kvm-qemu-virtualization-guide VS win10-gpu-passthrough

Also, I do remember that OVMF was buggy at the time, so I had to install a custom version for OVMF. Here are the necessary steps if you still need to do it (yes, it was only that): https://github.com/mateussouzaweb/kvm-qemu-virtualization-guide/commit/bba74a15a32fa89d783b57e977db5ff5bcd7a0d6

If you are using Fedora, I would recommend following my guide: https://github.com/mateussouzaweb/kvm-qemu-virtualization-guide

If you are interested, I have been maintaining a guide for it (https://github.com/mateussouzaweb/kvm-qemu-virtualization-guide), but there are many others guides on internet on that subject if you look into virtualization with QEMU/KVM...

Since I am experienced with bash scripts because of my work, I wrote one single hook script to rule all the needs that are common posted here like Single GPU passthrough, CPU scaling and USB redirection. Here is the link for my guide: https://github.com/mateussouzaweb/kvm-qemu-virtualization-guide

Finally. There is a problem where you cant re-attach the NVidia driver after the VM has started. You need to apply this kernel patch.

This kernel patch to solve this problem: https://www.reddit.com/r/VFIO/comments/11vvkn9/dynamic_bindingunbinding_of_vfio_almost_working/

And the custom kernel mod to prevent some memory error: https://github.com/Kinsteen/win10-gpu-passthrough ("remove the check").

There were maybe a couple other threads I referenced which pointed at using a kernel patch to disable some sort of check, but I'm not sure about this.