At the moment the virtio-iommu is not migratable. Besides the virtio generic stuff, its state is composed of - domains organized in a GTree (red black tree): each domain owns a list of end points using it and a gtree of iommu mappings. - a gtree of endpoints and each enpoint owns a gtree of reserved regions. There is currently no VMSTATE_* object that allows migrating Gtree elements. The closest thing seems to be VMSTATE_QTAILQ_V that allows to migrate a QTAILQ.
"[PATCH v2] migration: Support gtree migration" was sent to the ML. This is a pre-requisite for virtio-iommu migration as this latter has many gtrees
QEMU has been recently split into sub-components and as a one-time operation to avoid breakage of tools, we are setting the QEMU sub-component of this BZ to "General". Please review and change the sub-component if necessary the next time you review this BZ. Thanks
The code has landed upstream (will be in qemu 5.0) . Migration was tested on ARM.
Until non DT support is not supported we can claim the feature is available (despite the code is upstream). So let's move this to 8.4
Code is upstream and we now just wait for the ACPI VIOT integration. Let's close the BZ as a dup of the main virtio-iommu BZ. *** This bug has been marked as a duplicate of bug 1477099 ***