vbios_from_reg VS gvt-linux

Here's a nice little tool if you're having trouble extracting your own vbios.

I used this method: https://github.com/kyokenn/vbios_from_reg

Extra stuff you might want to try if you still have issues: - Extract the vbios from windows (GPU-Z or with this script to get it from registry in windows. - Check with GOPUpd script to see if you have a UEFI GOP (check post2 of the forum for the script, you may also want to read the first post to understand what the script does, basically just drop your vbios into the .bat executable and follow through) - In case your vbios contains a UEFI GOP, you do not need to update, just use it as it is - In case your vbios does not contain a UEFI GOP, install one according to your device's family (Pascal) - add just above the bottom of the passed device and change /path/to/vBIOS.rom to where your vBIOS is (note that sometimes, qemu may not be able to access the file because of apparmor, so use some other place that qemu can use like /usr/share/edk2-ovmf/x64/ or /usr/share/vbios or something or add an exception for it in apparmor)

The keys for my setup were the proper PCIe location in the VM and passing a GOP updated vBIOS, otherwise I'll be met with Error 43 or no display. To get the M2000M working properly: (I assume you've read Archlinux's wiki about OVMF GPU Passthrough and how to add stuff like qemu commands in a virsh xml) - Dump the vBIOS (there are many ways, through the windows registry (manually or with a script) or by using VBiosExtractor following this guide. Note that you will need to put the full path to the BIOS updater executable, the VBiosExtractor script doesn't seem to do well with relative paths. - Add UEFI GOP to it (as it's PC-AT only) (I used Maxwell MXM) - Patch the OVMF image with the regular (or GOP updated) vBIOS by building it automatically (thanks to @Ashymad's original gist) if you're on an Arch-based OS (you'll have to download the whole gist) or check the PKGBUILD and make the patching manually (this guide has the same steps too, don't forget the dos2unix commands, otherwise the patches wont be applied). You might want to add the path of the OVMF in /etc/libvirtd/qemu.conf accordingly or edit the PKGBUILD to copy them to /usr/share/edk2-ovmf/x64/ instead (where the edk2-ovmf official package store them). - Make a VM with virt-manager as you would normally, add the dGPU and the other hardware that comes with it (for me only Audio, for others it may be also USB or others peripherals that come with a GPU nowadays), but in the XML section (make sure you enable editing), remove all PCI addressing lines (to avoid conflicting addresses) and add the proper address for the dGPU to the hostdev section (note that it usually just means that the source address and the hostdev address match). When applying, virsh will properly populate the other addresses. xml

Yeah i read that too, I also found this.. https://github.com/intel/gvt-linux/issues/64

Intel has already confirmed that GVT-g is essentially dead and not supported on their Iris/Xe or anything newer graphics.. We can also confirm this via their own drivers source..

https://github.com/intel/gvt-linux/blob/gvt-staging/drivers/...

On Intel iGPUs, there are two methods: GVT-g and GVT-d. GVT-g is basically creating virtual instances of the iGPU for use in VMs, while GVT-d is passing through an entire iGPU to the guest in the same way you would do with a normal GPU.

From the Intel GVTg_Setup_Guide;

Not really pass through, no. If CONFIG_DRM_I915_GVT is enabled in your kernel, you can use Intel's graphics virtualization system... basically a virtio style virtual device that shares the GPU between VM and host. IMO this is way more convenient than real passthrough, where the device is only available either to the VM or the host. The downside is that you don't get full performance in the VM.

"Intel GVT-g is a full GPU virtualization solution with mediated pass-through (VFIO mediated device framework based), starting from 5th generation Intel Core(TM) processors with Intel Graphics processors. GVT-g supports both Xen and KVM (a.k.a XenGT & a.k.a KVMGT). A virtual GPU instance is maintained for each VM, with part of performance critical resources directly assigned. The capability of running native graphics driver inside a VM, without hypervisor intervention in performance critical paths, achieves a good balance among performance, feature, and sharing capability."

https://github.com/intel/gvt-linux/wiki/GVTg_Setup_Guide

Wait a minute. GVT-g with 11th gen iGPUs upwards does work in linux guest? Are you sure about that? See this github issue for reference.

To be clear, I never said it was dead, only a dead end.

As for GVT-g and Xe, according to a post in this[0] issue by one of the Intel devs, Rocket Lake (Xe) is not getting support and only does GVT-d.

Also in the same issue, someone pointed out that Intel themselves have states as much here[1].

I hope I am proven wrong in the end and GVT-g comes to then entire Xe and ARC lineup. Intel's communication on this matter has been...lacking.

0: https://github.com/intel/gvt-linux/issues/190

1: https://www.intel.com/content/www/us/en/support/articles/000...

After days of trial and error, I could not get it to work, maybe one of you knows it. Currently, I try to setup GVT-d with KVM on my Dell XPS 13 2 in 1 7390, which has a i7-1065G7. AFAIK GVT-g is not supported, so I gave GVT-d a chance. The virtual machine is booting without any errors, but the display stays black. I only found this guide, but couldn't get it to work...