+++ This bug was initially created as a clone of Bug #1762087 +++

Description of problem:

Libvirt and qemu have supported vIOMMU emulation for some time now.
nova curretly does not expose the vIOMMU as a configurable paramter.
using the q35 machine type and vIOMMU feature can enable better securty in
the guest when using dpdk or pci device passsthough by allowing the viommu
to work with the host iommu to restrict the the meory the passthough or dpdk
device can acess in the guest.

in some case this can also improve performance by reducing the cost of phyical page traslation.

by enableing support for vIOMMU emultaiton in the guest this will also enable
futrue work to allow nested sriov passhtough or other more advanced configurations that require a IOMMU to funciton correcly such as using vfio-pci in the guest.

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1.
2.
3.

Actual results:

Expected results:

Additional info:

— Additional comment from RHEL Program Management on 2019-10-16 00:59:14 UTC —

The keyword FutureFeature has been added. If this bug is not a FutureFeature, please remove from the Summary field any strings containing "RFE, rfe, FutureFeature, FEAT, Feat, feat". Additionally, if this feature is being backported to a previous release, clone this BZ to older releases and add the "FeatureBackport" keyword only to those cloned BZs.