Using sys-gui-gpu style pass through for other VMs

,

Hello everyone!

I recently picked up a Thinkpad T480 to run Qubes on, and I am loving the experience. However, I do wish I could use even the limited graphics performance of the integrated GPU for certain tasks (mainly games).

I managed to get sys-gui-gpu set up, but it seems like a bad idea to me to give that VM network access and install untrusted software directly into it as it has access to qubes management software.

So, I would like to create an HVM using the same techniques as sys-gui-gpu that acts independently from qubes- almost as if I was dual booting qubes and ie. fedora or windows. The only problem, however, is i do not even know where to begin.

My system only has the Intel iGPU and no Nvidia dGPU, so passing through the iGPU would mean that qubes would have no display out, and I am fine with this.

Does anyone know how I would begin setting this up?

Thanks for the help,
Will

Just wondering if anyone saw this