build: Release v43.0

Signed-off-by: Rob Bradford <rbradford@rivosinc.com>

.github/workflows/integration-arm64.yaml

@@ -6,7 +6,7 @@ concurrency:
 
 jobs:
   build:
-    timeout-minutes: 60
+    timeout-minutes: 120
     name: Tests (ARM64)
     runs-on: bookworm-arm64
     steps:
@@ -21,7 +21,7 @@ jobs:
       - name: Load openvswitch module
         run: sudo modprobe openvswitch
       - name: Run integration tests (musl)
-        timeout-minutes: 30
+        timeout-minutes: 60
         run: scripts/dev_cli.sh tests --integration --libc musl
       - name: Install Azure CLI
         if: ${{ github.event_name != 'pull_request' }}

.github/workflows/integration-rate-limiter.yaml

@@ -28,7 +28,7 @@ jobs:
           sudo apt install -y docker-ce docker-ce-cli
       - name: Run rate-limiter integration tests
         if: ${{ github.event_name != 'pull_request' }}
-        timeout-minutes: 10
+        timeout-minutes: 20
         run: scripts/dev_cli.sh tests --integration-rate-limiter
       - name: Skipping build for PR
         if: ${{ github.event_name == 'pull_request' }}

.github/workflows/package-consistency.yaml

@@ -0,0 +1,32 @@
+name: Cloud Hypervisor Consistency
+on: [pull_request, merge_group]
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  build:
+    name: Rust VMM Consistency Check
+    runs-on: ubuntu-latest
+    steps:
+      - name: Code checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Install dependencies
+        run: sudo apt install -y python3
+
+      - name: Install Rust toolchain stable
+        uses: dtolnay/rust-toolchain@stable
+        with:
+          toolchain: stable
+
+      - name: Check Rust VMM Package Consistency of root Workspace
+        run: python3 scripts/package-consistency-check.py github.com/rust-vmm
+
+      - name: Check Rust VMM Package Consistency of fuzz Workspace
+        run: |
+          pushd fuzz
+          python3 ../scripts/package-consistency-check.py github.com/rust-vmm
+          popd

.github/workflows/preview-riscv64.yaml

@@ -0,0 +1,39 @@
+name: Cloud Hypervisor RISC-V 64-bit Preview
+on: [pull_request, merge_group]
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  build:
+    name: Cargo
+    runs-on: riscv64-qemu-host
+    strategy:
+      fail-fast: false
+      matrix:
+        module:
+          - hypervisor
+          - arch
+          - vm-allocator
+          - devices
+
+    steps:
+      - name: Code checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Install Rust toolchain
+        run: /opt/scripts/exec-in-qemu.sh rustup default 1.77.0
+
+      - name: Build ${{ matrix.module }} Module (kvm)
+        run: /opt/scripts/exec-in-qemu.sh cargo rustc --locked -p ${{ matrix.module }} --no-default-features --features "kvm" -- -D warnings -D clippy::undocumented_unsafe_blocks -W clippy::assertions_on_result_states
+
+      - name: Clippy ${{ matrix.module }} Module (kvm)
+        run: /opt/scripts/exec-in-qemu.sh cargo clippy --locked -p ${{ matrix.module }} --no-default-features --features "kvm" -- -D warnings -D clippy::undocumented_unsafe_blocks -W clippy::assertions_on_result_states
+
+      - name: Test ${{ matrix.module }} Module (kvm)
+        run: /opt/scripts/exec-in-qemu.sh cargo test --locked -p ${{ matrix.module }} --no-default-features --features "kvm"
+
+      - name: Check no files were modified
+        run: test -z "$(git status --porcelain)"

Cargo.lock

@@ -7,23 +7,23 @@ name = "acpi_tables"
 version = "0.1.0"
 source = "git+https://github.com/rust-vmm/acpi_tables?branch=main#e268627630839bd22f1c13e7e81ec70c7e9b73d6"
 dependencies = [
- "zerocopy",
+ "zerocopy 0.7.35",
 ]
 
 [[package]]
 name = "addr2line"
-version = "0.22.0"
+version = "0.24.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6e4503c46a5c0c7844e948c9a4d6acd9f50cccb4de1c48eb9e291ea17470c678"
+checksum = "dfbe277e56a376000877090da837660b4427aad530e3028d44e0bffe4f89a1c1"
 dependencies = [
  "gimli",
 ]
 
 [[package]]
-name = "adler"
-version = "1.0.2"
+name = "adler2"
+version = "2.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f26201604c87b1e01bd3d98f8d5d9a8fcbb815e8cedb41ffccbeb4bf593a35fe"
+checksum = "512761e0bb2578dd7380c6baaa0f4ce03e84f95e960231d1dec8bf4d7d6e2627"
 
 [[package]]
 name = "aho-corasick"
@@ -57,9 +57,9 @@ checksum = "1bec1de6f59aedf83baf9ff929c98f2ad654b97c9510f4e70cf6f661d49fd5b1"
 
 [[package]]
 name = "anstyle-parse"
-version = "0.2.3"
+version = "0.2.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c75ac65da39e5fe5ab759307499ddad880d724eed2f6ce5b5e8a26f4f387928c"
+checksum = "3b2d16507662817a6a20a9ea92df6652ee4f94f914589377d69f3b21bc5798a9"
 dependencies = [
  "utf8parse",
 ]
@@ -85,15 +85,15 @@ dependencies = [
 
 [[package]]
 name = "anyhow"
-version = "1.0.87"
+version = "1.0.94"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "10f00e1f6e58a40e807377c75c6a7f97bf9044fab57816f2414e6f5f4499d7b8"
+checksum = "c1fd03a028ef38ba2276dce7e33fcd6369c158a1bca17946c4b1b701891c1ff7"
 
 [[package]]
 name = "api_client"
 version = "0.1.0"
 dependencies = [
- "thiserror",
+ "thiserror 1.0.62",
  "vmm-sys-util",
 ]
 
@@ -115,7 +115,7 @@ dependencies = [
  "linux-loader",
  "log",
  "serde",
- "thiserror",
+ "thiserror 1.0.62",
  "uuid",
  "vm-fdt",
  "vm-memory",
@@ -125,9 +125,9 @@ dependencies = [
 
 [[package]]
 name = "async-broadcast"
-version = "0.7.0"
+version = "0.7.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "258b52a1aa741b9f09783b2d86cf0aeeb617bbf847f6933340a39644227acbdb"
+checksum = "20cd0e2e25ea8e5f7e9df04578dc6cf5c83577fd09b1a46aaf5c85e1c33f2a7e"
 dependencies = [
  "event-listener",
  "event-listener-strategy",
@@ -204,9 +204,9 @@ dependencies = [
 
 [[package]]
 name = "async-process"
-version = "2.2.2"
+version = "2.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a53fc6301894e04a92cb2584fedde80cb25ba8e02d9dc39d4a87d036e22f397d"
+checksum = "63255f1dc2381611000436537bbedfe83183faa303a5a0edaf191edef06526bb"
 dependencies = [
  "async-channel",
  "async-io",
@@ -219,7 +219,6 @@ dependencies = [
  "futures-lite",
  "rustix",
  "tracing",
- "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -235,9 +234,9 @@ dependencies = [
 
 [[package]]
 name = "async-signal"
-version = "0.2.6"
+version = "0.2.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "afe66191c335039c7bb78f99dc7520b0cbb166b3a1cb33a03f53d8a1c6f2afda"
+checksum = "637e00349800c0bdf8bfc21ebbc0b6524abea702b0da4168ac00d070d0c0b9f3"
 dependencies = [
  "async-io",
  "async-lock",
@@ -248,7 +247,7 @@ dependencies = [
  "rustix",
  "signal-hook-registry",
  "slab",
- "windows-sys 0.52.0",
+ "windows-sys 0.59.0",
 ]
 
 [[package]]
@@ -259,9 +258,9 @@ checksum = "8b75356056920673b02621b35afd0f7dda9306d03c79a30f5c56c44cf256e3de"
 
 [[package]]
 name = "async-trait"
-version = "0.1.80"
+version = "0.1.83"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c6fa2087f2753a7da8cc1c0dbfcf89579dd57458e36769de5ac750b4671737ca"
+checksum = "721cae7de5c34fbb2acd27e21e6d2cf7b886dce0c27388d46c4e6c47ea4318dd"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -282,17 +281,17 @@ checksum = "0c4b4d0bd25bd0b74681c0ad21497610ce1b7c91b1022cd21c80c6fbdd9476b0"
 
 [[package]]
 name = "backtrace"
-version = "0.3.73"
+version = "0.3.74"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5cc23269a4f8976d0a4d2e7109211a419fe30e8d88d677cd60b6bc79c5732e0a"
+checksum = "8d82cb332cdfaed17ae235a638438ac4d4839913cc2af585c3c6746e8f8bee1a"
 dependencies = [
  "addr2line",
- "cc",
  "cfg-if",
  "libc",
  "miniz_oxide",
  "object",
  "rustc-demangle",
+ "windows-targets 0.52.6",
 ]
 
 [[package]]
@@ -330,7 +329,7 @@ dependencies = [
  "remain",
  "serde",
  "smallvec",
- "thiserror",
+ "thiserror 1.0.62",
  "uuid",
  "virtio-bindings",
  "virtio-queue",
@@ -421,7 +420,7 @@ checksum = "1462739cb27611015575c0c11df5df7601141071f07518d56fcc1be504cbec97"
 
 [[package]]
 name = "cloud-hypervisor"
-version = "42.0.0"
+version = "43.0.0"
 dependencies = [
  "anyhow",
  "api_client",
@@ -440,7 +439,7 @@ dependencies = [
  "serde_json",
  "signal-hook",
  "test_infra",
- "thiserror",
+ "thiserror 1.0.62",
  "tpm",
  "tracer",
  "vm-memory",
@@ -477,9 +476,9 @@ dependencies = [
 
 [[package]]
 name = "cpufeatures"
-version = "0.2.12"
+version = "0.2.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "53fe5e26ff1b7aef8bca9c6080520cfb8d9333c7568e1829cef191a9723e5504"
+checksum = "16b80225097f2e5ae4e7179dd2266824648f3e2f49d9134d584b76389d31c4c3"
 dependencies = [
  "libc",
 ]
@@ -575,7 +574,7 @@ dependencies = [
  "num_enum",
  "pci",
  "serde",
- "thiserror",
+ "thiserror 1.0.62",
  "tpm",
  "vm-allocator",
  "vm-device",
@@ -720,9 +719,9 @@ dependencies = [
 
 [[package]]
 name = "event-listener-strategy"
-version = "0.5.1"
+version = "0.5.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "332f51cb23d20b0de8458b86580878211da09bcd4503cb579c225b3d124cabb3"
+checksum = "3c3e4e0dd3673c1139bf041f3008816d9cf2946bbfac2945c09e523b8d7b05b2"
 dependencies = [
  "event-listener",
  "pin-project-lite",
@@ -741,9 +740,9 @@ dependencies = [
 
 [[package]]
 name = "fastrand"
-version = "2.1.1"
+version = "2.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e8c02a5121d4ea3eb16a80748c74f5549a5665e4c21333c6098f283870fbdea6"
+checksum = "486f806e73c5707928240ddc295403b1b93c96a02038563881c4a2fd84b81ac4"
 
 [[package]]
 name = "fdt"
@@ -753,9 +752,9 @@ checksum = "784a4df722dc6267a04af36895398f59d21d07dce47232adf31ec0ff2fa45e67"
 
 [[package]]
 name = "flume"
-version = "0.11.0"
+version = "0.11.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "55ac459de2512911e4b674ce33cf20befaba382d05b62b008afc1c8b57cbf181"
+checksum = "da0e4dd2a88388a1f4ccc7c9ce104604dab68d9f408dc34cd45823d5a9069095"
 dependencies = [
  "futures-core",
  "futures-sink",
@@ -796,9 +795,9 @@ dependencies = [
 
 [[package]]
 name = "futures-core"
-version = "0.3.30"
+version = "0.3.31"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dfc6580bb841c5a68e9ef15c77ccc837b40a7504914d52e47b8b0e9bbda25a1d"
+checksum = "05f29059c0c2090612e8d742178b0580d2dc940c837851ad723096f87af6663e"
 
 [[package]]
 name = "futures-executor"
@@ -907,9 +906,9 @@ dependencies = [
 
 [[package]]
 name = "getrandom"
-version = "0.2.14"
+version = "0.2.15"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "94b22e06ecb0110981051723910cbf0b5f5e09a2062dd7663334ee79a9d1286c"
+checksum = "c4567c8db10ae91089c99af84c68c38da3ec2f087c3f82960bcdbf3656b6f4d7"
 dependencies = [
  "cfg-if",
  "js-sys",
@@ -920,9 +919,9 @@ dependencies = [
 
 [[package]]
 name = "gimli"
-version = "0.29.0"
+version = "0.31.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "40ecd4077b5ae9fd2e9e169b102c6c330d0605168eb0e8bf79952b256dbefffd"
+checksum = "07e28edb80900c19c28f1072f2e8aeca7fa06b23cd4169cefe1af5aa3260783f"
 
 [[package]]
 name = "glob"
@@ -959,6 +958,7 @@ name = "hypervisor"
 version = "0.1.0"
 dependencies = [
  "anyhow",
+ "arc-swap",
  "byteorder",
  "cfg-if",
  "concat-idents",
@@ -974,7 +974,7 @@ dependencies = [
  "mshv-ioctls",
  "serde",
  "serde_with",
- "thiserror",
+ "thiserror 1.0.62",
  "vfio-ioctls",
  "vm-memory",
  "vmm-sys-util",
@@ -1007,9 +1007,9 @@ dependencies = [
  "igvm_defs",
  "open-enum",
  "range_map_vec",
- "thiserror",
+ "thiserror 1.0.62",
  "tracing",
- "zerocopy",
+ "zerocopy 0.7.35",
 ]
 
 [[package]]
@@ -1021,7 +1021,7 @@ dependencies = [
  "bitfield-struct",
  "open-enum",
  "static_assertions",
- "zerocopy",
+ "zerocopy 0.7.35",
 ]
 
 [[package]]
@@ -1085,20 +1085,20 @@ dependencies = [
 
 [[package]]
 name = "kvm-bindings"
-version = "0.9.1"
+version = "0.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2efe3f1a4437bffe000e6297a593b98184213cd27486776c335f95ab53d48e3a"
+checksum = "fa4933174d0cc4b77b958578cd45784071cc5ae212c2d78fbd755aaaa6dfa71a"
 dependencies = [
  "serde",
  "vmm-sys-util",
- "zerocopy",
+ "zerocopy 0.7.35",
 ]
 
 [[package]]
 name = "kvm-ioctls"
-version = "0.18.0"
+version = "0.19.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "92c2176b91f68903b54ac8c6185bada7d607ca6110998976ff15c032f88a7d39"
+checksum = "337d1afa126368bbd6a5c328048f71a69a737e9afe7e436b392a8f8d770c9171"
 dependencies = [
  "bitflags 2.6.0",
  "kvm-bindings",
@@ -1114,7 +1114,7 @@ checksum = "dafb8a4afee64f167eb2b52d32f0eea002e41a7a6450e68c799c8ec3a81a634c"
 dependencies = [
  "enumflags2",
  "libc",
- "thiserror",
+ "thiserror 1.0.62",
 ]
 
 [[package]]
@@ -1125,9 +1125,9 @@ checksum = "bbd2bcb4c963f2ddae06a2efc7e9f3591312473c50c6685e1f298068316e66fe"
 
 [[package]]
 name = "libc"
-version = "0.2.158"
+version = "0.2.167"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d8adc4bb1803a324070e64a98ae98f38934d91957a99cfb3a43dcbc01bc56439"
+checksum = "09d6582e104315a817dff97f75133544b2e094ee22447d2acf4a74e189ba06fc"
 
 [[package]]
 name = "libredox"
@@ -1167,9 +1167,9 @@ dependencies = [
 
 [[package]]
 name = "linux-loader"
-version = "0.12.0"
+version = "0.13.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3d379d0089d0fbf4161c35a4fdfd76125923f1a93632c49195f5372b4c0b1472"
+checksum = "870c3814345f050991f99869417779f6062542bcf4ed81db7a1b926ad1306638"
 dependencies = [
  "vm-memory",
 ]
@@ -1182,9 +1182,9 @@ checksum = "78b3ae25bc7c8c38cec158d1f2757ee79e9b3740fbc7ccf0e59e4b08d793fa89"
 
 [[package]]
 name = "lock_api"
-version = "0.4.11"
+version = "0.4.12"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3c168f8615b12bc01f9c17e2eb0cc07dcae1940121185446edc3744920e8ef45"
+checksum = "07af8b9cdd281b7915f413fa73f29ebd5d55d0d3f0155584dade1ff18cea1b17"
 dependencies = [
  "autocfg",
  "scopeguard",
@@ -1228,11 +1228,11 @@ dependencies = [
 
 [[package]]
 name = "miniz_oxide"
-version = "0.7.2"
+version = "0.8.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9d811f3e15f28568be3407c8e7fdb6514c1cda3cb30683f15b6a1a1dc4ea14a7"
+checksum = "e2d80299ef12ff69b16a84bb182e3b9df68b5a91574d3d4fa6e41b65deec4df1"
 dependencies = [
- "adler",
+ "adler2",
 ]
 
 [[package]]
@@ -1243,25 +1243,27 @@ checksum = "9bec4598fddb13cc7b528819e697852653252b760f1228b7642679bf2ff2cd07"
 
 [[package]]
 name = "mshv-bindings"
-version = "0.3.0"
-source = "git+https://github.com/rust-vmm/mshv?tag=v0.3.0#fda05380ea4c68b807996299d5ffb2854ca6d01d"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1e0cb5031f3243a7459b7c13d960d25420980874eebda816db24ce6077e21d43"
 dependencies = [
  "libc",
  "num_enum",
  "serde",
  "serde_derive",
  "vmm-sys-util",
- "zerocopy",
+ "zerocopy 0.8.8",
 ]
 
 [[package]]
 name = "mshv-ioctls"
-version = "0.3.0"
-source = "git+https://github.com/rust-vmm/mshv?tag=v0.3.0#fda05380ea4c68b807996299d5ffb2854ca6d01d"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "89abe853221fa6f14ad4066affb9abda241a03d65622887d5794e1422d0bd75a"
 dependencies = [
  "libc",
  "mshv-bindings",
- "thiserror",
+ "thiserror 2.0.6",
  "vmm-sys-util",
 ]
 
@@ -1296,7 +1298,7 @@ dependencies = [
  "rate_limiter",
  "serde",
  "serde_json",
- "thiserror",
+ "thiserror 1.0.62",
  "virtio-bindings",
  "virtio-queue",
  "vm-memory",
@@ -1355,9 +1357,9 @@ dependencies = [
 
 [[package]]
 name = "object"
-version = "0.36.1"
+version = "0.36.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "081b846d1d56ddfc18fdf1a922e4f6e07a11768ea1b92dec44e42b72712ccfce"
+checksum = "aedf0a2d09c573ed1d8d85b30c119153926a2b36dce0ab28322c09a117a4683e"
 dependencies = [
  "memchr",
 ]
@@ -1399,9 +1401,9 @@ dependencies = [
 
 [[package]]
 name = "openssl-sys"
-version = "0.9.102"
+version = "0.9.104"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c597637d56fbc83893a35eb0dd04b2b8e7a50c91e64e9493e398b5df4fb45fa2"
+checksum = "45abf306cbf99debc8195b66b7346498d7b10c210de50418b5ccd7ceba08c741"
 dependencies = [
  "cc",
  "libc",
@@ -1500,7 +1502,7 @@ dependencies = [
  "libc",
  "log",
  "serde",
- "thiserror",
+ "thiserror 1.0.62",
  "vfio-bindings",
  "vfio-ioctls",
  "vfio_user",
@@ -1520,15 +1522,15 @@ dependencies = [
  "serde",
  "serde_json",
  "test_infra",
- "thiserror",
+ "thiserror 1.0.62",
  "wait-timeout",
 ]
 
 [[package]]
 name = "pin-project-lite"
-version = "0.2.14"
+version = "0.2.15"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bda66fc9667c18cb2758a2ac84d1167245054bcf85d5d1aaa6923f45801bdd02"
+checksum = "915a1e146535de9163f3987b8944ed8cf49a18bb0056bcebcdcece385cece4ff"
 
 [[package]]
 name = "pin-utils"
@@ -1549,9 +1551,9 @@ dependencies = [
 
 [[package]]
 name = "pkg-config"
-version = "0.3.30"
+version = "0.3.31"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d231b230927b5e4ad203db57bbcbee2802f6bce620b1e4a9024a07d94e2907ec"
+checksum = "953ec861398dccce10c670dfeaf3ec4911ca479e9c02154b3a215178c5f566f2"
 
 [[package]]
 name = "pnet"
@@ -1665,7 +1667,7 @@ version = "0.2.20"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "77957b295656769bb8ad2b6a6b09d897d94f05c41b069aede1fcdaa675eaea04"
 dependencies = [
- "zerocopy",
+ "zerocopy 0.7.35",
 ]
 
 [[package]]
@@ -1738,7 +1740,7 @@ dependencies = [
  "epoll",
  "libc",
  "log",
- "thiserror",
+ "thiserror 1.0.62",
  "vmm-sys-util",
 ]
 
@@ -1768,14 +1770,14 @@ checksum = "ba009ff324d1fc1b900bd1fdb31564febe58a8ccc8a6fdbb93b543d33b13ca43"
 dependencies = [
  "getrandom",
  "libredox",
- "thiserror",
+ "thiserror 1.0.62",
 ]
 
 [[package]]
 name = "regex"
-version = "1.10.5"
+version = "1.11.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b91213439dad192326a0d7c6ee3955910425f441d7038e0d6933b0aec5c4517f"
+checksum = "b544ef1b4eac5dc2db33ea63606ae9ffcfac26c1416a2806ae0bf5f56b201191"
 dependencies = [
  "aho-corasick",
  "memchr",
@@ -1785,9 +1787,9 @@ dependencies = [
 
 [[package]]
 name = "regex-automata"
-version = "0.4.7"
+version = "0.4.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "38caf58cc5ef2fed281f89292ef23f6365465ed9a41b7a7754eb4e26496c92df"
+checksum = "809e8dc61f6de73b46c85f4c96486310fe304c434cfa43669d7b40f711150908"
 dependencies = [
  "aho-corasick",
  "memchr",
@@ -1796,9 +1798,9 @@ dependencies = [
 
 [[package]]
 name = "regex-syntax"
-version = "0.8.3"
+version = "0.8.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "adad44e29e4c806119491a7f06f03de4d1af22c3a680dd47f1e6e179439d1f56"
+checksum = "2b15c43186be67a4fd63bee50d0303afffcef381492ebe2c5d87f324e1b8815c"
 
 [[package]]
 name = "remain"
@@ -2006,9 +2008,9 @@ checksum = "7da8b5736845d9f2fcb837ea5d9e2628564b3b043a70948a3f0b778838c5fb4f"
 
 [[package]]
 name = "syn"
-version = "2.0.66"
+version = "2.0.87"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c42f3f41a2de00b01c0aaad383c5a45241efc8b2d1eda5661812fda5f3cdcff5"
+checksum = "25aa4ce346d03a6dcd68dd8b4010bcb74e54e62c90c573f394c46eae99aba32d"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -2059,7 +2061,16 @@ version = "1.0.62"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f2675633b1499176c2dff06b0856a27976a8f9d436737b4cf4f312d4d91d8bbb"
 dependencies = [
- "thiserror-impl",
+ "thiserror-impl 1.0.62",
+]
+
+[[package]]
+name = "thiserror"
+version = "2.0.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8fec2a1820ebd077e2b90c4df007bebf344cd394098a13c563957d0afc83ea47"
+dependencies = [
+ "thiserror-impl 2.0.6",
 ]
 
 [[package]]
@@ -2073,6 +2084,17 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "thiserror-impl"
+version = "2.0.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d65750cab40f4ff1929fb1ba509e9914eb756131cef4210da8d5d700d26f6312"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "thousands"
 version = "0.2.0"
@@ -2105,7 +2127,7 @@ dependencies = [
  "libc",
  "log",
  "net_gen",
- "thiserror",
+ "thiserror 1.0.62",
  "vmm-sys-util",
 ]
 
@@ -2210,14 +2232,14 @@ checksum = "accd4ea62f7bb7a82fe23066fb0957d48ef677f6eeb8215f372f52e48bb32426"
 
 [[package]]
 name = "version_check"
-version = "0.9.4"
+version = "0.9.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "49874b5167b65d7193b8aba1567f5c7d93d001cafc34600cee003eda787e483f"
+checksum = "0b928f33d975fc6ad9f86c8f283853ad26bdd5b10b7f1542aa2fa15e2289105a"
 
 [[package]]
 name = "vfio-bindings"
 version = "0.4.0"
-source = "git+https://github.com/rust-vmm/vfio?branch=main#a51a4746b0d317bfc21fa49d40f9287f3b8137fd"
+source = "git+https://github.com/rust-vmm/vfio?branch=main#e71efe41d59adf23377f7066a49903c6c19f5ac3"
 dependencies = [
  "vmm-sys-util",
 ]
@@ -2225,7 +2247,7 @@ dependencies = [
 [[package]]
 name = "vfio-ioctls"
 version = "0.2.0"
-source = "git+https://github.com/rust-vmm/vfio?branch=main#a51a4746b0d317bfc21fa49d40f9287f3b8137fd"
+source = "git+https://github.com/rust-vmm/vfio?branch=main#e71efe41d59adf23377f7066a49903c6c19f5ac3"
 dependencies = [
  "byteorder",
  "kvm-bindings",
@@ -2234,7 +2256,7 @@ dependencies = [
  "log",
  "mshv-bindings",
  "mshv-ioctls",
- "thiserror",
+ "thiserror 1.0.62",
  "vfio-bindings",
  "vm-memory",
  "vmm-sys-util",
@@ -2243,7 +2265,7 @@ dependencies = [
 [[package]]
 name = "vfio_user"
 version = "0.1.0"
-source = "git+https://github.com/rust-vmm/vfio-user?branch=main#bf7d7e851b604d8414a7960fb9137b59fc42421d"
+source = "git+https://github.com/rust-vmm/vfio-user?branch=main#3febcdd3fa2531623865663ca1721e1962ed9979"
 dependencies = [
  "bitflags 1.3.2",
  "libc",
@@ -2251,7 +2273,7 @@ dependencies = [
  "serde",
  "serde_derive",
  "serde_json",
- "thiserror",
+ "thiserror 1.0.62",
  "vfio-bindings",
  "vm-memory",
  "vmm-sys-util",
@@ -2259,9 +2281,8 @@ dependencies = [
 
 [[package]]
 name = "vhost"
-version = "0.12.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7c1c4c6c9f79fbe3150d9a403008ca416d34c489897effdda28b646f09900aad"
+version = "0.12.1"
+source = "git+https://github.com/rust-vmm/vhost?rev=d983ae0#d983ae07f78663b7d24059667376992460b571a2"
 dependencies = [
  "bitflags 2.6.0",
  "libc",
@@ -2272,9 +2293,8 @@ dependencies = [
 
 [[package]]
 name = "vhost-user-backend"
-version = "0.16.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "73768c8584e0be5ed8feb063785910cabe3f1af6661a5953fd3247fa611ddfaf"
+version = "0.16.1"
+source = "git+https://github.com/rust-vmm/vhost?rev=d983ae0#d983ae07f78663b7d24059667376992460b571a2"
 dependencies = [
  "libc",
  "log",
@@ -2324,9 +2344,9 @@ dependencies = [
 
 [[package]]
 name = "virtio-bindings"
-version = "0.2.3"
+version = "0.2.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "68d0df4f5ad79b1dc81b5913ac737e24a84dcd5100f36ed953a1faec18aba241"
+checksum = "1711e61c00f8cb450bd15368152a1e37a12ef195008ddc7d0f4812f9e2b30a68"
 
 [[package]]
 name = "virtio-devices"
@@ -2350,7 +2370,7 @@ dependencies = [
  "serde_json",
  "serde_with",
  "serial_buffer",
- "thiserror",
+ "thiserror 1.0.62",
  "vhost",
  "virtio-bindings",
  "virtio-queue",
@@ -2364,9 +2384,9 @@ dependencies = [
 
 [[package]]
 name = "virtio-queue"
-version = "0.13.0"
+version = "0.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ffb1761348d3b5e82131379b9373435b48dc8333100bff3f1cdf9cc541a0ad83"
+checksum = "872e2f3fbd70a7e6f01689720cce3d5c2c5efe52b484dd07b674246ada0e9a8d"
 dependencies = [
  "log",
  "virtio-bindings",
@@ -2390,7 +2410,7 @@ dependencies = [
  "anyhow",
  "hypervisor",
  "serde",
- "thiserror",
+ "thiserror 1.0.62",
  "vfio-ioctls",
  "vm-memory",
  "vmm-sys-util",
@@ -2403,13 +2423,13 @@ source = "git+https://github.com/rust-vmm/vm-fdt?branch=main#ef5bd734f5f66fb0772
 
 [[package]]
 name = "vm-memory"
-version = "0.15.0"
+version = "0.16.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a320fc11792e063174402ff444aa3c80363cbf1e31c47b5ef74124406c334ce6"
+checksum = "f1720e7240cdc739f935456eb77f370d7e9b2a3909204da1e2b47bef1137a013"
 dependencies = [
  "arc-swap",
  "libc",
- "thiserror",
+ "thiserror 1.0.62",
  "winapi",
 ]
 
@@ -2420,7 +2440,7 @@ dependencies = [
  "anyhow",
  "serde",
  "serde_json",
- "thiserror",
+ "thiserror 1.0.62",
  "vm-memory",
 ]
 
@@ -2475,7 +2495,7 @@ dependencies = [
  "serde_json",
  "serial_buffer",
  "signal-hook",
- "thiserror",
+ "thiserror 1.0.62",
  "tracer",
  "uuid",
  "vfio-ioctls",
@@ -2489,7 +2509,7 @@ dependencies = [
  "vm-virtio",
  "vmm-sys-util",
  "zbus",
- "zerocopy",
+ "zerocopy 0.7.35",
 ]
 
 [[package]]
@@ -2610,7 +2630,16 @@ version = "0.52.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "282be5f36a8ce781fad8c8ae18fa3f9beff57ec1b52cb3de0789201425d9a33d"
 dependencies = [
- "windows-targets 0.52.4",
+ "windows-targets 0.52.6",
+]
+
+[[package]]
+name = "windows-sys"
+version = "0.59.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1e38bc4d79ed67fd075bcc251a1c39b32a1776bbe92e5bef1f0bf1f8c531853b"
+dependencies = [
+ "windows-targets 0.52.6",
 ]
 
 [[package]]
@@ -2630,17 +2659,18 @@ dependencies = [
 
 [[package]]
 name = "windows-targets"
-version = "0.52.4"
+version = "0.52.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7dd37b7e5ab9018759f893a1952c9420d060016fc19a472b4bb20d1bdd694d1b"
+checksum = "9b724f72796e036ab90c1021d4780d4d3d648aca59e491e6b98e725b84e99973"
 dependencies = [
- "windows_aarch64_gnullvm 0.52.4",
- "windows_aarch64_msvc 0.52.4",
- "windows_i686_gnu 0.52.4",
- "windows_i686_msvc 0.52.4",
- "windows_x86_64_gnu 0.52.4",
- "windows_x86_64_gnullvm 0.52.4",
- "windows_x86_64_msvc 0.52.4",
+ "windows_aarch64_gnullvm 0.52.6",
+ "windows_aarch64_msvc 0.52.6",
+ "windows_i686_gnu 0.52.6",
+ "windows_i686_gnullvm",
+ "windows_i686_msvc 0.52.6",
+ "windows_x86_64_gnu 0.52.6",
+ "windows_x86_64_gnullvm 0.52.6",
+ "windows_x86_64_msvc 0.52.6",
 ]
 
 [[package]]
@@ -2651,9 +2681,9 @@ checksum = "2b38e32f0abccf9987a4e3079dfb67dcd799fb61361e53e2882c3cbaf0d905d8"
 
 [[package]]
 name = "windows_aarch64_gnullvm"
-version = "0.52.4"
+version = "0.52.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bcf46cf4c365c6f2d1cc93ce535f2c8b244591df96ceee75d8e83deb70a9cac9"
+checksum = "32a4622180e7a0ec044bb555404c800bc9fd9ec262ec147edd5989ccd0c02cd3"
 
 [[package]]
 name = "windows_aarch64_msvc"
@@ -2663,9 +2693,9 @@ checksum = "dc35310971f3b2dbbf3f0690a219f40e2d9afcf64f9ab7cc1be722937c26b4bc"
 
 [[package]]
 name = "windows_aarch64_msvc"
-version = "0.52.4"
+version = "0.52.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "da9f259dd3bcf6990b55bffd094c4f7235817ba4ceebde8e6d11cd0c5633b675"
+checksum = "09ec2a7bb152e2252b53fa7803150007879548bc709c039df7627cabbd05d469"
 
 [[package]]
 name = "windows_i686_gnu"
@@ -2675,9 +2705,15 @@ checksum = "a75915e7def60c94dcef72200b9a8e58e5091744960da64ec734a6c6e9b3743e"
 
 [[package]]
 name = "windows_i686_gnu"
-version = "0.52.4"
+version = "0.52.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b474d8268f99e0995f25b9f095bc7434632601028cf86590aea5c8a5cb7801d3"
+checksum = "8e9b5ad5ab802e97eb8e295ac6720e509ee4c243f69d781394014ebfe8bbfa0b"
+
+[[package]]
+name = "windows_i686_gnullvm"
+version = "0.52.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0eee52d38c090b3caa76c563b86c3a4bd71ef1a819287c19d586d7334ae8ed66"
 
 [[package]]
 name = "windows_i686_msvc"
@@ -2687,9 +2723,9 @@ checksum = "8f55c233f70c4b27f66c523580f78f1004e8b5a8b659e05a4eb49d4166cca406"
 
 [[package]]
 name = "windows_i686_msvc"
-version = "0.52.4"
+version = "0.52.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1515e9a29e5bed743cb4415a9ecf5dfca648ce85ee42e15873c3cd8610ff8e02"
+checksum = "240948bc05c5e7c6dabba28bf89d89ffce3e303022809e73deaefe4f6ec56c66"
 
 [[package]]
 name = "windows_x86_64_gnu"
@@ -2699,9 +2735,9 @@ checksum = "53d40abd2583d23e4718fddf1ebec84dbff8381c07cae67ff7768bbf19c6718e"
 
 [[package]]
 name = "windows_x86_64_gnu"
-version = "0.52.4"
+version = "0.52.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5eee091590e89cc02ad514ffe3ead9eb6b660aedca2183455434b93546371a03"
+checksum = "147a5c80aabfbf0c7d901cb5895d1de30ef2907eb21fbbab29ca94c5b08b1a78"
 
 [[package]]
 name = "windows_x86_64_gnullvm"
@@ -2711,9 +2747,9 @@ checksum = "0b7b52767868a23d5bab768e390dc5f5c55825b6d30b86c844ff2dc7414044cc"
 
 [[package]]
 name = "windows_x86_64_gnullvm"
-version = "0.52.4"
+version = "0.52.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "77ca79f2451b49fa9e2af39f0747fe999fcda4f5e241b2898624dca97a1f2177"
+checksum = "24d5b23dc417412679681396f2b49f3de8c1473deb516bd34410872eff51ed0d"
 
 [[package]]
 name = "windows_x86_64_msvc"
@@ -2723,9 +2759,9 @@ checksum = "ed94fce61571a4006852b7389a063ab983c02eb1bb37b47f8272ce92d06d9538"
 
 [[package]]
 name = "windows_x86_64_msvc"
-version = "0.52.4"
+version = "0.52.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "32b752e52a2da0ddfbdbcc6fceadfeede4c939ed16d13e648833a61dfb611ed8"
+checksum = "589f6da84c646204747d1270a2a5661ea66ed1cced2631d546fdfb155959f9ec"
 
 [[package]]
 name = "winnow"
@@ -2738,12 +2774,12 @@ dependencies = [
 
 [[package]]
 name = "xdg-home"
-version = "1.2.0"
+version = "1.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ca91dcf8f93db085f3a0a29358cd0b9d670915468f4290e8b85d118a34211ab8"
+checksum = "ec1cdab258fb55c0da61328dc52c8764709b249011b2cad0454c72f0bf10a1f6"
 dependencies = [
  "libc",
- "windows-sys 0.52.0",
+ "windows-sys 0.59.0",
 ]
 
 [[package]]
@@ -2815,7 +2851,16 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1b9b4fd18abc82b8136838da5d50bae7bdea537c574d8dc1a34ed098d6c166f0"
 dependencies = [
  "byteorder",
- "zerocopy-derive",
+ "zerocopy-derive 0.7.35",
+]
+
+[[package]]
+name = "zerocopy"
+version = "0.8.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5a4e33e6dce36f2adba29746927f8e848ba70989fdb61c772773bbdda8b5d6a7"
+dependencies = [
+ "zerocopy-derive 0.8.8",
 ]
 
 [[package]]
@@ -2829,6 +2874,17 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "zerocopy-derive"
+version = "0.8.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3cd137b4cc21bde6ecce3bbbb3350130872cda0be2c6888874279ea76e17d4c1"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "zvariant"
 version = "4.2.0"

Cargo.toml

@@ -7,7 +7,7 @@ edition = "2021"
 homepage = "https://github.com/cloud-hypervisor/cloud-hypervisor"
 license = "Apache-2.0 AND BSD-3-Clause"
 name = "cloud-hypervisor"
-version = "42.0.0"
+version = "43.0.0"
 # Minimum buildable version:
 # Keep in sync with version in .github/workflows/build.yaml
 # Policy on MSRV (see #4318):
@@ -29,14 +29,14 @@ inherits = "release"
 strip = false
 
 [dependencies]
-anyhow = "1.0.87"
+anyhow = "1.0.94"
 api_client = { path = "api_client" }
 clap = { version = "4.5.13", features = ["string"] }
 dhat = { version = "0.3.3", optional = true }
 epoll = "4.3.3"
 event_monitor = { path = "event_monitor" }
 hypervisor = { path = "hypervisor" }
-libc = "0.2.158"
+libc = "0.2.167"
 log = { version = "0.4.22", features = ["std"] }
 option_parser = { path = "option_parser" }
 seccompiler = { workspace = true }
@@ -103,18 +103,19 @@ members = [
 
 [workspace.dependencies]
 acpi_tables = { git = "https://github.com/rust-vmm/acpi_tables", branch = "main" }
-kvm-bindings = "0.9.1"
-kvm-ioctls = "0.18.0"
-linux-loader = "0.12.0"
-mshv-bindings = { git = "https://github.com/rust-vmm/mshv", tag = "v0.3.0" }
-mshv-ioctls = { git = "https://github.com/rust-vmm/mshv", tag = "v0.3.0" }
+kvm-bindings = "0.10.0"
+kvm-ioctls = "0.19.0"
+linux-loader = "0.13.0"
+mshv-bindings = "0.3.2"
+mshv-ioctls = "0.3.2"
 seccompiler = "0.4.0"
 vfio-bindings = { git = "https://github.com/rust-vmm/vfio", branch = "main" }
 vfio-ioctls = { git = "https://github.com/rust-vmm/vfio", branch = "main", default-features = false }
 vfio_user = { git = "https://github.com/rust-vmm/vfio-user", branch = "main" }
-vhost = "0.12.0"
-virtio-bindings = "0.2.2"
-virtio-queue = "0.13.0"
+vhost = { git = "https://github.com/rust-vmm/vhost", rev = "d983ae0" }
+vhost-user-backend = { git = "https://github.com/rust-vmm/vhost", rev = "d983ae0" }
+virtio-bindings = "0.2.4"
+virtio-queue = "0.14.0"
 vm-fdt = { git = "https://github.com/rust-vmm/vm-fdt", branch = "main" }
-vm-memory = "0.15.0"
+vm-memory = "0.16.1"
 vmm-sys-util = "0.12.1"

arch/Cargo.toml

@@ -11,10 +11,10 @@ sev_snp = []
 tdx = []
 
 [dependencies]
-anyhow = "1.0.87"
+anyhow = "1.0.94"
 byteorder = "1.5.0"
 hypervisor = { path = "../hypervisor" }
-libc = "0.2.158"
+libc = "0.2.167"
 linux-loader = { workspace = true, features = ["bzimage", "elf", "pe"] }
 log = "0.4.22"
 serde = { version = "1.0.208", features = ["derive", "rc"] }
@@ -24,6 +24,6 @@ vm-memory = { workspace = true, features = ["backend-bitmap", "backend-mmap"] }
 vm-migration = { path = "../vm-migration" }
 vmm-sys-util = { workspace = true, features = ["with-serde"] }
 
-[target.'cfg(target_arch = "aarch64")'.dependencies]
+[target.'cfg(any(target_arch = "aarch64", target_arch = "riscv64"))'.dependencies]
 fdt_parser = { version = "0.1.5", package = "fdt" }
 vm-fdt = { workspace = true }

arch/src/lib.rs

@@ -1,3 +1,4 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
 // Copyright 2020 Arm Limited (or its affiliates). All rights reserved.
 // Copyright © 2020, Oracle and/or its affiliates.
 //
@@ -5,7 +6,7 @@
 // SPDX-License-Identifier: Apache-2.0
 
 //! Implements platform specific functionality.
-//! Supported platforms: x86_64, aarch64.
+//! Supported platforms: x86_64, aarch64, riscv64.
 
 #[macro_use]
 extern crate log;
@@ -32,6 +33,9 @@ pub enum Error {
     #[cfg(target_arch = "aarch64")]
     #[error("Platform specific error (aarch64): {0:?}")]
     PlatformSpecific(aarch64::Error),
+    #[cfg(target_arch = "riscv64")]
+    #[error("Platform specific error (riscv64): {0:?}")]
+    PlatformSpecific(riscv64::Error),
     #[error("The memory map table extends past the end of guest memory")]
     MemmapTablePastRamEnd,
     #[error("Error writing memory map table to guest memory")]
@@ -85,6 +89,17 @@ pub use aarch64::{
     layout::IRQ_BASE, uefi, EntryPoint, _NSIG,
 };
 
+/// Module for riscv64 related functionality.
+#[cfg(target_arch = "riscv64")]
+pub mod riscv64;
+
+#[cfg(target_arch = "riscv64")]
+pub use riscv64::{
+    arch_memory_regions, configure_system, configure_vcpu, fdt::DeviceInfoForFdt,
+    get_host_cpu_phys_bits, initramfs_load_addr, layout, layout::CMDLINE_MAX_SIZE,
+    layout::IRQ_BASE, EntryPoint, _NSIG,
+};
+
 #[cfg(target_arch = "x86_64")]
 pub mod x86_64;
 
@@ -132,7 +147,7 @@ pub enum DeviceType {
     /// Device Type: Virtio.
     Virtio(u32),
     /// Device Type: Serial.
-    #[cfg(target_arch = "aarch64")]
+    #[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
     Serial,
     /// Device Type: RTC.
     #[cfg(target_arch = "aarch64")]
@@ -153,7 +168,7 @@ impl fmt::Display for DeviceType {
 
 /// Structure to describe MMIO device information
 #[derive(Clone, Debug)]
-#[cfg(target_arch = "aarch64")]
+#[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
 pub struct MmioDeviceInfo {
     pub addr: u64,
     pub len: u64,
@@ -162,7 +177,7 @@ pub struct MmioDeviceInfo {
 
 /// Structure to describe PCI space information
 #[derive(Clone, Debug)]
-#[cfg(target_arch = "aarch64")]
+#[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
 pub struct PciSpaceInfo {
     pub pci_segment_id: u16,
     pub mmio_config_address: u64,
@@ -170,7 +185,7 @@ pub struct PciSpaceInfo {
     pub pci_device_space_size: u64,
 }
 
-#[cfg(target_arch = "aarch64")]
+#[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
 impl DeviceInfoForFdt for MmioDeviceInfo {
     fn addr(&self) -> u64 {
         self.addr

arch/src/riscv64/fdt.rs

@@ -0,0 +1,484 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+// Copyright 2020 Arm Limited (or its affiliates). All rights reserved.
+// Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+// SPDX-License-Identifier: Apache-2.0
+//
+// Portions Copyright 2017 The Chromium OS Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the THIRD-PARTY file.
+
+use std::collections::HashMap;
+use std::ffi::CStr;
+use std::fmt::Debug;
+use std::sync::{Arc, Mutex};
+use std::{cmp, result, str};
+
+use byteorder::{BigEndian, ByteOrder};
+use hypervisor::arch::riscv64::aia::Vaia;
+use thiserror::Error;
+use vm_fdt::{FdtWriter, FdtWriterResult};
+use vm_memory::{Address, Bytes, GuestMemory, GuestMemoryError, GuestMemoryRegion};
+
+use super::super::{DeviceType, GuestMemoryMmap, InitramfsConfig};
+use super::layout::{
+    IRQ_BASE, MEM_32BIT_DEVICES_SIZE, MEM_32BIT_DEVICES_START, MEM_PCI_IO_SIZE, MEM_PCI_IO_START,
+    PCI_HIGH_BASE, PCI_MMIO_CONFIG_SIZE_PER_SEGMENT,
+};
+use crate::PciSpaceInfo;
+
+const AIA_APLIC_PHANDLE: u32 = 1;
+const AIA_IMSIC_PHANDLE: u32 = 2;
+const CPU_INTC_BASE_PHANDLE: u32 = 3;
+const CPU_BASE_PHANDLE: u32 = 256 + CPU_INTC_BASE_PHANDLE;
+// Read the documentation specified when appending the root node to the FDT.
+const ADDRESS_CELLS: u32 = 0x2;
+const SIZE_CELLS: u32 = 0x2;
+
+// From https://elixir.bootlin.com/linux/v6.10/source/include/dt-bindings/interrupt-controller/irq.h#L14
+const _IRQ_TYPE_EDGE_RISING: u32 = 1;
+const IRQ_TYPE_LEVEL_HI: u32 = 4;
+
+const S_MODE_EXT_IRQ: u32 = 9;
+
+/// Trait for devices to be added to the Flattened Device Tree.
+pub trait DeviceInfoForFdt {
+    /// Returns the address where this device will be loaded.
+    fn addr(&self) -> u64;
+    /// Returns the associated interrupt for this device.
+    fn irq(&self) -> u32;
+    /// Returns the amount of memory that needs to be reserved for this device.
+    fn length(&self) -> u64;
+}
+
+/// Errors thrown while configuring the Flattened Device Tree for riscv64.
+#[derive(Debug, Error)]
+pub enum Error {
+    /// Failure in writing FDT in memory.
+    #[error("Failure in writing FDT in memory: {0}")]
+    WriteFdtToMemory(GuestMemoryError),
+}
+type Result<T> = result::Result<T, Error>;
+
+/// Creates the flattened device tree for this riscv64 VM.
+#[allow(clippy::too_many_arguments)]
+pub fn create_fdt<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::BuildHasher>(
+    guest_mem: &GuestMemoryMmap,
+    cmdline: &str,
+    num_vcpu: u32,
+    device_info: &HashMap<(DeviceType, String), T, S>,
+    aia_device: &Arc<Mutex<dyn Vaia>>,
+    initrd: &Option<InitramfsConfig>,
+    pci_space_info: &[PciSpaceInfo],
+) -> FdtWriterResult<Vec<u8>> {
+    // Allocate stuff necessary for the holding the blob.
+    let mut fdt = FdtWriter::new()?;
+
+    // For an explanation why these nodes were introduced in the blob take a look at
+    // https://github.com/devicetree-org/devicetree-specification/releases/tag/v0.4
+    // In chapter 3.
+
+    // Header or the root node as per above mentioned documentation.
+    let root_node = fdt.begin_node("")?;
+    fdt.property_string("compatible", "linux,dummy-virt")?;
+    // For info on #address-cells and size-cells resort to Table 3.1 Root Node
+    // Properties
+    fdt.property_u32("#address-cells", ADDRESS_CELLS)?;
+    fdt.property_u32("#size-cells", SIZE_CELLS)?;
+    create_cpu_nodes(&mut fdt, num_vcpu)?;
+    create_memory_node(&mut fdt, guest_mem)?;
+    create_chosen_node(&mut fdt, cmdline, initrd)?;
+    create_aia_node(&mut fdt, aia_device)?;
+    create_devices_node(&mut fdt, device_info)?;
+    create_pci_nodes(&mut fdt, pci_space_info)?;
+
+    // End Header node.
+    fdt.end_node(root_node)?;
+
+    let fdt_final = fdt.finish()?;
+
+    Ok(fdt_final)
+}
+
+pub fn write_fdt_to_memory(fdt_final: Vec<u8>, guest_mem: &GuestMemoryMmap) -> Result<()> {
+    // Write FDT to memory.
+    guest_mem
+        .write_slice(fdt_final.as_slice(), super::layout::FDT_START)
+        .map_err(Error::WriteFdtToMemory)?;
+    Ok(())
+}
+
+// Following are the auxiliary function for creating the different nodes that we append to our FDT.
+fn create_cpu_nodes(fdt: &mut FdtWriter, num_cpus: u32) -> FdtWriterResult<()> {
+    // See https://elixir.bootlin.com/linux/v6.10/source/Documentation/devicetree/bindings/riscv/cpus.yaml
+    let cpus = fdt.begin_node("cpus")?;
+    // As per documentation, on RISC-V 64-bit systems value should be set to 1.
+    fdt.property_u32("#address-cells", 0x01)?;
+    fdt.property_u32("#size-cells", 0x0)?;
+    // TODO: Retrieve CPU frequency from cpu timer regs
+    let timebase_frequency: u32 = 0x989680;
+    fdt.property_u32("timebase-frequency", timebase_frequency)?;
+
+    for cpu_index in 0..num_cpus {
+        let cpu = fdt.begin_node(&format!("cpu@{:x}", cpu_index))?;
+        fdt.property_string("device_type", "cpu")?;
+        fdt.property_string("compatible", "riscv")?;
+        fdt.property_string("mmu-type", "sv48")?;
+        fdt.property_string("riscv,isa", "rv64imafdc_smaia_ssaia")?;
+        fdt.property_string("status", "okay")?;
+        fdt.property_u32("reg", cpu_index)?;
+        fdt.property_u32("phandle", CPU_BASE_PHANDLE + cpu_index)?;
+
+        // interrupt controller node
+        let intc_node = fdt.begin_node("interrupt-controller")?;
+        fdt.property_string("compatible", "riscv,cpu-intc")?;
+        fdt.property_u32("#interrupt-cells", 1u32)?;
+        fdt.property_null("#interrupt-controller")?;
+        fdt.property_u32("phandle", CPU_INTC_BASE_PHANDLE + cpu_index)?;
+        fdt.end_node(intc_node)?;
+
+        fdt.end_node(cpu)?;
+    }
+
+    fdt.end_node(cpus)?;
+
+    Ok(())
+}
+
+fn create_memory_node(fdt: &mut FdtWriter, guest_mem: &GuestMemoryMmap) -> FdtWriterResult<()> {
+    // Note: memory regions from "GuestMemory" are sorted and non-zero sized.
+    let ram_regions = {
+        let mut ram_regions = Vec::new();
+        let mut current_start = guest_mem
+            .iter()
+            .next()
+            .map(GuestMemoryRegion::start_addr)
+            .expect("GuestMemory must have one memory region at least")
+            .raw_value();
+        let mut current_end = current_start;
+
+        for (start, size) in guest_mem
+            .iter()
+            .map(|m| (m.start_addr().raw_value(), m.len()))
+        {
+            if current_end == start {
+                // This zone is continuous with the previous one.
+                current_end += size;
+            } else {
+                ram_regions.push((current_start, current_end));
+
+                current_start = start;
+                current_end = start + size;
+            }
+        }
+
+        ram_regions.push((current_start, current_end));
+
+        ram_regions
+    };
+
+    let mut mem_reg_property = Vec::new();
+    for region in ram_regions {
+        let mem_size = region.1 - region.0;
+        mem_reg_property.push(region.0);
+        mem_reg_property.push(mem_size);
+    }
+
+    let ram_start = super::layout::RAM_START.raw_value();
+    let memory_node_name = format!("memory@{:x}", ram_start);
+    let memory_node = fdt.begin_node(&memory_node_name)?;
+    fdt.property_string("device_type", "memory")?;
+    fdt.property_array_u64("reg", &mem_reg_property)?;
+    fdt.end_node(memory_node)?;
+
+    Ok(())
+}
+
+fn create_chosen_node(
+    fdt: &mut FdtWriter,
+    cmdline: &str,
+    initrd: &Option<InitramfsConfig>,
+) -> FdtWriterResult<()> {
+    let chosen_node = fdt.begin_node("chosen")?;
+    fdt.property_string("bootargs", cmdline)?;
+
+    if let Some(initrd_config) = initrd {
+        let initrd_start = initrd_config.address.raw_value();
+        let initrd_end = initrd_config.address.raw_value() + initrd_config.size as u64;
+        fdt.property_u64("linux,initrd-start", initrd_start)?;
+        fdt.property_u64("linux,initrd-end", initrd_end)?;
+    }
+
+    fdt.end_node(chosen_node)?;
+
+    Ok(())
+}
+
+fn create_aia_node(fdt: &mut FdtWriter, aia_device: &Arc<Mutex<dyn Vaia>>) -> FdtWriterResult<()> {
+    // IMSIC
+    if aia_device.lock().unwrap().msi_compatible() {
+        use super::layout::IMSIC_START;
+        let imsic_name = format!("imsics@{:x}", IMSIC_START.0);
+        let imsic_node = fdt.begin_node(&imsic_name)?;
+
+        fdt.property_string(
+            "compatible",
+            aia_device.lock().unwrap().imsic_compatibility(),
+        )?;
+        let imsic_reg_prop = aia_device.lock().unwrap().imsic_properties();
+        fdt.property_array_u32("reg", &imsic_reg_prop)?;
+        fdt.property_u32("#interrupt-cells", 0u32)?;
+        fdt.property_null("interrupt-controller")?;
+        fdt.property_null("msi-controller")?;
+        // TODO complete num-ids
+        fdt.property_u32("riscv,num-ids", 2047u32)?;
+        fdt.property_u32("phandle", AIA_IMSIC_PHANDLE)?;
+
+        let mut irq_cells = Vec::new();
+        let num_cpus = aia_device.lock().unwrap().vcpu_count();
+        for i in 0..num_cpus {
+            irq_cells.push(CPU_INTC_BASE_PHANDLE + i);
+            irq_cells.push(S_MODE_EXT_IRQ);
+        }
+        fdt.property_array_u32("interrupts-extended", &irq_cells)?;
+
+        fdt.end_node(imsic_node)?;
+    }
+
+    // APLIC
+    use super::layout::APLIC_START;
+    let aplic_name = format!("aplic@{:x}", APLIC_START.0);
+    let aplic_node = fdt.begin_node(&aplic_name)?;
+
+    fdt.property_string(
+        "compatible",
+        aia_device.lock().unwrap().aplic_compatibility(),
+    )?;
+    let reg_cells = aia_device.lock().unwrap().aplic_properties();
+    fdt.property_array_u32("reg", &reg_cells)?;
+    fdt.property_u32("#interrupt-cells", 2u32)?;
+    fdt.property_null("interrupt-controller")?;
+    // TODO complete num-srcs
+    fdt.property_u32("riscv,num-sources", 96u32)?;
+    fdt.property_u32("phandle", AIA_APLIC_PHANDLE)?;
+    fdt.property_u32("msi-parent", AIA_IMSIC_PHANDLE)?;
+
+    fdt.end_node(aplic_node)?;
+
+    Ok(())
+}
+
+fn create_serial_node<T: DeviceInfoForFdt + Clone + Debug>(
+    fdt: &mut FdtWriter,
+    dev_info: &T,
+) -> FdtWriterResult<()> {
+    let serial_reg_prop = [dev_info.addr(), dev_info.length()];
+    let irq = [dev_info.irq() - IRQ_BASE, IRQ_TYPE_LEVEL_HI];
+
+    let serial_node = fdt.begin_node(&format!("serial@{:x}", dev_info.addr()))?;
+    fdt.property_string("compatible", "ns16550a")?;
+    fdt.property_array_u64("reg", &serial_reg_prop)?;
+    fdt.property_u32("clock-frequency", 3686400)?;
+    fdt.property_u32("interrupt-parent", AIA_APLIC_PHANDLE)?;
+    fdt.property_array_u32("interrupts", &irq)?;
+    fdt.end_node(serial_node)?;
+
+    Ok(())
+}
+
+fn create_devices_node<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::BuildHasher>(
+    fdt: &mut FdtWriter,
+    dev_info: &HashMap<(DeviceType, String), T, S>,
+) -> FdtWriterResult<()> {
+    for ((device_type, _device_id), info) in dev_info {
+        match device_type {
+            DeviceType::Serial => create_serial_node(fdt, info)?,
+            DeviceType::Virtio(_) => unreachable!(),
+        }
+    }
+
+    Ok(())
+}
+
+fn create_pci_nodes(fdt: &mut FdtWriter, pci_device_info: &[PciSpaceInfo]) -> FdtWriterResult<()> {
+    // Add node for PCIe controller.
+    // See Documentation/devicetree/bindings/pci/host-generic-pci.txt in the kernel
+    // and https://elinux.org/Device_Tree_Usage.
+    // In multiple PCI segments setup, each PCI segment needs a PCI node.
+    for pci_device_info_elem in pci_device_info.iter() {
+        // EDK2 requires the PCIe high space above 4G address.
+        // The actual space in CLH follows the RAM. If the RAM space is small, the PCIe high space
+        // could fall below 4G.
+        // Here we cut off PCI device space below 8G in FDT to workaround the EDK2 check.
+        // But the address written in ACPI is not impacted.
+        let (pci_device_base_64bit, pci_device_size_64bit) =
+            if pci_device_info_elem.pci_device_space_start < PCI_HIGH_BASE.raw_value() {
+                (
+                    PCI_HIGH_BASE.raw_value(),
+                    pci_device_info_elem.pci_device_space_size
+                        - (PCI_HIGH_BASE.raw_value() - pci_device_info_elem.pci_device_space_start),
+                )
+            } else {
+                (
+                    pci_device_info_elem.pci_device_space_start,
+                    pci_device_info_elem.pci_device_space_size,
+                )
+            };
+        // There is no specific requirement of the 32bit MMIO range, and
+        // therefore at least we can make these ranges 4K aligned.
+        let pci_device_size_32bit: u64 =
+            MEM_32BIT_DEVICES_SIZE / ((1 << 12) * pci_device_info.len() as u64) * (1 << 12);
+        let pci_device_base_32bit: u64 = MEM_32BIT_DEVICES_START.0
+            + pci_device_size_32bit * pci_device_info_elem.pci_segment_id as u64;
+
+        let ranges = [
+            // io addresses. Since AArch64 will not use IO address,
+            // we can set the same IO address range for every segment.
+            0x1000000,
+            0_u32,
+            0_u32,
+            (MEM_PCI_IO_START.0 >> 32) as u32,
+            MEM_PCI_IO_START.0 as u32,
+            (MEM_PCI_IO_SIZE >> 32) as u32,
+            MEM_PCI_IO_SIZE as u32,
+            // mmio addresses
+            0x2000000,                            // (ss = 10: 32-bit memory space)
+            (pci_device_base_32bit >> 32) as u32, // PCI address
+            pci_device_base_32bit as u32,
+            (pci_device_base_32bit >> 32) as u32, // CPU address
+            pci_device_base_32bit as u32,
+            (pci_device_size_32bit >> 32) as u32, // size
+            pci_device_size_32bit as u32,
+            // device addresses
+            0x3000000,                            // (ss = 11: 64-bit memory space)
+            (pci_device_base_64bit >> 32) as u32, // PCI address
+            pci_device_base_64bit as u32,
+            (pci_device_base_64bit >> 32) as u32, // CPU address
+            pci_device_base_64bit as u32,
+            (pci_device_size_64bit >> 32) as u32, // size
+            pci_device_size_64bit as u32,
+        ];
+        let bus_range = [0, 0]; // Only bus 0
+        let reg = [
+            pci_device_info_elem.mmio_config_address,
+            PCI_MMIO_CONFIG_SIZE_PER_SEGMENT,
+        ];
+        // See kernel document Documentation/devicetree/bindings/pci/pci-msi.txt
+        let msi_map = [
+            // rid-base: A single cell describing the first RID matched by the entry.
+            0x0,
+            // msi-controller: A single phandle to an MSI controller.
+            AIA_IMSIC_PHANDLE,
+            // msi-base: An msi-specifier describing the msi-specifier produced for the
+            // first RID matched by the entry.
+            (pci_device_info_elem.pci_segment_id as u32) << 8,
+            // length: A single cell describing how many consecutive RIDs are matched
+            // following the rid-base.
+            0x100,
+        ];
+
+        let pci_node_name = format!("pci@{:x}", pci_device_info_elem.mmio_config_address);
+        let pci_node = fdt.begin_node(&pci_node_name)?;
+
+        fdt.property_string("compatible", "pci-host-ecam-generic")?;
+        fdt.property_string("device_type", "pci")?;
+        fdt.property_array_u32("ranges", &ranges)?;
+        fdt.property_array_u32("bus-range", &bus_range)?;
+        fdt.property_u32(
+            "linux,pci-domain",
+            pci_device_info_elem.pci_segment_id as u32,
+        )?;
+        fdt.property_u32("#address-cells", 3)?;
+        fdt.property_u32("#size-cells", 2)?;
+        fdt.property_array_u64("reg", &reg)?;
+        fdt.property_u32("#interrupt-cells", 1)?;
+        fdt.property_null("interrupt-map")?;
+        fdt.property_null("interrupt-map-mask")?;
+        fdt.property_null("dma-coherent")?;
+        fdt.property_array_u32("msi-map", &msi_map)?;
+        fdt.property_u32("msi-parent", AIA_IMSIC_PHANDLE)?;
+
+        fdt.end_node(pci_node)?;
+    }
+
+    Ok(())
+}
+
+// Parse the DTB binary and print for debugging
+pub fn print_fdt(dtb: &[u8]) {
+    match fdt_parser::Fdt::new(dtb) {
+        Ok(fdt) => {
+            if let Some(root) = fdt.find_node("/") {
+                debug!("Printing the FDT:");
+                print_node(root, 0);
+            } else {
+                debug!("Failed to find root node in FDT for debugging.");
+            }
+        }
+        Err(_) => debug!("Failed to parse FDT for debugging."),
+    }
+}
+
+fn print_node(node: fdt_parser::node::FdtNode<'_, '_>, n_spaces: usize) {
+    debug!("{:indent$}{}/", "", node.name, indent = n_spaces);
+    for property in node.properties() {
+        let name = property.name;
+
+        // If the property is 'compatible', its value requires special handling.
+        // The u8 array could contain multiple null-terminated strings.
+        // We copy the original array and simply replace all 'null' characters with spaces.
+        let value = if name == "compatible" {
+            let mut compatible = vec![0u8; 256];
+            let handled_value = property
+                .value
+                .iter()
+                .map(|&c| if c == 0 { b' ' } else { c })
+                .collect::<Vec<_>>();
+            let len = cmp::min(255, handled_value.len());
+            compatible[..len].copy_from_slice(&handled_value[..len]);
+            compatible[..(len + 1)].to_vec()
+        } else {
+            property.value.to_vec()
+        };
+        let value = &value;
+
+        // Now the value can be either:
+        //   - A null-terminated C string, or
+        //   - Binary data
+        // We follow a very simple logic to present the value:
+        //   - At first, try to convert it to CStr and print,
+        //   - If failed, print it as u32 array.
+        let value_result = match CStr::from_bytes_with_nul(value) {
+            Ok(value_cstr) => match value_cstr.to_str() {
+                Ok(value_str) => Some(value_str),
+                Err(_e) => None,
+            },
+            Err(_e) => None,
+        };
+
+        if let Some(value_str) = value_result {
+            debug!(
+                "{:indent$}{} : {:#?}",
+                "",
+                name,
+                value_str,
+                indent = (n_spaces + 2)
+            );
+        } else {
+            let mut array = Vec::with_capacity(256);
+            array.resize(value.len() / 4, 0u32);
+            BigEndian::read_u32_into(value, &mut array);
+            debug!(
+                "{:indent$}{} : {:X?}",
+                "",
+                name,
+                array,
+                indent = (n_spaces + 2)
+            );
+        };
+    }
+
+    // Print children nodes if there is any
+    for child in node.children() {
+        print_node(child, n_spaces + 2);
+    }
+}

arch/src/riscv64/layout.rs

@@ -0,0 +1,105 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+// Copyright 2020 Arm Limited (or its affiliates). All rights reserved.
+// Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+// SPDX-License-Identifier: Apache-2.0
+
+//
+// Memory layout of RISC-V 64-bit guest:
+//
+// Physical  +---------------------------------------------------------------+
+// address   |                                                               |
+// end       |                                                               |
+//           ~                   ~                       ~                   ~
+//           |                                                               |
+//           |                    Highmem PCI MMIO space                     |
+//           |                                                               |
+// RAM end   +---------------------------------------------------------------+
+// (dynamic, |                                                               |
+// including |                                                               |
+// hotplug   ~                   ~                       ~                   ~
+// memory)   |                                                               |
+//           |                             DRAM                              |
+//           |                                                               |
+//           |                                                               |
+//           |                                                               |
+//           |                                                               |
+//    1 GB   +---------------------------------------------------------------+
+//           |                                                               |
+//           |                      PCI MMCONFIG space                       |
+//           |                                                               |
+//  768 MB   +---------------------------------------------------------------+
+//           |                                                               |
+//           |                                                               |
+//           |                        PCI MMIO space                         |
+//           |                                                               |
+//  256 MB   +---------------------------------------------------------------|
+//           |                                                               |
+//           |                     Legacy devices space                      |
+//           |                                                               |
+//  128 MB   +---------------------------------------------------------------|
+//           |                                                               |
+//           |                            IMSICs                             |
+//           |                                                               |
+//   64 MB   +---------------------------------------------------------------+
+//           |                                                               |
+//           |                            APLICs                             |
+//           |                                                               |
+//    0 GB   +---------------------------------------------------------------+
+//
+//
+
+use vm_memory::GuestAddress;
+
+/// AIA related devices
+/// See https://elixir.bootlin.com/linux/v6.10/source/arch/riscv/include/uapi/asm/kvm.h
+/// 0x0 ~ 0x0400_0000 (64 MiB) resides APLICs
+pub const APLIC_START: GuestAddress = GuestAddress(0);
+pub const APLIC_SIZE: u64 = 0x4000;
+
+/// 0x0400_0000 ~ 0x0800_0000 (64 MiB) resides IMSICs
+pub const IMSIC_START: GuestAddress = GuestAddress(0x0400_0000);
+pub const IMSIC_SIZE: u64 = 0x1000;
+
+/// Below this address will reside the AIA, above this address will reside the MMIO devices.
+const MAPPED_IO_START: GuestAddress = GuestAddress(0x0800_0000);
+
+/// Space 0x0800_0000 ~ 0x1000_0000 is reserved for legacy devices.
+pub const LEGACY_SERIAL_MAPPED_IO_START: GuestAddress = MAPPED_IO_START;
+
+/// Space 0x0905_0000 ~ 0x0906_0000 is reserved for pcie io address
+pub const MEM_PCI_IO_START: GuestAddress = GuestAddress(0x0905_0000);
+pub const MEM_PCI_IO_SIZE: u64 = 0x1_0000;
+
+/// Starting from 0x1000_0000 (256MiB) to 0x3000_0000 (768MiB) is used for PCIE MMIO
+pub const MEM_32BIT_DEVICES_START: GuestAddress = GuestAddress(0x1000_0000);
+pub const MEM_32BIT_DEVICES_SIZE: u64 = 0x2000_0000;
+
+/// PCI MMCONFIG space (start: after the device space at 768MiB, length: 256MiB)
+pub const PCI_MMCONFIG_START: GuestAddress = GuestAddress(0x3000_0000);
+pub const PCI_MMCONFIG_SIZE: u64 = 256 << 20;
+// One bus with potentially 256 devices (32 slots x 8 functions).
+pub const PCI_MMIO_CONFIG_SIZE_PER_SEGMENT: u64 = 4096 * 256;
+
+/// Start of RAM.
+pub const RAM_START: GuestAddress = GuestAddress(0x4000_0000);
+
+/// Kernel command line maximum size on RISC-V.
+/// See https://elixir.bootlin.com/linux/v6.10/source/arch/riscv/include/uapi/asm/setup.h
+pub const CMDLINE_MAX_SIZE: usize = 1024;
+
+/// FDT is at the beginning of RAM.
+pub const FDT_START: GuestAddress = RAM_START;
+pub const FDT_MAX_SIZE: u64 = 0x1_0000;
+
+/// Kernel start after FDT
+pub const KERNEL_START: GuestAddress = GuestAddress(RAM_START.0 + FDT_MAX_SIZE);
+
+/// Pci high memory base
+pub const PCI_HIGH_BASE: GuestAddress = GuestAddress(0x2_0000_0000);
+
+/// First usable interrupt on riscv64
+pub const IRQ_BASE: u32 = 0;
+
+// As per https://elixir.bootlin.com/linux/v6.10/source/arch/riscv/include/asm/kvm_host.h#L31
+/// Number of supported interrupts
+pub const IRQ_NUM: u32 = 1023;

arch/src/riscv64/mod.rs

@@ -0,0 +1,173 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+// Copyright 2020 Arm Limited (or its affiliates). All rights reserved.
+// Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+// SPDX-License-Identifier: Apache-2.0
+
+/// Module for the flattened device tree.
+pub mod fdt;
+/// Layout for this riscv64 system.
+pub mod layout;
+
+use std::collections::HashMap;
+use std::fmt::Debug;
+use std::sync::{Arc, Mutex};
+
+use hypervisor::arch::riscv64::aia::Vaia;
+use log::{log_enabled, Level};
+use thiserror::Error;
+use vm_memory::{Address, GuestAddress, GuestMemory, GuestMemoryAtomic};
+
+pub use self::fdt::DeviceInfoForFdt;
+use crate::{DeviceType, GuestMemoryMmap, PciSpaceInfo, RegionType};
+
+pub const _NSIG: i32 = 65;
+
+/// Errors thrown while configuring riscv64 system.
+#[derive(Debug, Error)]
+pub enum Error {
+    /// Failed to create a FDT.
+    #[error("Failed to create a FDT")]
+    SetupFdt,
+
+    /// Failed to write FDT to memory.
+    #[error("Failed to write FDT to memory: {0}")]
+    WriteFdtToMemory(fdt::Error),
+
+    /// Failed to create a AIA.
+    #[error("Failed to create a AIA")]
+    SetupAia,
+
+    /// Failed to compute the initramfs address.
+    #[error("Failed to compute the initramfs address")]
+    InitramfsAddress,
+
+    /// Error configuring the general purpose registers
+    #[error("Error configuring the general purpose registers: {0}")]
+    RegsConfiguration(hypervisor::HypervisorCpuError),
+}
+
+impl From<Error> for super::Error {
+    fn from(e: Error) -> super::Error {
+        super::Error::PlatformSpecific(e)
+    }
+}
+
+#[derive(Debug, Copy, Clone)]
+/// Specifies the entry point address where the guest must start
+/// executing code.
+pub struct EntryPoint {
+    /// Address in guest memory where the guest must start execution
+    pub entry_addr: GuestAddress,
+}
+
+/// Configure the specified VCPU, and return its MPIDR.
+pub fn configure_vcpu(
+    vcpu: &Arc<dyn hypervisor::Vcpu>,
+    id: u8,
+    boot_setup: Option<(EntryPoint, &GuestMemoryAtomic<GuestMemoryMmap>)>,
+) -> super::Result<()> {
+    if let Some((kernel_entry_point, _guest_memory)) = boot_setup {
+        vcpu.setup_regs(
+            id,
+            kernel_entry_point.entry_addr.raw_value(),
+            layout::FDT_START.raw_value(),
+        )
+        .map_err(Error::RegsConfiguration)?;
+    }
+
+    Ok(())
+}
+
+pub fn arch_memory_regions() -> Vec<(GuestAddress, usize, RegionType)> {
+    vec![
+        // 0 MiB ~ 256 MiB: AIA and legacy devices
+        (
+            GuestAddress(0),
+            layout::MEM_32BIT_DEVICES_START.0 as usize,
+            RegionType::Reserved,
+        ),
+        // 256 MiB ~ 768 MiB: MMIO space
+        (
+            layout::MEM_32BIT_DEVICES_START,
+            layout::MEM_32BIT_DEVICES_SIZE as usize,
+            RegionType::SubRegion,
+        ),
+        // 768 MiB ~ 1 GiB: reserved. The leading 256M for PCIe MMCONFIG space
+        (
+            layout::PCI_MMCONFIG_START,
+            layout::PCI_MMCONFIG_SIZE as usize,
+            RegionType::Reserved,
+        ),
+        // 1GiB ~ inf: RAM
+        (layout::RAM_START, usize::MAX, RegionType::Ram),
+    ]
+}
+
+/// Configures the system and should be called once per vm before starting vcpu threads.
+#[allow(clippy::too_many_arguments)]
+pub fn configure_system<T: DeviceInfoForFdt + Clone + Debug, S: ::std::hash::BuildHasher>(
+    guest_mem: &GuestMemoryMmap,
+    cmdline: &str,
+    num_vcpu: u32,
+    device_info: &HashMap<(DeviceType, String), T, S>,
+    initrd: &Option<super::InitramfsConfig>,
+    pci_space_info: &[PciSpaceInfo],
+    aia_device: &Arc<Mutex<dyn Vaia>>,
+) -> super::Result<()> {
+    let fdt_final = fdt::create_fdt(
+        guest_mem,
+        cmdline,
+        num_vcpu,
+        device_info,
+        aia_device,
+        initrd,
+        pci_space_info,
+    )
+    .map_err(|_| Error::SetupFdt)?;
+
+    if log_enabled!(Level::Debug) {
+        fdt::print_fdt(&fdt_final);
+    }
+
+    fdt::write_fdt_to_memory(fdt_final, guest_mem).map_err(Error::WriteFdtToMemory)?;
+
+    Ok(())
+}
+
+/// Returns the memory address where the initramfs could be loaded.
+pub fn initramfs_load_addr(
+    guest_mem: &GuestMemoryMmap,
+    initramfs_size: usize,
+) -> super::Result<u64> {
+    let round_to_pagesize = |size| (size + (super::PAGE_SIZE - 1)) & !(super::PAGE_SIZE - 1);
+    match guest_mem
+        .last_addr()
+        .checked_sub(round_to_pagesize(initramfs_size) as u64 - 1)
+    {
+        Some(offset) => {
+            if guest_mem.address_in_range(offset) {
+                Ok(offset.raw_value())
+            } else {
+                Err(super::Error::PlatformSpecific(Error::InitramfsAddress))
+            }
+        }
+        None => Err(super::Error::PlatformSpecific(Error::InitramfsAddress)),
+    }
+}
+
+pub fn get_host_cpu_phys_bits(_hypervisor: &Arc<dyn hypervisor::Hypervisor>) -> u8 {
+    40
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_arch_memory_regions_dram() {
+        let regions = arch_memory_regions();
+        assert_eq!(4, regions.len());
+        assert_eq!(layout::RAM_START, regions[3].0);
+        assert_eq!(RegionType::Ram, regions[3].2);
+    }
+}

block/Cargo.toml

@@ -12,7 +12,7 @@ io_uring = ["dep:io-uring"]
 byteorder = "1.5.0"
 crc-any = "2.4.4"
 io-uring = { version = "0.6.4", optional = true }
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 remain = "0.2.14"
 serde = { version = "1.0.208", features = ["derive"] }

devices/Cargo.toml

@@ -6,13 +6,13 @@ version = "0.1.0"
 
 [dependencies]
 acpi_tables = { workspace = true }
-anyhow = "1.0.87"
+anyhow = "1.0.94"
 arch = { path = "../arch" }
 bitflags = "2.6.0"
 byteorder = "1.5.0"
 event_monitor = { path = "../event_monitor" }
 hypervisor = { path = "../hypervisor" }
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 num_enum = "0.7.2"
 pci = { path = "../pci" }
@@ -29,9 +29,10 @@ vm-memory = { workspace = true, features = [
 vm-migration = { path = "../vm-migration" }
 vmm-sys-util = { workspace = true }
 
-[target.'cfg(target_arch = "aarch64")'.dependencies]
+[target.'cfg(any(target_arch = "aarch64", target_arch = "riscv64"))'.dependencies]
 arch = { path = "../arch" }
 
 [features]
 default = []
+kvm = ["arch/kvm"]
 pvmemcontrol = []

devices/src/aia.rs

@@ -0,0 +1,144 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+// Copyright 2020, ARM Limited.
+//
+// SPDX-License-Identifier: Apache-2.0 AND BSD-3-Clause
+
+use super::interrupt_controller::{Error, InterruptController};
+extern crate arch;
+use std::result;
+use std::sync::{Arc, Mutex};
+
+use arch::layout;
+use hypervisor::arch::riscv64::aia::{Vaia, VaiaConfig};
+use hypervisor::{AiaState, CpuState};
+use vm_device::interrupt::{
+    InterruptIndex, InterruptManager, InterruptSourceConfig, InterruptSourceGroup,
+    LegacyIrqSourceConfig, MsiIrqGroupConfig,
+};
+use vm_memory::address::Address;
+use vm_migration::{Migratable, Pausable, Snapshottable, Transportable};
+use vmm_sys_util::eventfd::EventFd;
+
+type Result<T> = result::Result<T, Error>;
+
+// Reserve 32 IRQs for legacy devices.
+pub const IRQ_LEGACY_BASE: usize = layout::IRQ_BASE as usize;
+pub const IRQ_LEGACY_COUNT: usize = 32;
+// TODO: AIA snapshotting is not yet completed.
+pub const _AIA_SNAPSHOT_ID: &str = "";
+
+// Aia (Advance Interrupt Architecture) struct provides all the functionality of a
+// AIA device. It wraps a hypervisor-emulated AIA device (Vaia) provided by the
+// `hypervisor` crate.
+// Aia struct also implements InterruptController to provide interrupt delivery
+// service.
+pub struct Aia {
+    interrupt_source_group: Arc<dyn InterruptSourceGroup>,
+    // The hypervisor agnostic virtual AIA
+    vaia: Arc<Mutex<dyn Vaia>>,
+}
+
+impl Aia {
+    pub fn new(
+        vcpu_count: u8,
+        interrupt_manager: Arc<dyn InterruptManager<GroupConfig = MsiIrqGroupConfig>>,
+        vm: Arc<dyn hypervisor::Vm>,
+    ) -> Result<Aia> {
+        let interrupt_source_group = interrupt_manager
+            .create_group(MsiIrqGroupConfig {
+                base: IRQ_LEGACY_BASE as InterruptIndex,
+                count: IRQ_LEGACY_COUNT as InterruptIndex,
+            })
+            .map_err(Error::CreateInterruptSourceGroup)?;
+
+        let vaia = vm
+            .create_vaia(Aia::create_default_config(vcpu_count as u64))
+            .map_err(Error::CreateAia)?;
+
+        let aia = Aia {
+            interrupt_source_group,
+            vaia,
+        };
+        aia.enable()?;
+
+        Ok(aia)
+    }
+
+    pub fn restore_vaia(
+        &mut self,
+        state: Option<AiaState>,
+        _saved_vcpu_states: &[CpuState],
+    ) -> Result<()> {
+        self.vaia
+            .clone()
+            .lock()
+            .unwrap()
+            .set_state(&state.unwrap())
+            .map_err(Error::RestoreAia)
+    }
+
+    fn enable(&self) -> Result<()> {
+        // Set irqfd for legacy interrupts
+        self.interrupt_source_group
+            .enable()
+            .map_err(Error::EnableInterrupt)?;
+
+        // Set irq_routing for legacy interrupts.
+        //   irqchip: Hardcode to 0 as we support only 1 APLIC
+        //   pin: Use irq number as pin
+        for i in IRQ_LEGACY_BASE..(IRQ_LEGACY_BASE + IRQ_LEGACY_COUNT) {
+            let config = LegacyIrqSourceConfig {
+                irqchip: 0,
+                pin: (i - IRQ_LEGACY_BASE) as u32,
+            };
+            self.interrupt_source_group
+                .update(
+                    i as InterruptIndex,
+                    InterruptSourceConfig::LegacyIrq(config),
+                    false,
+                    false,
+                )
+                .map_err(Error::EnableInterrupt)?;
+        }
+
+        self.interrupt_source_group
+            .set_gsi()
+            .map_err(Error::EnableInterrupt)?;
+        Ok(())
+    }
+
+    /// Default config implied by arch::layout
+    pub fn create_default_config(vcpu_count: u64) -> VaiaConfig {
+        VaiaConfig {
+            vcpu_count: vcpu_count as u32,
+            aplic_addr: layout::APLIC_START.raw_value(),
+            imsic_addr: layout::IMSIC_START.raw_value(),
+            nr_irqs: layout::IRQ_NUM,
+        }
+    }
+
+    pub fn get_vaia(&mut self) -> Result<Arc<Mutex<dyn Vaia>>> {
+        Ok(self.vaia.clone())
+    }
+}
+
+impl InterruptController for Aia {
+    // This should be called anytime an interrupt needs to be injected into the
+    // running guest.
+    fn service_irq(&mut self, irq: usize) -> Result<()> {
+        self.interrupt_source_group
+            .trigger(irq as InterruptIndex)
+            .map_err(Error::TriggerInterrupt)?;
+
+        Ok(())
+    }
+
+    fn notifier(&self, irq: usize) -> Option<EventFd> {
+        self.interrupt_source_group.notifier(irq as InterruptIndex)
+    }
+}
+
+impl Snapshottable for Aia {}
+impl Pausable for Aia {}
+impl Transportable for Aia {}
+impl Migratable for Aia {}

devices/src/interrupt_controller.rs

@@ -1,3 +1,4 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
 // Copyright 2020, ARM Limited.
 //
 // SPDX-License-Identifier: Apache-2.0 AND BSD-3-Clause
@@ -41,6 +42,14 @@ pub enum Error {
     /// Failed restoring GIC device.
     #[error("Failed restoring GIC device: {0}")]
     RestoreGic(hypervisor::arch::aarch64::gic::Error),
+    #[cfg(target_arch = "riscv64")]
+    /// Failed creating AIA device.
+    #[error("Failed creating AIA device: {0}")]
+    CreateAia(hypervisor::HypervisorVmError),
+    #[cfg(target_arch = "riscv64")]
+    /// Failed restoring AIA device.
+    #[error("Failed restoring AIA device: {0}")]
+    RestoreAia(hypervisor::arch::riscv64::aia::Error),
 }
 
 type Result<T> = result::Result<T, Error>;
@@ -67,7 +76,7 @@ pub struct MsiMessage {
 // Introduce trait InterruptController to uniform the interrupt
 // service provided for devices.
 // Device manager uses this trait without caring whether it is a
-// IOAPIC (X86) or GIC (Arm).
+// IOAPIC (X86), GIC (Arm) or AIA (RISC-V).
 pub trait InterruptController: Send {
     fn service_irq(&mut self, irq: usize) -> Result<()>;
     #[cfg(target_arch = "x86_64")]

devices/src/lib.rs

@@ -15,6 +15,8 @@ extern crate event_monitor;
 extern crate log;
 
 pub mod acpi;
+#[cfg(target_arch = "riscv64")]
+pub mod aia;
 #[cfg(target_arch = "x86_64")]
 pub mod debug_console;
 #[cfg(target_arch = "aarch64")]
@@ -26,6 +28,8 @@ pub mod legacy;
 #[cfg(feature = "pvmemcontrol")]
 pub mod pvmemcontrol;
 pub mod pvpanic;
+// TODO: TPM is not yet supported
+#[cfg(not(target_arch = "riscv64"))]
 pub mod tpm;
 
 pub use self::acpi::{AcpiGedDevice, AcpiPmTimerDevice, AcpiShutdownDevice};

docs/api.md

@@ -50,17 +50,12 @@ for more details about the API payloads and responses.
 
 #### REST API Location and availability
 
-The REST API is available as soon as the Cloud Hypervisor binary is started,
-through a local UNIX socket.
-By default, it is located at `/run/user/{user ID}/cloud-hypervisor.{Cloud Hypervisor PID}`.
-For example, if you launched Cloud Hypervisor as user ID 1000 and its PID is
-123456, the Cloud Hypervisor REST API will be available at `/run/user/1000/cloud-hypervisor.123456`.
-
-The REST API default URL can be overridden through the Cloud Hypervisor
-option `--api-socket`:
+The REST API, if enabled, is available as soon as the Cloud Hypervisor binary is started,
+through either a local UNIX socket as given in the Cloud Hypervisor option `--api-socket path=...`
+or a fd with `--api-socket fd=...`.
 
 ```
-$ ./target/debug/cloud-hypervisor --api-socket /tmp/cloud-hypervisor.sock
+$ ./target/debug/cloud-hypervisor --api-socket path=/tmp/cloud-hypervisor.sock
 Cloud Hypervisor Guest
     API server: /tmp/cloud-hypervisor.sock
     vCPUs: 1

docs/live_migration.md

@@ -1,16 +1,19 @@
 # Live Migration
 
-This document gives two examples of how to use the live migration
-support in Cloud Hypervisor:
+This document gives examples of how to use the live migration support
+in Cloud Hypervisor:
 
-1. local migration - migrating between two VMs running on the same
-   machine;
-1. nested-vm migration - migrating between two nested VMs whose host VMs
-   are running on the same machine.
+1. local migration - migrating a VM from one Cloud Hypervisor instance to another on the same machine;
+1. remote migration - migrating a VM between two machines;
+
+> :warning: These examples place sockets /tmp. This is done for
+> simplicity and should not be done in production.
 
 ## Local Migration (Suitable for Live Upgrade of VMM)
+
 Launch the source VM (on the host machine):
-```bash
+
+```console
 $ target/release/cloud-hypervisor
     --kernel ~/workloads/vmlinux \
     --disk path=~/workloads/focal.raw \
@@ -20,17 +23,20 @@ $ target/release/cloud-hypervisor
 ```
 
 Launch the destination VM from the same directory (on the host machine):
-```bash
+
+```console
 $ target/release/cloud-hypervisor --api-socket=/tmp/api2
 ```
 
 Get ready for receiving migration for the destination VM (on the host machine):
-```bash
+
+```console
 $ target/release/ch-remote --api-socket=/tmp/api2 receive-migration unix:/tmp/sock
 ```
 
 Start to send migration for the source VM (on the host machine):
-```bash
+
+```console
 $ target/release/ch-remote --api-socket=/tmp/api1 send-migration --local unix:/tmp/sock
 ```
 
@@ -38,102 +44,139 @@ When the above commands completed, the source VM should be successfully
 migrated to the destination VM. Now the destination VM is running while
 the source VM is terminated gracefully.
 
-## Nested-VM Migration
+## Remote Migration
 
-Launch VM 1 (on the host machine) with an extra virtio-blk device for
-exposing a guest image for the nested source VM:
-> Note: the example below also attached an additional virtio-blk device
-> with a dummy image for testing purpose (which is optional).
-```bash
-$ head -c 1M < /dev/urandom > tmp.img # create a dummy image for testing
-$ sudo /target/release/cloud-hypervisor \
+In this example, we will migrate a VM from one machine (`src`) to
+another (`dst`) across the network. To keep it simple, we will use a
+minimal VM setup without storage.
+
+### Preparation
+
+Make sure that `src` and `dst` can reach each other via the
+network. You should be able to ping each machine. Also each machine
+should have an open TCP port.
+
+You will need a kernel and initramfs for a minimal Linux system. For
+this example, we will use the Debian netboot image.
+
+Place the kernel and initramfs into the _same directory_ on both
+machines. This is important for the migration to succeed. We will use
+`/var/images`:
+
+```console
+src $ export DEBIAN=https://ftp.debian.org/debian/dists/stable/main/installer-amd64/current/images/netboot/debian-installer/amd64
+src $ mkdir -p /var/images
+src $ curl $DEBIAN/linux > /var/images/linux
+src $ curl $DEBIAN/initrd.gz > /var/images/initrd
+```
+
+Repeat the above steps on the destination host.
+
+### Unix Socket Migration
+
+If Unix socket is selected for migration, we can tunnel traffic through "socat".
+
+#### Starting the Receiver VM
+
+On the receiver side, we prepare an empty VM:
+
+```console
+dst $ cloud-hypervisor --api-socket /tmp/api
+```
+
+In a different terminal, configure the VM as a migration target:
+
+```console
+dst $ ch-remote --api-socket=/tmp/api receive-migration unix:/tmp/sock
+```
+
+In yet another terminal, forward TCP connections to the Unix domain socket:
+
+```console
+dst $ socat TCP-LISTEN:{port},reuseaddr UNIX-CLIENT:/tmp/sock
+```
+
+#### Starting the Sender VM
+
+Let's start the VM on the source machine:
+
+```console
+src $ cloud-hypervisor \
         --serial tty --console off \
-        --cpus boot=1 --memory size=512M \
-        --kernel vmlinux \
-        --cmdline "root=/dev/vda1 console=ttyS0"  \
-        --disk path=focal-1.raw path=focal-nested.raw path=tmp.img\
-        --net ip=192.168.101.1
+        --cpus boot=2 --memory size=4G \
+        --kernel /var/images/linux \
+        --initramfs /var/images/initrd \
+        --cmdline "console=ttyS0" \
+        --api-socket /tmp/api
 ```
 
-Launch VM 2 (on the host machine) with an extra virtio-blk device for
-exposing the same guest image for the nested destination VM:
-```bash
-$ sudo /target/release/cloud-hypervisor \
+After a few seconds the VM should be up and you can interact with it.
+
+#### Performing the Migration
+
+First, we start `socat`:
+
+```console
+src $ socat UNIX-LISTEN:/tmp/sock,reuseaddr TCP:{dst}:{port}
+```
+
+> Replace {dst}:{port} with the actual IP address and port of your destination host.
+
+Then we kick-off the migration itself:
+
+```console
+src $ ch-remote --api-socket=/tmp/api send-migration unix:/tmp/sock
+```
+
+When the above commands completed, the VM should be successfully
+migrated to the destination machine without interrupting the workload.
+
+### TCP Socket Migration
+
+If TCP socket is selected for migration, we need to consider migrating
+in a trusted network.
+
+#### Starting the Receiver VM
+
+On the receiver side, we prepare an empty VM:
+
+```console
+dst $ cloud-hypervisor --api-socket /tmp/api
+```
+
+In a different terminal, prepare to receive the migration:
+
+```console
+dst $ ch-remote --api-socket=/tmp/api receive-migration tcp:0.0.0.0:{port}
+```
+
+#### Starting the Sender VM
+
+Let's start the VM on the source machine:
+
+```console
+src $ cloud-hypervisor \
         --serial tty --console off \
-        --cpus boot=1 --memory size=512M \
-        --kernel vmlinux \
-        --cmdline "root=/dev/vda1 console=ttyS0"  \
-        --disk path=focal-2.raw path=focal-nested.raw path=tmp.img\
-        --net ip=192.168.102.1
+        --cpus boot=2 --memory size=4G \
+        --kernel /var/images/linux \
+        --initramfs /var/images/initrd \
+        --cmdline "console=ttyS0" \
+        --api-socket /tmp/api
 ```
 
-Launch the nested source VM (inside the guest OS of the VM 1) :
-```bash
-vm-1:~$ sudo ./cloud-hypervisor \
-        --serial tty --console off \
-        --memory size=128M \
-        --kernel vmlinux \
-        --cmdline "console=ttyS0 root=/dev/vda1" \
-        --disk path=/dev/vdb path=/dev/vdc \
-        --api-socket=/tmp/api1 \
-        --net ip=192.168.100.1
-vm-1:~$ # setup the guest network if needed
-vm-1:~$ sudo ip addr add 192.168.101.2/24 dev ens4
-vm-1:~$ sudo ip link set up dev ens4
-vm-1:~$ sudo ip r add default via 192.168.101.1
-```
-Optional: Run the guest workload below (on the guest OS of the nested source VM),
-which performs intensive virtio-blk operations. Now the console of the nested
-source VM should repeatedly print `"equal"`, and our goal is migrating
-this VM and the running workload without interruption.
-```bash
-#/bin/bash
+After a few seconds the VM should be up and you can interact with it.
 
-# On the guest OS of the nested source VM
+#### Performing the Migration
 
-input="/dev/vdb"
-result=$(md5sum $input)
-tmp=$(md5sum $input)
+Initiate the Migration over TCP:
 
-while  [[ "$result" == "$tmp" ]]
-do
-    echo "equal"
-    tmp=$(md5sum $input)
-done
-
-echo "not equal"
-echo "result = $result"
-echo "tmp = $tmp"
+```console
+src $ ch-remote --api-socket=/tmp/api send-migration  tcp:{dst}:{port}
 ```
 
-Launch the nested destination VM (inside the guest OS of the VM 2):
-```bash
-vm-2:~$ sudo ./cloud-hypervisor --api-socket=/tmp/api2
-vm-2:~$ # setup the guest network with the following commands if needed
-vm-2:~$ sudo ip addr add 192.168.102.2/24 dev ens4
-vm-2:~$ sudo ip link set up dev ens4
-vm-2:~$ sudo ip r add default via 192.168.102.1
-vm-2:~$ ping 192.168.101.2 # This should succeed
-```
-> Note: If the above ping failed, please check the iptables rule on the
-> host machine, e.g. whether the policy for the `FORWARD` chain is set
-> to `DROP` (which is the default setting configured by Docker).
+> Replace {dst}:{port} with the actual IP address and port of your destination host.
 
-Get ready for receiving migration for the nested destination VM (inside
-the guest OS of the VM 2):
-```bash
-vm-2:~$ sudo ./ch-remote --api-socket=/tmp/api2 receive-migration unix:/tmp/sock2
-vm-2:~$ sudo socat TCP-LISTEN:6000,reuseaddr UNIX-CLIENT:/tmp/sock2
-```
-
-Start to send migration for the nested source VM (inside the guest OS of
-the VM 1):
-```bash
-vm-1:~$ sudo socat UNIX-LISTEN:/tmp/sock1,reuseaddr TCP:192.168.102.2:6000
-vm-1:~$ sudo ./ch-remote --api-socket=/tmp/api1 send-migration unix:/tmp/sock1
-```
-
-When the above commands completed, the source VM should be successfully
-migrated to the destination VM without interrupting our testing guest
-workload. Now the destination VM is running the testing guest workload
-while the source VM is terminated gracefully.
+After completing the above commands, the source VM will be migrated to
+the destination host and continue running there. The source VM instance
+will terminate normally. All ongoing processes and connections within
+the VM should remain intact after the migration.

event_monitor/Cargo.toml

@@ -5,8 +5,8 @@ name = "event_monitor"
 version = "0.1.0"
 
 [dependencies]
-flume = "0.11.0"
-libc = "0.2.158"
+flume = "0.11.1"
+libc = "0.2.167"
 once_cell = "1.20.2"
 serde = { version = "1.0.208", features = ["derive", "rc"] }
 serde_json = "1.0.120"

fuzz/Cargo.lock

@@ -5,7 +5,7 @@ version = 3
 [[package]]
 name = "acpi_tables"
 version = "0.1.0"
-source = "git+https://github.com/rust-vmm/acpi_tables?branch=main#ca1a473fe73cdd8eb49c1449faad7aaac06f32c2"
+source = "git+https://github.com/rust-vmm/acpi_tables?branch=main#849d5950196f66dd10f2b2606d8fe8c7cb39ec24"
 dependencies = [
  "zerocopy",
 ]
@@ -361,9 +361,9 @@ checksum = "784a4df722dc6267a04af36895398f59d21d07dce47232adf31ec0ff2fa45e67"
 
 [[package]]
 name = "flume"
-version = "0.11.0"
+version = "0.11.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "55ac459de2512911e4b674ce33cf20befaba382d05b62b008afc1c8b57cbf181"
+checksum = "da0e4dd2a88388a1f4ccc7c9ce104604dab68d9f408dc34cd45823d5a9069095"
 dependencies = [
  "futures-core",
  "futures-sink",
@@ -500,9 +500,9 @@ dependencies = [
 
 [[package]]
 name = "kvm-bindings"
-version = "0.9.1"
+version = "0.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2efe3f1a4437bffe000e6297a593b98184213cd27486776c335f95ab53d48e3a"
+checksum = "fa4933174d0cc4b77b958578cd45784071cc5ae212c2d78fbd755aaaa6dfa71a"
 dependencies = [
  "serde",
  "vmm-sys-util",
@@ -511,9 +511,9 @@ dependencies = [
 
 [[package]]
 name = "kvm-ioctls"
-version = "0.18.0"
+version = "0.19.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "92c2176b91f68903b54ac8c6185bada7d607ca6110998976ff15c032f88a7d39"
+checksum = "337d1afa126368bbd6a5c328048f71a69a737e9afe7e436b392a8f8d770c9171"
 dependencies = [
  "bitflags 2.6.0",
  "kvm-bindings",
@@ -551,9 +551,9 @@ dependencies = [
 
 [[package]]
 name = "linux-loader"
-version = "0.12.0"
+version = "0.13.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3d379d0089d0fbf4161c35a4fdfd76125923f1a93632c49195f5372b4c0b1472"
+checksum = "870c3814345f050991f99869417779f6062542bcf4ed81db7a1b926ad1306638"
 dependencies = [
  "vm-memory",
 ]
@@ -662,9 +662,9 @@ dependencies = [
 
 [[package]]
 name = "once_cell"
-version = "1.19.0"
+version = "1.20.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3fdb12b2476b595f9358c5161aa467c2438859caa136dec86c26fdd2efe17b92"
+checksum = "1261fe7e33c73b354eab43b1273a57c8f967d0391e80353e51f764ac02cf6775"
 
 [[package]]
 name = "option_parser"
@@ -1020,7 +1020,7 @@ dependencies = [
 [[package]]
 name = "vfio-bindings"
 version = "0.4.0"
-source = "git+https://github.com/rust-vmm/vfio?branch=main#a51a4746b0d317bfc21fa49d40f9287f3b8137fd"
+source = "git+https://github.com/rust-vmm/vfio?branch=main#b135b8305c2cc8ec333e0cf77a780445cc98dcee"
 dependencies = [
  "vmm-sys-util",
 ]
@@ -1028,7 +1028,7 @@ dependencies = [
 [[package]]
 name = "vfio-ioctls"
 version = "0.2.0"
-source = "git+https://github.com/rust-vmm/vfio?branch=main#a51a4746b0d317bfc21fa49d40f9287f3b8137fd"
+source = "git+https://github.com/rust-vmm/vfio?branch=main#b135b8305c2cc8ec333e0cf77a780445cc98dcee"
 dependencies = [
  "byteorder",
  "kvm-bindings",
@@ -1044,7 +1044,7 @@ dependencies = [
 [[package]]
 name = "vfio_user"
 version = "0.1.0"
-source = "git+https://github.com/rust-vmm/vfio-user?branch=main#bf7d7e851b604d8414a7960fb9137b59fc42421d"
+source = "git+https://github.com/rust-vmm/vfio-user?branch=main#3febcdd3fa2531623865663ca1721e1962ed9979"
 dependencies = [
  "bitflags 1.3.2",
  "libc",
@@ -1060,9 +1060,8 @@ dependencies = [
 
 [[package]]
 name = "vhost"
-version = "0.12.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7c1c4c6c9f79fbe3150d9a403008ca416d34c489897effdda28b646f09900aad"
+version = "0.12.1"
+source = "git+https://github.com/rust-vmm/vhost?rev=d983ae0#d983ae07f78663b7d24059667376992460b571a2"
 dependencies = [
  "bitflags 2.6.0",
  "libc",
@@ -1073,9 +1072,9 @@ dependencies = [
 
 [[package]]
 name = "virtio-bindings"
-version = "0.2.3"
+version = "0.2.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "68d0df4f5ad79b1dc81b5913ac737e24a84dcd5100f36ed953a1faec18aba241"
+checksum = "1711e61c00f8cb450bd15368152a1e37a12ef195008ddc7d0f4812f9e2b30a68"
 
 [[package]]
 name = "virtio-devices"
@@ -1112,9 +1111,9 @@ dependencies = [
 
 [[package]]
 name = "virtio-queue"
-version = "0.13.0"
+version = "0.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ffb1761348d3b5e82131379b9373435b48dc8333100bff3f1cdf9cc541a0ad83"
+checksum = "872e2f3fbd70a7e6f01689720cce3d5c2c5efe52b484dd07b674246ada0e9a8d"
 dependencies = [
  "log",
  "virtio-bindings",
@@ -1147,13 +1146,13 @@ dependencies = [
 [[package]]
 name = "vm-fdt"
 version = "0.3.0"
-source = "git+https://github.com/rust-vmm/vm-fdt?branch=main#982fb8d9c8cd7f53520d7e304b39ff307fa3a641"
+source = "git+https://github.com/rust-vmm/vm-fdt?branch=main#ef5bd734f5f66fb07722d766981adbc915f0d941"
 
 [[package]]
 name = "vm-memory"
-version = "0.15.0"
+version = "0.16.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a320fc11792e063174402ff444aa3c80363cbf1e31c47b5ef74124406c334ce6"
+checksum = "e2919f87420b6998a131eb7c78843890295e91a3f8f786ccc925c8d387b75121"
 dependencies = [
  "arc-swap",
  "libc",

fuzz/Cargo.toml

@@ -18,15 +18,15 @@ devices = { path = "../devices" }
 epoll = "4.3.3"
 libc = "0.2.155"
 libfuzzer-sys = "0.4.7"
-linux-loader = { version = "0.12.0", features = ["bzimage", "elf", "pe"] }
+linux-loader = { version = "0.13.0", features = ["bzimage", "elf", "pe"] }
 micro_http = { git = "https://github.com/firecracker-microvm/micro-http", branch = "main" }
 net_util = { path = "../net_util" }
 once_cell = "1.19.0"
 seccompiler = "0.4.0"
 virtio-devices = { path = "../virtio-devices" }
-virtio-queue = "0.13.0"
+virtio-queue = "0.14.0"
 vm-device = { path = "../vm-device" }
-vm-memory = "0.15.0"
+vm-memory = "0.16.0"
 vm-migration = { path = "../vm-migration" }
 vm-virtio = { path = "../vm-virtio" }
 vmm = { path = "../vmm", features = ["guest_debug"] }

hypervisor/Cargo.toml

@@ -12,7 +12,8 @@ sev_snp = ["igvm", "igvm_defs"]
 tdx = []
 
 [dependencies]
-anyhow = "1.0.87"
+anyhow = "1.0.94"
+arc-swap = "1.7.1"
 byteorder = "1.5.0"
 cfg-if = "1.0.0"
 concat-idents = "1.1.5"
@@ -20,7 +21,7 @@ igvm = { version = "0.3.3", optional = true }
 igvm_defs = { version = "0.3.1", optional = true }
 kvm-bindings = { workspace = true, optional = true, features = ["serde"] }
 kvm-ioctls = { workspace = true, optional = true }
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 mshv-bindings = { workspace = true, features = [
   "fam-wrappers",
@@ -33,7 +34,11 @@ serde_with = { version = "3.9.0", default-features = false, features = [
 ] }
 thiserror = "1.0.62"
 vfio-ioctls = { workspace = true, default-features = false }
-vm-memory = { workspace = true, features = ["backend-atomic", "backend-mmap"] }
+vm-memory = { workspace = true, features = [
+  "backend-atomic",
+  "backend-bitmap",
+  "backend-mmap",
+] }
 vmm-sys-util = { workspace = true, features = ["with-serde"] }
 
 [target.'cfg(target_arch = "x86_64")'.dependencies.iced-x86]

hypervisor/src/arch/mod.rs

@@ -21,3 +21,6 @@ pub mod x86;
 
 #[cfg(target_arch = "aarch64")]
 pub mod aarch64;
+
+#[cfg(target_arch = "riscv64")]
+pub mod riscv64;

hypervisor/src/arch/riscv64/aia.rs

@@ -0,0 +1,63 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+//
+// SPDX-License-Identifier: Apache-2.0
+
+use std::any::Any;
+use std::result;
+
+use thiserror::Error;
+
+use crate::{AiaState, HypervisorDeviceError, HypervisorVmError};
+
+/// Errors thrown while setting up the VAIA.
+#[derive(Debug, Error)]
+pub enum Error {
+    /// Error while calling KVM ioctl for setting up the global interrupt controller.
+    #[error("Failed creating AIA device: {0}")]
+    CreateAia(HypervisorVmError),
+    /// Error while setting device attributes for the AIA.
+    #[error("Failed setting device attributes for the AIA: {0}")]
+    SetDeviceAttribute(HypervisorDeviceError),
+    /// Error while getting device attributes for the AIA.
+    #[error("Failed getting device attributes for the AIA: {0}")]
+    GetDeviceAttribute(HypervisorDeviceError),
+}
+pub type Result<T> = result::Result<T, Error>;
+
+#[derive(Debug)]
+pub struct VaiaConfig {
+    pub vcpu_count: u32,
+    pub aplic_addr: u64,
+    pub imsic_addr: u64,
+    pub nr_irqs: u32,
+}
+
+/// Hypervisor agnostic interface for a virtualized AIA
+pub trait Vaia: Send + Sync {
+    /// Returns the compatibility property of APLIC
+    fn aplic_compatibility(&self) -> &str;
+
+    /// Returns an array with APLIC device properties
+    fn aplic_properties(&self) -> [u32; 4];
+
+    /// Returns the compatibility property of IMSIC
+    fn imsic_compatibility(&self) -> &str;
+
+    /// Returns an array with IMSIC device properties
+    fn imsic_properties(&self) -> [u32; 4];
+
+    /// Returns the number of vCPUs this AIA handles
+    fn vcpu_count(&self) -> u32;
+
+    /// Returns whether the AIA device is MSI compatible or not
+    fn msi_compatible(&self) -> bool;
+
+    /// Downcast the trait object to its concrete type.
+    fn as_any_concrete_mut(&mut self) -> &mut dyn Any;
+
+    /// Save the state of AiaImsics.
+    fn state(&self) -> Result<AiaState>;
+
+    /// Restore the state of AiaImsics.
+    fn set_state(&mut self, state: &AiaState) -> Result<()>;
+}

hypervisor/src/arch/riscv64/mod.rs

@@ -0,0 +1,5 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+//
+// SPDX-License-Identifier: Apache-2.0
+
+pub mod aia;

hypervisor/src/cpu.rs

@@ -1,3 +1,5 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+//
 // Copyright © 2019 Intel Corporation
 //
 // SPDX-License-Identifier: Apache-2.0 OR BSD-3-Clause
@@ -9,6 +11,7 @@
 //
 
 use thiserror::Error;
+#[cfg(not(target_arch = "riscv64"))]
 use vm_memory::GuestAddress;
 
 #[cfg(target_arch = "aarch64")]
@@ -17,6 +20,8 @@ use crate::aarch64::{RegList, VcpuInit};
 use crate::arch::x86::{CpuIdEntry, FpuState, LapicState, MsrEntry, SpecialRegisters};
 #[cfg(feature = "tdx")]
 use crate::kvm::{TdxExitDetails, TdxExitStatus};
+#[cfg(target_arch = "riscv64")]
+use crate::riscv64::RegList;
 use crate::{CpuState, MpState, StandardRegisters};
 
 #[cfg(target_arch = "x86_64")]
@@ -216,6 +221,16 @@ pub enum HypervisorCpuError {
     #[error("Failed to set aarch64 core register: {0}")]
     SetAarchCoreRegister(#[source] anyhow::Error),
     ///
+    /// Getting RISC-V 64-bit core register error
+    ///
+    #[error("Failed to get riscv64 core register: {0}")]
+    GetRiscvCoreRegister(#[source] anyhow::Error),
+    ///
+    /// Setting RISC-V 64-bit core register error
+    ///
+    #[error("Failed to set riscv64 core register: {0}")]
+    SetRiscvCoreRegister(#[source] anyhow::Error),
+    ///
     /// Getting registers list error
     ///
     #[error("Failed to retrieve list of registers: {0}")]
@@ -231,6 +246,16 @@ pub enum HypervisorCpuError {
     #[error("Failed to set system register: {0}")]
     SetSysRegister(#[source] anyhow::Error),
     ///
+    /// Getting RISC-V 64-bit non-core register error
+    ///
+    #[error("Failed to get non-core register: {0}")]
+    GetNonCoreRegister(#[source] anyhow::Error),
+    ///
+    /// Setting RISC-V 64-bit non-core register error
+    ///
+    #[error("Failed to set non-core register: {0}")]
+    SetNonCoreRegister(#[source] anyhow::Error),
+    ///
     /// GVA translation error
     ///
     #[error("Failed to translate GVA: {0}")]
@@ -412,6 +437,7 @@ pub trait Vcpu: Send + Sync {
     ///
     /// Sets debug registers to set hardware breakpoints and/or enable single step.
     ///
+    #[cfg(not(target_arch = "riscv64"))]
     fn set_guest_debug(&self, _addrs: &[GuestAddress], _singlestep: bool) -> Result<()> {
         Err(HypervisorCpuError::SetDebugRegs(anyhow!("unimplemented")))
     }
@@ -428,7 +454,7 @@ pub trait Vcpu: Send + Sync {
     /// Gets a list of the guest registers that are supported for the
     /// KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
     ///
-    #[cfg(target_arch = "aarch64")]
+    #[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
     fn get_reg_list(&self, reg_list: &mut RegList) -> Result<()>;
     ///
     /// Gets the value of a system register
@@ -436,9 +462,14 @@ pub trait Vcpu: Send + Sync {
     #[cfg(target_arch = "aarch64")]
     fn get_sys_reg(&self, sys_reg: u32) -> Result<u64>;
     ///
+    /// Gets the value of a non-core register on RISC-V 64-bit
+    ///
+    #[cfg(target_arch = "riscv64")]
+    fn get_non_core_reg(&self, non_core_reg: u32) -> Result<u64>;
+    ///
     /// Configure core registers for a given CPU.
     ///
-    #[cfg(target_arch = "aarch64")]
+    #[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
     fn setup_regs(&self, cpu_id: u8, boot_ip: u64, fdt_start: u64) -> Result<()>;
     ///
     /// Check if the CPU supports PMU

hypervisor/src/hypervisor.rs

@@ -118,6 +118,17 @@ pub trait Hypervisor: Send + Sync {
     fn create_vm_with_type(&self, _vm_type: u64) -> Result<Arc<dyn Vm>> {
         unreachable!()
     }
+    ///
+    /// Create a Vm of a specific type using the underlying hypervisor, passing memory size
+    /// Return a hypervisor-agnostic Vm trait object
+    ///
+    fn create_vm_with_type_and_memory(
+        &self,
+        _vm_type: u64,
+        #[cfg(feature = "sev_snp")] _mem_size: u64,
+    ) -> Result<Arc<dyn Vm>> {
+        unreachable!()
+    }
     #[cfg(target_arch = "x86_64")]
     ///
     /// Get the supported CpuID

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

@@ -103,9 +103,8 @@ fn dist_attr_get(gic: &DeviceFd, offset: u32) -> Result<u32> {
         flags: 0,
     };
 
-    // get_device_attr should be marked as unsafe, and will be in future.
     // SAFETY: gic_dist_attr.addr is safe to write to.
-    gic.get_device_attr(&mut gic_dist_attr).map_err(|e| {
+    unsafe { gic.get_device_attr(&mut gic_dist_attr) }.map_err(|e| {
         Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
     })?;
 
@@ -131,9 +130,8 @@ fn get_interrupts_num(gic: &DeviceFd) -> Result<u32> {
         addr: &mut num_irq as *mut u32 as u64,
         flags: 0,
     };
-    // get_device_attr should be marked as unsafe, and will be in future.
     // SAFETY: nr_irqs_attr.addr is safe to write to.
-    gic.get_device_attr(&mut nr_irqs_attr).map_err(|e| {
+    unsafe { gic.get_device_attr(&mut nr_irqs_attr) }.map_err(|e| {
         Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
     })?;
     Ok(num_irq)

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

@@ -105,9 +105,8 @@ fn icc_attr_get(gic: &DeviceFd, offset: u64, typer: u64) -> Result<u32> {
         flags: 0,
     };
 
-    // get_device_attr should be marked as unsafe, and will be in future.
     // SAFETY: gic_icc_attr.addr is safe to write to.
-    gic.get_device_attr(&mut gic_icc_attr).map_err(|e| {
+    unsafe { gic.get_device_attr(&mut gic_icc_attr) }.map_err(|e| {
         Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
     })?;
 

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

@@ -49,13 +49,10 @@ fn gicv3_its_attr_get(its_device: &DeviceFd, group: u32, attr: u32) -> Result<u6
         flags: 0,
     };
 
-    // get_device_attr should be marked as unsafe, and will be in future.
     // SAFETY: gicv3_its_attr.addr is safe to write to.
-    its_device
-        .get_device_attr(&mut gicv3_its_attr)
-        .map_err(|e| {
-            Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
-        })?;
+    unsafe { its_device.get_device_attr(&mut gicv3_its_attr) }.map_err(|e| {
+        Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
+    })?;
 
     Ok(val)
 }

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

@@ -121,9 +121,8 @@ fn redist_attr_get(gic: &DeviceFd, offset: u32, typer: u64) -> Result<u32> {
         flags: 0,
     };
 
-    // get_device_attr should be marked as unsafe, and will be in future.
     // SAFETY: gic_redist_attr.addr is safe to write to.
-    gic.get_device_attr(&mut gic_redist_attr).map_err(|e| {
+    unsafe { gic.get_device_attr(&mut gic_redist_attr) }.map_err(|e| {
         Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
     })?;
 

hypervisor/src/kvm/riscv64/aia.rs

@@ -0,0 +1,274 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+//
+// SPDX-License-Identifier: Apache-2.0
+
+use std::any::Any;
+
+use kvm_ioctls::DeviceFd;
+use serde::{Deserialize, Serialize};
+
+use crate::arch::riscv64::aia::{Error, Result, Vaia, VaiaConfig};
+use crate::device::HypervisorDeviceError;
+use crate::kvm::KvmVm;
+use crate::Vm;
+
+pub struct KvmAiaImsics {
+    /// The KVM device for the Aia
+    device: DeviceFd,
+
+    /// AIA APLIC address
+    aplic_addr: u64,
+
+    /// AIA IMSIC address
+    imsic_addr: u64,
+
+    /// Number of CPUs handled by the device
+    vcpu_count: u32,
+}
+
+#[derive(Clone, Default, Serialize, Deserialize)]
+pub struct AiaImsicsState {}
+
+impl KvmAiaImsics {
+    /// Device trees specific constants
+
+    fn version() -> u32 {
+        kvm_bindings::kvm_device_type_KVM_DEV_TYPE_RISCV_AIA
+    }
+
+    /// Setup the device-specific attributes
+    fn init_device_attributes(&mut self, nr_irqs: u32) -> Result<()> {
+        // AIA part attributes
+        // Getting the working mode of RISC-V AIA, defaults to EMUL, passible
+        // variants are EMUL, HW_ACCL, AUTO
+        let mut aia_mode = kvm_bindings::KVM_DEV_RISCV_AIA_MODE_EMUL;
+        Self::get_device_attribute(
+            &self.device,
+            kvm_bindings::KVM_DEV_RISCV_AIA_GRP_CONFIG,
+            u64::from(kvm_bindings::KVM_DEV_RISCV_AIA_CONFIG_MODE),
+            &mut aia_mode as *mut u32 as u64,
+            0,
+        )?;
+
+        // Report AIA MODE
+
+        // Setting up the number of wired interrupt sources
+        Self::set_device_attribute(
+            &self.device,
+            kvm_bindings::KVM_DEV_RISCV_AIA_GRP_CONFIG,
+            u64::from(kvm_bindings::KVM_DEV_RISCV_AIA_CONFIG_SRCS),
+            &nr_irqs as *const u32 as u64,
+            0,
+        )?;
+
+        // Getting the number of ids
+        let mut aia_nr_ids: u32 = 0;
+        Self::get_device_attribute(
+            &self.device,
+            kvm_bindings::KVM_DEV_RISCV_AIA_GRP_CONFIG,
+            u64::from(kvm_bindings::KVM_DEV_RISCV_AIA_CONFIG_IDS),
+            &mut aia_nr_ids as *mut u32 as u64,
+            0,
+        )?;
+
+        // Report NR_IDS
+
+        // Setting up hart_bits
+        let max_hart_index = self.vcpu_count as u64 - 1;
+        let hart_bits = std::cmp::max(64 - max_hart_index.leading_zeros(), 1);
+        Self::set_device_attribute(
+            &self.device,
+            kvm_bindings::KVM_DEV_RISCV_AIA_GRP_CONFIG,
+            u64::from(kvm_bindings::KVM_DEV_RISCV_AIA_CONFIG_HART_BITS),
+            &hart_bits as *const u32 as u64,
+            0,
+        )?;
+
+        // Designate addresses of APLIC and IMSICS
+
+        // Setting up RISC-V APLIC
+        Self::set_device_attribute(
+            &self.device,
+            kvm_bindings::KVM_DEV_RISCV_AIA_GRP_ADDR,
+            u64::from(kvm_bindings::KVM_DEV_RISCV_AIA_ADDR_APLIC),
+            &self.aplic_addr as *const u64 as u64,
+            0,
+        )?;
+
+        // Helpers to calculate address and attribute of IMSIC of each vCPU
+        let riscv_imsic_addr_of = |cpu_index: u32| -> u64 {
+            self.imsic_addr + (cpu_index * kvm_bindings::KVM_DEV_RISCV_IMSIC_SIZE) as u64
+        };
+        let riscv_imsic_attr_of = |cpu_index: u32| -> u64 { cpu_index as u64 };
+
+        // Setting up RISC-V IMSICs
+        for cpu_index in 0..self.vcpu_count {
+            let cpu_imsic_addr = riscv_imsic_addr_of(cpu_index);
+            Self::set_device_attribute(
+                &self.device,
+                kvm_bindings::KVM_DEV_RISCV_AIA_GRP_ADDR,
+                riscv_imsic_attr_of(cpu_index),
+                &cpu_imsic_addr as *const u64 as u64,
+                0,
+            )?;
+        }
+
+        // Finalizing the AIA device
+        Self::set_device_attribute(
+            &self.device,
+            kvm_bindings::KVM_DEV_RISCV_AIA_GRP_CTRL,
+            u64::from(kvm_bindings::KVM_DEV_RISCV_AIA_CTRL_INIT),
+            0,
+            0,
+        )
+    }
+
+    /// Create a KVM Vaia device
+    fn create_device(vm: &KvmVm) -> Result<DeviceFd> {
+        let mut aia_device = kvm_bindings::kvm_create_device {
+            type_: Self::version(),
+            fd: 0,
+            flags: 0,
+        };
+
+        let device_fd = vm
+            .create_device(&mut aia_device)
+            .map_err(Error::CreateAia)?;
+
+        // We know for sure this is a KVM fd
+        Ok(device_fd.to_kvm().unwrap())
+    }
+
+    /// Get an AIA device attribute
+    fn get_device_attribute(
+        device: &DeviceFd,
+        group: u32,
+        attr: u64,
+        addr: u64,
+        flags: u32,
+    ) -> Result<()> {
+        let mut attr = kvm_bindings::kvm_device_attr {
+            flags,
+            group,
+            attr,
+            addr,
+        };
+        // SAFETY: attr.addr is safe to write to.
+        unsafe {
+            device.get_device_attr(&mut attr).map_err(|e| {
+                Error::GetDeviceAttribute(HypervisorDeviceError::GetDeviceAttribute(e.into()))
+            })
+        }
+    }
+
+    /// Set an AIA device attribute
+    fn set_device_attribute(
+        device: &DeviceFd,
+        group: u32,
+        attr: u64,
+        addr: u64,
+        flags: u32,
+    ) -> Result<()> {
+        let attr = kvm_bindings::kvm_device_attr {
+            flags,
+            group,
+            attr,
+            addr,
+        };
+        device.set_device_attr(&attr).map_err(|e| {
+            Error::SetDeviceAttribute(HypervisorDeviceError::SetDeviceAttribute(e.into()))
+        })
+    }
+
+    /// Method to initialize the AIA device
+    pub fn new(vm: &dyn Vm, config: VaiaConfig) -> Result<KvmAiaImsics> {
+        // This is inside KVM module
+        let vm = vm.as_any().downcast_ref::<KvmVm>().expect("Wrong VM type?");
+
+        let vaia = Self::create_device(vm)?;
+
+        let mut aia_device = KvmAiaImsics {
+            device: vaia,
+            vcpu_count: config.vcpu_count,
+            aplic_addr: config.aplic_addr,
+            imsic_addr: config.imsic_addr,
+        };
+
+        aia_device.init_device_attributes(config.nr_irqs)?;
+
+        Ok(aia_device)
+    }
+}
+
+impl Vaia for KvmAiaImsics {
+    fn aplic_compatibility(&self) -> &str {
+        "riscv,aplic"
+    }
+
+    fn aplic_properties(&self) -> [u32; 4] {
+        [
+            0,
+            self.aplic_addr as u32,
+            0,
+            kvm_bindings::KVM_DEV_RISCV_APLIC_SIZE,
+        ]
+    }
+
+    fn imsic_compatibility(&self) -> &str {
+        "riscv,imsics"
+    }
+
+    fn imsic_properties(&self) -> [u32; 4] {
+        [
+            0,
+            self.imsic_addr as u32,
+            0,
+            kvm_bindings::KVM_DEV_RISCV_IMSIC_SIZE * self.vcpu_count,
+        ]
+    }
+
+    fn vcpu_count(&self) -> u32 {
+        self.vcpu_count
+    }
+
+    fn msi_compatible(&self) -> bool {
+        true
+    }
+
+    fn as_any_concrete_mut(&mut self) -> &mut dyn Any {
+        self
+    }
+
+    /// Save the state of AIA.
+    fn state(&self) -> Result<AiaImsicsState> {
+        unimplemented!()
+    }
+
+    /// Restore the state of AIA_IMSICs.
+    fn set_state(&mut self, _state: &AiaImsicsState) -> Result<()> {
+        unimplemented!()
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use crate::arch::riscv64::aia::VaiaConfig;
+    use crate::kvm::KvmAiaImsics;
+
+    fn create_test_vaia_config() -> VaiaConfig {
+        VaiaConfig {
+            vcpu_count: 1,
+            aplic_addr: 0xd000000,
+            imsic_addr: 0x2800000,
+            nr_irqs: 256,
+        }
+    }
+
+    #[test]
+    fn test_create_aia() {
+        let hv = crate::new().unwrap();
+        let vm = hv.create_vm().unwrap();
+
+        assert!(KvmAiaImsics::new(&*vm, create_test_vaia_config()).is_ok());
+    }
+}

hypervisor/src/kvm/riscv64/mod.rs

@@ -0,0 +1,163 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+//
+// SPDX-License-Identifier: Apache-2.0
+
+pub mod aia;
+
+use kvm_bindings::{
+    kvm_mp_state, kvm_one_reg, kvm_riscv_core, KVM_REG_RISCV_CORE, KVM_REG_RISCV_TYPE_MASK,
+    KVM_REG_SIZE_MASK, KVM_REG_SIZE_U64,
+};
+pub use kvm_bindings::{kvm_one_reg as Register, RegList};
+pub use kvm_ioctls::{Cap, Kvm};
+use serde::{Deserialize, Serialize};
+
+use crate::kvm::{KvmError, KvmResult};
+
+// This macro gets the offset of a structure (i.e `str`) member (i.e `field`) without having
+// an instance of that structure.
+#[macro_export]
+macro_rules! _offset_of {
+    ($str:ty, $field:ident) => {{
+        let tmp: std::mem::MaybeUninit<$str> = std::mem::MaybeUninit::uninit();
+        let base = tmp.as_ptr();
+
+        // Avoid warnings when nesting `unsafe` blocks.
+        #[allow(unused_unsafe)]
+        // SAFETY: The pointer is valid and aligned, just not initialised. Using `addr_of` ensures
+        // that we don't actually read from `base` (which would be UB) nor create an intermediate
+        // reference.
+        let member = unsafe { core::ptr::addr_of!((*base).$field) } as *const u8;
+
+        // Avoid warnings when nesting `unsafe` blocks.
+        #[allow(unused_unsafe)]
+        // SAFETY: The two pointers are within the same allocated object `tmp`. All requirements
+        // from offset_from are upheld.
+        unsafe {
+            member.offset_from(base as *const u8) as usize
+        }
+    }};
+}
+
+#[macro_export]
+macro_rules! offset_of {
+    ($reg_struct:ty, $field:ident) => {
+        $crate::_offset_of!($reg_struct, $field)
+    };
+    ($outer_reg_struct:ty, $outer_field:ident, $($inner_reg_struct:ty, $inner_field:ident), +) => {
+        $crate::_offset_of!($outer_reg_struct, $outer_field) + offset_of!($($inner_reg_struct, $inner_field), +)
+    };
+}
+
+// Following are macros that help with getting the ID of a riscv64 register, including config registers, core registers and timer registers.
+// The register of core registers are wrapped in the `user_regs_struct` structure. See:
+// https://elixir.bootlin.com/linux/v6.10/source/arch/riscv/include/uapi/asm/kvm.h#L62
+
+// Get the ID of a register
+#[macro_export]
+macro_rules! riscv64_reg_id {
+    ($reg_type: tt, $offset: tt) => {
+        // The core registers of an riscv64 machine are represented
+        // in kernel by the `kvm_riscv_core` structure:
+        //
+        // struct kvm_riscv_core {
+        //     struct user_regs_struct regs;
+        //     unsigned long mode;
+        // };
+        //
+        // struct user_regs_struct {
+        //     unsigned long pc;
+        //     unsigned long ra;
+        //     unsigned long sp;
+        //     unsigned long gp;
+        //     unsigned long tp;
+        //     unsigned long t0;
+        //     unsigned long t1;
+        //     unsigned long t2;
+        //     unsigned long s0;
+        //     unsigned long s1;
+        //     unsigned long a0;
+        //     unsigned long a1;
+        //     unsigned long a2;
+        //     unsigned long a3;
+        //     unsigned long a4;
+        //     unsigned long a5;
+        //     unsigned long a6;
+        //     unsigned long a7;
+        //     unsigned long s2;
+        //     unsigned long s3;
+        //     unsigned long s4;
+        //     unsigned long s5;
+        //     unsigned long s6;
+        //     unsigned long s7;
+        //     unsigned long s8;
+        //     unsigned long s9;
+        //     unsigned long s10;
+        //     unsigned long s11;
+        //     unsigned long t3;
+        //     unsigned long t4;
+        //     unsigned long t5;
+        //     unsigned long t6;
+        // };
+        // The id of a core register can be obtained like this: offset = id &
+        // ~(KVM_REG_ARCH_MASK | KVM_REG_SIZE_MASK | KVM_REG_RISCV_CORE). Thus,
+        // id = KVM_REG_RISCV | KVM_REG_SIZE_U64 | KVM_REG_RISCV_CORE | offset
+        //
+        // To generalize, the id of a register can be obtained by:
+        // id = KVM_REG_RISCV | KVM_REG_SIZE_U64 |
+        //      KVM_REG_RISCV_CORE/KVM_REG_RISCV_CONFIG/KVM_REG_RISCV_TIMER |
+        //      offset
+        kvm_bindings::KVM_REG_RISCV as u64
+            | u64::from($reg_type)
+            | u64::from(kvm_bindings::KVM_REG_SIZE_U64)
+            | (($offset / std::mem::size_of::<u64>()) as u64)
+    };
+}
+
+/// Specifies whether a particular register is a core register or not.
+///
+/// # Arguments
+///
+/// * `regid` - The index of the register we are checking.
+pub fn is_non_core_register(regid: u64) -> bool {
+    if (regid & KVM_REG_RISCV_TYPE_MASK as u64) == KVM_REG_RISCV_CORE as u64 {
+        return false;
+    }
+
+    let size = regid & KVM_REG_SIZE_MASK;
+
+    assert!(
+        size == KVM_REG_SIZE_U64,
+        "Unexpected register size for system register {size}"
+    );
+
+    true
+}
+
+pub fn check_required_kvm_extensions(kvm: &Kvm) -> KvmResult<()> {
+    macro_rules! check_extension {
+        ($cap:expr) => {
+            if !kvm.check_extension($cap) {
+                return Err(KvmError::CapabilityMissing($cap));
+            }
+        };
+    }
+
+    // SetGuestDebug is required but some kernels have it implemented without the capability flag.
+    check_extension!(Cap::ImmediateExit);
+    check_extension!(Cap::Ioeventfd);
+    check_extension!(Cap::Irqchip);
+    check_extension!(Cap::Irqfd);
+    check_extension!(Cap::IrqRouting);
+    check_extension!(Cap::MpState);
+    check_extension!(Cap::OneReg);
+    check_extension!(Cap::UserMemory);
+    Ok(())
+}
+
+#[derive(Clone, Default, Serialize, Deserialize)]
+pub struct VcpuKvmState {
+    pub mp_state: kvm_mp_state,
+    pub core_regs: kvm_riscv_core,
+    pub non_core_regs: Vec<kvm_one_reg>,
+}

hypervisor/src/lib.rs

@@ -1,3 +1,5 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+//
 // Copyright © 2019 Intel Corporation
 //
 // SPDX-License-Identifier: Apache-2.0 OR BSD-3-Clause
@@ -16,6 +18,7 @@
 //!
 //! - x86_64
 //! - arm64
+//! - riscv64 (experimental)
 //!
 
 #[macro_use]
@@ -57,6 +60,8 @@ pub use cpu::{HypervisorCpuError, Vcpu, VmExit};
 pub use device::HypervisorDeviceError;
 #[cfg(all(feature = "kvm", target_arch = "aarch64"))]
 pub use kvm::{aarch64, GicState};
+#[cfg(all(feature = "kvm", target_arch = "riscv64"))]
+pub use kvm::{riscv64, AiaState};
 pub use vm::{
     DataMatch, HypervisorVmError, InterruptSourceConfig, LegacyIrqSourceConfig, MsiIrqSourceConfig,
     Vm, VmOps,
@@ -193,8 +198,10 @@ pub enum IrqRoutingEntry {
 
 #[derive(Debug, Clone, Copy, PartialEq, serde::Serialize, serde::Deserialize)]
 pub enum StandardRegisters {
-    #[cfg(feature = "kvm")]
+    #[cfg(all(feature = "kvm", not(target_arch = "riscv64")))]
     Kvm(kvm_bindings::kvm_regs),
+    #[cfg(all(feature = "kvm", target_arch = "riscv64"))]
+    Kvm(kvm_bindings::kvm_riscv_core),
     #[cfg(all(feature = "mshv", target_arch = "x86_64"))]
     Mshv(mshv_bindings::StandardRegisters),
 }
@@ -314,3 +321,125 @@ get_aarch64_reg!(regs, [u64; 31usize]);
 get_aarch64_reg!(sp, u64);
 get_aarch64_reg!(pc, u64);
 get_aarch64_reg!(pstate, u64);
+
+macro_rules! set_riscv64_reg {
+    (mode) => {
+        #[cfg(target_arch = "riscv64")]
+        impl StandardRegisters {
+            pub fn set_mode(&mut self, val: u64) {
+                match self {
+                    #[cfg(feature = "kvm")]
+                    StandardRegisters::Kvm(s) => s.mode = val,
+                }
+            }
+        }
+    };
+    ($reg_name:ident) => {
+        concat_idents!(method_name = "set_", $reg_name {
+            #[cfg(target_arch = "riscv64")]
+            impl StandardRegisters {
+                pub fn method_name(&mut self, val: u64) {
+                    match self {
+                        #[cfg(feature = "kvm")]
+                        StandardRegisters::Kvm(s) => s.regs.$reg_name = val,
+                    }
+                }
+            }
+        });
+    }
+}
+
+macro_rules! get_riscv64_reg {
+    (mode) => {
+        #[cfg(target_arch = "riscv64")]
+        impl StandardRegisters {
+            pub fn get_mode(&self) -> u64 {
+                match self {
+                    #[cfg(feature = "kvm")]
+                    StandardRegisters::Kvm(s) => s.mode,
+                }
+            }
+        }
+    };
+    ($reg_name:ident) => {
+        concat_idents!(method_name = "get_", $reg_name {
+            #[cfg(target_arch = "riscv64")]
+            impl StandardRegisters {
+                pub fn method_name(&self) -> u64 {
+                    match self {
+                        #[cfg(feature = "kvm")]
+                        StandardRegisters::Kvm(s) => s.regs.$reg_name,
+                    }
+                }
+            }
+        });
+    }
+}
+
+set_riscv64_reg!(pc);
+set_riscv64_reg!(ra);
+set_riscv64_reg!(sp);
+set_riscv64_reg!(gp);
+set_riscv64_reg!(tp);
+set_riscv64_reg!(t0);
+set_riscv64_reg!(t1);
+set_riscv64_reg!(t2);
+set_riscv64_reg!(s0);
+set_riscv64_reg!(s1);
+set_riscv64_reg!(a0);
+set_riscv64_reg!(a1);
+set_riscv64_reg!(a2);
+set_riscv64_reg!(a3);
+set_riscv64_reg!(a4);
+set_riscv64_reg!(a5);
+set_riscv64_reg!(a6);
+set_riscv64_reg!(a7);
+set_riscv64_reg!(s2);
+set_riscv64_reg!(s3);
+set_riscv64_reg!(s4);
+set_riscv64_reg!(s5);
+set_riscv64_reg!(s6);
+set_riscv64_reg!(s7);
+set_riscv64_reg!(s8);
+set_riscv64_reg!(s9);
+set_riscv64_reg!(s10);
+set_riscv64_reg!(s11);
+set_riscv64_reg!(t3);
+set_riscv64_reg!(t4);
+set_riscv64_reg!(t5);
+set_riscv64_reg!(t6);
+set_riscv64_reg!(mode);
+
+get_riscv64_reg!(pc);
+get_riscv64_reg!(ra);
+get_riscv64_reg!(sp);
+get_riscv64_reg!(gp);
+get_riscv64_reg!(tp);
+get_riscv64_reg!(t0);
+get_riscv64_reg!(t1);
+get_riscv64_reg!(t2);
+get_riscv64_reg!(s0);
+get_riscv64_reg!(s1);
+get_riscv64_reg!(a0);
+get_riscv64_reg!(a1);
+get_riscv64_reg!(a2);
+get_riscv64_reg!(a3);
+get_riscv64_reg!(a4);
+get_riscv64_reg!(a5);
+get_riscv64_reg!(a6);
+get_riscv64_reg!(a7);
+get_riscv64_reg!(s2);
+get_riscv64_reg!(s3);
+get_riscv64_reg!(s4);
+get_riscv64_reg!(s5);
+get_riscv64_reg!(s6);
+get_riscv64_reg!(s7);
+get_riscv64_reg!(s8);
+get_riscv64_reg!(s9);
+get_riscv64_reg!(s10);
+get_riscv64_reg!(s11);
+get_riscv64_reg!(t3);
+get_riscv64_reg!(t4);
+get_riscv64_reg!(t5);
+get_riscv64_reg!(t6);
+get_riscv64_reg!(mode);

hypervisor/src/mshv/mod.rs

@@ -5,12 +5,18 @@
 
 use std::any::Any;
 use std::collections::HashMap;
+#[cfg(feature = "sev_snp")]
+use std::num::NonZeroUsize;
 use std::sync::{Arc, RwLock};
 
+#[cfg(feature = "sev_snp")]
+use arc_swap::ArcSwap;
 use mshv_bindings::*;
 use mshv_ioctls::{set_registers_64, InterruptRequest, Mshv, NoDatamatch, VcpuFd, VmFd, VmType};
 use vfio_ioctls::VfioDeviceFd;
 use vm::DataMatch;
+#[cfg(feature = "sev_snp")]
+use vm_memory::bitmap::AtomicBitmap;
 
 use crate::arch::emulator::PlatformEmulator;
 #[cfg(target_arch = "x86_64")]
@@ -205,48 +211,12 @@ impl MshvHypervisor {
             .get_msr_index_list()
             .map_err(|e| hypervisor::HypervisorError::GetMsrList(e.into()))
     }
-}
 
-impl MshvHypervisor {
-    /// Create a hypervisor based on Mshv
-    #[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(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
-///
-/// # Examples
-///
-/// ```
-/// # use hypervisor::mshv::MshvHypervisor;
-/// # use std::sync::Arc;
-/// let mshv = MshvHypervisor::new().unwrap();
-/// let hypervisor = Arc::new(mshv);
-/// 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
-    }
-
-    fn create_vm_with_type(&self, vm_type: u64) -> hypervisor::Result<Arc<dyn crate::Vm>> {
+    fn create_vm_with_type_and_memory_int(
+        &self,
+        vm_type: u64,
+        #[cfg(feature = "sev_snp")] _mem_size: Option<u64>,
+    ) -> hypervisor::Result<Arc<dyn crate::Vm>> {
         let mshv_vm_type: VmType = match VmType::try_from(vm_type) {
             Ok(vm_type) => vm_type,
             Err(_) => return Err(hypervisor::HypervisorError::UnsupportedVmType()),
@@ -339,6 +309,14 @@ impl hypervisor::Hypervisor for MshvHypervisor {
                 dirty_log_slots: Arc::new(RwLock::new(HashMap::new())),
                 #[cfg(feature = "sev_snp")]
                 sev_snp_enabled: mshv_vm_type == VmType::Snp,
+                #[cfg(feature = "sev_snp")]
+                host_access_pages: ArcSwap::new(
+                    AtomicBitmap::new(
+                        _mem_size.unwrap_or_default() as usize,
+                        NonZeroUsize::new(HV_PAGE_SIZE).unwrap(),
+                    )
+                    .into(),
+                ),
             }))
         }
 
@@ -350,6 +328,78 @@ impl hypervisor::Hypervisor for MshvHypervisor {
             }))
         }
     }
+}
+
+impl MshvHypervisor {
+    /// Create a hypervisor based on Mshv
+    #[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(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
+///
+/// # Examples
+///
+/// ```
+/// use hypervisor::mshv::MshvHypervisor;
+/// use std::sync::Arc;
+/// let mshv = MshvHypervisor::new().unwrap();
+/// let hypervisor = Arc::new(mshv);
+/// 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 Vm of a specific type using the underlying hypervisor, passing memory size
+    /// Return a hypervisor-agnostic Vm trait object
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use hypervisor::kvm::KvmHypervisor;
+    /// use hypervisor::kvm::KvmVm;
+    /// let hypervisor = KvmHypervisor::new().unwrap();
+    /// let vm = hypervisor.create_vm_with_type(0, 512*1024*1024).unwrap();
+    /// ```
+    fn create_vm_with_type_and_memory(
+        &self,
+        vm_type: u64,
+        #[cfg(feature = "sev_snp")] _mem_size: u64,
+    ) -> hypervisor::Result<Arc<dyn vm::Vm>> {
+        self.create_vm_with_type_and_memory_int(
+            vm_type,
+            #[cfg(feature = "sev_snp")]
+            Some(_mem_size),
+        )
+    }
+
+    fn create_vm_with_type(&self, vm_type: u64) -> hypervisor::Result<Arc<dyn crate::Vm>> {
+        self.create_vm_with_type_and_memory_int(
+            vm_type,
+            #[cfg(feature = "sev_snp")]
+            None,
+        )
+    }
 
     /// Create a mshv vm object and return the object as Vm trait object
     ///
@@ -357,7 +407,7 @@ impl hypervisor::Hypervisor for MshvHypervisor {
     ///
     /// ```
     /// # extern crate hypervisor;
-    /// # use hypervisor::mshv::MshvHypervisor;
+    /// use hypervisor::mshv::MshvHypervisor;
     /// use hypervisor::mshv::MshvVm;
     /// let hypervisor = MshvHypervisor::new().unwrap();
     /// let vm = hypervisor.create_vm().unwrap();
@@ -395,6 +445,20 @@ impl hypervisor::Hypervisor for MshvHypervisor {
     }
 }
 
+#[cfg(feature = "sev_snp")]
+struct Ghcb(*mut svm_ghcb_base);
+
+#[cfg(feature = "sev_snp")]
+// SAFETY: struct is based on GHCB page in the hypervisor,
+// safe to Send across threads
+unsafe impl Send for Ghcb {}
+
+#[cfg(feature = "sev_snp")]
+// SAFETY: struct is based on GHCB page in the hypervisor,
+// safe to Sync across threads as this is only required for Vcpu trait
+// functionally not used anyway
+unsafe impl Sync for Ghcb {}
+
 /// Vcpu struct for Microsoft Hypervisor
 pub struct MshvVcpu {
     fd: VcpuFd,
@@ -405,6 +469,10 @@ pub struct MshvVcpu {
     msrs: Vec<MsrEntry>,
     vm_ops: Option<Arc<dyn vm::VmOps>>,
     vm_fd: Arc<VmFd>,
+    #[cfg(feature = "sev_snp")]
+    ghcb: Option<Ghcb>,
+    #[cfg(feature = "sev_snp")]
+    host_access_pages: ArcSwap<AtomicBitmap>,
 }
 
 /// Implementation of Vcpu trait for Microsoft Hypervisor
@@ -412,8 +480,8 @@ pub struct MshvVcpu {
 /// # Examples
 ///
 /// ```
-/// # use hypervisor::mshv::MshvHypervisor;
-/// # use std::sync::Arc;
+/// use hypervisor::mshv::MshvHypervisor;
+/// use std::sync::Arc;
 /// let mshv = MshvHypervisor::new().unwrap();
 /// let hypervisor = Arc::new(mshv);
 /// let vm = hypervisor.create_vm().expect("new VM fd creation failed");
@@ -732,6 +800,12 @@ impl cpu::Vcpu for MshvVcpu {
                         .map_err(|e| cpu::HypervisorCpuError::RunVcpu(anyhow!(
                             "Unhandled VCPU exit: attribute intercept - couldn't modify host access {}", e
                         )))?;
+                    // Guest is revoking the shared access, so we need to update the bitmap
+                    self.host_access_pages.rcu(|_bitmap| {
+                        let bm = self.host_access_pages.load().as_ref().clone();
+                        bm.reset_addr_range(gpa_start as usize, gfn_count as usize);
+                        bm
+                    });
                     Ok(cpu::VmExit::Ignore)
                 }
                 #[cfg(target_arch = "x86_64")]
@@ -789,6 +863,10 @@ impl cpu::Vcpu for MshvVcpu {
                     let ghcb_msr = svm_ghcb_msr {
                         as_uint64: info.ghcb_msr,
                     };
+                    // Safe to use unwrap, for sev_snp guest we already have the
+                    // GHCB pointer wrapped in the option, otherwise this place is not reached.
+                    let ghcb = self.ghcb.as_ref().unwrap().0;
+
                     // SAFETY: Accessing a union element from bindgen generated bindings.
                     let ghcb_op = unsafe { ghcb_msr.__bindgen_anon_2.ghcb_info() as u32 };
                     // Sanity check on the header fields before handling other operations.
@@ -887,9 +965,7 @@ impl cpu::Vcpu for MshvVcpu {
                         GHCB_INFO_NORMAL => {
                             let exit_code =
                                 info.__bindgen_anon_2.__bindgen_anon_1.sw_exit_code as u32;
-                            // SAFETY: Accessing a union element from bindgen generated bindings.
-                            let pfn = unsafe { ghcb_msr.__bindgen_anon_2.gpa_page_number() };
-                            let ghcb_gpa = pfn << GHCB_INFO_BIT_WIDTH;
+
                             match exit_code {
                                 SVM_EXITCODE_HV_DOORBELL_PAGE => {
                                     let exit_info1 =
@@ -898,10 +974,11 @@ impl cpu::Vcpu for MshvVcpu {
                                         SVM_NAE_HV_DOORBELL_PAGE_GET_PREFERRED => {
                                             // Hypervisor does not have any preference for doorbell GPA.
                                             let preferred_doorbell_gpa: u64 = 0xFFFFFFFFFFFFFFFF;
-                                            self.gpa_write(
-                                                ghcb_gpa + GHCB_SW_EXITINFO2_OFFSET,
-                                                &preferred_doorbell_gpa.to_le_bytes(),
-                                            )?;
+                                            set_svm_field_u64_ptr!(
+                                                ghcb,
+                                                exit_info2,
+                                                preferred_doorbell_gpa
+                                            );
                                         }
                                         SVM_NAE_HV_DOORBELL_PAGE_SET => {
                                             let exit_info2 = info
@@ -928,13 +1005,10 @@ impl cpu::Vcpu for MshvVcpu {
                                                 cpu::HypervisorCpuError::SetRegister(e.into())
                                             })?;
 
-                                            self.gpa_write(
-                                                ghcb_gpa + GHCB_SW_EXITINFO2_OFFSET,
-                                                &exit_info2.to_le_bytes(),
-                                            )?;
+                                            set_svm_field_u64_ptr!(ghcb, exit_info2, exit_info2);
 
                                             // Clear the SW_EXIT_INFO1 register to indicate no error
-                                            self.clear_swexit_info1(ghcb_gpa)?;
+                                            self.clear_swexit_info1()?;
                                         }
                                         SVM_NAE_HV_DOORBELL_PAGE_QUERY => {
                                             let mut reg_assocs = [ hv_register_assoc {
@@ -945,19 +1019,13 @@ impl cpu::Vcpu for MshvVcpu {
                                             // SAFETY: Accessing a union element from bindgen generated bindings.
                                             let doorbell_gpa = unsafe { reg_assocs[0].value.reg64 };
 
-                                            self.gpa_write(
-                                                ghcb_gpa + GHCB_SW_EXITINFO2_OFFSET,
-                                                &doorbell_gpa.to_le_bytes(),
-                                            )?;
+                                            set_svm_field_u64_ptr!(ghcb, exit_info2, doorbell_gpa);
 
                                             // Clear the SW_EXIT_INFO1 register to indicate no error
-                                            self.clear_swexit_info1(ghcb_gpa)?;
+                                            self.clear_swexit_info1()?;
                                         }
                                         SVM_NAE_HV_DOORBELL_PAGE_CLEAR => {
-                                            self.gpa_write(
-                                                ghcb_gpa + GHCB_SW_EXITINFO2_OFFSET,
-                                                &[0; 8],
-                                            )?;
+                                            set_svm_field_u64_ptr!(ghcb, exit_info2, 0);
                                         }
                                         _ => {
                                             panic!(
@@ -991,9 +1059,8 @@ impl cpu::Vcpu for MshvVcpu {
                                     let is_write =
                                         // SAFETY: Accessing a union element from bindgen generated bindings.
                                         unsafe { port_info.__bindgen_anon_1.access_type() == 0 };
-
-                                    let mut data = [0; 8];
-                                    self.gpa_read(ghcb_gpa + GHCB_RAX_OFFSET, &mut data)?;
+                                    // SAFETY: Accessing the field from a mapped address
+                                    let mut data = unsafe { (*ghcb).rax.to_le_bytes() };
 
                                     if is_write {
                                         if let Some(vm_ops) = &self.vm_ops {
@@ -1009,17 +1076,15 @@ impl cpu::Vcpu for MshvVcpu {
                                                     cpu::HypervisorCpuError::RunVcpu(e.into())
                                                 })?;
                                         }
-
-                                        self.gpa_write(ghcb_gpa + GHCB_RAX_OFFSET, &data)?;
+                                        set_svm_field_u64_ptr!(ghcb, rax, u64::from_le_bytes(data));
                                     }
 
                                     // Clear the SW_EXIT_INFO1 register to indicate no error
-                                    self.clear_swexit_info1(ghcb_gpa)?;
+                                    self.clear_swexit_info1()?;
                                 }
                                 SVM_EXITCODE_MMIO_READ => {
                                     let src_gpa =
                                         info.__bindgen_anon_2.__bindgen_anon_1.sw_exit_info1;
-                                    let dst_gpa = info.__bindgen_anon_2.__bindgen_anon_1.sw_scratch;
                                     let data_len =
                                         info.__bindgen_anon_2.__bindgen_anon_1.sw_exit_info2
                                             as usize;
@@ -1032,16 +1097,18 @@ impl cpu::Vcpu for MshvVcpu {
                                             cpu::HypervisorCpuError::RunVcpu(e.into())
                                         })?;
                                     }
-
-                                    self.gpa_write(dst_gpa, &data)?;
+                                    // Copy the data to the shared buffer of the GHCB page
+                                    let mut buffer_data = [0; 8];
+                                    buffer_data[..data_len].copy_from_slice(&data[..data_len]);
+                                    // SAFETY: Updating the value of mapped area
+                                    unsafe { (*ghcb).shared[0] = u64::from_le_bytes(buffer_data) };
 
                                     // Clear the SW_EXIT_INFO1 register to indicate no error
-                                    self.clear_swexit_info1(ghcb_gpa)?;
+                                    self.clear_swexit_info1()?;
                                 }
                                 SVM_EXITCODE_MMIO_WRITE => {
                                     let dst_gpa =
                                         info.__bindgen_anon_2.__bindgen_anon_1.sw_exit_info1;
-                                    let src_gpa = info.__bindgen_anon_2.__bindgen_anon_1.sw_scratch;
                                     let data_len =
                                         info.__bindgen_anon_2.__bindgen_anon_1.sw_exit_info2
                                             as usize;
@@ -1049,7 +1116,10 @@ impl cpu::Vcpu for MshvVcpu {
                                     assert!(data_len <= 0x8);
 
                                     let mut data = vec![0; data_len];
-                                    self.gpa_read(src_gpa, &mut data)?;
+                                    // SAFETY: Accessing data from a mapped address
+                                    let bytes_shared_ghcb =
+                                        unsafe { (*ghcb).shared[0].to_le_bytes() };
+                                    data.copy_from_slice(&bytes_shared_ghcb[..data_len]);
 
                                     if let Some(vm_ops) = &self.vm_ops {
                                         vm_ops.mmio_write(dst_gpa, &data).map_err(|e| {
@@ -1058,7 +1128,7 @@ impl cpu::Vcpu for MshvVcpu {
                                     }
 
                                     // Clear the SW_EXIT_INFO1 register to indicate no error
-                                    self.clear_swexit_info1(ghcb_gpa)?;
+                                    self.clear_swexit_info1()?;
                                 }
                                 SVM_EXITCODE_SNP_GUEST_REQUEST
                                 | SVM_EXITCODE_SNP_EXTENDED_GUEST_REQUEST => {
@@ -1067,15 +1137,16 @@ impl cpu::Vcpu for MshvVcpu {
                                         // We don't support extended guest request, so we just write empty data.
                                         // This matches the behavior of KVM in Linux 6.11.
 
-                                        // Read RAX & RBX from the GHCB.
-                                        let mut data = [0; 8];
-                                        self.gpa_read(ghcb_gpa + GHCB_RAX_OFFSET, &mut data)?;
-                                        let data_gpa = u64::from_le_bytes(data);
-                                        self.gpa_read(ghcb_gpa + GHCB_RBX_OFFSET, &mut data)?;
-                                        let data_npages = u64::from_le_bytes(data);
+                                        // Read RBX from the GHCB.
+                                        // SAFETY: Accessing data from a mapped address
+                                        let data_gpa = unsafe { (*ghcb).rax };
+                                        // SAFETY: Accessing data from a mapped address
+                                        let data_npages = unsafe { (*ghcb).rbx };
 
                                         if data_npages > 0 {
                                             // The certificates are terminated by 24 zero bytes.
+                                            // TODO: Need to check if data_gpa is the address of the shared buffer in the GHCB page
+                                            // in that case we should clear the shared buffer(24 bytes)
                                             self.gpa_write(data_gpa, &[0; 24])?;
                                         }
                                     }
@@ -1096,7 +1167,7 @@ impl cpu::Vcpu for MshvVcpu {
                                         req_gpa, rsp_gpa
                                     );
 
-                                    self.gpa_write(ghcb_gpa + GHCB_SW_EXITINFO2_OFFSET, &[0; 8])?;
+                                    set_svm_field_u64_ptr!(ghcb, exit_info2, 0);
                                 }
                                 SVM_EXITCODE_SNP_AP_CREATION => {
                                     let vmsa_gpa =
@@ -1117,7 +1188,7 @@ impl cpu::Vcpu for MshvVcpu {
                                         .map_err(|e| cpu::HypervisorCpuError::RunVcpu(e.into()))?;
 
                                     // Clear the SW_EXIT_INFO1 register to indicate no error
-                                    self.clear_swexit_info1(ghcb_gpa)?;
+                                    self.clear_swexit_info1()?;
                                 }
                                 _ => panic!(
                                     "GHCB_INFO_NORMAL: Unhandled exit code: {:0x}",
@@ -1495,37 +1566,16 @@ impl MshvVcpu {
     /// Clear SW_EXIT_INFO1 register for SEV-SNP guests.
     ///
     #[cfg(feature = "sev_snp")]
-    fn clear_swexit_info1(
-        &self,
-        ghcb_gpa: u64,
-    ) -> std::result::Result<cpu::VmExit, cpu::HypervisorCpuError> {
+    fn clear_swexit_info1(&self) -> std::result::Result<cpu::VmExit, cpu::HypervisorCpuError> {
         // Clear the SW_EXIT_INFO1 register to indicate no error
-        self.gpa_write(ghcb_gpa + GHCB_SW_EXITINFO1_OFFSET, &[0; 4])?;
+        // Safe to use unwrap, for sev_snp guest we already have the
+        // GHCB pointer wrapped in the option, otherwise this place is not reached.
+        let ghcb = self.ghcb.as_ref().unwrap().0;
+        set_svm_field_u64_ptr!(ghcb, exit_info1, 0);
 
         Ok(cpu::VmExit::Ignore)
     }
 
-    #[cfg(feature = "sev_snp")]
-    fn gpa_read(&self, gpa: u64, data: &mut [u8]) -> cpu::Result<()> {
-        for (gpa, chunk) in (gpa..)
-            .step_by(HV_READ_WRITE_GPA_MAX_SIZE as usize)
-            .zip(data.chunks_mut(HV_READ_WRITE_GPA_MAX_SIZE as usize))
-        {
-            let mut rw_gpa_arg = mshv_bindings::mshv_read_write_gpa {
-                base_gpa: gpa,
-                byte_count: chunk.len() as u32,
-                ..Default::default()
-            };
-            self.fd
-                .gpa_read(&mut rw_gpa_arg)
-                .map_err(|e| cpu::HypervisorCpuError::GpaRead(e.into()))?;
-
-            chunk.copy_from_slice(&rw_gpa_arg.data[..chunk.len()]);
-        }
-
-        Ok(())
-    }
-
     #[cfg(feature = "sev_snp")]
     fn gpa_write(&self, gpa: u64, data: &[u8]) -> cpu::Result<()> {
         for (gpa, chunk) in (gpa..)
@@ -1558,6 +1608,8 @@ pub struct MshvVm {
     dirty_log_slots: Arc<RwLock<HashMap<u64, MshvDirtyLogSlot>>>,
     #[cfg(feature = "sev_snp")]
     sev_snp_enabled: bool,
+    #[cfg(feature = "sev_snp")]
+    host_access_pages: ArcSwap<AtomicBitmap>,
 }
 
 impl MshvVm {
@@ -1580,9 +1632,9 @@ impl MshvVm {
 /// # Examples
 ///
 /// ```
-/// # extern crate hypervisor;
-/// # use hypervisor::mshv::MshvHypervisor;
-/// # use std::sync::Arc;
+/// extern crate hypervisor;
+/// use hypervisor::mshv::MshvHypervisor;
+/// use std::sync::Arc;
 /// let mshv = MshvHypervisor::new().unwrap();
 /// let hypervisor = Arc::new(mshv);
 /// let vm = hypervisor.create_vm().expect("new VM fd creation failed");
@@ -1649,6 +1701,34 @@ impl vm::Vm for MshvVm {
             .fd
             .create_vcpu(id)
             .map_err(|e| vm::HypervisorVmError::CreateVcpu(e.into()))?;
+
+        /* Map the GHCB page to the VMM(root) address space
+         * The map is available after the vcpu creation. This address is mapped
+         * to the overlay ghcb page of the Microsoft Hypervisor, don't have
+         * to worry about the scenario when a guest changes the GHCB mapping.
+         */
+        #[cfg(feature = "sev_snp")]
+        let ghcb = if self.sev_snp_enabled {
+            // SAFETY: Safe to call as VCPU has this map already available upon creation
+            let addr = unsafe {
+                libc::mmap(
+                    std::ptr::null_mut(),
+                    HV_PAGE_SIZE,
+                    libc::PROT_READ | libc::PROT_WRITE,
+                    libc::MAP_SHARED,
+                    vcpu_fd.as_raw_fd(),
+                    MSHV_VP_MMAP_OFFSET_GHCB as i64 * libc::sysconf(libc::_SC_PAGE_SIZE),
+                )
+            };
+            if addr == libc::MAP_FAILED {
+                // No point of continuing, without this mmap VMGEXIT will fail anyway
+                // Return error
+                return Err(vm::HypervisorVmError::MmapToRoot);
+            }
+            Some(Ghcb(addr as *mut svm_ghcb_base))
+        } else {
+            None
+        };
         let vcpu = MshvVcpu {
             fd: vcpu_fd,
             vp_index: id,
@@ -1658,6 +1738,10 @@ impl vm::Vm for MshvVm {
             msrs: self.msrs.clone(),
             vm_ops,
             vm_fd: self.fd.clone(),
+            #[cfg(feature = "sev_snp")]
+            ghcb,
+            #[cfg(feature = "sev_snp")]
+            host_access_pages: ArcSwap::new(self.host_access_pages.load().clone()),
         };
         Ok(Arc::new(vcpu))
     }
@@ -2050,6 +2134,7 @@ impl vm::Vm for MshvVm {
     #[cfg(feature = "sev_snp")]
     fn gain_page_access(&self, gpa: u64, size: u32) -> vm::Result<()> {
         use mshv_ioctls::set_bits;
+        const ONE_GB: usize = 1024 * 1024 * 1024;
 
         if !self.sev_snp_enabled {
             return Ok(());
@@ -2058,7 +2143,25 @@ impl vm::Vm for MshvVm {
         let start_gpfn: u64 = gpa >> PAGE_SHIFT;
         let end_gpfn: u64 = (gpa + size as u64 - 1) >> PAGE_SHIFT;
 
-        let gpas: Vec<u64> = (start_gpfn..=end_gpfn).map(|x| x << PAGE_SHIFT).collect();
+        // Enlarge the bitmap if the PFN is greater than the bitmap length
+        if end_gpfn >= self.host_access_pages.load().as_ref().len() as u64 {
+            self.host_access_pages.rcu(|bitmap| {
+                let mut bm = bitmap.as_ref().clone();
+                bm.enlarge(ONE_GB);
+                bm
+            });
+        }
+
+        let gpas: Vec<u64> = (start_gpfn..=end_gpfn)
+            .filter(|x| {
+                !self
+                    .host_access_pages
+                    .load()
+                    .as_ref()
+                    .is_bit_set(*x as usize)
+            })
+            .map(|x| x << PAGE_SHIFT)
+            .collect();
 
         if !gpas.is_empty() {
             let mut gpa_list = vec_with_array_field::<mshv_modify_gpa_host_access, u64>(gpas.len());
@@ -2081,6 +2184,14 @@ impl vm::Vm for MshvVm {
             self.fd
                 .modify_gpa_host_access(&gpa_list[0])
                 .map_err(|e| vm::HypervisorVmError::ModifyGpaHostAccess(e.into()))?;
+
+            for acquired_gpa in gpas {
+                self.host_access_pages.rcu(|bitmap| {
+                    let bm = bitmap.clone();
+                    bm.set_bit((acquired_gpa >> PAGE_SHIFT) as usize);
+                    bm
+                });
+            }
         }
 
         Ok(())

hypervisor/src/mshv/snp_constants.rs

@@ -16,10 +16,3 @@ pub const ECDSA_SIG_X_COMPONENT_END: usize =
 pub const ECDSA_SIG_Y_COMPONENT_START: usize = ECDSA_SIG_X_COMPONENT_END;
 pub const ECDSA_SIG_Y_COMPONENT_END: usize =
     ECDSA_SIG_X_COMPONENT_END + ECDSA_SIG_Y_COMPONENT_SIZE_IN_BYTES;
-
-// These constants are derived from GHCB spec Sect. 2.6 Table 3 GHCB Layout
-// Link: https://www.amd.com/content/dam/amd/en/documents/epyc-technical-docs/specifications/56421.pdf
-pub const GHCB_RAX_OFFSET: u64 = 0x01F8;
-pub const GHCB_RBX_OFFSET: u64 = 0x0318;
-pub const GHCB_SW_EXITINFO1_OFFSET: u64 = 0x398;
-pub const GHCB_SW_EXITINFO2_OFFSET: u64 = 0x3A0;

hypervisor/src/vm.rs

@@ -1,3 +1,5 @@
+// Copyright © 2024 Institute of Software, CAS. All rights reserved.
+//
 // Copyright © 2019 Intel Corporation
 //
 // SPDX-License-Identifier: Apache-2.0 OR BSD-3-Clause
@@ -12,7 +14,7 @@ use std::any::Any;
 #[cfg(target_arch = "x86_64")]
 use std::fs::File;
 use std::sync::Arc;
-#[cfg(target_arch = "aarch64")]
+#[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
 use std::sync::Mutex;
 
 #[cfg(feature = "sev_snp")]
@@ -24,6 +26,8 @@ use vmm_sys_util::eventfd::EventFd;
 use crate::aarch64::VcpuInit;
 #[cfg(target_arch = "aarch64")]
 use crate::arch::aarch64::gic::{Vgic, VgicConfig};
+#[cfg(target_arch = "riscv64")]
+use crate::arch::riscv64::aia::{Vaia, VaiaConfig};
 #[cfg(feature = "tdx")]
 use crate::arch::x86::CpuIdEntry;
 use crate::cpu::Vcpu;
@@ -224,6 +228,11 @@ pub enum HypervisorVmError {
     #[error("Failed to create Vgic: {0}")]
     CreateVgic(#[source] anyhow::Error),
     ///
+    /// Create Vaia error
+    ///
+    #[error("Failed to create Vaia: {0}")]
+    CreateVaia(#[source] anyhow::Error),
+    ///
     /// Import isolated pages error
     ///
     #[error("Failed to import isolated pages: {0}")]
@@ -242,6 +251,12 @@ pub enum HypervisorVmError {
     #[cfg(feature = "sev_snp")]
     #[error("Failed to modify GPA host access: {0}")]
     ModifyGpaHostAccess(#[source] anyhow::Error),
+    ///
+    /// Failed to mmap
+    ///
+    #[cfg(feature = "sev_snp")]
+    #[error("Failed to mmap:")]
+    MmapToRoot,
 }
 ///
 /// Result type for returning from a function
@@ -293,6 +308,7 @@ pub trait Vm: Send + Sync + Any {
     #[cfg(target_arch = "x86_64")]
     /// Sets the address of the three-page region in the VM's address space.
     fn set_tss_address(&self, offset: usize) -> Result<()>;
+    #[cfg(not(target_arch = "riscv64"))]
     /// Creates an in-kernel interrupt controller.
     fn create_irq_chip(&self) -> Result<()>;
     /// Registers an event that will, when signaled, trigger the `gsi` IRQ.
@@ -303,6 +319,8 @@ pub trait Vm: Send + Sync + Any {
     fn create_vcpu(&self, id: u8, vm_ops: Option<Arc<dyn VmOps>>) -> Result<Arc<dyn Vcpu>>;
     #[cfg(target_arch = "aarch64")]
     fn create_vgic(&self, config: VgicConfig) -> Result<Arc<Mutex<dyn Vgic>>>;
+    #[cfg(target_arch = "riscv64")]
+    fn create_vaia(&self, config: VaiaConfig) -> Result<Arc<Mutex<dyn Vaia>>>;
 
     /// Registers an event to be signaled whenever a certain address is written to.
     fn register_ioevent(

net_util/Cargo.toml

@@ -6,8 +6,8 @@ version = "0.1.0"
 
 [dependencies]
 epoll = "4.3.3"
-getrandom = "0.2.14"
-libc = "0.2.158"
+getrandom = "0.2.15"
+libc = "0.2.167"
 log = "0.4.22"
 net_gen = { path = "../net_gen" }
 rate_limiter = { path = "../rate_limiter" }

net_util/src/open_tap.rs

@@ -75,9 +75,7 @@ pub fn open_tap(
     let mut ifname: String = String::new();
     let vnet_hdr_size = vnet_hdr_len() as i32;
     // Check if the given interface exists before we create it.
-    let tap_existed = if_name.map_or(false, |n| {
-        Path::new(&format!("/sys/class/net/{n}")).exists()
-    });
+    let tap_existed = if_name.is_some_and(|n| Path::new(&format!("/sys/class/net/{n}")).exists());
 
     // In case the tap interface already exists, check if the number of
     // queues is appropriate. The tap might not support multiqueue while

net_util/src/queue_pair.rs

@@ -468,7 +468,7 @@ impl NetQueuePair {
         let rate_limit_reached = self
             .rx_rate_limiter
             .as_ref()
-            .map_or(false, |r| r.is_blocked());
+            .is_some_and(|r| r.is_blocked());
 
         // Stop listening on the `RX_TAP_EVENT` when:
         // 1) there is no available describes, or

pci/Cargo.toml

@@ -10,10 +10,10 @@ kvm = ["vfio-ioctls/kvm"]
 mshv = ["vfio-ioctls/mshv"]
 
 [dependencies]
-anyhow = "1.0.87"
+anyhow = "1.0.94"
 byteorder = "1.5.0"
 hypervisor = { path = "../hypervisor" }
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 serde = { version = "1.0.208", features = ["derive"] }
 thiserror = "1.0.62"

pci/src/bus.rs

@@ -125,19 +125,16 @@ impl PciBus {
     pub fn register_mapping(
         &self,
         dev: Arc<dyn BusDeviceSync>,
-        #[cfg(target_arch = "x86_64")] io_bus: &Bus,
+        io_bus: &Bus,
         mmio_bus: &Bus,
         bars: Vec<PciBarConfiguration>,
     ) -> Result<()> {
         for bar in bars {
             match bar.region_type() {
                 PciBarRegionType::IoRegion => {
-                    #[cfg(target_arch = "x86_64")]
                     io_bus
                         .insert(dev.clone(), bar.addr(), bar.size())
                         .map_err(PciRootError::PioInsert)?;
-                    #[cfg(not(target_arch = "x86_64"))]
-                    error!("I/O region is not supported");
                 }
                 PciBarRegionType::Memory32BitRegion | PciBarRegionType::Memory64BitRegion => {
                     mmio_bus

pci/src/vfio.rs

@@ -23,7 +23,7 @@ use vfio_ioctls::{
 use vm_allocator::page_size::{
     align_page_size_down, align_page_size_up, is_4k_aligned, is_4k_multiple, is_page_size_aligned,
 };
-use vm_allocator::{AddressAllocator, SystemAllocator};
+use vm_allocator::{AddressAllocator, MemorySlotAllocator, SystemAllocator};
 use vm_device::dma_mapping::ExternalDmaMapping;
 use vm_device::interrupt::{
     InterruptIndex, InterruptManager, InterruptSourceGroup, MsiIrqGroupConfig,
@@ -713,11 +713,7 @@ impl VfioCommon {
 
             let bar_addr = match region_type {
                 PciBarRegionType::IoRegion => {
-                    #[cfg(not(target_arch = "x86_64"))]
-                    unimplemented!();
-
                     // The address needs to be 4 bytes aligned.
-                    #[cfg(target_arch = "x86_64")]
                     allocator
                         .lock()
                         .unwrap()
@@ -795,10 +791,7 @@ impl VfioCommon {
         for region in self.mmio_regions.iter() {
             match region.type_ {
                 PciBarRegionType::IoRegion => {
-                    #[cfg(target_arch = "x86_64")]
                     allocator.free_io_addresses(region.start, region.length);
-                    #[cfg(not(target_arch = "x86_64"))]
-                    error!("I/O region is not supported");
                 }
                 PciBarRegionType::Memory32BitRegion => {
                     mmio32_allocator.free(region.start, region.length);
@@ -1416,7 +1409,7 @@ pub struct VfioPciDevice {
     container: Arc<VfioContainer>,
     common: VfioCommon,
     iommu_attached: bool,
-    memory_slot: Arc<dyn Fn() -> u32 + Send + Sync>,
+    memory_slot_allocator: MemorySlotAllocator,
 }
 
 impl VfioPciDevice {
@@ -1431,7 +1424,7 @@ impl VfioPciDevice {
         legacy_interrupt_group: Option<Arc<dyn InterruptSourceGroup>>,
         iommu_attached: bool,
         bdf: PciBdf,
-        memory_slot: Arc<dyn Fn() -> u32 + Send + Sync>,
+        memory_slot_allocator: MemorySlotAllocator,
         snapshot: Option<Snapshot>,
         x_nv_gpudirect_clique: Option<u8>,
     ) -> Result<Self, VfioPciError> {
@@ -1457,7 +1450,7 @@ impl VfioPciDevice {
             container,
             common,
             iommu_attached,
-            memory_slot,
+            memory_slot_allocator,
         };
 
         Ok(vfio_pci_device)
@@ -1635,7 +1628,7 @@ impl VfioPciDevice {
                     }
 
                     let user_memory_region = UserMemoryRegion {
-                        slot: (self.memory_slot)(),
+                        slot: self.memory_slot_allocator.next_memory_slot(),
                         start: region.start.0 + area.offset,
                         size: area.size,
                         host_addr: host_addr as u64,
@@ -1699,6 +1692,9 @@ impl VfioPciDevice {
                     error!("Could not remove the userspace memory region: {}", e);
                 }
 
+                self.memory_slot_allocator
+                    .free_memory_slot(user_memory_region.slot);
+
                 // SAFETY: FFI call with correct arguments
                 let ret = unsafe {
                     libc::munmap(

pci/src/vfio_user.rs

@@ -13,7 +13,7 @@ use thiserror::Error;
 use vfio_bindings::bindings::vfio::*;
 use vfio_ioctls::VfioIrq;
 use vfio_user::{Client, Error as VfioUserError};
-use vm_allocator::{AddressAllocator, SystemAllocator};
+use vm_allocator::{AddressAllocator, MemorySlotAllocator, SystemAllocator};
 use vm_device::dma_mapping::ExternalDmaMapping;
 use vm_device::interrupt::{InterruptManager, InterruptSourceGroup, MsiIrqGroupConfig};
 use vm_device::{BusDevice, Resource};
@@ -35,7 +35,7 @@ pub struct VfioUserPciDevice {
     vm: Arc<dyn hypervisor::Vm>,
     client: Arc<Mutex<Client>>,
     common: VfioCommon,
-    memory_slot: Arc<dyn Fn() -> u32 + Send + Sync>,
+    memory_slot_allocator: MemorySlotAllocator,
 }
 
 #[derive(Error, Debug)]
@@ -74,7 +74,7 @@ impl VfioUserPciDevice {
         msi_interrupt_manager: Arc<dyn InterruptManager<GroupConfig = MsiIrqGroupConfig>>,
         legacy_interrupt_group: Option<Arc<dyn InterruptSourceGroup>>,
         bdf: PciBdf,
-        memory_slot: Arc<dyn Fn() -> u32 + Send + Sync>,
+        memory_slot_allocator: MemorySlotAllocator,
         snapshot: Option<Snapshot>,
     ) -> Result<Self, VfioUserPciDeviceError> {
         let resettable = client.lock().unwrap().resettable();
@@ -106,7 +106,7 @@ impl VfioUserPciDevice {
             vm: vm.clone(),
             client,
             common,
-            memory_slot,
+            memory_slot_allocator,
         })
     }
 
@@ -178,7 +178,7 @@ impl VfioUserPciDevice {
                     }
 
                     let user_memory_region = UserMemoryRegion {
-                        slot: (self.memory_slot)(),
+                        slot: self.memory_slot_allocator.next_memory_slot(),
                         start: mmio_region.start.0 + s.offset,
                         size: s.size,
                         host_addr: host_addr as u64,
@@ -222,6 +222,9 @@ impl VfioUserPciDevice {
                     error!("Could not remove the userspace memory region: {}", e);
                 }
 
+                self.memory_slot_allocator
+                    .free_memory_slot(user_memory_region.slot);
+
                 // Remove mmaps
                 // SAFETY: FFI call with correct arguments
                 let ret = unsafe {

rate_limiter/Cargo.toml

@@ -5,7 +5,7 @@ version = "0.1.0"
 
 [dependencies]
 epoll = "4.3.3"
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 thiserror = "1.0.62"
 vmm-sys-util = { workspace = true }

release-notes.md

@@ -1,25 +1,30 @@
+- [v43.0](#v430)
+    - [Live Migration over TCP Connections](#live-migration-over-tcp-connections)
+    - [Notable Performance Improvements](#notable-performance-improvements)
+    - [Notable Bug Fixes](#notable-bug-fixes)
+    - [Contributors](#contributors)
 - [v42.0](#v420)
     - [SVE/SVE2 Support on AArch64](#svesve2-support-on-aarch64)
-    - [Notable Bug Fixes](#notable-bug-fixes)
+    - [Notable Bug Fixes](#notable-bug-fixes-1)
     - [Sponsorships](#sponsorships)
-    - [Contributors](#contributors)
+    - [Contributors](#contributors-1)
 - [v41.0](#v410)
     - [Experimental "Pvmemcontrol" Support](#experimental-pvmemcontrol-support)
     - [Sandboxing With Landlock Support](#sandboxing-with-landlock-support)
-    - [Notable Performance Improvements](#notable-performance-improvements)
-    - [Notable Bug Fixes](#notable-bug-fixes-1)
-    - [Contributors](#contributors-1)
-- [v40.0](#v400)
-    - [Support for Restoring File Descriptor Backed Network Devices](#support-for-restoring-file-descriptor-backed-network-devices)
+    - [Notable Performance Improvements](#notable-performance-improvements-1)
     - [Notable Bug Fixes](#notable-bug-fixes-2)
     - [Contributors](#contributors-2)
+- [v40.0](#v400)
+    - [Support for Restoring File Descriptor Backed Network Devices](#support-for-restoring-file-descriptor-backed-network-devices)
+    - [Notable Bug Fixes](#notable-bug-fixes-3)
+    - [Contributors](#contributors-3)
 - [v39.0](#v390)
     - [Variable Sizing of PCI Apertures for Segments](#variable-sizing-of-pci-apertures-for-segments)
     - [Direct Booting with bzImages](#direct-booting-with-bzimages)
     - [Support for NVIDIA GPUDirect P2P Support](#support-for-nvidia-gpudirect-p2p-support)
     - [Guest NMI Injection Support](#guest-nmi-injection-support)
-    - [Notable Bug Fixes](#notable-bug-fixes-3)
-    - [Contributors](#contributors-3)
+    - [Notable Bug Fixes](#notable-bug-fixes-4)
+    - [Contributors](#contributors-4)
 - [v38.0](#v380)
     - [Group Rate Limiter on Block Devices](#group-rate-limiter-on-block-devices)
     - [CPU Pinning Support for Block Device Worker Thread](#cpu-pinning-support-for-block-device-worker-thread)
@@ -27,16 +32,16 @@
     - [New 'debug-console' Device](#new-debug-console-device)
     - [Improved VFIO Device Support](#improved-vfio-device-support)
     - [Extended CPU Affinity Support](#extended-cpu-affinity-support)
-    - [Notable Bug Fixes](#notable-bug-fixes-4)
-    - [Contributors](#contributors-4)
+    - [Notable Bug Fixes](#notable-bug-fixes-5)
+    - [Contributors](#contributors-5)
 - [v37.0](#v370)
     - [Long Term Support (LTS) Release](#long-term-support-lts-release)
     - [Multiple PCI segments Support for 32-bit VFIO devices](#multiple-pci-segments-support-for-32-bit-vfio-devices)
     - [Configurable Named TAP Devices](#configurable-named-tap-devices)
     - [TTY Output from Both Serial Device and Virtio Console](#tty-output-from-both-serial-device-and-virtio-console)
     - [Faster VM Restoration from Snapshots](#faster-vm-restoration-from-snapshots)
-    - [Notable Bug Fixes](#notable-bug-fixes-5)
-    - [Contributors](#contributors-5)
+    - [Notable Bug Fixes](#notable-bug-fixes-6)
+    - [Contributors](#contributors-6)
 - [v36.0](#v360)
     - [Command Line Changes](#command-line-changes)
     - [Enabled Features Reported via API Endpoint and CLI](#enabled-features-reported-via-api-endpoint-and-cli)
@@ -45,31 +50,31 @@
     - [Unix Socket Backend for Serial Port](#unix-socket-backend-for-serial-port)
     - [AIO Backend for Block Devices](#aio-backend-for-block-devices)
     - [Documentation Improvements](#documentation-improvements)
-    - [Notable Bug Fixes](#notable-bug-fixes-6)
-    - [Contributors](#contributors-6)
+    - [Notable Bug Fixes](#notable-bug-fixes-7)
+    - [Contributors](#contributors-7)
 - [v35.0](#v350)
     - [`virtio-vsock` Support for Linux Guest Kernel v6.3+](#virtio-vsock-support-for-linux-guest-kernel-v63)
     - [User Specified Serial Number for `virtio-block`](#user-specified-serial-number-for-virtio-block)
     - [vCPU TSC Frequency Included in Migration State](#vcpu-tsc-frequency-included-in-migration-state)
-    - [Notable Bug Fixes](#notable-bug-fixes-7)
-    - [Contributors](#contributors-7)
+    - [Notable Bug Fixes](#notable-bug-fixes-8)
+    - [Contributors](#contributors-8)
 - [v34.0](#v340)
     - [Paravirtualised Panic Device Support](#paravirtualised-panic-device-support)
     - [Improvements to VM Core Dump](#improvements-to-vm-core-dump)
     - [QCOW2 Support for Backing Files](#qcow2-support-for-backing-files)
     - [Minimum Host Kernel Bump](#minimum-host-kernel-bump)
-    - [Notable Bug Fixes](#notable-bug-fixes-8)
-    - [Contributors](#contributors-8)
+    - [Notable Bug Fixes](#notable-bug-fixes-9)
+    - [Contributors](#contributors-9)
 - [v33.0](#v330)
     - [D-Bus based API](#d-bus-based-api)
     - [Expose Host CPU Cache Details for AArch64](#expose-host-cpu-cache-details-for-aarch64)
-    - [Notable Bug Fixes](#notable-bug-fixes-9)
-    - [Contributors](#contributors-9)
+    - [Notable Bug Fixes](#notable-bug-fixes-10)
+    - [Contributors](#contributors-10)
 - [v32.0](#v320)
     - [Increased PCI Segment Limit](#increased-pci-segment-limit)
     - [API Changes](#api-changes)
-    - [Notable Bug Fixes](#notable-bug-fixes-10)
-    - [Contributors](#contributors-10)
+    - [Notable Bug Fixes](#notable-bug-fixes-11)
+    - [Contributors](#contributors-11)
 - [v31.1](#v311)
 - [v31.0](#v310)
     - [Update to Latest `acpi_tables`](#update-to-latest-acpi_tables)
@@ -77,15 +82,15 @@
     - [Improvements on Console `SIGWINCH` Handler](#improvements-on-console-sigwinch-handler)
     - [Remove Directory Support from `MemoryZoneConfig::file`](#remove-directory-support-from-memoryzoneconfigfile)
     - [Documentation Improvements](#documentation-improvements-1)
-    - [Notable Bug Fixes](#notable-bug-fixes-11)
-    - [Contributors](#contributors-11)
+    - [Notable Bug Fixes](#notable-bug-fixes-12)
+    - [Contributors](#contributors-12)
 - [v30.0](#v300)
     - [Command Line Changes for Reduced Binary Size](#command-line-changes-for-reduced-binary-size)
     - [Basic vfio-user Server Support](#basic-vfio-user-server-support)
     - [Heap Profiling Support](#heap-profiling-support)
     - [Documentation Improvements](#documentation-improvements-2)
-    - [Notable Bug Fixes](#notable-bug-fixes-12)
-    - [Contributors](#contributors-12)
+    - [Notable Bug Fixes](#notable-bug-fixes-13)
+    - [Contributors](#contributors-13)
 - [v28.2](#v282)
 - [v29.0](#v290)
     - [Release Binary Supports Both MSHV and KVM](#release-binary-supports-both-mshv-and-kvm)
@@ -95,10 +100,10 @@
     - [`AArch64` Documentation Integration](#aarch64-documentation-integration)
     - [`virtio-block` Counters Enhancement](#virtio-block-counters-enhancement)
     - [TCP Offload Control](#tcp-offload-control)
-    - [Notable Bug Fixes](#notable-bug-fixes-13)
+    - [Notable Bug Fixes](#notable-bug-fixes-14)
     - [Removals](#removals)
     - [Deprecations](#deprecations)
-    - [Contributors](#contributors-13)
+    - [Contributors](#contributors-14)
 - [v28.1](#v281)
 - [v28.0](#v280)
     - [Community Engagement (Reminder)](#community-engagement-reminder)
@@ -106,9 +111,9 @@
     - [Virtualised TPM Support](#virtualised-tpm-support)
     - [Transparent Huge Page Support](#transparent-huge-page-support)
     - [README Quick Start Improved](#readme-quick-start-improved)
-    - [Notable Bug Fixes](#notable-bug-fixes-14)
+    - [Notable Bug Fixes](#notable-bug-fixes-15)
     - [Removals](#removals-1)
-    - [Contributors](#contributors-14)
+    - [Contributors](#contributors-15)
 - [v27.0](#v270)
     - [Community Engagement](#community-engagement)
     - [Prebuilt Packages](#prebuilt-packages)
@@ -117,41 +122,41 @@
     - [Simplified Build Feature Flags](#simplified-build-feature-flags)
     - [Asynchronous Kernel Loading](#asynchronous-kernel-loading)
     - [GDB Support for AArch64](#gdb-support-for-aarch64)
-    - [Notable Bug Fixes](#notable-bug-fixes-15)
+    - [Notable Bug Fixes](#notable-bug-fixes-16)
     - [Deprecations](#deprecations-1)
-    - [Contributors](#contributors-15)
+    - [Contributors](#contributors-16)
 - [v26.0](#v260)
     - [SMBIOS Improvements via `--platform`](#smbios-improvements-via---platform)
     - [Unified Binary MSHV and KVM Support](#unified-binary-mshv-and-kvm-support)
-    - [Notable Bug Fixes](#notable-bug-fixes-16)
+    - [Notable Bug Fixes](#notable-bug-fixes-17)
     - [Deprecations](#deprecations-2)
     - [Removals](#removals-2)
-    - [Contributors](#contributors-16)
+    - [Contributors](#contributors-17)
 - [v25.0](#v250)
     - [`ch-remote` Improvements](#ch-remote-improvements-1)
     - [VM "Coredump" Support](#vm-coredump-support)
-    - [Notable Bug Fixes](#notable-bug-fixes-17)
+    - [Notable Bug Fixes](#notable-bug-fixes-18)
     - [Removals](#removals-3)
-    - [Contributors](#contributors-17)
+    - [Contributors](#contributors-18)
 - [v24.0](#v240)
     - [Bypass Mode for `virtio-iommu`](#bypass-mode-for-virtio-iommu)
     - [Ensure Identifiers Uniqueness](#ensure-identifiers-uniqueness)
     - [Sparse Mmap support](#sparse-mmap-support)
     - [Expose Platform Serial Number](#expose-platform-serial-number)
-    - [Notable Bug Fixes](#notable-bug-fixes-18)
+    - [Notable Bug Fixes](#notable-bug-fixes-19)
     - [Notable Improvements](#notable-improvements)
     - [Deprecations](#deprecations-3)
     - [New on the Website](#new-on-the-website)
-    - [Contributors](#contributors-18)
+    - [Contributors](#contributors-19)
 - [v23.1](#v231)
 - [v23.0](#v230)
     - [vDPA Support](#vdpa-support)
     - [Updated OS Support list](#updated-os-support-list)
     - [`AArch64` Memory Map Improvements](#aarch64-memory-map-improvements)
     - [`AMX` Support](#amx-support)
-    - [Notable Bug Fixes](#notable-bug-fixes-19)
+    - [Notable Bug Fixes](#notable-bug-fixes-20)
     - [Deprecations](#deprecations-4)
-    - [Contributors](#contributors-19)
+    - [Contributors](#contributors-20)
 - [v22.1](#v221)
 - [v22.0](#v220)
     - [GDB Debug Stub Support](#gdb-debug-stub-support)
@@ -162,13 +167,13 @@
     - [PMU Support for AArch64](#pmu-support-for-aarch64)
     - [Documentation Under CC-BY-4.0 License](#documentation-under-cc-by-40-license)
     - [Deprecation of "Classic" `virtiofsd`](#deprecation-of-classic-virtiofsd)
-    - [Notable Bug Fixes](#notable-bug-fixes-20)
-    - [Contributors](#contributors-20)
+    - [Notable Bug Fixes](#notable-bug-fixes-21)
+    - [Contributors](#contributors-21)
 - [v21.0](#v210)
     - [Efficient Local Live Migration (for Live Upgrade)](#efficient-local-live-migration-for-live-upgrade)
     - [Recommended Kernel is Now 5.15](#recommended-kernel-is-now-515)
-    - [Notable Bug fixes](#notable-bug-fixes-21)
-    - [Contributors](#contributors-21)
+    - [Notable Bug fixes](#notable-bug-fixes-22)
+    - [Contributors](#contributors-22)
 - [v20.2](#v202)
 - [v20.1](#v201)
 - [v20.0](#v200)
@@ -177,8 +182,8 @@
     - [Improved VFIO support](#improved-vfio-support)
     - [Safer code](#safer-code)
     - [Extended documentation](#extended-documentation)
-    - [Notable bug fixes](#notable-bug-fixes-22)
-    - [Contributors](#contributors-22)
+    - [Notable bug fixes](#notable-bug-fixes-23)
+    - [Contributors](#contributors-23)
 - [v19.0](#v190)
     - [Improved PTY handling for serial and `virtio-console`](#improved-pty-handling-for-serial-and-virtio-console)
     - [PCI boot time optimisations](#pci-boot-time-optimisations)
@@ -186,8 +191,8 @@
     - [Live migration enhancements](#live-migration-enhancements)
     - [`virtio-mem` support with `vfio-user`](#virtio-mem-support-with-vfio-user)
     - [AArch64 for `virtio-iommu`](#aarch64-for-virtio-iommu)
-    - [Notable bug fixes](#notable-bug-fixes-23)
-    - [Contributors](#contributors-23)
+    - [Notable bug fixes](#notable-bug-fixes-24)
+    - [Contributors](#contributors-24)
 - [v18.0](#v180)
     - [Experimental User Device (`vfio-user`) support](#experimental-user-device-vfio-user-support)
     - [Migration support for `vhost-user` devices](#migration-support-for-vhost-user-devices)
@@ -197,23 +202,23 @@
     - [Live migration on MSHV hypervisor](#live-migration-on-mshv-hypervisor)
     - [AArch64 CPU topology support](#aarch64-cpu-topology-support)
     - [Power button support on AArch64](#power-button-support-on-aarch64)
-    - [Notable bug fixes](#notable-bug-fixes-24)
-    - [Contributors](#contributors-24)
+    - [Notable bug fixes](#notable-bug-fixes-25)
+    - [Contributors](#contributors-25)
 - [v17.0](#v170)
     - [ARM64 NUMA support using ACPI](#arm64-numa-support-using-acpi)
     - [`Seccomp` support for MSHV backend](#seccomp-support-for-mshv-backend)
     - [Hotplug of `macvtap` devices](#hotplug-of-macvtap-devices)
     - [Improved SGX support](#improved-sgx-support)
     - [Inflight tracking for `vhost-user` devices](#inflight-tracking-for-vhost-user-devices)
-    - [Notable bug fixes](#notable-bug-fixes-25)
-    - [Contributors](#contributors-25)
+    - [Notable bug fixes](#notable-bug-fixes-26)
+    - [Contributors](#contributors-26)
 - [v16.0](#v160)
     - [Improved live migration support](#improved-live-migration-support)
     - [Improved `vhost-user` support](#improved-vhost-user-support)
     - [ARM64 ACPI and UEFI support](#arm64-acpi-and-uefi-support)
-    - [Notable bug fixes](#notable-bug-fixes-26)
+    - [Notable bug fixes](#notable-bug-fixes-27)
     - [Removed functionality](#removed-functionality)
-    - [Contributors](#contributors-26)
+    - [Contributors](#contributors-27)
 - [v15.0](#v150)
     - [Version numbering and stability guarantees](#version-numbering-and-stability-guarantees)
     - [Network device rate limiting](#network-device-rate-limiting)
@@ -221,7 +226,7 @@
     - [`--api-socket` supports file descriptor parameter](#--api-socket-supports-file-descriptor-parameter)
     - [Bug fixes](#bug-fixes)
     - [Deprecations](#deprecations-5)
-    - [Contributors](#contributors-27)
+    - [Contributors](#contributors-28)
 - [v0.14.1](#v0141)
 - [v0.14.0](#v0140)
     - [Structured event monitoring](#structured-event-monitoring)
@@ -231,7 +236,7 @@
     - [PTY control for serial and `virtio-console`](#pty-control-for-serial-and-virtio-console)
     - [Block device rate limiting](#block-device-rate-limiting)
     - [Deprecations](#deprecations-6)
-    - [Contributors](#contributors-28)
+    - [Contributors](#contributors-29)
 - [v0.13.0](#v0130)
     - [Wider VFIO device support](#wider-vfio-device-support)
     - [Improved huge page support](#improved-huge-page-support)
@@ -239,13 +244,13 @@
     - [VHD disk image support](#vhd-disk-image-support)
     - [Improved Virtio device threading](#improved-virtio-device-threading)
     - [Clean shutdown support via synthetic power button](#clean-shutdown-support-via-synthetic-power-button)
-    - [Contributors](#contributors-29)
+    - [Contributors](#contributors-30)
 - [v0.12.0](#v0120)
     - [ARM64 enhancements](#arm64-enhancements)
     - [Removal of `vhost-user-net` and `vhost-user-block` self spawning](#removal-of-vhost-user-net-and-vhost-user-block-self-spawning)
     - [Migration of `vhost-user-fs` backend](#migration-of-vhost-user-fs-backend)
     - [Enhanced "info" API](#enhanced-info-api)
-    - [Contributors](#contributors-30)
+    - [Contributors](#contributors-31)
 - [v0.11.0](#v0110)
     - [`io_uring` support by default for `virtio-block`](#io_uring-support-by-default-for-virtio-block)
     - [Windows Guest Support](#windows-guest-support)
@@ -257,15 +262,15 @@
     - [Default Log Level Changed](#default-log-level-changed)
     - [New `--balloon` Parameter Added](#new---balloon-parameter-added)
     - [Experimental `virtio-watchdog` Support](#experimental-virtio-watchdog-support)
-    - [Notable Bug Fixes](#notable-bug-fixes-27)
-    - [Contributors](#contributors-31)
+    - [Notable Bug Fixes](#notable-bug-fixes-28)
+    - [Contributors](#contributors-32)
 - [v0.10.0](#v0100)
     - [`virtio-block` Support for Multiple Descriptors](#virtio-block-support-for-multiple-descriptors)
     - [Memory Zones](#memory-zones)
     - [`Seccomp` Sandbox Improvements](#seccomp-sandbox-improvements)
     - [Preliminary KVM HyperV Emulation Control](#preliminary-kvm-hyperv-emulation-control)
-    - [Notable Bug Fixes](#notable-bug-fixes-28)
-    - [Contributors](#contributors-32)
+    - [Notable Bug Fixes](#notable-bug-fixes-29)
+    - [Contributors](#contributors-33)
 - [v0.9.0](#v090)
     - [`io_uring` Based Block Device Support](#io_uring-based-block-device-support)
     - [Block and Network Device Statistics](#block-and-network-device-statistics)
@@ -278,17 +283,17 @@
     - [Enhancements to ARM64 Support](#enhancements-to-arm64-support)
     - [Intel SGX Support](#intel-sgx-support)
     - [`Seccomp` Sandbox Improvements](#seccomp-sandbox-improvements-1)
-    - [Notable Bug Fixes](#notable-bug-fixes-29)
-    - [Contributors](#contributors-33)
+    - [Notable Bug Fixes](#notable-bug-fixes-30)
+    - [Contributors](#contributors-34)
 - [v0.8.0](#v080)
     - [Experimental Snapshot and Restore Support](#experimental-snapshot-and-restore-support)
     - [Experimental ARM64 Support](#experimental-arm64-support)
     - [Support for Using 5-level Paging in Guests](#support-for-using-5-level-paging-in-guests)
     - [Virtio Device Interrupt Suppression for Network Devices](#virtio-device-interrupt-suppression-for-network-devices)
     - [`vhost_user_fs` Improvements](#vhost_user_fs-improvements)
-    - [Notable Bug Fixes](#notable-bug-fixes-30)
+    - [Notable Bug Fixes](#notable-bug-fixes-31)
     - [Command Line and API Changes](#command-line-and-api-changes)
-    - [Contributors](#contributors-34)
+    - [Contributors](#contributors-35)
 - [v0.7.0](#v070)
     - [Block, Network, Persistent Memory (PMEM), VirtioFS and Vsock hotplug](#block-network-persistent-memory-pmem-virtiofs-and-vsock-hotplug)
     - [Alternative `libc` Support](#alternative-libc-support)
@@ -298,14 +303,14 @@
     - [`Seccomp` Sandboxing](#seccomp-sandboxing)
     - [Updated Distribution Support](#updated-distribution-support)
     - [Command Line and API Changes](#command-line-and-api-changes-1)
-    - [Contributors](#contributors-35)
+    - [Contributors](#contributors-36)
 - [v0.6.0](#v060)
     - [Directly Assigned Devices Hotplug](#directly-assigned-devices-hotplug)
     - [Shared Filesystem Improvements](#shared-filesystem-improvements)
     - [Block and Networking IO Self Offloading](#block-and-networking-io-self-offloading)
     - [Command Line Interface](#command-line-interface)
     - [PVH Boot](#pvh-boot)
-    - [Contributors](#contributors-36)
+    - [Contributors](#contributors-37)
 - [v0.5.1](#v051)
 - [v0.5.0](#v050)
     - [Virtual Machine Dynamic Resizing](#virtual-machine-dynamic-resizing)
@@ -313,7 +318,7 @@
     - [New Interrupt Management Framework](#new-interrupt-management-framework)
     - [Development Tools](#development-tools)
     - [Kata Containers Integration](#kata-containers-integration)
-    - [Contributors](#contributors-37)
+    - [Contributors](#contributors-38)
 - [v0.4.0](#v040)
     - [Dynamic virtual CPUs addition](#dynamic-virtual-cpus-addition)
     - [Programmatic firmware tables generation](#programmatic-firmware-tables-generation)
@@ -322,7 +327,7 @@
     - [Userspace IOAPIC by default](#userspace-ioapic-by-default)
     - [PCI BAR reprogramming](#pci-bar-reprogramming)
     - [New `cloud-hypervisor` organization](#new-cloud-hypervisor-organization)
-    - [Contributors](#contributors-38)
+    - [Contributors](#contributors-39)
 - [v0.3.0](#v030)
     - [Block device offloading](#block-device-offloading)
     - [Network device backend](#network-device-backend)
@@ -349,6 +354,48 @@
     - [Unit testing](#unit-testing)
     - [Integration tests parallelization](#integration-tests-parallelization)
 
+# v43.0
+
+This release has been tracked in our [roadmap
+project](https://github.com/orgs/cloud-hypervisor/projects/6) as iteration
+v43.0. The following user visible changes have been made:
+
+### Live Migration over TCP Connections
+
+Support has been added to enable direct live migration from two hosts via TCP
+connections. This supplements the existing support for migrating over a UNIX
+socket which can then be tunnelled as desired. The documentation has been
+updated. (#6850)
+
+### Notable Performance Improvements
+
+The `VIRTIO_RING_F_INDIRECT_DESC` feature has been enabled for `virtio-block`
+devices. This significantly increases the throughput of the devices with a
+small negative impact on latency. (#6826)
+
+### Notable Bug Fixes
+
+* Cloud Hypervisor now accepts VFIO devices that use I/O PCI BARs on non x86-64
+  architectures. Whether they function depends on the host PCI host bridge
+  support - previously they would be rejected even if the driver did not use
+  these BARs. (#6871)
+* Command line groups were adjusted to ensure that at least one payload
+  parameter was provided if any other VM parameters provided. (#6832)
+
+### Contributors
+
+Many thanks to everyone who has contributed to our release:
+
+* Alyssa Ross <hi@alyssa.is>
+* Andrew Consroe <aconz2@gmail.com>
+* Bo Chen <bo.arvin.chen@gmail.com>
+* Jinrong Liang <cloudliang@tencent.com>
+* Julian Stecklina <julian.stecklina@cyberus-technology.de>
+* Muminul Islam <muislam@microsoft.com>
+* Rob Bradford <rbradford@rivosinc.com>
+* Ruoqing He <heruoqing@iscas.ac.cn>
+* Wojtek Czekalski <wczekalski@me.com>
+
 # v42.0
 
 This release has been tracked in our [roadmap

scripts/package-consistency-check.py

@@ -0,0 +1,83 @@
+#!/bin/env python3
+#
+# Copyright © 2024 Institute of Software, CAS. All rights reserved.
+#
+# SPDX-License-Identifier: Apache-2.0
+#
+
+import subprocess
+import json
+from argparse import ArgumentParser
+from collections import defaultdict
+
+def get_cargo_metadata():
+    result = subprocess.run(
+        ['cargo', 'metadata', '--format-version=1'],
+        capture_output=True,
+        text=True
+    )
+    if result.returncode != 0:
+        exit(1)
+
+    metadata = json.loads(result.stdout)
+    return metadata
+
+def find_dependents_of_package(metadata, package_source):
+    """Find dependencies based on the provided source identifier and return related package info."""
+    packages = defaultdict(list)
+    direct_dependents = defaultdict(list)
+
+    # Identify packages from the given package source and record version
+    for pkg in metadata['packages']:
+        repository = pkg['repository'] or ''
+        if package_source in repository:
+            packages[pkg['name']].append(pkg['version'])
+
+    # Find packages that immediately depend on the identified source packages
+    for node in metadata['resolve']['nodes']:
+        current_pkg = next(pkg for pkg in metadata['packages'] if pkg['id'] == node['id'])
+        current_pkg_name = current_pkg['name']
+        current_pkg_version = current_pkg['version']
+
+        for dep_id in node['dependencies']:
+            dep_pkg = next(pkg for pkg in metadata['packages'] if pkg['id'] == dep_id)
+            dep_name = dep_pkg['name']
+            dep_version = dep_pkg['version']
+
+            if dep_name in packages:
+                direct_dependents[(dep_name, dep_version)].append((current_pkg_name, current_pkg_version))
+
+    return packages, direct_dependents
+
+def check_for_version_conflicts(packages, direct_dependents):
+    """Check if there are multiple versions of dependencies, and return True if conflicts are found."""
+    has_conflicts = False
+
+    for pkg_name, versions in packages.items():
+        if len(set(versions)) > 1:
+            has_conflicts = True
+            print(f"Error: Multiple versions detected for {pkg_name}: {set(versions)}")
+            for version in set(versions):
+                print(f"  Version {version} used by:")
+                for dependent, dep_version in direct_dependents[(pkg_name, version)]:
+                    print(f"          - {dependent} v{dep_version}")
+
+    return has_conflicts
+
+if __name__ == '__main__':
+    parser = ArgumentParser(description='Cargo dependency conflict checker.')
+    parser.add_argument('package_source', type=str, help='A keyword used to match the repository URL field')
+    args = parser.parse_args()
+
+    metadata = get_cargo_metadata()
+    if metadata is None:
+        print("Error: Metadata is empty")
+        exit(1)
+
+    packages, direct_dependents = find_dependents_of_package(metadata, args.package_source)
+
+    has_conflicts = check_for_version_conflicts(packages, direct_dependents)
+
+    if has_conflicts:
+        exit(1)
+

scripts/run_integration_tests_aarch64.sh

@@ -154,8 +154,8 @@ update_workloads() {
     fi
     popd || exit
 
-    # Download prebuild linux binaries
-    download_linux
+    # Prepare linux image (build from source or download pre-built)
+    prepare_linux
 
     # Update the kernel in the cloud image for some tests that requires recent kernel version
     FOCAL_OS_RAW_IMAGE_UPDATE_KERNEL_NAME="focal-server-cloudimg-arm64-custom-20210929-0-update-kernel.raw"

scripts/run_integration_tests_live_migration.sh

@@ -56,8 +56,8 @@ popd || exit
 # Build custom kernel based on virtio-pmem and virtio-fs upstream patches
 VMLINUX_IMAGE="$WORKLOADS_DIR/vmlinux"
 if [ ! -f "$VMLINUX_IMAGE" ]; then
-    # Download prebuild linux binaries
-    download_linux
+    # Prepare linux image (build from source or download pre-built)
+    prepare_linux
 fi
 
 CFLAGS=""

scripts/run_integration_tests_rate_limiter.sh

@@ -44,8 +44,8 @@ if ! grep focal sha1sums-x86_64 | sha1sum --check; then
 fi
 popd || exit
 
-# Download prebuild linux binaries
-download_linux
+# Prepare linux image (build from source or download pre-built)
+prepare_linux
 
 CFLAGS=""
 if [[ "${BUILD_TARGET}" == "x86_64-unknown-linux-musl" ]]; then

scripts/run_integration_tests_x86_64.sh

@@ -102,8 +102,8 @@ popd || exit
 # Build custom kernel based on virtio-pmem and virtio-fs upstream patches
 VMLINUX_IMAGE="$WORKLOADS_DIR/vmlinux"
 if [ ! -f "$VMLINUX_IMAGE" ]; then
-    # Download prebuild linux binaries
-    download_linux
+    # Prepare linux image (build from source or download pre-built)
+    prepare_linux
 fi
 
 VIRTIOFSD="$WORKLOADS_DIR/virtiofsd"

scripts/run_metrics.sh

@@ -80,8 +80,8 @@ if [ "${TEST_ARCH}" == "aarch64" ]; then
     guestunmount "$FOCAL_OS_RAW_IMAGE_UPDATE_TOOL_ROOT_DIR"
 fi
 
-# Download prebuild linux binaries
-download_linux
+# Prepare linux image (build from source or download pre-built)
+prepare_linux
 
 CFLAGS=""
 if [[ "${BUILD_TARGET}" == "${TEST_ARCH}-unknown-linux-musl" ]]; then

scripts/test-util.sh

@@ -1,6 +1,7 @@
 #!/usr/bin/env bash
 hypervisor="kvm"
 test_filter=""
+build_kernel=false
 
 # Checkout source code of a GIT repo with specified branch and commit
 # Args:
@@ -75,11 +76,13 @@ cmd_help() {
     echo ""
     echo "    --hypervisor  Underlying hypervisor. Options kvm, mshv"
     echo "    --test-filter Tests to run"
+    echo "    --build-guest-kernel Build guest kernel from source instead of downloading pre-built"
     echo ""
     echo "    --help        Display this help message."
     echo ""
 }
 
+# shellcheck disable=SC2034
 process_common_args() {
     while [ $# -gt 0 ]; do
         case "$1" in
@@ -93,9 +96,11 @@ process_common_args() {
             ;;
         "--test-filter")
             shift
-            # shellcheck disable=SC2034
             test_filter="$1"
             ;;
+        "--build-guest-kernel")
+            build_kernel=true
+            ;;
         "--") {
             shift
             break
@@ -115,14 +120,17 @@ process_common_args() {
 }
 
 download_hypervisor_fw() {
+    FW_TAG="0.5.0"
     if [ -n "$AUTH_DOWNLOAD_TOKEN" ]; then
         echo "Using authenticated download from GitHub"
-        FW_URL=$(curl --silent https://api.github.com/repos/cloud-hypervisor/rust-hypervisor-firmware/releases/latest \
+        FW_URL=$(curl --silent https://api.github.com/repos/cloud-hypervisor/rust-hypervisor-firmware/releases/tags/${FW_TAG} \
             --header "Authorization: Token $AUTH_DOWNLOAD_TOKEN" \
-            --header "X-GitHub-Api-Version: 2022-11-28" | grep "browser_download_url" | grep -o 'https://.*[^ "]')
+            --header "X-GitHub-Api-Version: 2022-11-28" | grep "browser_download_url" |
+            grep -oP '"https://[^"]*hypervisor-fw"' | sed -e 's/^"//' -e 's/"$//')
     else
         echo "Using anonymous download from GitHub"
-        FW_URL=$(curl --silent https://api.github.com/repos/cloud-hypervisor/rust-hypervisor-firmware/releases/latest | grep "browser_download_url" | grep -o 'https://.*[^ "]')
+        FW_URL=$(curl --silent https://api.github.com/repos/cloud-hypervisor/rust-hypervisor-firmware/releases/tags/${FW_TAG} |
+            grep "browser_download_url" | grep -oP '"https://[^"]*hypervisor-fw"' | sed -e 's/^"//' -e 's/"$//')
     fi
     FW="$WORKLOADS_DIR/hypervisor-fw"
     pushd "$WORKLOADS_DIR" || exit
@@ -149,6 +157,18 @@ download_linux() {
     popd || exit
 }
 
+prepare_linux() {
+    if [ "$build_kernel" = true ]; then
+        echo "Building kernel from source"
+        build_custom_linux
+        echo "Using kernel built from source"
+    else
+        echo "Downloading pre-built kernel from GitHub"
+        download_linux
+        echo "Using kernel downloaded from GitHub"
+    fi
+}
+
 download_ovmf() {
     OVMF_FW_TAG="ch-6624aa331f"
     OVMF_FW_URL="https://github.com/cloud-hypervisor/edk2/releases/download/$OVMF_FW_TAG/CLOUDHV.fd"

src/main.rs

@@ -184,9 +184,10 @@ fn create_app(default_vcpus: String, default_memory: String, default_rng: String
         .author(env!("CARGO_PKG_AUTHORS"))
         .about("Launch a cloud-hypervisor VMM.")
         .arg_required_else_help(true)
-        .group(ArgGroup::new("vm-config").multiple(true))
+        .group(ArgGroup::new("vm-config").multiple(true).requires("vm-payload"))
         .group(ArgGroup::new("vmm-config").multiple(true))
         .group(ArgGroup::new("logging").multiple(true))
+        .group(ArgGroup::new("vm-payload").multiple(true))
         .arg(
             Arg::new("cpus")
                 .long("cpus")
@@ -243,7 +244,7 @@ fn create_app(default_vcpus: String, default_memory: String, default_rng: String
                 .long("firmware")
                 .help("Path to firmware that is loaded in an architectural specific way")
                 .num_args(1)
-                .group("vm-config"),
+                .group("vm-payload"),
         )
         .arg(
             Arg::new("kernel")
@@ -253,7 +254,7 @@ fn create_app(default_vcpus: String, default_memory: String, default_rng: String
                 entry point (e.g. vmlinux) or architecture equivalent",
                 )
                 .num_args(1)
-                .group("vm-config"),
+                .group("vm-payload"),
         )
         .arg(
             Arg::new("initramfs")
@@ -459,7 +460,7 @@ fn create_app(default_vcpus: String, default_memory: String, default_rng: String
                 .long("tpm")
                 .num_args(1)
                 .help(TpmConfig::SYNTAX)
-                .group("vmm-config"),
+                .group("vm-config"),
         );
 
     #[cfg(target_arch = "x86_64")]
@@ -519,7 +520,7 @@ fn create_app(default_vcpus: String, default_memory: String, default_rng: String
             .long("igvm")
             .help("Path to IGVM file to load.")
             .num_args(1)
-            .group("vm-config"),
+            .group("vm-payload"),
     );
     #[cfg(feature = "sev_snp")]
     let app = app.arg(

test_infra/Cargo.toml

@@ -7,7 +7,7 @@ version = "0.1.0"
 [dependencies]
 dirs = "5.0.1"
 epoll = "4.3.3"
-libc = "0.2.158"
+libc = "0.2.167"
 once_cell = "1.20.2"
 serde = { version = "1.0.208", features = ["derive", "rc"] }
 serde_json = "1.0.120"

tests/integration.rs

@@ -12,12 +12,14 @@ extern crate test_infra;
 
 use std::collections::HashMap;
 use std::io::{BufRead, Read, Seek, Write};
+use std::net::TcpListener;
 use std::os::unix::io::AsRawFd;
 use std::path::PathBuf;
 use std::process::{Child, Command, Stdio};
 use std::string::String;
 use std::sync::mpsc::Receiver;
 use std::sync::{mpsc, Mutex};
+use std::time::Duration;
 use std::{fs, io, thread};
 
 use net_util::MacAddr;
@@ -9401,7 +9403,7 @@ mod live_migration {
 
         // Check the source vm has been terminated successful (give it '3s' to settle)
         thread::sleep(std::time::Duration::new(3, 0));
-        if !src_child.try_wait().unwrap().map_or(false, |s| s.success()) {
+        if !src_child.try_wait().unwrap().is_some_and(|s| s.success()) {
             print_and_panic(
                 src_child,
                 dest_child,
@@ -9575,7 +9577,7 @@ mod live_migration {
 
         // Check the source vm has been terminated successful (give it '3s' to settle)
         thread::sleep(std::time::Duration::new(3, 0));
-        if !src_child.try_wait().unwrap().map_or(false, |s| s.success()) {
+        if !src_child.try_wait().unwrap().is_some_and(|s| s.success()) {
             print_and_panic(
                 src_child,
                 dest_child,
@@ -9793,7 +9795,7 @@ mod live_migration {
 
         // Check the source vm has been terminated successful (give it '3s' to settle)
         thread::sleep(std::time::Duration::new(3, 0));
-        if !src_child.try_wait().unwrap().map_or(false, |s| s.success()) {
+        if !src_child.try_wait().unwrap().is_some_and(|s| s.success()) {
             print_and_panic(
                 src_child,
                 dest_child,
@@ -10009,7 +10011,7 @@ mod live_migration {
 
         // Check the source vm has been terminated successful (give it '3s' to settle)
         thread::sleep(std::time::Duration::new(3, 0));
-        if !src_child.try_wait().unwrap().map_or(false, |s| s.success()) {
+        if !src_child.try_wait().unwrap().is_some_and(|s| s.success()) {
             print_and_panic(
                 src_child,
                 dest_child,
@@ -10119,7 +10121,7 @@ mod live_migration {
 
         // Check the source vm has been terminated successful (give it '3s' to settle)
         thread::sleep(std::time::Duration::new(3, 0));
-        if !src_child.try_wait().unwrap().map_or(false, |s| s.success()) {
+        if !src_child.try_wait().unwrap().is_some_and(|s| s.success()) {
             print_and_panic(
                 src_child,
                 dest_child,
@@ -10266,7 +10268,7 @@ mod live_migration {
 
         // Check the source vm has been terminated successful (give it '3s' to settle)
         thread::sleep(std::time::Duration::new(3, 0));
-        if !src_child.try_wait().unwrap().map_or(false, |s| s.success()) {
+        if !src_child.try_wait().unwrap().is_some_and(|s| s.success()) {
             print_and_panic(
                 src_child,
                 dest_child,
@@ -10295,6 +10297,220 @@ mod live_migration {
         handle_child_output(r, &dest_output);
     }
 
+    // Function to get an available port
+    fn get_available_port() -> u16 {
+        TcpListener::bind("127.0.0.1:0")
+            .expect("Failed to bind to address")
+            .local_addr()
+            .unwrap()
+            .port()
+    }
+
+    fn start_live_migration_tcp(src_api_socket: &str, dest_api_socket: &str) -> bool {
+        // Get an available TCP port
+        let migration_port = get_available_port();
+        let host_ip = "127.0.0.1";
+
+        // Start the 'receive-migration' command on the destination
+        let mut receive_migration = Command::new(clh_command("ch-remote"))
+            .args([
+                &format!("--api-socket={}", dest_api_socket),
+                "receive-migration",
+                &format!("tcp:0.0.0.0:{}", migration_port),
+            ])
+            .stdin(Stdio::null())
+            .stderr(Stdio::piped())
+            .stdout(Stdio::piped())
+            .spawn()
+            .unwrap();
+
+        // Give the destination some time to start listening
+        thread::sleep(Duration::from_secs(1));
+
+        // Start the 'send-migration' command on the source
+        let mut send_migration = Command::new(clh_command("ch-remote"))
+            .args([
+                &format!("--api-socket={}", src_api_socket),
+                "send-migration",
+                &format!("tcp:{}:{}", host_ip, migration_port),
+            ])
+            .stdin(Stdio::null())
+            .stderr(Stdio::piped())
+            .stdout(Stdio::piped())
+            .spawn()
+            .unwrap();
+
+        // Check if the 'send-migration' command executed successfully
+        let send_success = if let Some(status) = send_migration
+            .wait_timeout(Duration::from_secs(60))
+            .unwrap()
+        {
+            status.success()
+        } else {
+            false
+        };
+
+        if !send_success {
+            let _ = send_migration.kill();
+            let output = send_migration.wait_with_output().unwrap();
+            eprintln!(
+                "\n\n==== Start 'send_migration' output ====\n\n---stdout---\n{}\n\n---stderr---\n{}\n\n==== End 'send_migration' output ====\n\n",
+                String::from_utf8_lossy(&output.stdout),
+                String::from_utf8_lossy(&output.stderr)
+            );
+        }
+
+        // Check if the 'receive-migration' command executed successfully
+        let receive_success = if let Some(status) = receive_migration
+            .wait_timeout(Duration::from_secs(60))
+            .unwrap()
+        {
+            status.success()
+        } else {
+            false
+        };
+
+        if !receive_success {
+            let _ = receive_migration.kill();
+            let output = receive_migration.wait_with_output().unwrap();
+            eprintln!(
+                "\n\n==== Start 'receive_migration' output ====\n\n---stdout---\n{}\n\n---stderr---\n{}\n\n==== End 'receive_migration' output ====\n\n",
+                String::from_utf8_lossy(&output.stdout),
+                String::from_utf8_lossy(&output.stderr)
+            );
+        }
+
+        send_success && receive_success
+    }
+
+    fn _test_live_migration_tcp() {
+        let focal = UbuntuDiskConfig::new(FOCAL_IMAGE_NAME.to_string());
+        let guest = Guest::new(Box::new(focal));
+        let kernel_path = direct_kernel_boot_path();
+        let console_text = String::from("On a branch floating down river a cricket, singing.");
+        let net_id = "net123";
+        let net_params = format!(
+            "id={},tap=,mac={},ip={},mask=255.255.255.0",
+            net_id, guest.network.guest_mac, guest.network.host_ip
+        );
+        let memory_param: &[&str] = &["--memory", "size=4G,shared=on"];
+        let boot_vcpus = 2;
+        let max_vcpus = 4;
+        let pmem_temp_file = TempFile::new().unwrap();
+        pmem_temp_file.as_file().set_len(128 << 20).unwrap();
+        std::process::Command::new("mkfs.ext4")
+            .arg(pmem_temp_file.as_path())
+            .output()
+            .expect("Expect creating disk image to succeed");
+        let pmem_path = String::from("/dev/pmem0");
+
+        // Start the source VM
+        let src_vm_path = clh_command("cloud-hypervisor");
+        let src_api_socket = temp_api_path(&guest.tmp_dir);
+        let mut src_vm_cmd = GuestCommand::new_with_binary_path(&guest, &src_vm_path);
+        src_vm_cmd
+            .args([
+                "--cpus",
+                format!("boot={},max={}", boot_vcpus, max_vcpus).as_str(),
+            ])
+            .args(memory_param)
+            .args(["--kernel", kernel_path.to_str().unwrap()])
+            .args(["--cmdline", DIRECT_KERNEL_BOOT_CMDLINE])
+            .default_disks()
+            .args(["--net", net_params.as_str()])
+            .args(["--api-socket", &src_api_socket])
+            .args([
+                "--pmem",
+                format!(
+                    "file={},discard_writes=on",
+                    pmem_temp_file.as_path().to_str().unwrap(),
+                )
+                .as_str(),
+            ])
+            .capture_output();
+        let mut src_child = src_vm_cmd.spawn().unwrap();
+
+        // Start the destination VM
+        let mut dest_api_socket = temp_api_path(&guest.tmp_dir);
+        dest_api_socket.push_str(".dest");
+        let mut dest_child = GuestCommand::new(&guest)
+            .args(["--api-socket", &dest_api_socket])
+            .capture_output()
+            .spawn()
+            .unwrap();
+
+        let r = std::panic::catch_unwind(|| {
+            guest.wait_vm_boot(None).unwrap();
+            // Ensure the source VM is running normally
+            assert_eq!(guest.get_cpu_count().unwrap_or_default(), boot_vcpus);
+            assert!(guest.get_total_memory().unwrap_or_default() > 3_840_000);
+            guest.check_devices_common(None, Some(&console_text), Some(&pmem_path));
+
+            // On x86_64 architecture, remove and re-add the virtio-net device
+            #[cfg(target_arch = "x86_64")]
+            {
+                assert!(remote_command(
+                    &src_api_socket,
+                    "remove-device",
+                    Some(net_id),
+                ));
+                thread::sleep(Duration::new(10, 0));
+                // Re-add the virtio-net device
+                assert!(remote_command(
+                    &src_api_socket,
+                    "add-net",
+                    Some(net_params.as_str()),
+                ));
+                thread::sleep(Duration::new(10, 0));
+            }
+            // Start TCP live migration
+            assert!(
+                start_live_migration_tcp(&src_api_socket, &dest_api_socket),
+                "Unsuccessful command: 'send-migration' or 'receive-migration'."
+            );
+        });
+
+        // Check and report any errors that occurred during live migration
+        if r.is_err() {
+            print_and_panic(
+                src_child,
+                dest_child,
+                None,
+                "Error occurred during live-migration",
+            );
+        }
+
+        // Check the source vm has been terminated successful (give it '3s' to settle)
+        thread::sleep(std::time::Duration::new(3, 0));
+        if !src_child.try_wait().unwrap().is_some_and(|s| s.success()) {
+            print_and_panic(
+                src_child,
+                dest_child,
+                None,
+                "Source VM was not terminated successfully.",
+            );
+        };
+
+        // After live migration, ensure the destination VM is running normally
+        let r = std::panic::catch_unwind(|| {
+            // Perform the same checks to ensure the VM has migrated correctly
+            assert_eq!(guest.get_cpu_count().unwrap_or_default(), boot_vcpus);
+            assert!(guest.get_total_memory().unwrap_or_default() > 3_840_000);
+            guest.check_devices_common(None, Some(&console_text), Some(&pmem_path));
+        });
+
+        // Clean up the destination VM and ensure it terminates properly
+        let _ = dest_child.kill();
+        let dest_output = dest_child.wait_with_output().unwrap();
+        handle_child_output(r, &dest_output);
+
+        // Check if the expected `console_text` is present in the destination VM's output
+        let r = std::panic::catch_unwind(|| {
+            assert!(String::from_utf8_lossy(&dest_output.stdout).contains(&console_text));
+        });
+        handle_child_output(r, &dest_output);
+    }
+
     mod live_migration_parallel {
         use super::*;
         #[test]
@@ -10307,6 +10523,11 @@ mod live_migration {
             _test_live_migration(false, true)
         }
 
+        #[test]
+        fn test_live_migration_tcp() {
+            _test_live_migration_tcp();
+        }
+
         #[test]
         fn test_live_migration_watchdog() {
             _test_live_migration_watchdog(false, false)

tracer/Cargo.toml

@@ -5,7 +5,7 @@ name = "tracer"
 version = "0.1.0"
 
 [dependencies]
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 once_cell = "1.20.2"
 serde = { version = "1.0.208", features = ["derive", "rc"] }

vhost_user_block/Cargo.toml

@@ -10,11 +10,11 @@ block = { path = "../block" }
 clap = { version = "4.5.13", features = ["cargo", "wrap_help"] }
 env_logger = "0.11.3"
 epoll = "4.3.3"
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 option_parser = { path = "../option_parser" }
 vhost = { workspace = true, features = ["vhost-user-backend"] }
-vhost-user-backend = "0.16.0"
+vhost-user-backend = { workspace = true }
 virtio-bindings = { workspace = true }
 virtio-queue = { workspace = true }
 vm-memory = { workspace = true }

vhost_user_net/Cargo.toml

@@ -9,12 +9,12 @@ version = "0.1.0"
 clap = { version = "4.5.13", features = ["cargo", "wrap_help"] }
 env_logger = "0.11.3"
 epoll = "4.3.3"
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 net_util = { path = "../net_util" }
 option_parser = { path = "../option_parser" }
 vhost = { workspace = true, features = ["vhost-user-backend"] }
-vhost-user-backend = "0.16.0"
+vhost-user-backend = { workspace = true }
 virtio-bindings = { workspace = true }
 vm-memory = { workspace = true }
 vmm-sys-util = { workspace = true }

virtio-devices/Cargo.toml

@@ -9,13 +9,13 @@ default = []
 sev_snp = ["mshv-ioctls"]
 
 [dependencies]
-anyhow = "1.0.87"
+anyhow = "1.0.94"
 arc-swap = "1.7.1"
 block = { path = "../block" }
 byteorder = "1.5.0"
 epoll = "4.3.3"
 event_monitor = { path = "../event_monitor" }
-libc = "0.2.158"
+libc = "0.2.167"
 log = "0.4.22"
 mshv-ioctls = { workspace = true, optional = true }
 net_gen = { path = "../net_gen" }

virtio-devices/src/block.rs

@@ -27,7 +27,7 @@ use serde::{Deserialize, Serialize};
 use thiserror::Error;
 use virtio_bindings::virtio_blk::*;
 use virtio_bindings::virtio_config::*;
-use virtio_bindings::virtio_ring::VIRTIO_RING_F_EVENT_IDX;
+use virtio_bindings::virtio_ring::{VIRTIO_RING_F_EVENT_IDX, VIRTIO_RING_F_INDIRECT_DESC};
 use virtio_queue::{Queue, QueueOwnedT, QueueT};
 use vm_memory::{ByteValued, Bytes, GuestAddressSpace, GuestMemoryAtomic, GuestMemoryError};
 use vm_migration::{Migratable, MigratableError, Pausable, Snapshot, Snapshottable, Transportable};
@@ -489,8 +489,7 @@ impl EpollHelperHandler for BlockEpollHandler {
                     EpollHelperError::HandleEvent(anyhow!("Failed to get queue event: {:?}", e))
                 })?;
 
-                let rate_limit_reached =
-                    self.rate_limiter.as_ref().map_or(false, |r| r.is_blocked());
+                let rate_limit_reached = self.rate_limiter.as_ref().is_some_and(|r| r.is_blocked());
 
                 // Process the queue only when the rate limit is not reached
                 if !rate_limit_reached {
@@ -509,8 +508,7 @@ impl EpollHelperHandler for BlockEpollHandler {
                     ))
                 })?;
 
-                let rate_limit_reached =
-                    self.rate_limiter.as_ref().map_or(false, |r| r.is_blocked());
+                let rate_limit_reached = self.rate_limiter.as_ref().is_some_and(|r| r.is_blocked());
 
                 // Process the queue only when the rate limit is not reached
                 if !rate_limit_reached {
@@ -627,8 +625,8 @@ impl Block {
                     | (1u64 << VIRTIO_BLK_F_CONFIG_WCE)
                     | (1u64 << VIRTIO_BLK_F_BLK_SIZE)
                     | (1u64 << VIRTIO_BLK_F_TOPOLOGY)
-                    | (1u64 << VIRTIO_RING_F_EVENT_IDX);
-
+                    | (1u64 << VIRTIO_RING_F_EVENT_IDX)
+                    | (1u64 << VIRTIO_RING_F_INDIRECT_DESC);
                 if iommu {
                     avail_features |= 1u64 << VIRTIO_F_IOMMU_PLATFORM;
                 }

virtio-devices/src/net.rs

@@ -201,7 +201,7 @@ impl NetEpollHandler {
             .net
             .rx_rate_limiter
             .as_ref()
-            .map_or(false, |r| r.is_blocked());
+            .is_some_and(|r| r.is_blocked());
 
         // Start to listen on RX_TAP_EVENT only when the rate limit is not reached
         if !self.net.rx_tap_listening && !rate_limit_reached {
@@ -238,7 +238,7 @@ impl NetEpollHandler {
             .net
             .tx_rate_limiter
             .as_ref()
-            .map_or(false, |r| r.is_blocked());
+            .is_some_and(|r| r.is_blocked());
 
         if !rate_limit_reached {
             self.process_tx()?;

vm-allocator/Cargo.toml

@@ -4,9 +4,13 @@ edition = "2021"
 name = "vm-allocator"
 version = "0.1.0"
 
+[features]
+default = []
+kvm = ["arch/kvm"]
+
 [dependencies]
-libc = "0.2.158"
+libc = "0.2.167"
 vm-memory = { workspace = true }
 
-[target.'cfg(target_arch = "aarch64")'.dependencies]
+[target.'cfg(any(target_arch = "aarch64", target_arch = "riscv64"))'.dependencies]
 arch = { path = "../arch" }

vm-allocator/src/gsi.rs

@@ -62,7 +62,7 @@ impl GsiAllocator {
         allocator
     }
 
-    #[cfg(target_arch = "aarch64")]
+    #[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
     /// New GSI allocator
     pub fn new() -> Self {
         GsiAllocator {
@@ -97,7 +97,7 @@ impl GsiAllocator {
         Ok(irq)
     }
 
-    #[cfg(target_arch = "aarch64")]
+    #[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
     /// Allocate an IRQ
     pub fn allocate_irq(&mut self) -> Result<u32> {
         let irq = self.next_irq;
@@ -106,7 +106,7 @@ impl GsiAllocator {
     }
 }
 
-#[cfg(target_arch = "aarch64")]
+#[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
 impl Default for GsiAllocator {
     fn default() -> Self {
         GsiAllocator::new()

vm-allocator/src/lib.rs

@@ -21,3 +21,5 @@ pub use crate::gsi::GsiAllocator;
 #[cfg(target_arch = "x86_64")]
 pub use crate::gsi::GsiApic;
 pub use crate::system::SystemAllocator;
+mod memory_slot;
+pub use memory_slot::MemorySlotAllocator;

vm-allocator/src/memory_slot.rs

@@ -0,0 +1,39 @@
+// Copyright © 2024 Rivos Inc
+//
+// SPDX-License-Identifier: Apache-2.0
+//
+
+use std::sync::atomic::{AtomicU32, Ordering};
+use std::sync::{Arc, Mutex};
+
+/// Allocator for KVM memory slots
+pub struct MemorySlotAllocator {
+    next_memory_slot: Arc<AtomicU32>,
+    memory_slot_free_list: Arc<Mutex<Vec<u32>>>,
+}
+
+impl MemorySlotAllocator {
+    /// Next free memory slot
+    pub fn next_memory_slot(&self) -> u32 {
+        if let Some(slot_id) = self.memory_slot_free_list.lock().unwrap().pop() {
+            return slot_id;
+        }
+        self.next_memory_slot.fetch_add(1, Ordering::SeqCst)
+    }
+
+    /// Release memory slot for reuse
+    pub fn free_memory_slot(&mut self, slot: u32) {
+        self.memory_slot_free_list.lock().unwrap().push(slot)
+    }
+
+    /// Instantiate struct
+    pub fn new(
+        next_memory_slot: Arc<AtomicU32>,
+        memory_slot_free_list: Arc<Mutex<Vec<u32>>>,
+    ) -> Self {
+        Self {
+            next_memory_slot,
+            memory_slot_free_list,
+        }
+    }
+}

vm-allocator/src/system.rs

@@ -22,27 +22,30 @@ use crate::page_size::get_page_size;
 /// ```
 /// # #[cfg(target_arch = "x86_64")]
 /// # use vm_allocator::{GsiApic, SystemAllocator};
-/// # #[cfg(target_arch = "aarch64")]
+/// # #[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
 /// # use vm_allocator::SystemAllocator;
 /// # use vm_memory::{Address, GuestAddress, GuestUsize};
 ///   let mut allocator = SystemAllocator::new(
-///           #[cfg(target_arch = "x86_64")] GuestAddress(0x1000),
-///           #[cfg(target_arch = "x86_64")] 0x10000,
+///           GuestAddress(0x1000),
+///           0x10000,
 ///           GuestAddress(0x10000000), 0x10000000,
 ///           #[cfg(target_arch = "x86_64")] vec![GsiApic::new(5, 19)]).unwrap();
 ///   #[cfg(target_arch = "x86_64")]
 ///   assert_eq!(allocator.allocate_irq(), Some(5));
 ///   #[cfg(target_arch = "aarch64")]
 ///   assert_eq!(allocator.allocate_irq(), Some(32));
+///   #[cfg(target_arch = "riscv64")]
+///   assert_eq!(allocator.allocate_irq(), Some(0));
 ///   #[cfg(target_arch = "x86_64")]
 ///   assert_eq!(allocator.allocate_irq(), Some(6));
 ///   #[cfg(target_arch = "aarch64")]
 ///   assert_eq!(allocator.allocate_irq(), Some(33));
+///   #[cfg(target_arch = "riscv64")]
+///   assert_eq!(allocator.allocate_irq(), Some(1));
 ///   assert_eq!(allocator.allocate_platform_mmio_addresses(None, 0x1000, Some(0x1000)), Some(GuestAddress(0x1fff_f000)));
 ///
 /// ```
 pub struct SystemAllocator {
-    #[cfg(target_arch = "x86_64")]
     io_address_space: AddressAllocator,
     platform_mmio_address_space: AddressAllocator,
     gsi_allocator: GsiAllocator,
@@ -59,14 +62,13 @@ impl SystemAllocator {
     /// * `apics` - (X86) Vector of APIC's.
     ///
     pub fn new(
-        #[cfg(target_arch = "x86_64")] io_base: GuestAddress,
-        #[cfg(target_arch = "x86_64")] io_size: GuestUsize,
+        io_base: GuestAddress,
+        io_size: GuestUsize,
         platform_mmio_base: GuestAddress,
         platform_mmio_size: GuestUsize,
         #[cfg(target_arch = "x86_64")] apics: Vec<GsiApic>,
     ) -> Option<Self> {
         Some(SystemAllocator {
-            #[cfg(target_arch = "x86_64")]
             io_address_space: AddressAllocator::new(io_base, io_size)?,
             platform_mmio_address_space: AddressAllocator::new(
                 platform_mmio_base,
@@ -74,7 +76,7 @@ impl SystemAllocator {
             )?,
             #[cfg(target_arch = "x86_64")]
             gsi_allocator: GsiAllocator::new(apics),
-            #[cfg(target_arch = "aarch64")]
+            #[cfg(any(target_arch = "aarch64", target_arch = "riscv64"))]
             gsi_allocator: GsiAllocator::new(),
         })
     }
@@ -89,7 +91,6 @@ impl SystemAllocator {
         self.gsi_allocator.allocate_gsi().ok()
     }
 
-    #[cfg(target_arch = "x86_64")]
     /// Reserves a section of `size` bytes of IO address space.
     pub fn allocate_io_addresses(
         &mut self,
@@ -115,7 +116,6 @@ impl SystemAllocator {
         )
     }
 
-    #[cfg(target_arch = "x86_64")]
     /// Free an IO address range.
     /// We can only free a range if it matches exactly an already allocated range.
     pub fn free_io_addresses(&mut self, address: GuestAddress, size: GuestUsize) {

vm-device/Cargo.toml

@@ -10,7 +10,7 @@ kvm = ["vfio-ioctls/kvm"]
 mshv = ["vfio-ioctls/mshv"]
 
 [dependencies]
-anyhow = "1.0.87"
+anyhow = "1.0.94"
 hypervisor = { path = "../hypervisor" }
 serde = { version = "1.0.208", features = ["derive", "rc"] }
 thiserror = "1.0.62"

vm-migration/Cargo.toml

@@ -5,7 +5,7 @@ name = "vm-migration"
 version = "0.1.0"
 
 [dependencies]
-anyhow = "1.0.87"
+anyhow = "1.0.94"
 serde = { version = "1.0.208", features = ["derive", "rc"] }
 serde_json = "1.0.120"
 thiserror = "1.0.62"

vmm/Cargo.toml

@@ -26,7 +26,7 @@ tracing = ["tracer/tracing"]
 
 [dependencies]
 acpi_tables = { workspace = true }
-anyhow = "1.0.87"
+anyhow = "1.0.94"
 arc-swap = "1.7.1"
 arch = { path = "../arch" }
 bitflags = "2.6.0"
@@ -38,7 +38,7 @@ devices = { path = "../devices" }
 dhat = { version = "0.3.3", optional = true }
 epoll = "4.3.3"
 event_monitor = { path = "../event_monitor" }
-flume = "0.11.0"
+flume = "0.11.1"
 futures = { version = "0.3.30", optional = true }
 gdbstub = { version = "0.7.1", optional = true }
 gdbstub_arch = { version = "0.3.0", optional = true }
@@ -47,7 +47,7 @@ hypervisor = { path = "../hypervisor" }
 igvm = { version = "0.3.3", optional = true }
 igvm_defs = { version = "0.3.1", optional = true }
 landlock = "0.4.0"
-libc = "0.2.158"
+libc = "0.2.167"
 linux-loader = { workspace = true, features = ["bzimage", "elf", "pe"] }
 log = "0.4.22"
 micro_http = { git = "https://github.com/firecracker-microvm/micro-http", branch = "main" }

vmm/src/config.rs

@@ -1144,7 +1144,8 @@ impl DiskConfig {
          bw_size=<bytes>,bw_one_time_burst=<bytes>,bw_refill_time=<ms>,\
          ops_size=<io_ops>,ops_one_time_burst=<io_ops>,ops_refill_time=<ms>,\
          id=<device_id>,pci_segment=<segment_id>,rate_limit_group=<group_id>,\
-         queue_affinity=<list_of_queue_indices_with_their_associated_cpuset>";
+         queue_affinity=<list_of_queue_indices_with_their_associated_cpuset>,\
+         serial=<serial_number>";
 
     pub fn parse(disk: &str) -> Result<Self> {
         let mut parser = OptionParser::new();

vmm/src/device_manager.rs

@@ -536,7 +536,6 @@ impl Console {
 
 pub(crate) struct AddressManager {
     pub(crate) allocator: Arc<Mutex<SystemAllocator>>,
-    #[cfg(target_arch = "x86_64")]
     pub(crate) io_bus: Arc<Bus>,
     pub(crate) mmio_bus: Arc<Bus>,
     pub(crate) vm: Arc<dyn hypervisor::Vm>,
@@ -556,33 +555,24 @@ impl DeviceRelocation for AddressManager {
     ) -> std::result::Result<(), std::io::Error> {
         match region_type {
             PciBarRegionType::IoRegion => {
-                #[cfg(target_arch = "x86_64")]
-                {
-                    // Update system allocator
-                    self.allocator
-                        .lock()
-                        .unwrap()
-                        .free_io_addresses(GuestAddress(old_base), len as GuestUsize);
+                // Update system allocator
+                self.allocator
+                    .lock()
+                    .unwrap()
+                    .free_io_addresses(GuestAddress(old_base), len as GuestUsize);
 
-                    self.allocator
-                        .lock()
-                        .unwrap()
-                        .allocate_io_addresses(
-                            Some(GuestAddress(new_base)),
-                            len as GuestUsize,
-                            None,
-                        )
-                        .ok_or_else(|| {
-                            io::Error::new(io::ErrorKind::Other, "failed allocating new IO range")
-                        })?;
+                self.allocator
+                    .lock()
+                    .unwrap()
+                    .allocate_io_addresses(Some(GuestAddress(new_base)), len as GuestUsize, None)
+                    .ok_or_else(|| {
+                        io::Error::new(io::ErrorKind::Other, "failed allocating new IO range")
+                    })?;
 
-                    // Update PIO bus
-                    self.io_bus
-                        .update_range(old_base, len, new_base, len)
-                        .map_err(|e| io::Error::new(io::ErrorKind::Other, e))?;
-                }
-                #[cfg(target_arch = "aarch64")]
-                error!("I/O region is not supported");
+                // Update PIO bus
+                self.io_bus
+                    .update_range(old_base, len, new_base, len)
+                    .map_err(|e| io::Error::new(io::ErrorKind::Other, e))?;
             }
             PciBarRegionType::Memory32BitRegion | PciBarRegionType::Memory64BitRegion => {
                 let allocators = if region_type == PciBarRegionType::Memory32BitRegion {
@@ -992,7 +982,7 @@ fn create_mmio_allocators(
 impl DeviceManager {
     #[allow(clippy::too_many_arguments)]
     pub fn new(
-        #[cfg(target_arch = "x86_64")] io_bus: Arc<Bus>,
+        io_bus: Arc<Bus>,
         mmio_bus: Arc<Bus>,
         vm: Arc<dyn hypervisor::Vm>,
         config: Arc<Mutex<VmConfig>>,
@@ -1072,7 +1062,6 @@ impl DeviceManager {
 
         let address_manager = Arc::new(AddressManager {
             allocator: memory_manager.lock().unwrap().allocator(),
-            #[cfg(target_arch = "x86_64")]
             io_bus,
             mmio_bus,
             vm: vm.clone(),
@@ -3432,7 +3421,7 @@ impl DeviceManager {
             legacy_interrupt_group,
             device_cfg.iommu,
             pci_device_bdf,
-            Arc::new(move || memory_manager.lock().unwrap().allocate_memory_slot()),
+            memory_manager.lock().unwrap().memory_slot_allocator(),
             vm_migration::snapshot_from_id(self.snapshot.as_ref(), vfio_name.as_str()),
             device_cfg.x_nv_gpudirect_clique,
         )
@@ -3512,7 +3501,6 @@ impl DeviceManager {
         pci_bus
             .register_mapping(
                 bus_device,
-                #[cfg(target_arch = "x86_64")]
                 self.address_manager.io_bus.as_ref(),
                 self.address_manager.mmio_bus.as_ref(),
                 bars.clone(),
@@ -3596,7 +3584,7 @@ impl DeviceManager {
             self.msi_interrupt_manager.clone(),
             legacy_interrupt_group,
             pci_device_bdf,
-            Arc::new(move || memory_manager.lock().unwrap().allocate_memory_slot()),
+            memory_manager.lock().unwrap().memory_slot_allocator(),
             vm_migration::snapshot_from_id(self.snapshot.as_ref(), vfio_user_name.as_str()),
         )
         .map_err(DeviceManagerError::VfioUserCreate)?;

vmm/src/lib.rs

@@ -11,6 +11,7 @@ extern crate log;
 use std::collections::HashMap;
 use std::fs::File;
 use std::io::{stdout, Read, Write};
+use std::net::{TcpListener, TcpStream};
 use std::os::unix::io::{AsRawFd, FromRawFd, RawFd};
 use std::os::unix::net::{UnixListener, UnixStream};
 use std::panic::AssertUnwindSafe;
@@ -36,8 +37,8 @@ use serde::{Deserialize, Serialize};
 use signal_hook::iterator::{Handle, Signals};
 use thiserror::Error;
 use tracer::trace_scoped;
-use vm_memory::bitmap::AtomicBitmap;
-use vm_memory::{ReadVolatile, WriteVolatile};
+use vm_memory::bitmap::{AtomicBitmap, BitmapSlice};
+use vm_memory::{ReadVolatile, VolatileMemoryError, VolatileSlice, WriteVolatile};
 use vm_migration::protocol::*;
 use vm_migration::{Migratable, MigratableError, Pausable, Snapshot, Snapshottable, Transportable};
 use vmm_sys_util::eventfd::EventFd;
@@ -235,6 +236,89 @@ impl From<u64> for EpollDispatch {
     }
 }
 
+enum SocketStream {
+    Unix(UnixStream),
+    Tcp(TcpStream),
+}
+
+impl Read for SocketStream {
+    fn read(&mut self, buf: &mut [u8]) -> std::io::Result<usize> {
+        match self {
+            SocketStream::Unix(stream) => stream.read(buf),
+            SocketStream::Tcp(stream) => stream.read(buf),
+        }
+    }
+}
+
+impl Write for SocketStream {
+    fn write(&mut self, buf: &[u8]) -> std::io::Result<usize> {
+        match self {
+            SocketStream::Unix(stream) => stream.write(buf),
+            SocketStream::Tcp(stream) => stream.write(buf),
+        }
+    }
+
+    fn flush(&mut self) -> std::io::Result<()> {
+        match self {
+            SocketStream::Unix(stream) => stream.flush(),
+            SocketStream::Tcp(stream) => stream.flush(),
+        }
+    }
+}
+
+impl AsRawFd for SocketStream {
+    fn as_raw_fd(&self) -> RawFd {
+        match self {
+            SocketStream::Unix(s) => s.as_raw_fd(),
+            SocketStream::Tcp(s) => s.as_raw_fd(),
+        }
+    }
+}
+
+impl ReadVolatile for SocketStream {
+    fn read_volatile<B: BitmapSlice>(
+        &mut self,
+        buf: &mut VolatileSlice<B>,
+    ) -> std::result::Result<usize, VolatileMemoryError> {
+        match self {
+            SocketStream::Unix(s) => s.read_volatile(buf),
+            SocketStream::Tcp(s) => s.read_volatile(buf),
+        }
+    }
+
+    fn read_exact_volatile<B: BitmapSlice>(
+        &mut self,
+        buf: &mut VolatileSlice<B>,
+    ) -> std::result::Result<(), VolatileMemoryError> {
+        match self {
+            SocketStream::Unix(s) => s.read_exact_volatile(buf),
+            SocketStream::Tcp(s) => s.read_exact_volatile(buf),
+        }
+    }
+}
+
+impl WriteVolatile for SocketStream {
+    fn write_volatile<B: BitmapSlice>(
+        &mut self,
+        buf: &VolatileSlice<B>,
+    ) -> std::result::Result<usize, VolatileMemoryError> {
+        match self {
+            SocketStream::Unix(s) => s.write_volatile(buf),
+            SocketStream::Tcp(s) => s.write_volatile(buf),
+        }
+    }
+
+    fn write_all_volatile<B: BitmapSlice>(
+        &mut self,
+        buf: &VolatileSlice<B>,
+    ) -> std::result::Result<(), VolatileMemoryError> {
+        match self {
+            SocketStream::Unix(s) => s.write_all_volatile(buf),
+            SocketStream::Tcp(s) => s.write_all_volatile(buf),
+        }
+    }
+}
+
 pub struct EpollContext {
     epoll_file: File,
 }
@@ -786,6 +870,8 @@ impl Vmm {
             false,
             #[cfg(feature = "sev_snp")]
             false,
+            #[cfg(feature = "sev_snp")]
+            config.lock().unwrap().memory.total_size(),
         )
         .map_err(|e| {
             MigratableError::MigrateReceive(anyhow!(
@@ -919,14 +1005,72 @@ impl Vmm {
             .map(|s| s.into())
     }
 
+    fn send_migration_socket(
+        destination_url: &str,
+    ) -> std::result::Result<SocketStream, MigratableError> {
+        if let Some(address) = destination_url.strip_prefix("tcp:") {
+            info!("Connecting to TCP socket at {}", address);
+
+            let socket = TcpStream::connect(address).map_err(|e| {
+                MigratableError::MigrateSend(anyhow!("Error connecting to TCP socket: {}", e))
+            })?;
+
+            Ok(SocketStream::Tcp(socket))
+        } else {
+            let path = Vmm::socket_url_to_path(destination_url)?;
+            info!("Connecting to UNIX socket at {:?}", path);
+
+            let socket = UnixStream::connect(&path).map_err(|e| {
+                MigratableError::MigrateSend(anyhow!("Error connecting to UNIX socket: {}", e))
+            })?;
+
+            Ok(SocketStream::Unix(socket))
+        }
+    }
+
+    fn receive_migration_socket(
+        receiver_url: &str,
+    ) -> std::result::Result<SocketStream, MigratableError> {
+        if let Some(address) = receiver_url.strip_prefix("tcp:") {
+            let listener = TcpListener::bind(address).map_err(|e| {
+                MigratableError::MigrateReceive(anyhow!("Error binding to TCP socket: {}", e))
+            })?;
+
+            let (socket, _addr) = listener.accept().map_err(|e| {
+                MigratableError::MigrateReceive(anyhow!(
+                    "Error accepting connection on TCP socket: {}",
+                    e
+                ))
+            })?;
+
+            Ok(SocketStream::Tcp(socket))
+        } else {
+            let path = Vmm::socket_url_to_path(receiver_url)?;
+            let listener = UnixListener::bind(&path).map_err(|e| {
+                MigratableError::MigrateReceive(anyhow!("Error binding to UNIX socket: {}", e))
+            })?;
+
+            let (socket, _addr) = listener.accept().map_err(|e| {
+                MigratableError::MigrateReceive(anyhow!(
+                    "Error accepting connection on UNIX socket: {}",
+                    e
+                ))
+            })?;
+
+            // Remove the UNIX socket file after accepting the connection
+            std::fs::remove_file(&path).map_err(|e| {
+                MigratableError::MigrateReceive(anyhow!("Error removing UNIX socket file: {}", e))
+            })?;
+
+            Ok(SocketStream::Unix(socket))
+        }
+    }
+
     // Returns true if there were dirty pages to send
-    fn vm_maybe_send_dirty_pages<T>(
+    fn vm_maybe_send_dirty_pages(
         vm: &mut Vm,
-        socket: &mut T,
-    ) -> result::Result<bool, MigratableError>
-    where
-        T: Read + Write + WriteVolatile,
-    {
+        socket: &mut SocketStream,
+    ) -> result::Result<bool, MigratableError> {
         // Send (dirty) memory table
         let table = vm.dirty_log()?;
 
@@ -954,10 +1098,8 @@ impl Vmm {
         >,
         send_data_migration: VmSendMigrationData,
     ) -> result::Result<(), MigratableError> {
-        let path = Self::socket_url_to_path(&send_data_migration.destination_url)?;
-        let mut socket = UnixStream::connect(path).map_err(|e| {
-            MigratableError::MigrateSend(anyhow!("Error connecting to UNIX socket: {}", e))
-        })?;
+        // Set up the socket connection
+        let mut socket = Self::send_migration_socket(&send_data_migration.destination_url)?;
 
         // Start the migration
         Request::start().write_to(&mut socket)?;
@@ -997,7 +1139,17 @@ impl Vmm {
         };
 
         if send_data_migration.local {
-            vm.send_memory_fds(&mut socket)?;
+            match &mut socket {
+                SocketStream::Unix(unix_socket) => {
+                    // Proceed with sending memory file descriptors over UNIX socket
+                    vm.send_memory_fds(unix_socket)?;
+                }
+                SocketStream::Tcp(_tcp_socket) => {
+                    return Err(MigratableError::MigrateSend(anyhow!(
+                        "--local option is not supported with TCP sockets",
+                    )));
+                }
+            }
         }
 
         let vm_migration_config = VmMigrationConfig {
@@ -1958,16 +2110,8 @@ impl RequestHandler for Vmm {
             receive_data_migration.receiver_url
         );
 
-        let path = Self::socket_url_to_path(&receive_data_migration.receiver_url)?;
-        let listener = UnixListener::bind(&path).map_err(|e| {
-            MigratableError::MigrateReceive(anyhow!("Error binding to UNIX socket: {}", e))
-        })?;
-        let (mut socket, _addr) = listener.accept().map_err(|e| {
-            MigratableError::MigrateReceive(anyhow!("Error accepting on UNIX socket: {}", e))
-        })?;
-        std::fs::remove_file(&path).map_err(|e| {
-            MigratableError::MigrateReceive(anyhow!("Error unlinking UNIX socket: {}", e))
-        })?;
+        // Accept the connection and get the socket
+        let mut socket = Vmm::receive_migration_socket(&receive_data_migration.receiver_url)?;
 
         let mut started = false;
         let mut memory_manager: Option<Arc<Mutex<MemoryManager>>> = None;
@@ -2035,24 +2179,35 @@ impl RequestHandler for Vmm {
                         continue;
                     }
 
-                    let mut buf = [0u8; 4];
-                    let (_, file) = socket.recv_with_fd(&mut buf).map_err(|e| {
-                        MigratableError::MigrateReceive(anyhow!(
-                            "Error receiving slot from socket: {}",
-                            e
-                        ))
-                    })?;
+                    match &mut socket {
+                        SocketStream::Unix(unix_socket) => {
+                            let mut buf = [0u8; 4];
+                            let (_, file) = unix_socket.recv_with_fd(&mut buf).map_err(|e| {
+                                MigratableError::MigrateReceive(anyhow!(
+                                    "Error receiving slot from socket: {}",
+                                    e
+                                ))
+                            })?;
 
-                    if existing_memory_files.is_none() {
-                        existing_memory_files = Some(HashMap::default())
+                            if existing_memory_files.is_none() {
+                                existing_memory_files = Some(HashMap::default())
+                            }
+
+                            if let Some(ref mut existing_memory_files) = existing_memory_files {
+                                let slot = u32::from_le_bytes(buf);
+                                existing_memory_files.insert(slot, file.unwrap());
+                            }
+
+                            Response::ok().write_to(&mut socket)?;
+                        }
+                        SocketStream::Tcp(_tcp_socket) => {
+                            // For TCP sockets, we cannot transfer file descriptors
+                            warn!(
+                                "MemoryFd command received over TCP socket, which is not supported"
+                            );
+                            Response::error().write_to(&mut socket)?;
+                        }
                     }
-
-                    if let Some(ref mut existing_memory_files) = existing_memory_files {
-                        let slot = u32::from_le_bytes(buf);
-                        existing_memory_files.insert(slot, file.unwrap());
-                    }
-
-                    Response::ok().write_to(&mut socket)?;
                 }
                 Command::Complete => {
                     info!("Complete Command Received");

vmm/src/memory_manager.rs

@@ -13,6 +13,7 @@ use std::ops::{BitAnd, Deref, Not, Sub};
 use std::os::fd::AsFd;
 use std::os::unix::io::{AsRawFd, FromRawFd, RawFd};
 use std::path::PathBuf;
+use std::sync::atomic::{AtomicU32, Ordering};
 use std::sync::{Arc, Barrier, Mutex};
 use std::{ffi, result, thread};
 
@@ -33,7 +34,7 @@ use tracer::trace_scoped;
 use virtio_devices::BlocksState;
 #[cfg(target_arch = "x86_64")]
 use vm_allocator::GsiApic;
-use vm_allocator::{AddressAllocator, SystemAllocator};
+use vm_allocator::{AddressAllocator, MemorySlotAllocator, SystemAllocator};
 use vm_device::BusDevice;
 use vm_memory::bitmap::AtomicBitmap;
 use vm_memory::guest_memory::FileOffset;
@@ -162,7 +163,8 @@ struct ArchMemRegion {
 pub struct MemoryManager {
     boot_guest_memory: GuestMemoryMmap,
     guest_memory: GuestMemoryAtomic<GuestMemoryMmap>,
-    next_memory_slot: u32,
+    next_memory_slot: Arc<AtomicU32>,
+    memory_slot_free_list: Arc<Mutex<Vec<u32>>>,
     start_of_device_area: GuestAddress,
     end_of_device_area: GuestAddress,
     end_of_ram_area: GuestAddress,
@@ -1155,17 +1157,10 @@ impl MemoryManager {
 
         let guest_memory = GuestMemoryAtomic::new(guest_memory);
 
-        // Both MMIO and PIO address spaces start at address 0.
         let allocator = Arc::new(Mutex::new(
             SystemAllocator::new(
-                #[cfg(target_arch = "x86_64")]
-                {
-                    GuestAddress(0)
-                },
-                #[cfg(target_arch = "x86_64")]
-                {
-                    1 << 16
-                },
+                GuestAddress(0),
+                1 << 16,
                 start_of_platform_device_area,
                 PLATFORM_DEVICE_AREA_SIZE,
                 #[cfg(target_arch = "x86_64")]
@@ -1205,7 +1200,8 @@ impl MemoryManager {
         let mut memory_manager = MemoryManager {
             boot_guest_memory,
             guest_memory,
-            next_memory_slot,
+            next_memory_slot: Arc::new(AtomicU32::new(next_memory_slot)),
+            memory_slot_free_list: Arc::new(Mutex::new(Vec::new())),
             start_of_device_area,
             end_of_device_area,
             end_of_ram_area,
@@ -1729,10 +1725,14 @@ impl MemoryManager {
         self.end_of_device_area
     }
 
+    pub fn memory_slot_allocator(&mut self) -> MemorySlotAllocator {
+        let memory_slot_free_list = Arc::clone(&self.memory_slot_free_list);
+        let next_memory_slot = Arc::clone(&self.next_memory_slot);
+        MemorySlotAllocator::new(next_memory_slot, memory_slot_free_list)
+    }
+
     pub fn allocate_memory_slot(&mut self) -> u32 {
-        let slot_id = self.next_memory_slot;
-        self.next_memory_slot += 1;
-        slot_id
+        self.memory_slot_allocator().next_memory_slot()
     }
 
     pub fn create_userspace_mapping(
@@ -2120,7 +2120,7 @@ impl MemoryManager {
             current_ram: self.current_ram,
             arch_mem_regions: self.arch_mem_regions.clone(),
             hotplug_slots: self.hotplug_slots.clone(),
-            next_memory_slot: self.next_memory_slot,
+            next_memory_slot: self.next_memory_slot.load(Ordering::SeqCst),
             selected_slot: self.selected_slot,
             next_hotplug_slot: self.next_hotplug_slot,
         }

vmm/src/vm.rs

@@ -531,7 +531,6 @@ impl Vm {
         let stop_on_boot = false;
 
         let memory = memory_manager.lock().unwrap().guest_memory();
-        #[cfg(target_arch = "x86_64")]
         let io_bus = Arc::new(Bus::new());
         let mmio_bus = Arc::new(Bus::new());
 
@@ -622,7 +621,6 @@ impl Vm {
         let dynamic = true;
 
         let device_manager = DeviceManager::new(
-            #[cfg(target_arch = "x86_64")]
             io_bus,
             mmio_bus,
             vm.clone(),
@@ -827,6 +825,8 @@ impl Vm {
             tdx_enabled,
             #[cfg(feature = "sev_snp")]
             sev_snp_enabled,
+            #[cfg(feature = "sev_snp")]
+            vm_config.lock().unwrap().memory.total_size(),
         )?;
 
         let phys_bits = physical_bits(&hypervisor, vm_config.lock().unwrap().cpus.max_phys_bits);
@@ -885,6 +885,7 @@ impl Vm {
         hypervisor: &Arc<dyn hypervisor::Hypervisor>,
         #[cfg(feature = "tdx")] tdx_enabled: bool,
         #[cfg(feature = "sev_snp")] sev_snp_enabled: bool,
+        #[cfg(feature = "sev_snp")] mem_size: u64,
     ) -> Result<Arc<dyn hypervisor::Vm>> {
         hypervisor.check_required_extensions().unwrap();
 
@@ -901,7 +902,7 @@ impl Vm {
                 // Otherwise SEV_SNP_DISABLED: 0
                 // value of sev_snp_enabled is mapped to SEV_SNP_ENABLED for true or SEV_SNP_DISABLED for false
                 let vm = hypervisor
-                    .create_vm_with_type(u64::from(sev_snp_enabled))
+                    .create_vm_with_type_and_memory(u64::from(sev_snp_enabled), mem_size)
                     .unwrap();
             } else {
                 let vm = hypervisor.create_vm().unwrap();