Differences

This shows you the differences between two versions of the page.

--- public:gpu-passthrough [2015/11/11 07:49] – squires
+++ public:gpu-passthrough [2015/11/13 20:29] (current) – [References] admin
@@ Line 1: / Line 1: @@
 ====== GPU Passthrough ======
-GPU passthrough refers to configuration where a graphics card is reserved for use by a virtual machine. Making use of VT-d, IOMMU, and related technologies, the graphics card can be used at native performance.
+GPU passthrough refers to configurations where a graphics card is reserved for use by a virtual machine. Making use of IOMMU, the graphics card can be used at native performance.
-This page describes my specific
+This page first describes the process in general and then my specific system configurations.
+====== Information ======
+===== Hardware Requirements =====
+The focus will be on Intel + NVIDIA systems, because I am unfamiliar with how GPU passthrough works with AMD processors and GPUs.
+==== VT-d ====
+The CPU, motherboard, and BIOS **must** support VT-d. Check [[http://ark.intel.com/|ARK]] for CPU support. Note that K series (unlocked multiplier) CPUs historically did not have VT-d support, but Haswell Refresh and Skylake do.
+==== IOMMU Groups and ACS ====
+The guest graphics card needs proper IOMMU isolation from other devices on the PCIe bus. It should be alone in its own IOMMU group. This ensures no DMA transfers will occur between devices without being translated by the IOMMU.
+If you have a Xeon E5 processor, it will probably support ACS. Your PCIe devices will be isolated in their own IOMMU groups. Some socket 2011 CPUs also support ACS (according to the datasheets).
+If your processor doesn't support ACS, but your chipset is affected by the root port isolation quirk, then your graphics cards might still be isolated. You can use <q>find /sys/kernel/iommu_groups/ -type l</q> or the iommu_groups.sh script to view what devices are in each group.
+++++ iommu_groups.sh |
+<code>
+#!/bin/sh
+# List the devices in each IOMMU group, from AW at
+# https://bbs.archlinux.org/viewtopic.php?id=162768&p=29
+BASE="/sys/kernel/iommu_groups"
+for i in $(find $BASE -maxdepth 1 -mindepth 1 -type d); do
+        GROUP=$(basename $i)
+        echo "### Group $GROUP ###"
+        for j in $(find $i/devices -type l); do
+                DEV=$(basename $j)
+                echo -n "    "
+                lspci -s $DEV
+        done
+done
+</code>
+++++
+If your guest graphics card is not isolated, you can try moving it to a different slot on your motherboard to see if that slot has better isolation. Some slots are connected to the processor PCI lanes, which only have isolation if your processor supports ACS. Other slots are connected to the PCH PCI lanes, which may have isolation on root ports.
+If you don't have isolation on any PCI slot and you don't need the devices that share the guest graphics card's IOMMU group, you can bind the extra device to either of the vfio-pci or pci-stub drivers.
+As a last resort, you can try using the ACS override kernel patch. However, this does not create isolation. It merely hides the fact that isolation doesn't exist. Devices in the same real IOMMU group could still perform DMA transfers within another's address space.
+==== Graphics Cards ====
+You must have at least one GPU for the host and one GPU for the guest.
+You may use an IGP (CPU integrated graphics) for the host, but you cannot pass it through to the guest. If you use an IGP for the host and a guest graphics card that doesn't have a UEFI firmware, then you must apply the vga arbitration kernel patch. If you use a guest graphics card that has a UEFI firmware, then you can use OVMF virtual machine firmware and not apply the vga arbitration patch.
+^ Host Device ^ Guest Device ^ Guest Device UEFI-enabled ^ Result                                ^
+| GPU         | GPU          | Yes or No                 | OK                                   |
+| GPU         | IGP          | Yes or No                 | Invalid                              |
+| IGP         | GPU          | No                        | Requires VGA arbitration kernel patch; DRI will be disabled |
+| IGP         | GPU          | Yes                       | Use OVMF firmware; can still use DRI |
+Virtualization is fully supported by NVIDIA Quadro cards. GeForce cards will work, but require that the virtualization be hidden from NVIDIA drivers.
+GeForce 600 series cards probably did not ship with a UEFI firmware, but one can be flashed onto them. Check the manufacturer's site for an update, or use [[http://www.win-raid.com/t892f16-AMD-and-Nvidia-GOP-update-No-requests-DIY.html|GOPupd]].
+====== Xeon E3-1275 ======
 ===== Hardware =====
@@ Line 98: / Line 160: @@
 ===== Kernel =====
-Because the C206 chipset is affected by the Intel root port isolation quirk ([[https://lkml.org/lkml/2014/2/3/487|1]], [[http://lxr.free-electrons.com/ident?i=pci_quirk_intel_pch_acs_ids|2]]), both graphics cards are isolated in their own IOMMU group.
+Because the C206 chipset is affected by the Intel root port isolation quirk ([[https://lkml.org/lkml/2014/2/3/487|1]], [[http://lxr.free-electrons.com/ident?i=pci_quirk_intel_pch_acs_ids|2]]), both graphics cards are isolated in their own IOMMU group. The pcie_acs_override patch is not needed. (To check if the root port isolation quirk exists, grep dmesg for "Intel PCH root port ACS workaround enabled".)
-===== References =====
+If we wanted to use the integrated graphics of the E3-1275, then we would need to either:
+  * Use the i914.enable_hd_vgaarb patch (and lose DRI).
+  * Use a guest GPU that supports UEFI and use the OVMF boot firmware to start the virtual machine. (OVMF can initialize the GPU using legacy-free UEFI, keeping DRI and bypassing VGA and the need for arbitration.)
+====== Core i5-4690K ======
+===== Hardware =====
+===== Kernel =====
+====== References ======
+[[http://vfio.blogspot.com/|VFIO tips and tricks]]
+[[https://wiki.archlinux.org/index.php/PCI_passthrough_via_OVMF|ArchWiki: PCI passthrough via OVMF]]
+[[https://www.reddit.com/r/linux_gaming/comments/3lnpg6/gpu_passthrough_revisited_an_updated_guide_on_how/|GPU Passthrough Revisited - An Updated Guide On How To Game In A Virtual Machine]]
+[[http://www.se7ensins.com/forums/threads/how-to-setup-a-gaming-virtual-machine-with-gpu-passthrough-qemu-kvm-libvirt-and-vfio.1371980/|How to setup a gaming virtual machine with GPU passthrough (QEMU, KVM, libvirt, and VFIO)]]
+[[http://ubuntuforums.org/showthread.php?t=2266916|Windows Gaming VM - KVM / UEFI Version - HowTo]]
 [[https://wiki.debian.org/VGAPassthrough|Debian Wiki: VGAPassthrough]]