libvirt/docs/kbase/virtiofs.rst

144 lines
3.3 KiB
ReStructuredText
Raw Normal View History

============================
Sharing files with Virtio-FS
============================
.. contents::
Virtio-FS
=========
Virtio-FS is a shared file system that lets virtual machines access
a directory tree on the host. Unlike existing approaches, it
is designed to offer local file system semantics and performance.
See https://virtio-fs.gitlab.io/
Host setup
==========
The host-side virtiofsd daemon, like other vhost-user backed devices,
requires shared memory between the host and the guest. As of QEMU 4.2, this
requires specifying a NUMA topology for the guest and explicitly specifying
a memory backend. Multiple options are available:
Either of the following:
* Use file-backed memory
Configure the directory where the files backing the memory will be stored
with the ``memory_backing_dir`` option in ``/etc/libvirt/qemu.conf``
::
# This directory is used for memoryBacking source if configured as file.
# NOTE: big files will be stored here
memory_backing_dir = "/dev/shm/"
* Use hugepage-backed memory
Make sure there are enough huge pages allocated for the requested guest memory.
For example, for one guest with 2 GiB of RAM backed by 2 MiB hugepages:
::
# virsh allocpages 2M 1024
Guest setup
===========
#. Specify the NUMA topology
in the domain XML of the guest.
For the simplest one-node topology for a guest with 2GiB of RAM and 8 vCPUs:
::
<domain>
...
<cpu ...>
<numa>
<cell id='0' cpus='0-7' memory='2' unit='GiB' memAccess='shared'/>
</numa>
</cpu>
...
</domain>
Note that the CPU element might already be specified and only one is allowed.
#. Specify the memory backend
Either of the following:
* File-backed memory
::
<domain>
...
<memoryBacking>
<access mode='shared'/>
</memoryBacking>
...
</domain>
This will create a file in the directory specified in ``qemu.conf``
* Hugepage-backed memory
::
<domain>
...
<memoryBacking>
<hugepages>
<page size='2' unit='M'/>
</hugepages>
<access mode='shared'/>
</memoryBacking>
...
</domain>
#. Add the ``vhost-user-fs`` QEMU device via the ``filesystem`` element
::
<domain>
...
<devices>
...
<filesystem type='mount' accessmode='passthrough'>
<driver type='virtiofs'/>
<source dir='/path'/>
<target dir='mount_tag'/>
</filesystem>
...
</devices>
</domain>
Note that despite its name, the ``target dir`` is actually a mount tag and does
not have to correspond to the desired mount point in the guest.
So far, ``passthrough`` is the only supported access mode and it requires
running the ``virtiofsd`` daemon as root.
#. Boot the guest and mount the filesystem
::
guest# mount -t virtiofs mount_tag /mnt/mount/path
Note: this requires virtiofs support in the guest kernel (Linux v5.4 or later)
Optional parameters
===================
More optional elements can be specified
::
<driver type='virtiofs' queue='1024'/>
<binary path='/usr/libexec/virtiofsd' xattr='on'>
<cache mode='always'/>
<lock posix='on' flock='on'/>
</binary>