April 11, 20224 yr Is it possible to use a GPU configured for passthrough to VMs by Docker containers when no VMs are running, without changing IOMMU bindings or rebooting? Based on my research it seems this isn't possible, but I figured I'd ask in case I'm misunderstanding something. I can't run a 2nd GPU in my system and only use my GPU for my gaming VM. I'd love to be able to setup automated video encoding that only runs when that VM is shutdown. I already built the container for this, only diff is it checks whether other containers are runnning. If I can figure out how to use the GPU when the IOMMU group is bound, then it's only a small change to make it only run when the VM isn't running My backup plan is a Linux VM that runs the scripts then shuts down when complete, but I'd rather not have to architect another solution Edited April 11, 20224 yr by c0d3m0nk3y punctuation
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.