Running Linux without a Graphics Card

SgtAwesomesauce · February 10, 2017, 8:09pm

Sure thing. What follows will assume you have a preliminary understanding of how IOMMU and IOVA works. An overview (albeit, brief) can be found on part 3 of my Pragmatic Neckbeard discussion, linked below:

The Pragmatic Neckbeard 3: VFIO, IOMMU and PCIe Linux

Intro In this installation we are going to be discussing the technology behind PCI Passthrough to VMs. The concept of passthrough is relatively simple. You take a physical device and forward it's memory registers to the VM. A simple idea, however, doesn't make an implementation simple. There's a lot that goes into passthrough, and a bunch of extremely talented people have put a lot of time into software to bring passthrough to the point where it's a lot easier than it was before, but still not quite plug 'n' play. Conceptual Discussion Now, what goes into it exactly? To give a brief overview, we've got the hardware support for passthrough, the IOMMU or Input-Output Memory Management Unit, supported by both the motherboard and CPU. (more info on that here) The Linux driver, VFIO, is assigned to the device at boot, preventing the device from being initialized. This will help us when it comes to passing our GPU into the vm. If we've got the GPU bound to another driver, we won't achieve successful passthrough, because you won't be able to exclusively lock the GPU's resources to the QEMU vm. Now, let's talk about the PCIe bus. The machine I'm going to be using as a reference has an ASUS Z170-a and a 6700k. This gives me 16 PCIe lanes on the CPU to play with. Most GPU's will be happy with 8 lanes, so we shouldn't have bandwith issues here. I am going to be passing two devices to my VM: GPU and USB-3 controller. The passthrough GPU will be using an 8x connection on the PCH and the USB controller will be using a 4x connection on the CPU. This will allow the GPU I'm using for Linux output to be connected by an 8x connection as well. Handling PCIe passthrough isn't 100% straightforward. There are limitations and rules about how the IOMMU sees a device and its IO Virtual Addresses (IOVA). Some devices will alias to the same IOVA space which makes the IOMMU unable to destinguish between the two devices. This becomes problematic when dealing with tr…

So, when you pass through a GPU to a VM, you're attaching all the IOVA space (it's input/output registers) to the VM. This gives said VM exclusive access to it. The VM isn't aware of the X11 server that's trying to access it and the X11 server (for example) isn't aware of the VM that's got access to it. They're both theoretically able to access it, but when the VM tries to execute a PCI call after the X server does, you're going to have the device in an inconsistent state. Imagine trying to have two people trying to assemble the same ikea desk with a blindfold while being entirely unaware of the other person that's working on the desk. You go to pick up part D and find that screw H is no longer where you'd left it. It's the same idea if you were to share them. This would require a rewrite of the fundamental principals of the GPU drivers on both Windows and Linux for GPU sharing to be capable. At least, that's my understanding of it.

I hope this wasn't too out there. If you've read through the conceptual discussion on the link and still have questions, feel free to call me out for being a terrible writer.