Who made this implementation? (OpenGL/Xen/Qubes Team/… ?)
Where I can read about how it works?
Is it singlethreaded or maxed to two threads or utilizes maximum cores in vm? (some people say it uses “sync” eg singlethreaded, some people say its multithreaded, mine tests shows load maximum at 200% in xentop (two threads while 8vcpus in vm).