On Fri, Jul 13, 2018 at 03:09:16PM +0200, Marc-André Lureau wrote: > Add new virtio-gpu devices with a "vhost-user" property. Tthe > associated vhost-user backend is used to handle the virtio rings and > provide rendering results thanks to the vhost-user-gpu protocol. > > Example usage: > -object vhost-user-backend,id=vug,cmd="./vhost-user-gpu" > -device vhost-user-vga,vhost-user=vug
This is a pretty different approach from what QEMU has done for modelling in the past. We've always had separation between the frontend device and the backend provider, with a single frontend device impl being plugged into an arbitrary backend type. This though is introducing a frontend device that actually has the name of a backend which is rather bizarre and quite misleading as it on first sight looks like a new device type. If we must introduce new device types, at very least the confusion could be reduced by calling it "virtio-vga-vhost" so it is clear its the same device type as virtio-vga, likewise 'virtio-gpu-pci-vhost' Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
