I would like to run a Windows 10 VM that has access to a directory on my host and I would like to use virtiofs[1] for that. I don't want to run a whole virtualization solution like libvirt since I only need a single VM so I'll try to set this up manually[2]. [1] https://virtio-fs.gitlab.io/ [2] https://virtio-fs.gitlab.io/howto-qemu.html I should use Windows 10 Pro since that runs better in an virtualized environment. System disk should be at least 120GB. qemu-img create -f qcow2 -o compression_type=zstd \ system.qcow2 $(( 120 * 1024 ** 3 )) We run virtiofsd by hand as root for now: /usr/libexec/virtiofsd --socket-path /tmp/virtiofs \ --shared-dir /path/to/shared/dir/ --cache auto And the VM is started like this qemu-system-x86_64 --enable-kvm -name windows-vm \ -cpu host -machine q35,accel=kvm \ -bios /usr/share/OVMF/x64/OVMF_CODE.fd \ -smp 12,sockets=12,cores=1,threads=1 -m 8G \ -object memory-backend-file,id=mem,size=8G,mem-path=/dev/shm,share=on \ -numa node,memdev=mem \ -device qemu-xhci \ -device usb-tablet \ -device virtio-scsi-pci,id=scsi0 \ -drive file=system.qcow2,if=none,discard=unmap,aio=native,cache=none,id=drive0 \ -device scsi-hd,drive=drive0,bus=scsi0.0 \ -chardev socket,id=char0,path=/tmp/virtiofs \ -device vhost-user-fs-pci,queue-size=1024,chardev=char0,tag=shared \ -netdev user,id=netdev0,ipv6=off \ -device virtio-net-pci,netdev=netdev0 \ -monitor unix:/tmp/windows-vm-monitor,server=on,wait=off \ -vga std -display none -vnc 127.0.0.1:0 This way we have a monitor we can use by doing socat unix:/tmp/windows-vm-monitor readline and we could connect via vnc if we wanted to. On the Windows machine we had to install virtio drivers for SCSI and networking and we also had to do the procedure to get virtiofs running on Windows[3]. [3] https://virtio-fs.gitlab.io/howto-windows.html I copied virtiofs.exe to C:\ and set up the service there. Somehow the service doesn't correctly start when rebooting the VM but I've since set the service to start automatically (delayed). I didn't test whether that is a fix for that.