how to implement gpu operator to support kubevirt and pod infrence at the same time

[amghazanfari]

Hello there!

I want to create a kubernetes environment that both pods and kubevirt vms can access to gpu. for kubevirt i used vfio so i can passthrough gpu. in pods i wanted to use kata container (as nvidia recommended) but the problem is that it only supports one gpu per pod. is there any way to fix this probelm?