From 2173def64373a50a13390ee342fb95abc901f203 Mon Sep 17 00:00:00 2001 From: Markus Frank Date: Fri, 26 Jul 2024 10:21:12 +0200 Subject: [PATCH] passthrough: viommu: replace host requirement with reference to pcie passthrough Enabling IOMMU on the host is not a requirement for vIOMMU. It is only a requirement for passthrough. Add a sentence to clarify the need for a configured PCI(e) passthrough on the host for passthrough to nested VMs. While at it also clarify that this does not only applies to PCI but also PCIe here. Suggested-by: Dominik Csapak Signed-off-by: Markus Frank [ TL: mention all changes and avoid overly long line with link ] Signed-off-by: Thomas Lamprecht --- qm-pci-passthrough.adoc | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/qm-pci-passthrough.adoc b/qm-pci-passthrough.adoc index bbd6b85..4cbaadd 100644 --- a/qm-pci-passthrough.adoc +++ b/qm-pci-passthrough.adoc @@ -505,15 +505,14 @@ vIOMMU (emulated IOMMU) vIOMMU is the emulation of a hardware IOMMU within a virtual machine, providing improved memory access control and security for virtualized I/O devices. Using -the vIOMMU option also allows you to pass through PCI devices to level-2 VMs in -level-1 VMs via https://pve.proxmox.com/wiki/Nested_Virtualization[Nested Virtualization]. +the vIOMMU option also allows you to pass through PCI(e) devices to level-2 VMs +in level-1 VMs via +https://pve.proxmox.com/wiki/Nested_Virtualization[Nested Virtualization]. +To pass through physical PCI(e) devices from the host to nested VMs, follow the +PCI(e) passthrough instructions. + There are currently two vIOMMU implementations available: Intel and VirtIO. -Host requirement: - -* Add `intel_iommu=on` or `amd_iommu=on` depending on your CPU to your kernel -command line. - Intel vIOMMU ^^^^^^^^^^^^