docs/system/arm/virt.rst

   1 'virt' generic virtual platform (``virt``)
   2 ==========================================
   3
   4 The ``virt`` board is a platform which does not correspond to any
   5 real hardware; it is designed for use in virtual machines.
   6 It is the recommended board type if you simply want to run
   7 a guest such as Linux and do not care about reproducing the
   8 idiosyncrasies and limitations of a particular bit of real-world
   9 hardware.
  10
  11 This is a "versioned" board model, so as well as the ``virt`` machine
  12 type itself (which may have improvements, bugfixes and other minor
  13 changes between QEMU versions) a version is provided that guarantees
  14 to have the same behaviour as that of previous QEMU releases, so
  15 that VM migration will work between QEMU versions. For instance the
  16 ``virt-5.0`` machine type will behave like the ``virt`` machine from
  17 the QEMU 5.0 release, and migration should work between ``virt-5.0``
  18 of the 5.0 release and ``virt-5.0`` of the 5.1 release. Migration
  19 is not guaranteed to work between different QEMU releases for
  20 the non-versioned ``virt`` machine type.
  21
  22 Supported devices
  23 """""""""""""""""
  24
  25 The virt board supports:
  26
  27 - PCI/PCIe devices
  28 - Flash memory
  29 - One PL011 UART
  30 - An RTC
  31 - The fw_cfg device that allows a guest to obtain data from QEMU
  32 - A PL061 GPIO controller
  33 - An optional SMMUv3 IOMMU
  34 - hotpluggable DIMMs
  35 - hotpluggable NVDIMMs
  36 - An MSI controller (GICv2M or ITS). GICv2M is selected by default along
  37   with GICv2. ITS is selected by default with GICv3 (>= virt-2.7). Note
  38   that ITS is not modeled in TCG mode.
  39 - 32 virtio-mmio transport devices
  40 - running guests using the KVM accelerator on aarch64 hardware
  41 - large amounts of RAM (at least 255GB, and more if using highmem)
  42 - many CPUs (up to 512 if using a GICv3 and highmem)
  43 - Secure-World-only devices if the CPU has TrustZone:
  44
  45   - A second PL011 UART
  46   - A second PL061 GPIO controller, with GPIO lines for triggering
  47     a system reset or system poweroff
  48   - A secure flash memory
  49   - 16MB of secure RAM
  50
  51 Supported guest CPU types:
  52
  53 - ``cortex-a7`` (32-bit)
  54 - ``cortex-a15`` (32-bit; the default)
  55 - ``cortex-a35`` (64-bit)
  56 - ``cortex-a53`` (64-bit)
  57 - ``cortex-a55`` (64-bit)
  58 - ``cortex-a57`` (64-bit)
  59 - ``cortex-a72`` (64-bit)
  60 - ``cortex-a76`` (64-bit)
  61 - ``cortex-a710`` (64-bit)
  62 - ``a64fx`` (64-bit)
  63 - ``host`` (with KVM only)
  64 - ``neoverse-n1`` (64-bit)
  65 - ``neoverse-v1`` (64-bit)
  66 - ``neoverse-n2`` (64-bit)
  67 - ``max`` (same as ``host`` for KVM; best possible emulation with TCG)
  68
  69 Note that the default is ``cortex-a15``, so for an AArch64 guest you must
  70 specify a CPU type.
  71
  72 Also, please note that passing ``max`` CPU (i.e. ``-cpu max``) won't
  73 enable all the CPU features for a given ``virt`` machine. Where a CPU
  74 architectural feature requires support in both the CPU itself and in the
  75 wider system (e.g. the MTE feature), it may not be enabled by default,
  76 but instead requires a machine option to enable it.
  77
  78 For example, MTE support must be enabled with ``-machine virt,mte=on``,
  79 as well as by selecting an MTE-capable CPU (e.g., ``max``) with the
  80 ``-cpu`` option.
  81
  82 See the machine-specific options below, or check them for a given machine
  83 by passing the ``help`` suboption, like: ``-machine virt-9.0,help``.
  84
  85 Graphics output is available, but unlike the x86 PC machine types
  86 there is no default display device enabled: you should select one from
  87 the Display devices section of "-device help". The recommended option
  88 is ``virtio-gpu-pci``; this is the only one which will work correctly
  89 with KVM. You may also need to ensure your guest kernel is configured
  90 with support for this; see below.
  91
  92 Machine-specific options
  93 """"""""""""""""""""""""
  94
  95 The following machine-specific options are supported:
  96
  97 secure
  98   Set ``on``/``off`` to enable/disable emulating a guest CPU which implements the
  99   Arm Security Extensions (TrustZone). The default is ``off``.
 100
 101 virtualization
 102   Set ``on``/``off`` to enable/disable emulating a guest CPU which implements the
 103   Arm Virtualization Extensions. The default is ``off``.
 104
 105 mte
 106   Set ``on``/``off`` to enable/disable emulating a guest CPU which implements the
 107   Arm Memory Tagging Extensions. The default is ``off``.
 108
 109 highmem
 110   Set ``on``/``off`` to enable/disable placing devices and RAM in physical
 111   address space above 32 bits. The default is ``on`` for machine types
 112   later than ``virt-2.12`` when the CPU supports an address space
 113   bigger than 32 bits (i.e. 64-bit CPUs, and 32-bit CPUs with the
 114   Large Physical Address Extension (LPAE) feature). If you want to
 115   boot a 32-bit kernel which does not have ``CONFIG_LPAE`` enabled on
 116   a CPU type which implements LPAE, you will need to manually set
 117   this to ``off``; otherwise some devices, such as the PCI controller,
 118   will not be accessible.
 119
 120 compact-highmem
 121   Set ``on``/``off`` to enable/disable the compact layout for high memory regions.
 122   The default is ``on`` for machine types later than ``virt-7.2``.
 123
 124 highmem-redists
 125   Set ``on``/``off`` to enable/disable the high memory region for GICv3 or
 126   GICv4 redistributor. The default is ``on``. Setting this to ``off`` will
 127   limit the maximum number of CPUs when GICv3 or GICv4 is used.
 128
 129 highmem-ecam
 130   Set ``on``/``off`` to enable/disable the high memory region for PCI ECAM.
 131   The default is ``on`` for machine types later than ``virt-3.0``.
 132
 133 highmem-mmio
 134   Set ``on``/``off`` to enable/disable the high memory region for PCI MMIO.
 135   The default is ``on``.
 136
 137 gic-version
 138   Specify the version of the Generic Interrupt Controller (GIC) to provide.
 139   Valid values are:
 140
 141   ``2``
 142     GICv2. Note that this limits the number of CPUs to 8.
 143   ``3``
 144     GICv3. This allows up to 512 CPUs.
 145   ``4``
 146     GICv4. Requires ``virtualization`` to be ``on``; allows up to 317 CPUs.
 147   ``host``
 148     Use the same GIC version the host provides, when using KVM
 149   ``max``
 150     Use the best GIC version possible (same as host when using KVM;
 151     with TCG this is currently ``3`` if ``virtualization`` is ``off`` and
 152     ``4`` if ``virtualization`` is ``on``, but this may change in future)
 153
 154 its
 155   Set ``on``/``off`` to enable/disable ITS instantiation. The default is ``on``
 156   for machine types later than ``virt-2.7``.
 157
 158 iommu
 159   Set the IOMMU type to create for the guest. Valid values are:
 160
 161   ``none``
 162     Don't create an IOMMU (the default)
 163   ``smmuv3``
 164     Create an SMMUv3
 165
 166 ras
 167   Set ``on``/``off`` to enable/disable reporting host memory errors to a guest
 168   using ACPI and guest external abort exceptions. The default is off.
 169
 170 dtb-randomness
 171   Set ``on``/``off`` to pass random seeds via the guest DTB
 172   rng-seed and kaslr-seed nodes (in both "/chosen" and
 173   "/secure-chosen") to use for features like the random number
 174   generator and address space randomisation. The default is
 175   ``on``. You will want to disable it if your trusted boot chain
 176   will verify the DTB it is passed, since this option causes the
 177   DTB to be non-deterministic. It would be the responsibility of
 178   the firmware to come up with a seed and pass it on if it wants to.
 179
 180 dtb-kaslr-seed
 181   A deprecated synonym for dtb-randomness.
 182
 183 Linux guest kernel configuration
 184 """"""""""""""""""""""""""""""""
 185
 186 The 'defconfig' for Linux arm and arm64 kernels should include the
 187 right device drivers for virtio and the PCI controller; however some older
 188 kernel versions, especially for 32-bit Arm, did not have everything
 189 enabled by default. If you're not seeing PCI devices that you expect,
 190 then check that your guest config has::
 191
 192   CONFIG_PCI=y
 193   CONFIG_VIRTIO_PCI=y
 194   CONFIG_PCI_HOST_GENERIC=y
 195
 196 If you want to use the ``virtio-gpu-pci`` graphics device you will also
 197 need::
 198
 199   CONFIG_DRM=y
 200   CONFIG_DRM_VIRTIO_GPU=y
 201
 202 Hardware configuration information for bare-metal programming
 203 """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
 204
 205 The ``virt`` board automatically generates a device tree blob ("dtb")
 206 which it passes to the guest. This provides information about the
 207 addresses, interrupt lines and other configuration of the various devices
 208 in the system. Guest code can rely on and hard-code the following
 209 addresses:
 210
 211 - Flash memory starts at address 0x0000_0000
 212
 213 - RAM starts at 0x4000_0000
 214
 215 All other information about device locations may change between
 216 QEMU versions, so guest code must look in the DTB.
 217
 218 QEMU supports two types of guest image boot for ``virt``, and
 219 the way for the guest code to locate the dtb binary differs:
 220
 221 - For guests using the Linux kernel boot protocol (this means any
 222   non-ELF file passed to the QEMU ``-kernel`` option) the address
 223   of the DTB is passed in a register (``r2`` for 32-bit guests,
 224   or ``x0`` for 64-bit guests)
 225
 226 - For guests booting as "bare-metal" (any other kind of boot),
 227   the DTB is at the start of RAM (0x4000_0000)