.into(),
+            idt: s.idt.into(),
+            cr0: s.cr0,
+            cr2: s.cr2,
+            cr3: s.cr3,
+            cr4: s.cr4,
+            cr8: s.cr8,
+            efer: s.efer,
+            apic_base: s.apic_base,
+            interrupt_bitmap: s.interrupt_bitmap,
+        }
+    }
+}
+
+impl From<CpuIdEntry> for kvm_cpuid_entry2 {
+    fn from(e: CpuIdEntry) -> Self {
+        let flags = if e.flags & CPUID_FLAG_VALID_INDEX != 0 {
+            KVM_CPUID_FLAG_SIGNIFCANT_INDEX
+        } else {
+            0
+        };
+        Self {
+            function: e.function,
+            index: e.index,
+            flags,
+            eax: e.eax,
+            ebx: e.ebx,
+            ecx: e.ecx,
+            edx: e.edx,
+            ..Default::default()
+        }
+    }
+}
+
+impl From<kvm_cpuid_entry2> for CpuIdEntry {
+    fn from(e: kvm_cpuid_entry2) -> Self {
+        let flags = if e.flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX != 0 {
+            CPUID_FLAG_VALID_INDEX
+        } else {
+            0
+        };
+        Self {
+            function: e.function,
+            index: e.index,
+            flags,
+            eax: e.eax,
+            ebx: e.ebx,
+            ecx: e.ecx,
+            edx: e.edx,
+        }
+    }
+}
+
+impl From<kvm_fpu> for FpuState {
+    fn from(s: kvm_fpu) -> Self {
+        Self {
+            fpr: s.fpr,
+            fcw: s.fcw,
+            fsw: s.fsw,
+            ftwx: s.ftwx,
+            last_opcode: s.last_opcode,
+            last_ip: s.last_ip,
+            last_dp: s.last_dp,
+            xmm: s.xmm,
+            mxcsr: s.mxcsr,
+        }
+    }
+}
+
+impl From<FpuState> for kvm_fpu {
+    fn from(s: FpuState) -> Self {
+        Self {
+            fpr: s.fpr,
+            fcw: s.fcw,
+            fsw: s.fsw,
+            ftwx: s.ftwx,
+            last_opcode: s.last_opcode,
+            last_ip: s.last_ip,
+            last_dp: s.last_dp,
+            xmm: s.xmm,
+            mxcsr: s.mxcsr,
+            ..Default::default()
+        }
+    }
+}
+
+impl From<LapicState> for kvm_lapic_state {
+    fn from(s: LapicState) -> Self {
+        Self { regs: s.regs }
+    }
+}
+
+impl From<kvm_lapic_state> for LapicState {
+    fn from(s: kvm_lapic_state) -> Self {
+        Self { regs: s.regs }
+    }
+}
+
+impl From<kvm_msr_entry> for MsrEntry {
+    fn from(e: kvm_msr_entry) -> Self {
+        Self {
+            index: e.index,
+            data: e.data,
+        }
+    }
+}
+
+impl From<MsrEntry> for kvm_msr_entry {
+    fn from(e: MsrEntry) -> Self {
+        Self {
+            index: e.index,
+            data: e.data,
+            ..Default::default()
+        }
+    }
+}

hypervisor/src/lib.rs

@@ -51,40 +51,36 @@ mod cpu;
 mod device;
 
 pub use cpu::{HypervisorCpuError, Vcpu, VmExit};
-pub use device::{Device, HypervisorDeviceError};
+pub use device::HypervisorDeviceError;
 pub use hypervisor::{Hypervisor, HypervisorError};
-#[cfg(all(feature = "kvm", target_arch = "x86_64"))]
-pub use kvm::x86_64;
 #[cfg(all(feature = "kvm", target_arch = "aarch64"))]
 pub use kvm::{aarch64, GicState};
-// Aliased types exposed from both hypervisors
-#[cfg(feature = "kvm")]
-pub use kvm::{
-    ClockData, CpuState, CreateDevice, DeviceAttr, DeviceFd, IoEventAddress, IrqRoutingEntry,
-    MemoryRegion, MpState, VcpuEvents, VcpuExit, VmState,
-};
-#[cfg(all(feature = "mshv", target_arch = "x86_64"))]
-pub use mshv::x86_64;
-// Aliased types exposed from both hypervisors
-#[cfg(all(feature = "mshv", target_arch = "x86_64"))]
-pub use mshv::{
-    CpuState, CreateDevice, DeviceAttr, DeviceFd, IoEventAddress, IrqRoutingEntry, MemoryRegion,
-    MpState, VcpuEvents, VcpuExit, VmState,
-};
 use std::sync::Arc;
 pub use vm::{
     DataMatch, HypervisorVmError, InterruptSourceConfig, LegacyIrqSourceConfig, MsiIrqSourceConfig,
     Vm, VmOps,
 };
 
+#[derive(Debug, Copy, Clone, Eq, PartialEq)]
+pub enum HypervisorType {
+    Kvm,
+    Mshv,
+}
+
 pub fn new() -> std::result::Result<Arc<dyn Hypervisor>, HypervisorError> {
     #[cfg(feature = "kvm")]
-    let hv = kvm::KvmHypervisor::new()?;
+    if kvm::KvmHypervisor::is_available()? {
+        return kvm::KvmHypervisor::new();
+    }
 
     #[cfg(feature = "mshv")]
-    let hv = mshv::MshvHypervisor::new()?;
+    if mshv::MshvHypervisor::is_available()? {
+        return mshv::MshvHypervisor::new();
+    }
 
-    Ok(Arc::new(hv))
+    Err(HypervisorError::HypervisorCreate(anyhow!(
+        "no supported hypervisor"
+    )))
 }
 
 // Returns a `Vec<T>` with a size in bytes at least as large as `size_in_bytes`.
@@ -116,3 +112,75 @@ pub fn vec_with_array_field<T: Default, F>(count: usize) -> Vec<T> {
     let vec_size_bytes = size_of::<T>() + element_space;
     vec_with_size_in_bytes(vec_size_bytes)
 }
+
+///
+/// User memory region structure
+///
+#[derive(Debug, Default, Eq, PartialEq)]
+pub struct UserMemoryRegion {
+    pub slot: u32,
+    pub guest_phys_addr: u64,
+    pub memory_size: u64,
+    pub userspace_addr: u64,
+    pub flags: u32,
+}
+
+///
+/// Flags for user memory region
+///
+pub const USER_MEMORY_REGION_READ: u32 = 1;
+pub const USER_MEMORY_REGION_WRITE: u32 = 1 << 1;
+pub const USER_MEMORY_REGION_EXECUTE: u32 = 1 << 2;
+pub const USER_MEMORY_REGION_LOG_DIRTY: u32 = 1 << 3;
+
+#[derive(Debug)]
+pub enum MpState {
+    #[cfg(feature = "kvm")]
+    Kvm(kvm_bindings::kvm_mp_state),
+    #[cfg(all(feature = "mshv", target_arch = "x86_64"))]
+    Mshv, /* MSHV does not supprt MpState yet */
+}
+
+#[derive(Debug, Clone, Copy)]
+pub enum IoEventAddress {
+    Pio(u64),
+    Mmio(u64),
+}
+
+#[derive(Clone, serde::Serialize, serde::Deserialize)]
+#[allow(clippy::large_enum_variant)]
+pub enum CpuState {
+    #[cfg(feature = "kvm")]
+    Kvm(kvm::VcpuKvmState),
+    #[cfg(all(feature = "mshv", target_arch = "x86_64"))]
+    Mshv(mshv::VcpuMshvState),
+}
+
+#[derive(Debug, Clone, Copy, serde::Serialize, serde::Deserialize)]
+#[cfg(target_arch = "x86_64")]
+pub enum ClockData {
+    #[cfg(feature = "kvm")]
+    Kvm(kvm_bindings::kvm_clock_data),
+    #[cfg(feature = "mshv")]
+    Mshv, /* MSHV does not supprt ClockData yet */
+}
+
+#[cfg(target_arch = "x86_64")]
+impl ClockData {
+    pub fn reset_flags(&mut self) {
+        match self {
+            #[cfg(feature = "kvm")]
+            ClockData::Kvm(s) => s.flags = 0,
+            #[allow(unreachable_patterns)]
+            _ => {}
+        }
+    }
+}
+
+#[derive(Copy, Clone)]
+pub enum IrqRoutingEntry {
+    #[cfg(feature = "kvm")]
+    Kvm(kvm_bindings::kvm_irq_routing_entry),
+    #[cfg(feature = "mshv")]
+    Mshv(mshv_bindings::mshv_msi_routing_entry),
+}

hypervisor/src/mshv/mod.rs

@@ -12,26 +12,35 @@ use crate::cpu::Vcpu;
 use crate::hypervisor;
 use crate::vec_with_array_field;
 use crate::vm::{self, InterruptSourceConfig, VmOps};
+use crate::HypervisorType;
 pub use mshv_bindings::*;
-pub use mshv_ioctls::IoEventAddress;
 use mshv_ioctls::{set_registers_64, Mshv, NoDatamatch, VcpuFd, VmFd};
-use serde::{Deserialize, Serialize};
+use std::any::Any;
 use std::collections::HashMap;
 use std::sync::{Arc, RwLock};
+use vfio_ioctls::VfioDeviceFd;
 use vm::DataMatch;
 // x86_64 dependencies
 #[cfg(target_arch = "x86_64")]
 pub mod x86_64;
-use crate::device;
+use crate::{
+    ClockData, CpuState, IoEventAddress, IrqRoutingEntry, MpState, UserMemoryRegion,
+    USER_MEMORY_REGION_EXECUTE, USER_MEMORY_REGION_READ, USER_MEMORY_REGION_WRITE,
+};
 use vmm_sys_util::eventfd::EventFd;
 #[cfg(target_arch = "x86_64")]
-pub use x86_64::VcpuMshvState as CpuState;
+pub use x86_64::VcpuMshvState;
 #[cfg(target_arch = "x86_64")]
 pub use x86_64::*;
 
 #[cfg(target_arch = "x86_64")]
 use std::fs::File;
-use std::os::unix::io::{AsRawFd, RawFd};
+use std::os::unix::io::AsRawFd;
+
+#[cfg(target_arch = "x86_64")]
+use crate::arch::x86::{
+    CpuIdEntry, FpuState, LapicState, MsrEntry, SpecialRegisters, StandardRegisters,
+};
 
 const DIRTY_BITMAP_CLEAR_DIRTY: u64 = 0x4;
 const DIRTY_BITMAP_SET_DIRTY: u64 = 0x8;
@@ -41,18 +50,108 @@ const DIRTY_BITMAP_SET_DIRTY: u64 = 0x8;
 ///
 pub use {
     mshv_bindings::mshv_create_device as CreateDevice,
-    mshv_bindings::mshv_device_attr as DeviceAttr,
-    mshv_bindings::mshv_msi_routing_entry as IrqRoutingEntry, mshv_ioctls::DeviceFd,
+    mshv_bindings::mshv_device_attr as DeviceAttr, mshv_ioctls::DeviceFd,
 };
 
 pub const PAGE_SHIFT: usize = 12;
 
-#[derive(Debug, Default, Copy, Clone, Serialize, Deserialize)]
-pub struct HvState {
-    hypercall_page: u64,
+impl From<mshv_user_mem_region> for UserMemoryRegion {
+    fn from(region: mshv_user_mem_region) -> Self {
+        let mut flags: u32 = 0;
+        if region.flags & HV_MAP_GPA_READABLE != 0 {
+            flags |= USER_MEMORY_REGION_READ;
+        }
+        if region.flags & HV_MAP_GPA_WRITABLE != 0 {
+            flags |= USER_MEMORY_REGION_WRITE;
+        }
+        if region.flags & HV_MAP_GPA_EXECUTABLE != 0 {
+            flags |= USER_MEMORY_REGION_EXECUTE;
+        }
+
+        UserMemoryRegion {
+            guest_phys_addr: (region.guest_pfn << PAGE_SHIFT as u64)
+                + (region.userspace_addr & ((1 << PAGE_SHIFT) - 1)),
+            memory_size: region.size,
+            userspace_addr: region.userspace_addr,
+            flags,
+            ..Default::default()
+        }
+    }
 }
 
-pub use HvState as VmState;
+impl From<UserMemoryRegion> for mshv_user_mem_region {
+    fn from(region: UserMemoryRegion) -> Self {
+        let mut flags: u32 = 0;
+        if region.flags & USER_MEMORY_REGION_READ != 0 {
+            flags |= HV_MAP_GPA_READABLE;
+        }
+        if region.flags & USER_MEMORY_REGION_WRITE != 0 {
+            flags |= HV_MAP_GPA_WRITABLE;
+        }
+        if region.flags & USER_MEMORY_REGION_EXECUTE != 0 {
+            flags |= HV_MAP_GPA_EXECUTABLE;
+        }
+
+        mshv_user_mem_region {
+            guest_pfn: region.guest_phys_addr >> PAGE_SHIFT,
+            size: region.memory_size,
+            userspace_addr: region.userspace_addr,
+            flags,
+        }
+    }
+}
+
+impl From<mshv_ioctls::IoEventAddress> for IoEventAddress {
+    fn from(a: mshv_ioctls::IoEventAddress) -> Self {
+        match a {
+            mshv_ioctls::IoEventAddress::Pio(x) => Self::Pio(x),
+            mshv_ioctls::IoEventAddress::Mmio(x) => Self::Mmio(x),
+        }
+    }
+}
+
+impl From<IoEventAddress> for mshv_ioctls::IoEventAddress {
+    fn from(a: IoEventAddress) -> Self {
+        match a {
+            IoEventAddress::Pio(x) => Self::Pio(x),
+            IoEventAddress::Mmio(x) => Self::Mmio(x),
+        }
+    }
+}
+
+impl From<VcpuMshvState> for CpuState {
+    fn from(s: VcpuMshvState) -> Self {
+        CpuState::Mshv(s)
+    }
+}
+
+impl From<CpuState> for VcpuMshvState {
+    fn from(s: CpuState) -> Self {
+        match s {
+            CpuState::Mshv(s) => s,
+            /* Needed in case other hypervisors are enabled */
+            #[allow(unreachable_patterns)]
+            _ => panic!("CpuState is not valid"),
+        }
+    }
+}
+
+impl From<mshv_msi_routing_entry> for IrqRoutingEntry {
+    fn from(s: mshv_msi_routing_entry) -> Self {
+        IrqRoutingEntry::Mshv(s)
+    }
+}
+
+impl From<IrqRoutingEntry> for mshv_msi_routing_entry {
+    fn from(e: IrqRoutingEntry) -> Self {
+        match e {
+            IrqRoutingEntry::Mshv(e) => e,
+            /* Needed in case other hypervisors are enabled */
+            #[allow(unreachable_patterns)]
+            _ => panic!("IrqRoutingEntry is not valid"),
+        }
+    }
+}
 
 struct MshvDirtyLogSlot {
     guest_pfn: u64,
@@ -64,12 +163,35 @@ pub struct MshvHypervisor {
     mshv: Mshv,
 }
 
+impl MshvHypervisor {
+    #[cfg(target_arch = "x86_64")]
+    ///
+    /// Retrieve the list of MSRs supported by MSHV.
+    ///
+    fn get_msr_list(&self) -> hypervisor::Result<MsrList> {
+        self.mshv
+            .get_msr_index_list()
+            .map_err(|e| hypervisor::HypervisorError::GetMsrList(e.into()))
+    }
+}
+
 impl MshvHypervisor {
     /// Create a hypervisor based on Mshv
-    pub fn new() -> hypervisor::Result<MshvHypervisor> {
+    #[allow(clippy::new_ret_no_self)]
+    pub fn new() -> hypervisor::Result<Arc<dyn hypervisor::Hypervisor>> {
         let mshv_obj =
             Mshv::new().map_err(|e| hypervisor::HypervisorError::HypervisorCreate(e.into()))?;
-        Ok(MshvHypervisor { mshv: mshv_obj })
+        Ok(Arc::new(MshvHypervisor { mshv: mshv_obj }))
+    }
+    /// Check if the hypervisor is available
+    pub fn is_available() -> hypervisor::Result<bool> {
+        match std::fs::metadata("/dev/mshv") {
+            Ok(_) => Ok(true),
+            Err(err) if err.kind() == std::io::ErrorKind::NotFound => Ok(false),
+            Err(err) => Err(hypervisor::HypervisorError::HypervisorAvailableCheck(
+                err.into(),
+            )),
+        }
     }
 }
 /// Implementation of Hypervisor trait for Mshv
@@ -81,6 +203,12 @@ impl MshvHypervisor {
 /// let vm = hypervisor.create_vm().expect("new VM fd creation failed");
 ///
 impl hypervisor::Hypervisor for MshvHypervisor {
+    ///
+    /// Returns the type of the hypervisor
+    ///
+    fn hypervisor_type(&self) -> HypervisorType {
+        HypervisorType::Mshv
+    }
     /// Create a mshv vm object and return the object as Vm trait object
     /// Example
     /// # extern crate hypervisor;
@@ -121,47 +249,38 @@ impl hypervisor::Hypervisor for MshvHypervisor {
 
         let msr_list = self.get_msr_list()?;
         let num_msrs = msr_list.as_fam_struct_ref().nmsrs as usize;
-        let mut msrs = MsrEntries::new(num_msrs).unwrap();
+        let mut msrs: Vec<MsrEntry> = vec![
+            MsrEntry {
+                ..Default::default()
+            };
+            num_msrs
+        ];
         let indices = msr_list.as_slice();
-        let msr_entries = msrs.as_mut_slice();
         for (pos, index) in indices.iter().enumerate() {
-            msr_entries[pos].index = *index;
+            msrs[pos].index = *index;
         }
         let vm_fd = Arc::new(fd);
 
         Ok(Arc::new(MshvVm {
             fd: vm_fd,
             msrs,
-            hv_state: hv_state_init(),
-            vm_ops: None,
             dirty_log_slots: Arc::new(RwLock::new(HashMap::new())),
         }))
     }
     ///
     /// Get the supported CpuID
     ///
-    fn get_cpuid(&self) -> hypervisor::Result<CpuId> {
-        Ok(CpuId::new(1).unwrap())
-    }
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// Retrieve the list of MSRs supported by MSHV.
-    ///
-    fn get_msr_list(&self) -> hypervisor::Result<MsrList> {
-        self.mshv
-            .get_msr_index_list()
-            .map_err(|e| hypervisor::HypervisorError::GetMsrList(e.into()))
+    fn get_cpuid(&self) -> hypervisor::Result<Vec<CpuIdEntry>> {
+        Ok(Vec::new())
     }
 }
 
-#[allow(dead_code)]
 /// Vcpu struct for Microsoft Hypervisor
 pub struct MshvVcpu {
     fd: VcpuFd,
     vp_index: u8,
-    cpuid: CpuId,
-    msrs: MsrEntries,
-    hv_state: Arc<RwLock<HvState>>, // Mshv State
+    cpuid: Vec<CpuIdEntry>,
+    msrs: Vec<MsrEntry>,
     vm_ops: Option<Arc<dyn vm::VmOps>>,
 }
 
@@ -181,17 +300,20 @@ impl cpu::Vcpu for MshvVcpu {
     /// Returns the vCPU general purpose registers.
     ///
     fn get_regs(&self) -> cpu::Result<StandardRegisters> {
-        self.fd
+        Ok(self
+            .fd
             .get_regs()
-            .map_err(|e| cpu::HypervisorCpuError::GetStandardRegs(e.into()))
+            .map_err(|e| cpu::HypervisorCpuError::GetStandardRegs(e.into()))?
+            .into())
     }
     #[cfg(target_arch = "x86_64")]
     ///
     /// Sets the vCPU general purpose registers.
     ///
     fn set_regs(&self, regs: &StandardRegisters) -> cpu::Result<()> {
+        let regs = (*regs).into();
         self.fd
-            .set_regs(regs)
+            .set_regs(&regs)
             .map_err(|e| cpu::HypervisorCpuError::SetStandardRegs(e.into()))
     }
     #[cfg(target_arch = "x86_64")]
@@ -199,17 +321,20 @@ impl cpu::Vcpu for MshvVcpu {
     /// Returns the vCPU special registers.
     ///
     fn get_sregs(&self) -> cpu::Result<SpecialRegisters> {
-        self.fd
+        Ok(self
+            .fd
             .get_sregs()
-            .map_err(|e| cpu::HypervisorCpuError::GetSpecialRegs(e.into()))
+            .map_err(|e| cpu::HypervisorCpuError::GetSpecialRegs(e.into()))?
+            .into())
     }
     #[cfg(target_arch = "x86_64")]
     ///
     /// Sets the vCPU special registers.
     ///
     fn set_sregs(&self, sregs: &SpecialRegisters) -> cpu::Result<()> {
+        let sregs = (*sregs).into();
         self.fd
-            .set_sregs(sregs)
+            .set_sregs(&sregs)
             .map_err(|e| cpu::HypervisorCpuError::SetSpecialRegs(e.into()))
     }
     #[cfg(target_arch = "x86_64")]
@@ -217,17 +342,20 @@ impl cpu::Vcpu for MshvVcpu {
     /// Returns the floating point state (FPU) from the vCPU.
     ///
     fn get_fpu(&self) -> cpu::Result<FpuState> {
-        self.fd
+        Ok(self
+            .fd
             .get_fpu()
-            .map_err(|e| cpu::HypervisorCpuError::GetFloatingPointRegs(e.into()))
+            .map_err(|e| cpu::HypervisorCpuError::GetFloatingPointRegs(e.into()))?
+            .into())
     }
     #[cfg(target_arch = "x86_64")]
     ///
     /// Set the floating point state (FPU) of a vCPU.
     ///
     fn set_fpu(&self, fpu: &FpuState) -> cpu::Result<()> {
+        let fpu: mshv_bindings::FloatingPointUnit = (*fpu).clone().into();
         self.fd
-            .set_fpu(fpu)
+            .set_fpu(&fpu)
             .map_err(|e| cpu::HypervisorCpuError::SetFloatingPointRegs(e.into()))
     }
 
@@ -235,60 +363,36 @@ impl cpu::Vcpu for MshvVcpu {
     ///
     /// Returns the model-specific registers (MSR) for this vCPU.
     ///
-    fn get_msrs(&self, msrs: &mut MsrEntries) -> cpu::Result<usize> {
-        self.fd
-            .get_msrs(msrs)
-            .map_err(|e| cpu::HypervisorCpuError::GetMsrEntries(e.into()))
+    fn get_msrs(&self, msrs: &mut Vec<MsrEntry>) -> cpu::Result<usize> {
+        let mshv_msrs: Vec<msr_entry> = msrs.iter().map(|e| (*e).into()).collect();
+        let mut mshv_msrs = MsrEntries::from_entries(&mshv_msrs).unwrap();
+        let succ = self
+            .fd
+            .get_msrs(&mut mshv_msrs)
+            .map_err(|e| cpu::HypervisorCpuError::GetMsrEntries(e.into()))?;
+
+        msrs[..succ].copy_from_slice(
+            &mshv_msrs.as_slice()[..succ]
+                .iter()
+                .map(|e| (*e).into())
+                .collect::<Vec<MsrEntry>>(),
+        );
+
+        Ok(succ)
     }
     #[cfg(target_arch = "x86_64")]
     ///
     /// Setup the model-specific registers (MSR) for this vCPU.
     /// Returns the number of MSR entries actually written.
     ///
-    fn set_msrs(&self, msrs: &MsrEntries) -> cpu::Result<usize> {
+    fn set_msrs(&self, msrs: &[MsrEntry]) -> cpu::Result<usize> {
+        let mshv_msrs: Vec<msr_entry> = msrs.iter().map(|e| (*e).into()).collect();
+        let mshv_msrs = MsrEntries::from_entries(&mshv_msrs).unwrap();
         self.fd
-            .set_msrs(msrs)
+            .set_msrs(&mshv_msrs)
             .map_err(|e| cpu::HypervisorCpuError::SetMsrEntries(e.into()))
     }
 
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// X86 specific call that returns the vcpu's current "xcrs".
-    ///
-    fn get_xcrs(&self) -> cpu::Result<ExtendedControlRegisters> {
-        self.fd
-            .get_xcrs()
-            .map_err(|e| cpu::HypervisorCpuError::GetXcsr(e.into()))
-    }
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// X86 specific call that sets the vcpu's current "xcrs".
-    ///
-    fn set_xcrs(&self, xcrs: &ExtendedControlRegisters) -> cpu::Result<()> {
-        self.fd
-            .set_xcrs(xcrs)
-            .map_err(|e| cpu::HypervisorCpuError::SetXcsr(e.into()))
-    }
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// Returns currently pending exceptions, interrupts, and NMIs as well as related
-    /// states of the vcpu.
-    ///
-    fn get_vcpu_events(&self) -> cpu::Result<VcpuEvents> {
-        self.fd
-            .get_vcpu_events()
-            .map_err(|e| cpu::HypervisorCpuError::GetVcpuEvents(e.into()))
-    }
-    #[cfg(target_arch = "x86_64")]
-    ///
-    /// Sets pending exceptions, interrupts, and NMIs as well as related states
-    /// of the vcpu.
-    ///
-    fn set_vcpu_events(&self, events: &VcpuEvents) -> cpu::Result<()> {
-        self.fd
-            .set_vcpu_events(events)
-            .map_err(|e| cpu::HypervisorCpuError::SetVcpuEvents(e.into()))
-    }
     #[cfg(target_arch = "x86_64")]
     ///
     /// X86 specific call to enable HyperV SynIC
@@ -463,14 +567,14 @@ impl cpu::Vcpu for MshvVcpu {
     ///
     /// X86 specific call to setup the CPUID registers.
     ///
-    fn set_cpuid2(&self, _cpuid: &CpuId) -> cpu::Result<()> {
+    fn set_cpuid2(&self, _cpuid: &[CpuIdEntry]) -> cpu::Result<()> {
         Ok(())
     }
     #[cfg(target_arch = "x86_64")]
     ///
     /// X86 specific call to retrieve the CPUID registers.
     ///
-    fn get_cpuid2(&self, _num_entries: usize) -> cpu::Result<CpuId> {
+    fn get_cpuid2(&self, _num_entries: usize) -> cpu::Result<Vec<CpuIdEntry>> {
         Ok(self.cpuid.clone())
     }
     #[cfg(target_arch = "x86_64")]
@@ -478,45 +582,43 @@ impl cpu::Vcpu for MshvVcpu {
     /// Returns the state of the LAPIC (Local Advanced Programmable Interrupt Controller).
     ///
     fn get_lapic(&self) -> cpu::Result<LapicState> {
-        self.fd
+        Ok(self
+            .fd
             .get_lapic()
-            .map_err(|e| cpu::HypervisorCpuError::GetlapicState(e.into()))
+            .map_err(|e| cpu::HypervisorCpuError::GetlapicState(e.into()))?
+            .into())
     }
     #[cfg(target_arch = "x86_64")]
     ///
     /// Sets the state of the LAPIC (Local Advanced Programmable Interrupt Controller).
     ///
     fn set_lapic(&self, lapic: &LapicState) -> cpu::Result<()> {
+        let lapic: mshv_bindings::LapicState = (*lapic).clone().into();
         self.fd
-            .set_lapic(lapic)
+            .set_lapic(&lapic)
             .map_err(|e| cpu::HypervisorCpuError::SetLapicState(e.into()))
     }
-    #[cfg(target_arch = "x86_64")]
     ///
-    /// X86 specific call that returns the vcpu's current "xsave struct".
+    /// Returns the vcpu's current "multiprocessing state".
     ///
-    fn get_xsave(&self) -> cpu::Result<Xsave> {
-        self.fd
-            .get_xsave()
-            .map_err(|e| cpu::HypervisorCpuError::GetXsaveState(e.into()))
+    fn get_mp_state(&self) -> cpu::Result<MpState> {
+        Ok(MpState::Mshv)
     }
-    #[cfg(target_arch = "x86_64")]
     ///
-    /// X86 specific call that sets the vcpu's current "xsave struct".
+    /// Sets the vcpu's current "multiprocessing state".
     ///
-    fn set_xsave(&self, xsave: &Xsave) -> cpu::Result<()> {
-        self.fd
-            .set_xsave(xsave)
-            .map_err(|e| cpu::HypervisorCpuError::SetXsaveState(e.into()))
+    fn set_mp_state(&self, _mp_state: MpState) -> cpu::Result<()> {
+        Ok(())
     }
     ///
     /// Set CPU state
     ///
     fn set_state(&self, state: &CpuState) -> cpu::Result<()> {
+        let state: VcpuMshvState = state.clone().into();
         self.set_msrs(&state.msrs)?;
         self.set_vcpu_events(&state.vcpu_events)?;
-        self.set_regs(&state.regs)?;
-        self.set_sregs(&state.sregs)?;
+        self.set_regs(&state.regs.into())?;
+        self.set_sregs(&state.sregs.into())?;
         self.set_fpu(&state.fpu)?;
         self.set_xcrs(&state.xcrs)?;
         self.set_lapic(&state.lapic)?;
@@ -555,18 +657,19 @@ impl cpu::Vcpu for MshvVcpu {
             .get_debug_regs()
             .map_err(|e| cpu::HypervisorCpuError::GetDebugRegs(e.into()))?;
 
-        Ok(CpuState {
+        Ok(VcpuMshvState {
             msrs,
             vcpu_events,
-            regs,
-            sregs,
+            regs: regs.into(),
+            sregs: sregs.into(),
             fpu,
             xcrs,
             lapic,
             dbg,
             xsave,
             misc,
-        })
+        }
+        .into())
     }
     #[cfg(target_arch = "x86_64")]
     ///
@@ -586,42 +689,82 @@ impl cpu::Vcpu for MshvVcpu {
     }
     #[cfg(target_arch = "x86_64")]
     ///
-    /// X86 specific call that returns the vcpu's current "suspend registers".
+    /// Return the list of initial MSR entries for a VCPU
     ///
-    fn get_suspend_regs(&self) -> cpu::Result<SuspendRegisters> {
-        self.fd
-            .get_suspend_regs()
-            .map_err(|e| cpu::HypervisorCpuError::GetSuspendRegs(e.into()))
+    fn boot_msr_entries(&self) -> Vec<MsrEntry> {
+        use crate::arch::x86::{msr_index, MTRR_ENABLE, MTRR_MEM_TYPE_WB};
+
+        [
+            msr!(msr_index::MSR_IA32_SYSENTER_CS),
+            msr!(msr_index::MSR_IA32_SYSENTER_ESP),
+            msr!(msr_index::MSR_IA32_SYSENTER_EIP),
+            msr!(msr_index::MSR_STAR),
+            msr!(msr_index::MSR_CSTAR),
+            msr!(msr_index::MSR_LSTAR),
+            msr!(msr_index::MSR_KERNEL_GS_BASE),
+            msr!(msr_index::MSR_SYSCALL_MASK),
+            msr_data!(msr_index::MSR_MTRRdefType, MTRR_ENABLE | MTRR_MEM_TYPE_WB),
+        ]
+        .to_vec()
     }
 }
 
-/// Device struct for MSHV
-pub struct MshvDevice {
-    fd: DeviceFd,
-}
-
-impl device::Device for MshvDevice {
+impl MshvVcpu {
+    #[cfg(target_arch = "x86_64")]
     ///
-    /// Set device attribute
+    /// X86 specific call that returns the vcpu's current "xsave struct".
     ///
-    fn set_device_attr(&self, attr: &DeviceAttr) -> device::Result<()> {
+    fn get_xsave(&self) -> cpu::Result<Xsave> {
         self.fd
-            .set_device_attr(attr)
-            .map_err(|e| device::HypervisorDeviceError::SetDeviceAttribute(e.into()))
+            .get_xsave()
+            .map_err(|e| cpu::HypervisorCpuError::GetXsaveState(e.into()))
     }
+    #[cfg(target_arch = "x86_64")]
     ///
-    /// Get device attribute
+    /// X86 specific call that sets the vcpu's current "xsave struct".
     ///
-    fn get_device_attr(&self, attr: &mut DeviceAttr) -> device::Result<()> {
+    fn set_xsave(&self, xsave: &Xsave) -> cpu::Result<()> {
         self.fd
-            .get_device_attr(attr)
-            .map_err(|e| device::HypervisorDeviceError::GetDeviceAttribute(e.into()))
+            .set_xsave(xsave)
+            .map_err(|e| cpu::HypervisorCpuError::SetXsaveState(e.into()))
     }
-}
-
-impl AsRawFd for MshvDevice {
-    fn as_raw_fd(&self) -> RawFd {
-        self.fd.as_raw_fd()
+    #[cfg(target_arch = "x86_64")]
+    ///
+    /// X86 specific call that returns the vcpu's current "xcrs".
+    ///
+    fn get_xcrs(&self) -> cpu::Result<ExtendedControlRegisters> {
+        self.fd
+            .get_xcrs()
+            .map_err(|e| cpu::HypervisorCpuError::GetXcsr(e.into()))
+    }
+    #[cfg(target_arch = "x86_64")]
+    ///
+    /// X86 specific call that sets the vcpu's current "xcrs".
+    ///
+    fn set_xcrs(&self, xcrs: &ExtendedControlRegisters) -> cpu::Result<()> {
+        self.fd
+            .set_xcrs(xcrs)
+            .map_err(|e| cpu::HypervisorCpuError::SetXcsr(e.into()))
+    }
+    #[cfg(target_arch = "x86_64")]
+    ///
+    /// Returns currently pending exceptions, interrupts, and NMIs as well as related
+    /// states of the vcpu.
+    ///
+    fn get_vcpu_events(&self) -> cpu::Result<VcpuEvents> {
+        self.fd
+            .get_vcpu_events()
+            .map_err(|e| cpu::HypervisorCpuError::GetVcpuEvents(e.into()))
+    }
+    #[cfg(target_arch = "x86_64")]
+    ///
+    /// Sets pending exceptions, interrupts, and NMIs as well as related states
+    /// of the vcpu.
+    ///
+    fn set_vcpu_events(&self, events: &VcpuEvents) -> cpu::Result<()> {
+        self.fd
+            .set_vcpu_events(events)
+            .map_err(|e| cpu::HypervisorCpuError::SetVcpuEvents(e.into()))
     }
 }
 
@@ -750,19 +893,25 @@ impl<'a> PlatformEmulator for MshvEmulatorContext<'a> {
     }
 }
 
-#[allow(dead_code)]
 /// Wrapper over Mshv VM ioctls.
 pub struct MshvVm {
     fd: Arc<VmFd>,
-    msrs: MsrEntries,
-    // Hypervisor State
-    hv_state: Arc<RwLock<HvState>>,
-    vm_ops: Option<Arc<dyn vm::VmOps>>,
+    msrs: Vec<MsrEntry>,
     dirty_log_slots: Arc<RwLock<HashMap<u64, MshvDirtyLogSlot>>>,
 }
 
-fn hv_state_init() -> Arc<RwLock<HvState>> {
-    Arc::new(RwLock::new(HvState { hypercall_page: 0 }))
+impl MshvVm {
+    ///
+    /// Creates an in-kernel device.
+    ///
+    /// See the documentation for `MSHV_CREATE_DEVICE`.
+    fn create_device(&self, device: &mut CreateDevice) -> vm::Result<VfioDeviceFd> {
+        let device_fd = self
+            .fd
+            .create_device(device)
+            .map_err(|e| vm::HypervisorVmError::CreateDevice(e.into()))?;
+        Ok(VfioDeviceFd::new_from_mshv(device_fd))
+    }
 }
 
 ///
@@ -836,9 +985,8 @@ impl vm::Vm for MshvVm {
         let vcpu = MshvVcpu {
             fd: vcpu_fd,
             vp_index: id,
-            cpuid: CpuId::new(1).unwrap(),
+            cpuid: Vec::new(),
             msrs: self.msrs.clone(),
-            hv_state: self.hv_state.clone(),
             vm_ops,
         };
         Ok(Arc::new(vcpu))
@@ -857,6 +1005,7 @@ impl vm::Vm for MshvVm {
         addr: &IoEventAddress,
         datamatch: Option<DataMatch>,
     ) -> vm::Result<()> {
+        let addr = &mshv_ioctls::IoEventAddress::from(*addr);
         debug!(
             "register_ioevent fd {} addr {:x?} datamatch {:?}",
             fd.as_raw_fd(),
@@ -882,6 +1031,7 @@ impl vm::Vm for MshvVm {
     }
     /// Unregister an event from a certain address it has been previously registered to.
     fn unregister_ioevent(&self, fd: &EventFd, addr: &IoEventAddress) -> vm::Result<()> {
+        let addr = &mshv_ioctls::IoEventAddress::from(*addr);
         debug!("unregister_ioevent fd {} addr {:x?}", fd.as_raw_fd(), addr);
 
         self.fd
@@ -890,7 +1040,8 @@ impl vm::Vm for MshvVm {
     }
 
     /// Creates a guest physical memory region.
-    fn create_user_memory_region(&self, user_memory_region: MemoryRegion) -> vm::Result<()> {
+    fn create_user_memory_region(&self, user_memory_region: UserMemoryRegion) -> vm::Result<()> {
+        let user_memory_region: mshv_user_mem_region = user_memory_region.into();
         // No matter read only or not we keep track the slots.
         // For readonly hypervisor can enable the dirty bits,
         // but a VM exit happens before setting the dirty bits
@@ -909,7 +1060,8 @@ impl vm::Vm for MshvVm {
     }
 
     /// Removes a guest physical memory region.
-    fn remove_user_memory_region(&self, user_memory_region: MemoryRegion) -> vm::Result<()> {
+    fn remove_user_memory_region(&self, user_memory_region: UserMemoryRegion) -> vm::Result<()> {
+        let user_memory_region: mshv_user_mem_region = user_memory_region.into();
         // Remove the corresponding entry from "self.dirty_log_slots" if needed
         self.dirty_log_slots
             .write()
@@ -930,7 +1082,7 @@ impl vm::Vm for MshvVm {
         userspace_addr: u64,
         readonly: bool,
         _log_dirty_pages: bool,
-    ) -> MemoryRegion {
+    ) -> UserMemoryRegion {
         let mut flags = HV_MAP_GPA_READABLE | HV_MAP_GPA_EXECUTABLE;
         if !readonly {
             flags |= HV_MAP_GPA_WRITABLE;
@@ -940,24 +1092,12 @@ 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()
     }
 
-    ///
-    /// Creates an in-kernel device.
-    ///
-    /// See the documentation for `MSHV_CREATE_DEVICE`.
-    fn create_device(&self, device: &mut CreateDevice) -> vm::Result<Arc<dyn device::Device>> {
-        let fd = self
-            .fd
-            .create_device(device)
-            .map_err(|e| vm::HypervisorVmError::CreateDevice(e.into()))?;
-        let device = MshvDevice { fd };
-        Ok(Arc::new(device))
-    }
-
-    fn create_passthrough_device(&self) -> vm::Result<Arc<dyn device::Device>> {
+    fn create_passthrough_device(&self) -> vm::Result<VfioDeviceFd> {
         let mut vfio_dev = mshv_create_device {
             type_: mshv_device_type_MSHV_DEV_TYPE_VFIO,
             fd: 0,
@@ -971,18 +1111,15 @@ impl vm::Vm for MshvVm {
     ///
     /// Constructs a routing entry
     ///
-    fn make_routing_entry(
-        &self,
-        gsi: u32,
-        config: &InterruptSourceConfig,
-    ) -> mshv_msi_routing_entry {
+    fn make_routing_entry(&self, gsi: u32, config: &InterruptSourceConfig) -> IrqRoutingEntry {
         match config {
             InterruptSourceConfig::MsiIrq(cfg) => mshv_msi_routing_entry {
                 gsi,
                 address_lo: cfg.low_addr,
                 address_hi: cfg.high_addr,
                 data: cfg.data,
-            },
+            }
+            .into(),
             _ => {
                 unreachable!()
             }
@@ -994,13 +1131,22 @@ impl vm::Vm for MshvVm {
             vec_with_array_field::<mshv_msi_routing, mshv_msi_routing_entry>(entries.len());
         msi_routing[0].nr = entries.len() as u32;
 
+        let entries: Vec<mshv_msi_routing_entry> = entries
+            .iter()
+            .map(|entry| match entry {
+                IrqRoutingEntry::Mshv(e) => *e,
+                #[allow(unreachable_patterns)]
+                _ => panic!("IrqRoutingEntry type is wrong"),
+            })
+            .collect();
+
         // SAFETY: msi_routing initialized with entries.len() and now it is being turned into
         // entries_slice with entries.len() again. It is guaranteed to be large enough to hold
         // everything from entries.
         unsafe {
             let entries_slice: &mut [mshv_msi_routing_entry] =
                 msi_routing[0].entries.as_mut_slice(entries.len());
-            entries_slice.copy_from_slice(entries);
+            entries_slice.copy_from_slice(&entries);
         }
 
         self.fd
@@ -1008,19 +1154,6 @@ impl vm::Vm for MshvVm {
             .map_err(|e| vm::HypervisorVmError::SetGsiRouting(e.into()))
     }
     ///
-    /// Get the Vm state. Return VM specific data
-    ///
-    fn state(&self) -> vm::Result<VmState> {
-        Ok(*self.hv_state.read().unwrap())
-    }
-    ///
-    /// Set the VM state
-    ///
-    fn set_state(&self, state: VmState) -> vm::Result<()> {
-        self.hv_state.write().unwrap().hypercall_page = state.hypercall_page;
-        Ok(())
-    }
-    ///
     /// Start logging dirty pages
     ///
     fn start_dirty_log(&self) -> vm::Result<()> {
@@ -1058,4 +1191,18 @@ impl vm::Vm for MshvVm {
             )
             .map_err(|e| vm::HypervisorVmError::GetDirtyLog(e.into()))
     }
+    /// Retrieve guest clock.
+    #[cfg(target_arch = "x86_64")]
+    fn get_clock(&self) -> vm::Result<ClockData> {
+        Ok(ClockData::Mshv)
+    }
+    /// Set guest clock.
+    #[cfg(target_arch = "x86_64")]
+    fn set_clock(&self, _data: &ClockData) -> vm::Result<()> {
+        Ok(())
+    }
+    /// Downcast to the underlying MshvVm type
+    fn as_any(&self) -> &dyn Any {
+        self
+    }
 }

hypervisor/src/mshv/x86_64/mod.rs

@@ -7,7 +7,10 @@
 // Copyright 2018-2019 CrowdStrike, Inc.
 //
 //
-use crate::arch::x86::{msr_index, SegmentRegisterOps, MTRR_ENABLE, MTRR_MEM_TYPE_WB};
+use crate::arch::x86::{
+    CpuIdEntry, DescriptorTable, FpuState, LapicState, MsrEntry, SegmentRegister, SpecialRegisters,
+    StandardRegisters,
+};
 use serde::{Deserialize, Serialize};
 use std::fmt;
 
@@ -15,25 +18,24 @@ use std::fmt;
 /// Export generically-named wrappers of mshv_bindings for Unix-based platforms
 ///
 pub use {
-    mshv_bindings::hv_cpuid_entry as CpuIdEntry,
-    mshv_bindings::mshv_user_mem_region as MemoryRegion, mshv_bindings::msr_entry as MsrEntry,
-    mshv_bindings::CpuId, mshv_bindings::DebugRegisters,
-    mshv_bindings::FloatingPointUnit as FpuState, mshv_bindings::LapicState,
+    mshv_bindings::hv_cpuid_entry, mshv_bindings::mshv_user_mem_region as MemoryRegion,
+    mshv_bindings::msr_entry, mshv_bindings::CpuId, mshv_bindings::DebugRegisters,
+    mshv_bindings::FloatingPointUnit, mshv_bindings::LapicState as MshvLapicState,
     mshv_bindings::MiscRegs as MiscRegisters, mshv_bindings::MsrList,
-    mshv_bindings::Msrs as MsrEntries, mshv_bindings::Msrs, mshv_bindings::SegmentRegister,
-    mshv_bindings::SpecialRegisters, mshv_bindings::StandardRegisters,
-    mshv_bindings::SuspendRegisters, mshv_bindings::VcpuEvents, mshv_bindings::XSave as Xsave,
+    mshv_bindings::Msrs as MsrEntries, mshv_bindings::Msrs,
+    mshv_bindings::SegmentRegister as MshvSegmentRegister,
+    mshv_bindings::SpecialRegisters as MshvSpecialRegisters,
+    mshv_bindings::StandardRegisters as MshvStandardRegisters, mshv_bindings::SuspendRegisters,
+    mshv_bindings::TableRegister, mshv_bindings::VcpuEvents, mshv_bindings::XSave as Xsave,
     mshv_bindings::Xcrs as ExtendedControlRegisters,
 };
 
-pub const CPUID_FLAG_VALID_INDEX: u32 = 0;
-
 #[derive(Clone, Serialize, Deserialize)]
 pub struct VcpuMshvState {
-    pub msrs: MsrEntries,
+    pub msrs: Vec<MsrEntry>,
     pub vcpu_events: VcpuEvents,
-    pub regs: StandardRegisters,
-    pub sregs: SpecialRegisters,
+    pub regs: MshvStandardRegisters,
+    pub sregs: MshvSpecialRegisters,
     pub fpu: FpuState,
     pub xcrs: ExtendedControlRegisters,
     pub lapic: LapicState,
@@ -44,10 +46,10 @@ pub struct VcpuMshvState {
 
 impl fmt::Display for VcpuMshvState {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        let expected_num_msrs = self.msrs.as_fam_struct_ref().nmsrs as usize;
+        let expected_num_msrs = self.msrs.len();
         let mut msr_entries = vec![vec![0; 2]; expected_num_msrs];
 
-        for (i, entry) in self.msrs.as_slice().iter().enumerate() {
+        for (i, entry) in self.msrs.iter().enumerate() {
             msr_entries[i][1] = entry.data;
             msr_entries[i][0] = entry.index as u64;
         }
@@ -66,87 +68,252 @@ impl fmt::Display for VcpuMshvState {
     }
 }
 
-pub struct IrqRouting {}
-pub enum VcpuExit {}
-pub struct MpState {}
-
-impl SegmentRegisterOps for SegmentRegister {
-    fn segment_type(&self) -> u8 {
-        self.type_
-    }
-    fn set_segment_type(&mut self, val: u8) {
-        self.type_ = val;
-    }
-
-    fn dpl(&self) -> u8 {
-        self.dpl
-    }
-
-    fn set_dpl(&mut self, val: u8) {
-        self.dpl = val;
-    }
-
-    fn present(&self) -> u8 {
-        self.present
-    }
-
-    fn set_present(&mut self, val: u8) {
-        self.present = val;
-    }
-
-    fn long(&self) -> u8 {
-        self.l
-    }
-
-    fn set_long(&mut self, val: u8) {
-        self.l = val;
-    }
-
-    fn avl(&self) -> u8 {
-        self.avl
-    }
-
-    fn set_avl(&mut self, val: u8) {
-        self.avl = val;
-    }
-
-    fn desc_type(&self) -> u8 {
-        self.s
-    }
-
-    fn set_desc_type(&mut self, val: u8) {
-        self.s = val;
-    }
-
-    fn granularity(&self) -> u8 {
-        self.g
-    }
-
-    fn set_granularity(&mut self, val: u8) {
-        self.g = val;
-    }
-
-    fn db(&self) -> u8 {
-        self.db
-    }
-
-    fn set_db(&mut self, val: u8) {
-        self.db = val;
+impl From<StandardRegisters> for MshvStandardRegisters {
+    fn from(regs: StandardRegisters) -> Self {
+        Self {
+            rax: regs.rax,
+            rbx: regs.rbx,
+            rcx: regs.rcx,
+            rdx: regs.rdx,
+            rsi: regs.rsi,
+            rdi: regs.rdi,
+            rsp: regs.rsp,
+            rbp: regs.rbp,
+            r8: regs.r8,
+            r9: regs.r9,
+            r10: regs.r10,
+            r11: regs.r11,
+            r12: regs.r12,
+            r13: regs.r13,
+            r14: regs.r14,
+            r15: regs.r15,
+            rip: regs.rip,
+            rflags: regs.rflags,
+        }
     }
 }
 
-pub fn boot_msr_entries() -> MsrEntries {
-    MsrEntries::from_entries(&[
-        msr!(msr_index::MSR_IA32_SYSENTER_CS),
-        msr!(msr_index::MSR_IA32_SYSENTER_ESP),
-        msr!(msr_index::MSR_IA32_SYSENTER_EIP),
-        msr!(msr_index::MSR_STAR),
-        msr!(msr_index::MSR_CSTAR),
-        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),
-    ])
-    .unwrap()
+impl From<MshvStandardRegisters> for StandardRegisters {
+    fn from(regs: MshvStandardRegisters) -> Self {
+        Self {
+            rax: regs.rax,
+            rbx: regs.rbx,
+            rcx: regs.rcx,
+            rdx: regs.rdx,
+            rsi: regs.rsi,
+            rdi: regs.rdi,
+            rsp: regs.rsp,
+            rbp: regs.rbp,
+            r8: regs.r8,
+            r9: regs.r9,
+            r10: regs.r10,
+            r11: regs.r11,
+            r12: regs.r12,
+            r13: regs.r13,
+            r14: regs.r14,
+            r15: regs.r15,
+            rip: regs.rip,
+            rflags: regs.rflags,
+        }
+    }
+}
+
+impl From<SegmentRegister> for MshvSegmentRegister {
+    fn from(s: SegmentRegister) -> Self {
+        Self {
+            base: s.base,
+            limit: s.limit,
+            selector: s.selector,
+            type_: s.type_,
+            present: s.present,
+            dpl: s.dpl,
+            db: s.db,
+            s: s.s,
+            l: s.l,
+            g: s.g,
+            avl: s.avl,
+            unusable: s.unusable,
+            ..Default::default()
+        }
+    }
+}
+
+impl From<MshvSegmentRegister> for SegmentRegister {
+    fn from(s: MshvSegmentRegister) -> Self {
+        Self {
+            base: s.base,
+            limit: s.limit,
+            selector: s.selector,
+            type_: s.type_,
+            present: s.present,
+            dpl: s.dpl,
+            db: s.db,
+            s: s.s,
+            l: s.l,
+            g: s.g,
+            avl: s.avl,
+            unusable: s.unusable,
+        }
+    }
+}
+
+impl From<DescriptorTable> for TableRegister {
+    fn from(dt: DescriptorTable) -> Self {
+        Self {
+            base: dt.base,
+            limit: dt.limit,
+        }
+    }
+}
+
+impl From<TableRegister> for DescriptorTable {
+    fn from(dt: TableRegister) -> Self {
+        Self {
+            base: dt.base,
+            limit: dt.limit,
+        }
+    }
+}
+
+impl From<SpecialRegisters> for MshvSpecialRegisters {
+    fn from(s: SpecialRegisters) -> Self {
+        Self {
+            cs: s.cs.into(),
+            ds: s.ds.into(),
+            es: s.es.into(),
+            fs: s.fs.into(),
+            gs: s.gs.into(),
+            ss: s.ss.into(),
+            tr: s.tr.into(),
+            ldt: s.ldt.into(),
+            gdt: s.gdt.into(),
+            idt: s.idt.into(),
+            cr0: s.cr0,
+            cr2: s.cr2,
+            cr3: s.cr3,
+            cr4: s.cr4,
+            cr8: s.cr8,
+            efer: s.efer,
+            apic_base: s.apic_base,
+            interrupt_bitmap: s.interrupt_bitmap,
+        }
+    }
+}
+
+impl From<MshvSpecialRegisters> for SpecialRegisters {
+    fn from(s: MshvSpecialRegisters) -> Self {
+        Self {
+            cs: s.cs.into(),
+            ds: s.ds.into(),
+            es: s.es.into(),
+            fs: s.fs.into(),
+            gs: s.gs.into(),
+            ss: s.ss.into(),
+            tr: s.tr.into(),
+            ldt: s.ldt.into(),
+            gdt: s.gdt.into(),
+            idt: s.idt.into(),
+            cr0: s.cr0,
+            cr2: s.cr2,
+            cr3: s.cr3,
+            cr4: s.cr4,
+            cr8: s.cr8,
+            efer: s.efer,
+            apic_base: s.apic_base,
+            interrupt_bitmap: s.interrupt_bitmap,
+        }
+    }
+}
+
+impl From<CpuIdEntry> for hv_cpuid_entry {
+    fn from(e: CpuIdEntry) -> Self {
+        Self {
+            function: e.function,
+            index: e.index,
+            flags: e.flags,
+            eax: e.eax,
+            ebx: e.ebx,
+            ecx: e.ecx,
+            edx: e.edx,
+            ..Default::default()
+        }
+    }
+}
+
+impl From<hv_cpuid_entry> for CpuIdEntry {
+    fn from(e: hv_cpuid_entry) -> Self {
+        Self {
+            function: e.function,
+            index: e.index,
+            flags: e.flags,
+            eax: e.eax,
+            ebx: e.ebx,
+            ecx: e.ecx,
+            edx: e.edx,
+        }
+    }
+}
+
+impl From<FloatingPointUnit> for FpuState {
+    fn from(s: FloatingPointUnit) -> Self {
+        Self {
+            fpr: s.fpr,
+            fcw: s.fcw,
+            fsw: s.fsw,
+            ftwx: s.ftwx,
+            last_opcode: s.last_opcode,
+            last_ip: s.last_ip,
+            last_dp: s.last_dp,
+            xmm: s.xmm,
+            mxcsr: s.mxcsr,
+        }
+    }
+}
+
+impl From<FpuState> for FloatingPointUnit {
+    fn from(s: FpuState) -> Self {
+        Self {
+            fpr: s.fpr,
+            fcw: s.fcw,
+            fsw: s.fsw,
+            ftwx: s.ftwx,
+            last_opcode: s.last_opcode,
+            last_ip: s.last_ip,
+            last_dp: s.last_dp,
+            xmm: s.xmm,
+            mxcsr: s.mxcsr,
+            ..Default::default()
+        }
+    }
+}
+
+impl From<LapicState> for MshvLapicState {
+    fn from(s: LapicState) -> Self {
+        Self { regs: s.regs }
+    }
+}
+
+impl From<MshvLapicState> for LapicState {
+    fn from(s: MshvLapicState) -> Self {
+        Self { regs: s.regs }
+    }
+}
+
+impl From<msr_entry> for MsrEntry {
+    fn from(e: msr_entry) -> Self {
+        Self {
+            index: e.index,
+            data: e.data,
+        }
+    }
+}
+
+impl From<MsrEntry> for msr_entry {
+    fn from(e: MsrEntry) -> Self {
+        Self {
+            index: e.index,
+            data: e.data,
+            ..Default::default()
+        }
+    }
 }

hypervisor/src/vm.rs

@@ -11,21 +11,15 @@
 #[cfg(target_arch = "aarch64")]
 use crate::aarch64::VcpuInit;
 #[cfg(target_arch = "aarch64")]
-use crate::arch::aarch64::gic::Vgic;
-use crate::cpu::Vcpu;
-use crate::device::Device;
-#[cfg(feature = "kvm")]
-use crate::kvm::KvmVmState as VmState;
-#[cfg(feature = "mshv")]
-use crate::mshv::HvState as VmState;
+use crate::arch::aarch64::gic::{Vgic, VgicConfig};
 #[cfg(feature = "tdx")]
-use crate::x86_64::CpuId;
-#[cfg(all(feature = "kvm", target_arch = "x86_64"))]
+use crate::arch::x86::CpuIdEntry;
+use crate::cpu::Vcpu;
+#[cfg(target_arch = "x86_64")]
 use crate::ClockData;
-use crate::CreateDevice;
-use crate::{IoEventAddress, IrqRoutingEntry, MemoryRegion};
-#[cfg(feature = "kvm")]
-use kvm_ioctls::Cap;
+use crate::UserMemoryRegion;
+use crate::{IoEventAddress, IrqRoutingEntry};
+use std::any::Any;
 #[cfg(target_arch = "x86_64")]
 use std::fs::File;
 use std::sync::Arc;
@@ -263,7 +257,7 @@ pub enum InterruptSourceConfig {
 ///
 /// This crate provides a hypervisor-agnostic interfaces for Vm
 ///
-pub trait Vm: Send + Sync {
+pub trait Vm: Send + Sync + Any {
     #[cfg(target_arch = "x86_64")]
     /// Sets the address of the one-page region in the VM's address space.
     fn set_identity_map_address(&self, address: u64) -> Result<()>;
@@ -279,15 +273,7 @@ pub trait Vm: Send + Sync {
     /// 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(
@@ -311,15 +297,13 @@ pub trait Vm: Send + Sync {
         userspace_addr: u64,
         readonly: bool,
         log_dirty_pages: bool,
-    ) -> MemoryRegion;
+    ) -> UserMemoryRegion;
     /// Creates a guest physical memory slot.
-    fn create_user_memory_region(&self, user_memory_region: MemoryRegion) -> Result<()>;
+    fn create_user_memory_region(&self, user_memory_region: UserMemoryRegion) -> Result<()>;
     /// Removes a guest physical memory slot.
-    fn remove_user_memory_region(&self, user_memory_region: MemoryRegion) -> Result<()>;
-    /// Creates an emulated device in the kernel.
-    fn create_device(&self, device: &mut CreateDevice) -> Result<Arc<dyn Device>>;
+    fn remove_user_memory_region(&self, user_memory_region: UserMemoryRegion) -> Result<()>;
     /// Returns the preferred CPU target type which can be emulated by KVM on underlying host.
-    #[cfg(any(target_arch = "arm", target_arch = "aarch64"))]
+    #[cfg(target_arch = "aarch64")]
     fn get_preferred_target(&self, kvi: &mut VcpuInit) -> Result<()>;
     /// Enable split Irq capability
     #[cfg(target_arch = "x86_64")]
@@ -327,20 +311,13 @@ pub trait Vm: Send + Sync {
     #[cfg(target_arch = "x86_64")]
     fn enable_sgx_attribute(&self, file: File) -> Result<()>;
     /// Retrieve guest clock.
-    #[cfg(all(feature = "kvm", target_arch = "x86_64"))]
+    #[cfg(target_arch = "x86_64")]
     fn get_clock(&self) -> Result<ClockData>;
     /// Set guest clock.
-    #[cfg(all(feature = "kvm", target_arch = "x86_64"))]
+    #[cfg(target_arch = "x86_64")]
     fn set_clock(&self, data: &ClockData) -> Result<()>;
-    #[cfg(feature = "kvm")]
-    /// Checks if a particular `Cap` is available.
-    fn check_extension(&self, c: Cap) -> bool;
     /// Create a device that is used for passthrough
-    fn create_passthrough_device(&self) -> Result<Arc<dyn Device>>;
-    /// Get the Vm state. Return VM specific data
-    fn state(&self) -> Result<VmState>;
-    /// Set the VM state
-    fn set_state(&self, state: VmState) -> Result<()>;
+    fn create_passthrough_device(&self) -> Result<vfio_ioctls::VfioDeviceFd>;
     /// Start logging dirty pages
     fn start_dirty_log(&self) -> Result<()>;
     /// Stop logging dirty pages
@@ -349,19 +326,27 @@ pub trait Vm: Send + Sync {
     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: &CpuId, 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;
 }
 
 pub trait VmOps: Send + Sync {

net_gen/Cargo.toml

@@ -5,4 +5,4 @@ authors = ["The Chromium OS Authors"]
 edition = "2021"
 
 [dependencies]
-vmm-sys-util = "0.9.0"
+vmm-sys-util = "0.11.0"

net_util/Cargo.toml

@@ -6,22 +6,23 @@ edition = "2021"
 
 [dependencies]
 epoll = "4.3.1"
-getrandom = "0.2"
-libc = "0.2.126"
+getrandom = "0.2.8"
+libc = "0.2.138"
 log = "0.4.17"
 net_gen = { path = "../net_gen" }
 rate_limiter = { path = "../rate_limiter" }
-serde = "1.0.138"
-versionize = "0.1.6"
+serde = "1.0.150"
+thiserror = "1.0.37"
+versionize = "0.1.9"
 versionize_derive = "0.1.4"
 virtio-bindings = "0.1.0"
-virtio-queue = "0.4.0"
-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 = "0.9.0"
+vmm-sys-util = "0.11.0"
 
 [dev-dependencies]
-once_cell = "1.13.0"
+once_cell = "1.16.0"
 pnet = "0.31.0"
 pnet_datalink = "0.31.0"
-serde_json = "1.0.82"
+serde_json = "1.0.89"

net_util/src/ctrl_queue.rs

@@ -13,8 +13,8 @@ use virtio_bindings::bindings::virtio_net::{
     VIRTIO_NET_F_GUEST_ECN, VIRTIO_NET_F_GUEST_TSO4, VIRTIO_NET_F_GUEST_TSO6,
     VIRTIO_NET_F_GUEST_UFO, VIRTIO_NET_OK,
 };
-use virtio_queue::Queue;
-use vm_memory::{ByteValued, Bytes, GuestMemoryAtomic, GuestMemoryError};
+use virtio_queue::{Queue, QueueT};
+use vm_memory::{ByteValued, Bytes, GuestMemoryError};
 use vm_virtio::{AccessPlatform, Translatable};
 
 #[derive(Debug)]
@@ -58,98 +58,93 @@ impl CtrlQueue {
 
     pub fn process(
         &mut self,
-        queue: &mut Queue<GuestMemoryAtomic<GuestMemoryMmap>>,
+        mem: &GuestMemoryMmap,
+        queue: &mut Queue,
         access_platform: Option<&Arc<dyn AccessPlatform>>,
     ) -> Result<()> {
-        let mut used_desc_heads = Vec::new();
-        loop {
-            for mut desc_chain in queue.iter().map_err(Error::QueueIterator)? {
-                let ctrl_desc = desc_chain.next().ok_or(Error::NoControlHeaderDescriptor)?;
+        while let Some(mut desc_chain) = queue.pop_descriptor_chain(mem) {
+            let ctrl_desc = desc_chain.next().ok_or(Error::NoControlHeaderDescriptor)?;
 
-                let ctrl_hdr: ControlHeader = desc_chain
-                    .memory()
-                    .read_obj(
-                        ctrl_desc
-                            .addr()
-                            .translate_gva(access_platform, ctrl_desc.len() as usize),
-                    )
-                    .map_err(Error::GuestMemory)?;
-                let data_desc = desc_chain.next().ok_or(Error::NoDataDescriptor)?;
+            let ctrl_hdr: ControlHeader = desc_chain
+                .memory()
+                .read_obj(
+                    ctrl_desc
+                        .addr()
+                        .translate_gva(access_platform, ctrl_desc.len() as usize),
+                )
+                .map_err(Error::GuestMemory)?;
+            let data_desc = desc_chain.next().ok_or(Error::NoDataDescriptor)?;
 
-                let data_desc_addr = data_desc
-                    .addr()
-                    .translate_gva(access_platform, data_desc.len() as usize);
+            let data_desc_addr = data_desc
+                .addr()
+                .translate_gva(access_platform, data_desc.len() as usize);
 
-                let status_desc = desc_chain.next().ok_or(Error::NoStatusDescriptor)?;
+            let status_desc = desc_chain.next().ok_or(Error::NoStatusDescriptor)?;
 
-                let ok = match u32::from(ctrl_hdr.class) {
-                    VIRTIO_NET_CTRL_MQ => {
-                        let queue_pairs = desc_chain
-                            .memory()
-                            .read_obj::<u16>(data_desc_addr)
-                            .map_err(Error::GuestMemory)?;
-                        if u32::from(ctrl_hdr.cmd) != VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET {
-                            warn!("Unsupported command: {}", ctrl_hdr.cmd);
-                            false
-                        } else if (queue_pairs < VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MIN as u16)
-                            || (queue_pairs > VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX as u16)
-                        {
-                            warn!("Number of MQ pairs out of range: {}", queue_pairs);
-                            false
-                        } else {
-                            info!("Number of MQ pairs requested: {}", queue_pairs);
-                            true
-                        }
-                    }
-                    VIRTIO_NET_CTRL_GUEST_OFFLOADS => {
-                        let features = desc_chain
-                            .memory()
-                            .read_obj::<u64>(data_desc_addr)
-                            .map_err(Error::GuestMemory)?;
-                        if u32::from(ctrl_hdr.cmd) != VIRTIO_NET_CTRL_GUEST_OFFLOADS_SET {
-                            warn!("Unsupported command: {}", ctrl_hdr.cmd);
-                            false
-                        } else {
-                            let mut ok = true;
-                            for tap in self.taps.iter_mut() {
-                                info!("Reprogramming tap offload with features: {}", features);
-                                tap.set_offload(virtio_features_to_tap_offload(features))
-                                    .map_err(|e| {
-                                        error!("Error programming tap offload: {:?}", e);
-                                        ok = false
-                                    })
-                                    .ok();
-                            }
-                            ok
-                        }
-                    }
-                    _ => {
-                        warn!("Unsupported command {:?}", ctrl_hdr);
+            let ok = match u32::from(ctrl_hdr.class) {
+                VIRTIO_NET_CTRL_MQ => {
+                    let queue_pairs = desc_chain
+                        .memory()
+                        .read_obj::<u16>(data_desc_addr)
+                        .map_err(Error::GuestMemory)?;
+                    if u32::from(ctrl_hdr.cmd) != VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET {
+                        warn!("Unsupported command: {}", ctrl_hdr.cmd);
                         false
+                    } else if (queue_pairs < VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MIN as u16)
+                        || (queue_pairs > VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX as u16)
+                    {
+                        warn!("Number of MQ pairs out of range: {}", queue_pairs);
+                        false
+                    } else {
+                        info!("Number of MQ pairs requested: {}", queue_pairs);
+                        true
                     }
-                };
+                }
+                VIRTIO_NET_CTRL_GUEST_OFFLOADS => {
+                    let features = desc_chain
+                        .memory()
+                        .read_obj::<u64>(data_desc_addr)
+                        .map_err(Error::GuestMemory)?;
+                    if u32::from(ctrl_hdr.cmd) != VIRTIO_NET_CTRL_GUEST_OFFLOADS_SET {
+                        warn!("Unsupported command: {}", ctrl_hdr.cmd);
+                        false
+                    } else {
+                        let mut ok = true;
+                        for tap in self.taps.iter_mut() {
+                            info!("Reprogramming tap offload with features: {}", features);
+                            tap.set_offload(virtio_features_to_tap_offload(features))
+                                .map_err(|e| {
+                                    error!("Error programming tap offload: {:?}", e);
+                                    ok = false
+                                })
+                                .ok();
+                        }
+                        ok
+                    }
+                }
+                _ => {
+                    warn!("Unsupported command {:?}", ctrl_hdr);
+                    false
+                }
+            };
 
-                desc_chain
-                    .memory()
-                    .write_obj(
-                        if ok { VIRTIO_NET_OK } else { VIRTIO_NET_ERR } as u8,
-                        status_desc
-                            .addr()
-                            .translate_gva(access_platform, status_desc.len() as usize),
-                    )
-                    .map_err(Error::GuestMemory)?;
-                let len = ctrl_desc.len() + data_desc.len() + status_desc.len();
-                used_desc_heads.push((desc_chain.head_index(), len));
-            }
+            desc_chain
+                .memory()
+                .write_obj(
+                    if ok { VIRTIO_NET_OK } else { VIRTIO_NET_ERR } as u8,
+                    status_desc
+                        .addr()
+                        .translate_gva(access_platform, status_desc.len() as usize),
+                )
+                .map_err(Error::GuestMemory)?;
+            let len = ctrl_desc.len() + data_desc.len() + status_desc.len();
 
-            for (desc_index, len) in used_desc_heads.iter() {
-                queue
-                    .add_used(*desc_index, *len)
-                    .map_err(Error::QueueAddUsed)?;
-            }
+            queue
+                .add_used(desc_chain.memory(), desc_chain.head_index(), len)
+                .map_err(Error::QueueAddUsed)?;
 
             if !queue
-                .enable_notification()
+                .enable_notification(mem)
                 .map_err(Error::QueueEnableNotification)?
             {
                 break;

net_util/src/lib.rs

@@ -18,6 +18,7 @@ use std::io::Error as IoError;
 use std::os::raw::c_uint;
 use std::os::unix::io::{FromRawFd, RawFd};
 use std::{io, mem, net};
+use thiserror::Error;
 use versionize::{VersionMap, Versionize, VersionizeResult};
 use versionize_derive::Versionize;
 use virtio_bindings::bindings::virtio_net::{
@@ -35,9 +36,9 @@ pub use open_tap::{open_tap, Error as OpenTapError};
 pub use queue_pair::{NetCounters, NetQueuePair, NetQueuePairError, RxVirtio, TxVirtio};
 pub use tap::{Error as TapError, Tap};
 
-#[derive(Debug)]
+#[derive(Error, Debug)]
 pub enum Error {
-    /// Failed to create a socket.
+    #[error("Failed to create a socket: {0}")]
     CreateSocket(IoError),
 }
 
@@ -130,17 +131,19 @@ 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;
@@ -150,6 +153,9 @@ 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

@@ -6,30 +6,33 @@ use super::{vnet_hdr_len, MacAddr, Tap, TapError};
 use std::net::Ipv4Addr;
 use std::path::Path;
 use std::{fs, io};
+use thiserror::Error;
 
-#[derive(Debug)]
+#[derive(Error, Debug)]
 pub enum Error {
-    /// Failed to convert an hexadecimal string into an integer.
+    #[error("Failed to convert an hexadecimal string into an integer: {0}")]
     ConvertHexStringToInt(std::num::ParseIntError),
-    /// Error related to the multiqueue support (no support TAP side).
+    #[error("Error related to the multiqueue support (no support TAP side)")]
     MultiQueueNoTapSupport,
-    /// Error related to the multiqueue support (no support device side).
+    #[error("Error related to the multiqueue support (no support device side)")]
     MultiQueueNoDeviceSupport,
-    /// Failed to read the TAP flags from sysfs.
+    #[error("Failed to read the TAP flags from sysfs: {0}")]
     ReadSysfsTunFlags(io::Error),
-    /// Open tap device failed.
+    #[error("Open tap device failed: {0}")]
     TapOpen(TapError),
-    /// Setting tap IP failed.
+    #[error("Setting tap IP failed: {0}")]
     TapSetIp(TapError),
-    /// Setting tap netmask failed.
+    #[error("Setting tap netmask failed: {0}")]
     TapSetNetmask(TapError),
-    /// Setting MAC address failed
+    #[error("Setting MAC address failed: {0}")]
     TapSetMac(TapError),
-    /// Getting MAC address failed
+    #[error("Getting MAC address failed: {0}")]
     TapGetMac(TapError),
-    /// Setting vnet header size failed.
+    #[error("Setting vnet header size failed: {0}")]
     TapSetVnetHdrSize(TapError),
-    /// Enabling tap interface failed.
+    #[error("Setting MTU failed: {0}")]
+    TapSetMtu(TapError),
+    #[error("Enabling tap interface failed: {0}")]
     TapEnable(TapError),
 }
 
@@ -38,7 +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(());
@@ -62,6 +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>> {
@@ -94,6 +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

@@ -10,8 +10,9 @@ use std::num::Wrapping;
 use std::os::unix::io::{AsRawFd, RawFd};
 use std::sync::atomic::{AtomicU64, Ordering};
 use std::sync::Arc;
-use virtio_queue::Queue;
-use vm_memory::{Bytes, GuestMemory, GuestMemoryAtomic};
+use thiserror::Error;
+use virtio_queue::{Queue, QueueOwnedT, QueueT};
+use vm_memory::{Bytes, GuestMemory};
 use vm_virtio::{AccessPlatform, Translatable};
 
 #[derive(Clone)]
@@ -36,104 +37,99 @@ impl TxVirtio {
 
     pub fn process_desc_chain(
         &mut self,
+        mem: &GuestMemoryMmap,
         tap: &mut Tap,
-        queue: &mut Queue<GuestMemoryAtomic<GuestMemoryMmap>>,
+        queue: &mut Queue,
         rate_limiter: &mut Option<RateLimiter>,
         access_platform: Option<&Arc<dyn AccessPlatform>>,
     ) -> Result<bool, NetQueuePairError> {
         let mut retry_write = false;
         let mut rate_limit_reached = false;
 
-        loop {
-            let used_desc_head: (u16, u32);
-            let mut avail_iter = queue
-                .iter()
-                .map_err(NetQueuePairError::QueueIteratorFailed)?;
-
-            if let Some(mut desc_chain) = avail_iter.next() {
-                if rate_limit_reached {
-                    avail_iter.go_to_previous_position();
-                    break;
-                }
-
-                let mut next_desc = desc_chain.next();
-
-                let mut iovecs = Vec::new();
-                while let Some(desc) = next_desc {
-                    let desc_addr = desc
-                        .addr()
-                        .translate_gva(access_platform, desc.len() as usize);
-                    if !desc.is_write_only() && desc.len() > 0 {
-                        let buf = desc_chain
-                            .memory()
-                            .get_slice(desc_addr, desc.len() as usize)
-                            .map_err(NetQueuePairError::GuestMemory)?
-                            .as_ptr();
-                        let iovec = libc::iovec {
-                            iov_base: buf as *mut libc::c_void,
-                            iov_len: desc.len() as libc::size_t,
-                        };
-                        iovecs.push(iovec);
-                    } else {
-                        error!(
-                            "Invalid descriptor chain: address = 0x{:x} length = {} write_only = {}",
-                            desc_addr.0,
-                            desc.len(),
-                            desc.is_write_only()
-                        );
-                        return Err(NetQueuePairError::DescriptorChainInvalid);
-                    }
-                    next_desc = desc_chain.next();
-                }
-
-                let len = if !iovecs.is_empty() {
-                    let result = unsafe {
-                        libc::writev(
-                            tap.as_raw_fd() as libc::c_int,
-                            iovecs.as_ptr() as *const libc::iovec,
-                            iovecs.len() as libc::c_int,
-                        )
-                    };
-
-                    if result < 0 {
-                        let e = std::io::Error::last_os_error();
-
-                        /* EAGAIN */
-                        if e.kind() == std::io::ErrorKind::WouldBlock {
-                            avail_iter.go_to_previous_position();
-                            retry_write = true;
-                            break;
-                        }
-                        error!("net: tx: failed writing to tap: {}", e);
-                        return Err(NetQueuePairError::WriteTap(e));
-                    }
-
-                    self.counter_bytes += Wrapping(result as u64 - vnet_hdr_len() as u64);
-                    self.counter_frames += Wrapping(1);
-
-                    result as u32
-                } else {
-                    0
-                };
-
-                used_desc_head = (desc_chain.head_index(), len);
-
-                // For the sake of simplicity (similar to the RX rate limiting), we always
-                // let the 'last' descriptor chain go-through even if it was over the rate
-                // limit, and simply stop processing oncoming `avail_desc` if any.
-                if let Some(rate_limiter) = rate_limiter {
-                    rate_limit_reached = !rate_limiter.consume(1, TokenType::Ops)
-                        || !rate_limiter.consume(len as u64, TokenType::Bytes);
-                }
-            } else {
+        while let Some(mut desc_chain) = queue.pop_descriptor_chain(mem) {
+            if rate_limit_reached {
+                queue.go_to_previous_position();
                 break;
             }
 
+            let mut next_desc = desc_chain.next();
+
+            let mut iovecs = Vec::new();
+            while let Some(desc) = next_desc {
+                let desc_addr = desc
+                    .addr()
+                    .translate_gva(access_platform, desc.len() as usize);
+                if !desc.is_write_only() && desc.len() > 0 {
+                    let buf = desc_chain
+                        .memory()
+                        .get_slice(desc_addr, desc.len() as usize)
+                        .map_err(NetQueuePairError::GuestMemory)?
+                        .as_ptr();
+                    let iovec = libc::iovec {
+                        iov_base: buf as *mut libc::c_void,
+                        iov_len: desc.len() as libc::size_t,
+                    };
+                    iovecs.push(iovec);
+                } else {
+                    error!(
+                        "Invalid descriptor chain: address = 0x{:x} length = {} write_only = {}",
+                        desc_addr.0,
+                        desc.len(),
+                        desc.is_write_only()
+                    );
+                    return Err(NetQueuePairError::DescriptorChainInvalid);
+                }
+                next_desc = desc_chain.next();
+            }
+
+            let len = if !iovecs.is_empty() {
+                let result = unsafe {
+                    libc::writev(
+                        tap.as_raw_fd() as libc::c_int,
+                        iovecs.as_ptr() as *const libc::iovec,
+                        iovecs.len() as libc::c_int,
+                    )
+                };
+
+                if result < 0 {
+                    let e = std::io::Error::last_os_error();
+
+                    /* EAGAIN */
+                    if e.kind() == std::io::ErrorKind::WouldBlock {
+                        queue.go_to_previous_position();
+                        retry_write = true;
+                        break;
+                    }
+                    error!("net: tx: failed writing to tap: {}", e);
+                    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);
+
+                result as u32
+            } else {
+                0
+            };
+
+            // For the sake of simplicity (similar to the RX rate limiting), we always
+            // let the 'last' descriptor chain go-through even if it was over the rate
+            // limit, and simply stop processing oncoming `avail_desc` if any.
+            if let Some(rate_limiter) = rate_limiter {
+                rate_limit_reached = !rate_limiter.consume(1, TokenType::Ops)
+                    || !rate_limiter.consume(len as u64, TokenType::Bytes);
+            }
+
             queue
-                .add_used(used_desc_head.0, used_desc_head.1)
+                .add_used(desc_chain.memory(), desc_chain.head_index(), len)
                 .map_err(NetQueuePairError::QueueAddUsed)?;
+
             if !queue
-                .enable_notification()
+                .enable_notification(mem)
                 .map_err(NetQueuePairError::QueueEnableNotification)?
             {
                 break;
@@ -166,125 +162,120 @@ impl RxVirtio {
 
     pub fn process_desc_chain(
         &mut self,
+        mem: &GuestMemoryMmap,
         tap: &mut Tap,
-        queue: &mut Queue<GuestMemoryAtomic<GuestMemoryMmap>>,
+        queue: &mut Queue,
         rate_limiter: &mut Option<RateLimiter>,
         access_platform: Option<&Arc<dyn AccessPlatform>>,
     ) -> Result<bool, NetQueuePairError> {
         let mut exhausted_descs = true;
         let mut rate_limit_reached = false;
 
-        loop {
-            let used_desc_head: (u16, u32);
-            let mut avail_iter = queue
-                .iter()
-                .map_err(NetQueuePairError::QueueIteratorFailed)?;
-
-            if let Some(mut desc_chain) = avail_iter.next() {
-                if rate_limit_reached {
-                    exhausted_descs = false;
-                    avail_iter.go_to_previous_position();
-                    break;
-                }
-
-                let desc = desc_chain
-                    .next()
-                    .ok_or(NetQueuePairError::DescriptorChainTooShort)?;
-
-                let num_buffers_addr = desc_chain
-                    .memory()
-                    .checked_offset(
-                        desc.addr()
-                            .translate_gva(access_platform, desc.len() as usize),
-                        10,
-                    )
-                    .unwrap();
-                let mut next_desc = Some(desc);
-
-                let mut iovecs = Vec::new();
-                while let Some(desc) = next_desc {
-                    let desc_addr = desc
-                        .addr()
-                        .translate_gva(access_platform, desc.len() as usize);
-                    if desc.is_write_only() && desc.len() > 0 {
-                        let buf = desc_chain
-                            .memory()
-                            .get_slice(desc_addr, desc.len() as usize)
-                            .map_err(NetQueuePairError::GuestMemory)?
-                            .as_ptr();
-                        let iovec = libc::iovec {
-                            iov_base: buf as *mut libc::c_void,
-                            iov_len: desc.len() as libc::size_t,
-                        };
-                        iovecs.push(iovec);
-                    } else {
-                        error!(
-                            "Invalid descriptor chain: address = 0x{:x} length = {} write_only = {}",
-                            desc_addr.0,
-                            desc.len(),
-                            desc.is_write_only()
-                        );
-                        return Err(NetQueuePairError::DescriptorChainInvalid);
-                    }
-                    next_desc = desc_chain.next();
-                }
-
-                let len = if !iovecs.is_empty() {
-                    let result = unsafe {
-                        libc::readv(
-                            tap.as_raw_fd() as libc::c_int,
-                            iovecs.as_ptr() as *const libc::iovec,
-                            iovecs.len() as libc::c_int,
-                        )
-                    };
-                    if result < 0 {
-                        let e = std::io::Error::last_os_error();
-                        exhausted_descs = false;
-                        avail_iter.go_to_previous_position();
-
-                        /* EAGAIN */
-                        if e.kind() == std::io::ErrorKind::WouldBlock {
-                            break;
-                        }
-
-                        error!("net: rx: failed reading from tap: {}", e);
-                        return Err(NetQueuePairError::ReadTap(e));
-                    }
-
-                    // Write num_buffers to guest memory. We simply write 1 as we
-                    // never spread the frame over more than one descriptor chain.
-                    desc_chain
-                        .memory()
-                        .write_obj(1u16, num_buffers_addr)
-                        .map_err(NetQueuePairError::GuestMemory)?;
-
-                    self.counter_bytes += Wrapping(result as u64 - vnet_hdr_len() as u64);
-                    self.counter_frames += Wrapping(1);
-
-                    result as u32
-                } else {
-                    0
-                };
-
-                used_desc_head = (desc_chain.head_index(), len);
-
-                // For the sake of simplicity (keeping the handling of RX_QUEUE_EVENT and
-                // RX_TAP_EVENT totally asynchronous), we always let the 'last' descriptor
-                // chain go-through even if it was over the rate limit, and simply stop
-                // processing oncoming `avail_desc` if any.
-                if let Some(rate_limiter) = rate_limiter {
-                    rate_limit_reached = !rate_limiter.consume(1, TokenType::Ops)
-                        || !rate_limiter.consume(len as u64, TokenType::Bytes);
-                }
-            } else {
+        while let Some(mut desc_chain) = queue.pop_descriptor_chain(mem) {
+            if rate_limit_reached {
+                exhausted_descs = false;
+                queue.go_to_previous_position();
                 break;
             }
 
+            let desc = desc_chain
+                .next()
+                .ok_or(NetQueuePairError::DescriptorChainTooShort)?;
+
+            let num_buffers_addr = desc_chain
+                .memory()
+                .checked_offset(
+                    desc.addr()
+                        .translate_gva(access_platform, desc.len() as usize),
+                    10,
+                )
+                .ok_or(NetQueuePairError::DescriptorInvalidHeader)?;
+            let mut next_desc = Some(desc);
+
+            let mut iovecs = Vec::new();
+            while let Some(desc) = next_desc {
+                let desc_addr = desc
+                    .addr()
+                    .translate_gva(access_platform, desc.len() as usize);
+                if desc.is_write_only() && desc.len() > 0 {
+                    let buf = desc_chain
+                        .memory()
+                        .get_slice(desc_addr, desc.len() as usize)
+                        .map_err(NetQueuePairError::GuestMemory)?
+                        .as_ptr();
+                    let iovec = libc::iovec {
+                        iov_base: buf as *mut libc::c_void,
+                        iov_len: desc.len() as libc::size_t,
+                    };
+                    iovecs.push(iovec);
+                } else {
+                    error!(
+                        "Invalid descriptor chain: address = 0x{:x} length = {} write_only = {}",
+                        desc_addr.0,
+                        desc.len(),
+                        desc.is_write_only()
+                    );
+                    return Err(NetQueuePairError::DescriptorChainInvalid);
+                }
+                next_desc = desc_chain.next();
+            }
+
+            let len = if !iovecs.is_empty() {
+                let result = unsafe {
+                    libc::readv(
+                        tap.as_raw_fd() as libc::c_int,
+                        iovecs.as_ptr() as *const libc::iovec,
+                        iovecs.len() as libc::c_int,
+                    )
+                };
+                if result < 0 {
+                    let e = std::io::Error::last_os_error();
+                    exhausted_descs = false;
+                    queue.go_to_previous_position();
+
+                    /* EAGAIN */
+                    if e.kind() == std::io::ErrorKind::WouldBlock {
+                        break;
+                    }
+
+                    error!("net: rx: failed reading from tap: {}", e);
+                    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
+                    .memory()
+                    .write_obj(1u16, num_buffers_addr)
+                    .map_err(NetQueuePairError::GuestMemory)?;
+
+                self.counter_bytes += Wrapping(result as u64 - vnet_hdr_len() as u64);
+                self.counter_frames += Wrapping(1);
+
+                result as u32
+            } else {
+                0
+            };
+
+            // For the sake of simplicity (keeping the handling of RX_QUEUE_EVENT and
+            // RX_TAP_EVENT totally asynchronous), we always let the 'last' descriptor
+            // chain go-through even if it was over the rate limit, and simply stop
+            // processing oncoming `avail_desc` if any.
+            if let Some(rate_limiter) = rate_limiter {
+                rate_limit_reached = !rate_limiter.consume(1, TokenType::Ops)
+                    || !rate_limiter.consume(len as u64, TokenType::Bytes);
+            }
+
             queue
-                .add_used(used_desc_head.0, used_desc_head.1)
+                .add_used(desc_chain.memory(), desc_chain.head_index(), len)
                 .map_err(NetQueuePairError::QueueAddUsed)?;
+
             if !queue
-                .enable_notification()
+                .enable_notification(mem)
                 .map_err(NetQueuePairError::QueueEnableNotification)?
             {
                 break;
@@ -303,32 +294,36 @@ pub struct NetCounters {
     pub rx_frames: Arc<AtomicU64>,
 }
 
-#[derive(Debug)]
+#[derive(Error, Debug)]
 pub enum NetQueuePairError {
-    /// No memory configured
+    #[error("No memory configured")]
     NoMemoryConfigured,
-    /// Error registering listener
+    #[error("Error registering listener: {0}")]
     RegisterListener(io::Error),
-    /// Error unregistering listener
+    #[error("Error unregistering listener: {0}")]
     UnregisterListener(io::Error),
-    /// Error writing to the TAP device
+    #[error("Error writing to the TAP device: {0}")]
     WriteTap(io::Error),
-    /// Error reading from the TAP device
+    #[error("Error reading from the TAP device: {0}")]
     ReadTap(io::Error),
-    /// Error related to guest memory
+    #[error("Error related to guest memory: {0}")]
     GuestMemory(vm_memory::GuestMemoryError),
-    /// Returned an error while iterating through the queue
+    #[error("Returned an error while iterating through the queue: {0}")]
     QueueIteratorFailed(virtio_queue::Error),
-    /// Descriptor chain is too short
+    #[error("Descriptor chain is too short")]
     DescriptorChainTooShort,
-    /// Descriptor chain does not contain valid descriptors
+    #[error("Descriptor chain does not contain valid descriptors")]
     DescriptorChainInvalid,
-    /// Failed to determine if queue needed notification
+    #[error("Failed to determine if queue needed notification: {0}")]
     QueueNeedsNotification(virtio_queue::Error),
-    /// Failed to enable notification on the queue
+    #[error("Failed to enable notification on the queue: {0}")]
     QueueEnableNotification(virtio_queue::Error),
-    /// Failed to add used index to the queue
+    #[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 {
@@ -355,9 +350,11 @@ pub struct NetQueuePair {
 impl NetQueuePair {
     pub fn process_tx(
         &mut self,
-        queue: &mut Queue<GuestMemoryAtomic<GuestMemoryMmap>>,
+        mem: &GuestMemoryMmap,
+        queue: &mut Queue,
     ) -> Result<bool, NetQueuePairError> {
         let tx_tap_retry = self.tx.process_desc_chain(
+            mem,
             &mut self.tap,
             queue,
             &mut self.tx_rate_limiter,
@@ -397,15 +394,17 @@ impl NetQueuePair {
         self.tx.counter_frames = Wrapping(0);
 
         queue
-            .needs_notification()
+            .needs_notification(mem)
             .map_err(NetQueuePairError::QueueNeedsNotification)
     }
 
     pub fn process_rx(
         &mut self,
-        queue: &mut Queue<GuestMemoryAtomic<GuestMemoryMmap>>,
+        mem: &GuestMemoryMmap,
+        queue: &mut Queue,
     ) -> Result<bool, NetQueuePairError> {
         self.rx_desc_avail = !self.rx.process_desc_chain(
+            mem,
             &mut self.tap,
             queue,
             &mut self.rx_rate_limiter,
@@ -440,7 +439,7 @@ impl NetQueuePair {
         self.rx.counter_frames = Wrapping(0);
 
         queue
-            .needs_notification()
+            .needs_notification(mem)
             .map_err(NetQueuePairError::QueueNeedsNotification)
     }
 }