dsg/vhost-device
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
vhost-device/vhost-device-gpio
History
Exact
Davíð Steinn Geirsson 1c021aaea6 Use our vhost fork for all crates and add GPU flake package 
Use our vhost fork for all crates. It has SHMEM and GPU backend support.
Add a Nix flake package for vhost-device-gpu with the virgl backend enabled.

Also fixes duplicate imports in vhost-device-gpu/src/backend/virgl.rs.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-22 19:10:55 +00:00
..
src chore: remove pub visibility from exit event fields 2025-11-19 12:09:37 +02:00
Cargo.toml Use our vhost fork for all crates and add GPU flake package 2026-03-22 19:10:55 +00:00
CHANGELOG.md chore: standardize changelogs 2024-07-24 19:37:19 +05:30
LICENSE-APACHE Move all crates to workspace root 2023-10-16 12:03:57 +05:30
LICENSE-BSD-3-Clause Move all crates to workspace root 2023-10-16 12:03:57 +05:30
README.md vhost-device-gpio: reformat synopsis and options 2025-09-12 14:22:27 +03:00

README.md

vhost-device-gpio - GPIO emulation backend daemon

Description

This program is a vhost-user backend that emulates a VirtIO GPIO device. This program takes a list of gpio devices on the host OS and then talks to them via the /dev/gpiochip{X} interface when a request comes from the guest OS for an GPIO device.

This program is tested with QEMU's -device vhost-user-gpio-pci but should work with any virtual machine monitor (VMM) that supports vhost-user. See the Examples section below.

Synopsis

vhost-device-gpio [OPTIONS] --socket-path <SOCKET_PATH> --device-list <DEVICE_LIST>

Options

 vhost-device-gpio

 -h, --help

  Print help.

 -s, --socket-path=PATH

  Location of vhost-user Unix domain sockets, this path will be suffixed with
  0,1,2..socket_count-1.

 -c, --socket-count=INT

  Number of guests (sockets) to attach to, default set to 1.

 -l, --device-list=GPIO-DEVICES

  GPIO device list at the host OS in the format:
      <device1>[:<device2>]

      Example: --device-list "2:4:7"

  Here, each GPIO devices correspond to a separate guest instance, i.e. the
  number of devices in the device-list must match the number of sockets in the
  --socket-count. For example, the GPIO device 0 will be allocated to the guest
  with "<socket-path>0" path.

MockGpioDevice support

As connecting VM guests to random GPIO pins on your host is generally asking for trouble you can enable the "mock_gpio" feature in your build:

cargo build --features "mock_gpio"

You can then enable simulated GPIOs using the 's' prefix:

--device-list s4:s8

Which will create two gpio devices, the first with 4 pins and the second with 8. By default updates are display via env logger:

vhost-device-gpio -s /tmp/vus.sock -c 1 -l s4
[2023-09-14T14:15:14Z INFO  vhost_device_gpio::mock_gpio] gpio dummy0 set value to 1
[2023-09-14T14:15:14Z INFO  vhost_device_gpio::mock_gpio] gpio dummy0 set direction to 1
[2023-09-14T14:15:14Z INFO  vhost_device_gpio::mock_gpio] gpio dummy0 set direction to 0
[2023-09-14T14:15:19Z INFO  vhost_device_gpio::mock_gpio] gpio dummy1 set value to 1
[2023-09-14T14:15:19Z INFO  vhost_device_gpio::mock_gpio] gpio dummy1 set direction to 1
[2023-09-14T14:15:19Z INFO  vhost_device_gpio::mock_gpio] gpio dummy1 set direction to 0

Examples

The daemon should be started first:

::

host# vhost-device-gpio --socket-path=gpio.sock --socket-count=2 --device-list 0:3

The QEMU invocation needs to create a chardev socket the device can use to communicate as well as share the guests memory over a memfd.

::

host# qemu-system
-chardev socket,path=vgpio.sock,id=vgpio
-device vhost-user-gpio-pci,chardev=vgpio,id=gpio
-m 4096
-object memory-backend-file,id=mem,size=4G,mem-path=/dev/shm,share=on
-numa node,memdev=mem
...

License

This project is licensed under either of