On Fri, Jan 30, 2026 at 10:59:11AM -0800, Nathan Chen via Devel wrote:
> Hi,
>
> This is a follow up to the fifth patch series [0] for using iommufd
> to propagate DMA mappings to the kernel for VM-assigned host
> devices in a qemu VM.
>
> We add a new 'iommufd' attribute for hostdev devices to be
> associated with the iommufd object.
>
> For instance, specifying the iommufd object and associated hostdev in a
> VM definition:
>
> <devices>
> ...
> <hostdev mode='subsystem' type='pci' managed='no'>
> <driver iommufd='yes'/>
> <source>
> <address domain='0x0009' bus='0x01' slot='0x00' function='0x0'/>
> </source>
> <address type='pci' domain='0x0000' bus='0x15' slot='0x00' function='0x0'/>
> </hostdev>
> <hostdev mode='subsystem' type='pci' managed='no'>
> <driver iommufd='yes'/>
> <source>
> <address domain='0x0019' bus='0x01' slot='0x00' function='0x0'/>
> </source>
> <address type='pci' domain='0x0000' bus='0x16' slot='0x00' function='0x0'/>
> </hostdev>
> ...
> </devices>
>
> This would get translated to a qemu command line with the arguments below.
> Note that libvirt will open the /dev/iommu and VFIO cdev, passing the
> associated fd number to qemu:
>
> -object '{"qom-type":"iommufd","id":"iommufd0","fd":"24"}' \
> -device '{"driver":"vfio-pci","host":"0009:01:00.0","id":"hostdev0","iommufd":"iommufd0","fd":"22","bus":"pci.21","addr":"0x0"}' \
> -device '{"driver":"vfio-pci","host":"0019:01:00.0","id":"hostdev1","iommufd":"iommufd0","fd":"25","bus":"pci.22","addr":"0x0"}' \
>
> Changes from v5:
> - Updated documentation to point to 12.1.0 (QEMU and KVM only)
> - Move define and VIR_LOG_INIT out of #ifdef __linux__ block
> in viriommufd.h
> - Remove unneeded virPCIDeviceFree() from AppArmorSetSecurityHostdevLabel
> - Always error out if per-process accounting failed
>
> Changes from v4:
> - Move qemuProcessOpenVfioFds() to qemu_process.h in patch 4/7
> - Add `linux/iommufd.h` into headers list in the meson.build file
> - Correct viriommufd.c pre-processor definition indentations
> - Restore call to virIOMMUFDSetRLimitMode() in patch 5/7
> - Fix indentations for various print function calls
> - Move the qemuProcessOpenVfioFds() call to qemuProcessPrepareHost()
> - Make qemuProcessOpenVfioFds() static
> - Remove unnecessary formatting and comment in virPCIDeviceGetVfioPath()
> - Clean up namespace and cgroup changes with conditional check for
> iommufd
> - Remove virIOMMUFDSupported()
> - Fix seclabel return logic
> - In Apparmor seclabel logic, use g_autoptr for virPCIDevice pointer
>
> Changes from v3:
> - Resolved issue from v2 where stale FD from previous VM boot was
> in use
> - Remove second approach for retrieving VFIO device path in
> virPCIDeviceGetVfioPath()
> - Resolve broken build of libvirt on non-Linux platforms
> - Conditionally define iommufd headers and use system headers where
> possible
> - Add non-fatal handling + warning print for EPERM for the
> IOMMU_OPTION_RLIMIT_MODE ioctl
> - Replace references to /dev/iommu with VIR_IOMMU_DEV_PATH
> - Implement virIOMMUFDSupported(void) to check for existence of
> /dev/iommu on host
> - Include tests for multiple hostdevs
>
> Changes from v2:
> - Set per-process memory accounting mode for iommufd
> - Separated out formatting of iommufd object from qemuBuildHostdevCommandLine
> - Placed hostdev private data implementation in a separate commit
> - Allocate hostdev private data unconditionally
> - Compare FDs against -1
> - Integrated callback function in virQEMUDriverPrivateDataCallbacks for qemuDomainHostdevPrivateNew
> - Dropped qemuProcessCloseVfioFds
> - Addressed other feedback from v2 (formatting, includes, etc.)
> - Revised seclabel logic to be device-specific for AppArmor and to allow paths for SELinux/DAC
>
> This series is on Github:
> https://github.com/NathanChenNVIDIA/libvirt/commits/iommufd-v6-01-26
>
> Thanks,
> Nathan
>
> [0] https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/thread/GKSDPK5VSPUKE5C73GI5IPDPUHO37D55/
>
> Nathan Chen (7):
> qemu: Implement support for associating iommufd to hostdev
> qemu: Introduce privateData for hostdevs
> qemu: Support per-process memory accounting for iommufd
> qemu: open VFIO FDs from libvirt backend
> qemu: open iommufd FD from libvirt backend
> qemu: Update Cgroup, namespace, and seclabel for iommufd
> tests: qemuxmlconfdata: provide iommufd sample XML and CLI args
Once we decide what error reporting to use in PATCH 3.
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>