October 15, 20232 yr I am struggling to get my gpu passed through to vms. I feel like I may be trying to do something that is not possible. I have nvidia plugin/driver installed and I have jellyfin pointed at gpu for hw transcodes. Is this something i can do simultaneously? Can i have the same gpu working with dockers/plugin driver and also pass it to vms? I cannot find anything specifcally saying it cannot be done, but I think i may be over extending it. Thank you for any help offered.
October 16, 20232 yr Solution 1 hour ago, thedudeabides79 said: Is this something i can do simultaneously? No.
October 16, 20232 yr You can do it with the script, so when you start the VM you stop the docker container, unload drivers and disable GPU in the system, then passthrough it to VM, but simultaneously I'm not so sure. Another variant is some USB GPU(external case, or just usb tiny usb gpu for docker and iGPU(or if you have dedicated) for VM).
October 16, 20232 yr https://forums.unraid.net/topic/136621-unraid-plugin-for-igpu-sr-iov-support/ This works for intel integrated GPU or nvidia, so you can use it in multiple VM/docker containers
October 17, 20232 yr Author I kind of thought this was going to be the case. thank you all for the replies and support. So i have ordered an amd gpu and will hopefully have it friday. what i am thinking is that i can then keep the nvidia with the existing plugin/driver config working for jellyfin, and then set up new amd gpu for passthrough to vms. Does this sound like something that should bear more tasty fruit?? Thanks again. Really appreciate it.
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.