docs/system/arm/virt.rst

.. _arm-virt:

'virt' generic virtual platform (``virt``)
==========================================

The ``virt`` board is a platform which does not correspond to any
real hardware; it is designed for use in virtual machines.
It is the recommended board type if you simply want to run
a guest such as Linux and do not care about reproducing the
idiosyncrasies and limitations of a particular bit of real-world
hardware.

This is a "versioned" board model, so as well as the ``virt`` machine
type itself (which may have improvements, bugfixes and other minor
changes between QEMU versions) a version is provided that guarantees
to have the same behaviour as that of previous QEMU releases, so
that VM migration will work between QEMU versions. For instance the
``virt-5.0`` machine type will behave like the ``virt`` machine from
the QEMU 5.0 release, and migration should work between ``virt-5.0``
of the 5.0 release and ``virt-5.0`` of the 5.1 release. Migration
is not guaranteed to work between different QEMU releases for
the non-versioned ``virt`` machine type.

VM migration is not guaranteed when using ``-cpu max``, as features
supported may change between QEMU versions.  To ensure your VM can be
migrated, it is recommended to use another cpu model instead.

Supported devices
"""""""""""""""""

The virt board supports:

- PCI/PCIe devices
- CXL Fixed memory windows, root bridges and devices.
- Flash memory
- Either one or two PL011 UARTs for the NonSecure World
- An RTC
- The fw_cfg device that allows a guest to obtain data from QEMU
- A PL061 GPIO controller
- An optional machine-wide SMMUv3 IOMMU
- User-creatable SMMUv3 devices (see below for example)
- An optional SBSA Generic Watchdog Timer (see below)
- hotpluggable DIMMs
- hotpluggable NVDIMMs
- An MSI controller (GICv2m or ITS).
  - When using GICv3, ITS is selected by default when available on the platform.
  - If using GICv2, a GICv2m is provided by default instead.
  - When ITS is not available on a GICv3 platform, a GICv2m is provided by default.
- 32 virtio-mmio transport devices
- running guests using the KVM accelerator on aarch64 hardware
- large amounts of RAM (at least 255GB, and more if using highmem)
- many CPUs (up to 512 if using a GICv3 and highmem)
- Secure-World-only devices if the CPU has TrustZone:

  - A second PL011 UART
  - A second PL061 GPIO controller, with GPIO lines for triggering
    a system reset or system poweroff
  - A secure flash memory
  - 16MB of secure RAM

The second NonSecure UART only exists if a backend is configured
explicitly (e.g. with a second -serial command line option) and
TrustZone emulation is not enabled.

Supported guest CPU types:

- ``cortex-a7`` (32-bit)
- ``cortex-a15`` (32-bit; the default)
- ``cortex-a35`` (64-bit)
- ``cortex-a53`` (64-bit)
- ``cortex-a55`` (64-bit)
- ``cortex-a57`` (64-bit)
- ``cortex-a72`` (64-bit)
- ``cortex-a76`` (64-bit)
- ``cortex-a710`` (64-bit)
- ``a64fx`` (64-bit)
- ``host`` (with KVM and HVF only)
- ``neoverse-n1`` (64-bit)
- ``neoverse-v1`` (64-bit)
- ``neoverse-n2`` (64-bit)
- ``max`` (same as ``host`` for KVM and HVF; best possible emulation with TCG)

Note that the default is ``cortex-a15``, so for an AArch64 guest you must
specify a CPU type.

Also, please note that passing ``max`` CPU (i.e. ``-cpu max``) won't
enable all the CPU features for a given ``virt`` machine. Where a CPU
architectural feature requires support in both the CPU itself and in the
wider system (e.g. the MTE feature), it may not be enabled by default,
but instead requires a machine option to enable it.

For example, MTE support must be enabled with ``-machine virt,mte=on``,
as well as by selecting an MTE-capable CPU (e.g., ``max``) with the
``-cpu`` option.

See the machine-specific options below, or check them for a given machine
by passing the ``help`` suboption, like: ``-machine virt-9.0,help``.

Graphics output is available, but unlike the x86 PC machine types
there is no default display device enabled: you should select one from
the Display devices section of "-device help". The recommended option
is ``virtio-gpu-pci``; this is the only one which will work correctly
with KVM. You may also need to ensure your guest kernel is configured
with support for this; see below.

Machine-specific options
""""""""""""""""""""""""

The following machine-specific options are supported:

secure
  Set ``on``/``off`` to enable/disable emulating a guest CPU which implements the
  Arm Security Extensions (TrustZone). The default is ``off``.

virtualization
  Set ``on``/``off`` to enable/disable emulating a guest CPU which implements the
  Arm Virtualization Extensions. The default is ``off``.

mte
  Set ``on``/``off`` to enable/disable emulating a guest CPU which implements the
  Arm Memory Tagging Extensions. The default is ``off``.

highmem
  Set ``on``/``off`` to enable/disable placing devices and RAM in physical
  address space above 32 bits. The default is ``on`` for machine types
  later than ``virt-2.12`` when the CPU supports an address space
  bigger than 32 bits (i.e. 64-bit CPUs, and 32-bit CPUs with the
  Large Physical Address Extension (LPAE) feature). If you want to
  boot a 32-bit kernel which does not have ``CONFIG_LPAE`` enabled on
  a CPU type which implements LPAE, you will need to manually set
  this to ``off``; otherwise some devices, such as the PCI controller,
  will not be accessible.

compact-highmem
  Set ``on``/``off`` to enable/disable the compact layout for high memory regions.
  The default is ``on`` for machine types later than ``virt-7.2``.

highmem-redists
  Set ``on``/``off`` to enable/disable the high memory region for GICv3 or
  GICv4 redistributor. The default is ``on``. Setting this to ``off`` will
  limit the maximum number of CPUs when GICv3 or GICv4 is used.

highmem-ecam
  Set ``on``/``off`` to enable/disable the high memory region for PCI ECAM.
  The default is ``on`` for machine types later than ``virt-3.0``.

highmem-mmio
  Set ``on``/``off`` to enable/disable the high memory region for PCI MMIO.
  The default is ``on``.

highmem-mmio-size
  Set the high memory region size for PCI MMIO. Must be a power of 2 and
  greater than or equal to the default size (512G).

gic-version
  Specify the version of the Generic Interrupt Controller (GIC) to provide.
  Valid values are:

  ``2``
    GICv2. Note that this limits the number of CPUs to 8.
  ``3``
    GICv3. This allows up to 512 CPUs.
  ``4``
    GICv4. Requires ``virtualization`` to be ``on``; allows up to 317 CPUs.
  ``x-5``
    GICv5 (experimental). This is an experimental emulation of the GICv5,
    based on the EAC release of the GICv5 architecture specification.
    Experimental means:

    - guest-visible behaviour may change when the final version of
      the specification is released and QEMU implements it
    - migration support is not yet implemented
    - the GICv5 is not exposed to the guest via ACPI tables, only via DTB
    - the way the interrupt controller is exposed to the guest and the
      command line syntax for enabling it may change

    The current implementation supports only an EL1 guest (no EL2 or
    EL3 and no Realm support), and does not implement the ITS (no
    MSI support).

    Note that as the GICv5 is an Armv9 feature, enabling it will
    automatically disable support for AArch32 at all exception levels
    except for EL0 (userspace).
  ``host``
    Use the same GIC version the host provides, when using KVM
  ``max``
    Use the best GIC version possible (same as host when using KVM;
    with TCG this is currently ``3`` if ``virtualization`` is ``off`` and
    ``4`` if ``virtualization`` is ``on``, but this may change in future)

msi
  Specify the MSI and MSI-X controller (GIC) to provide.
  Valid values are:

  ``auto``
    Use the best available MSI-X controller option. ITS when supported, GICv2m otherwise.
  ``gicv2m``
    GICv2m. Typically used with a GICv2. Also available with a newer GIC.
  ``its``
    GICv3 ITS. This is the default option when using a GICv3 or GICv4 with a supported
    accelerator.
  ``off``
    Disable support for MSI/MSI-X interrupts.

its
  Set ``on``/``off`` to control ITS instantiation. This is a deprecated option, use ``msi`` instead.

iommu
  Set the IOMMU type to create for the guest. Valid values are:

  ``none``
    Don't create an IOMMU (the default)
  ``smmuv3``
    Create a machine-wide SMMUv3.

default-bus-bypass-iommu
  Set ``on``/``off`` to enable/disable `bypass_iommu
  <https://gitlab.com/qemu-project/qemu/-/blob/master/docs/bypass-iommu.txt>`_
  for default root bus.

ras
  Set ``on``/``off`` to enable/disable reporting host memory errors to a guest
  using ACPI and guest external abort exceptions. The default is off.

acpi
  Set ``on``/``off``/``auto`` to enable/disable ACPI.

cxl
  Set  ``on``/``off`` to enable/disable CXL. More details in
  :doc:`../devices/cxl`. The default is off.

cxl-fmw
  Array of CXL fixed memory windows describing fixed address routing to
  target CXL host bridges. See :doc:`../devices/cxl`.

dtb-randomness
  Set ``on``/``off`` to pass random seeds via the guest DTB
  rng-seed and kaslr-seed nodes (in both "/chosen" and
  "/secure-chosen") to use for features like the random number
  generator and address space randomisation. The default is
  ``on``. You will want to disable it if your trusted boot chain
  will verify the DTB it is passed, since this option causes the
  DTB to be non-deterministic. It would be the responsibility of
  the firmware to come up with a seed and pass it on if it wants to.

dtb-kaslr-seed
  A deprecated synonym for dtb-randomness.

virtio-mmio-transports
  Set the number of virtio-mmio transports to create (between 0 and 32;
  the default is 32).  Unused transports are harmless, but you can
  use this property to avoid exposing them to the guest if you wish.

x-oem-id
  Set string (up to 6 bytes) to override the default value of field OEMID in ACPI
  table header.

x-oem-table-id
  Set string (up to 8 bytes) to override the default value of field OEM Table ID
  in ACPI table header.

SMMU configuration
""""""""""""""""""

Machine-wide SMMUv3 IOMMU
  Setting the machine-specific option ``iommu=smmuv3`` causes QEMU to
  create a single, machine-wide SMMUv3 instance that applies to all
  devices in the PCIe topology.

  For information about selectively bypassing devices, refer to
  ``docs/bypass-iommu.txt``.

User-creatable SMMUv3 devices
  You can use the ``-device arm-smmuv3`` option to create multiple
  user-defined SMMUv3 devices, each associated with a separate PCIe
  root complex. This is only permitted if the machine-wide SMMUv3
  (``iommu=smmuv3``) option is not used. Each ``arm-smmuv3`` device
  uses the ``primary-bus`` sub-option to specify which PCIe root
  complex it is associated with.

  This model is useful when you want to mirror a host configuration where
  each NUMA node typically has its own SMMU, allowing the VM topology to
  align more closely with the host’s hardware layout.

  Example::

      -device arm-smmuv3,primary-bus=pcie.0,id=smmuv3.0
      ...
      -device pxb-pcie,id=pcie.1,numa_node=1
      -device arm-smmuv3,primary-bus=pcie.1,id=smmuv3.1

  *Accelerated SMMUv3 (nested translation)*

  The ``accel=on`` option enables hardware-accelerated nested translation
  for vfio-pci passthrough devices. In this mode the guest SMMU driver
  programs its own Stage-1 page tables, with the host SMMUv3 handling both
  Stage-1 (guest) and Stage-2 (host) translations in hardware. The host
  SMMUv3 must support nested translation. This mode requires the iommufd
  backend and is only supported when booting with ACPI (not device tree).

  When ``accel=on``, QEMU automatically derives the values for the
  ``ril``, ``ats``, ``oas``, ``ssidsize`` and ``cmdqv`` sub-options
  from the host SMMUv3 capabilities unless they are set explicitly.

  Example::

      -device arm-smmuv3,primary-bus=pcie.0,id=smmuv3.0,accel=on

  *Accelerated SMMUv3 command queues (Tegra241 CMDQV)*

  The ``cmdqv`` sub-option enables NVIDIA Tegra241 Command Queue
  Virtualization (CMDQV) on supported hosts. With CMDQV, each accelerated
  ``arm-smmuv3`` device gets dedicated hardware command queues and the
  guest issues SMMU invalidation commands directly to real hardware,
  bypassing QEMU and improving throughput for workloads that issue many
  invalidations. Without it, every invalidation command traps into QEMU.

SBSA Generic Watchdog
"""""""""""""""""""""

The SBSA Generic Watchdog Timer (GWDT) can be added to the virt machine
using ``-device sbsa-gwdt``. It is only supported on the virt machine,
which wires up statically assigned MMIO regions and IRQs via
machine-specific plug handlers.

Two modes are available:

Native mode (default)
  The watchdog is described via the ACPI GTDT table and FDT, using
  the system counter frequency. Example::

    -device sbsa-gwdt

WDAT mode
  The watchdog is described via the ACPI WDAT table (no FDT node),
  using a 1 kHz timer frequency. WDAT and GTDT watchdog entries are
  mutually exclusive. Example::

    -device sbsa-gwdt,wdat=on

Linux guest kernel configuration
""""""""""""""""""""""""""""""""

The 'defconfig' for Linux arm and arm64 kernels should include the
right device drivers for virtio and the PCI controller; however some older
kernel versions, especially for 32-bit Arm, did not have everything
enabled by default. If you're not seeing PCI devices that you expect,
then check that your guest config has::

  CONFIG_PCI=y
  CONFIG_VIRTIO_PCI=y
  CONFIG_PCI_HOST_GENERIC=y

If you want to use the ``virtio-gpu-pci`` graphics device you will also
need::

  CONFIG_DRM=y
  CONFIG_DRM_VIRTIO_GPU=y

Hardware configuration information for bare-metal programming
"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""

The ``virt`` board automatically generates a device tree blob ("dtb")
which it passes to the guest. This provides information about the
addresses, interrupt lines and other configuration of the various devices
in the system. Guest code can rely on and hard-code the following
addresses:

- Flash memory starts at address 0x0000_0000

- RAM starts at 0x4000_0000

All other information about device locations may change between
QEMU versions, so guest code must look in the DTB.

QEMU supports two types of guest image boot for ``virt``, and
the way for the guest code to locate the dtb binary differs:

- For guests using the Linux kernel boot protocol (this means any
  non-ELF file passed to the QEMU ``-kernel`` option) the address
  of the DTB is passed in a register (``r2`` for 32-bit guests,
  or ``x0`` for 64-bit guests)

- For guests booting as "bare-metal" (any other kind of boot),
  the DTB is at the start of RAM (0x4000_0000)