Re: [PATCH v9 00/13] PCI: liveupdate: PCI core support for Live Update

[Samiullah Khawaja <skhawaja@google.com>]

On Thu, Sep 24, 2026 at 09:47:03PM -0700, Zhu Yanjun wrote:
>
>在 2026/9/22 11:53, David Matlack 写道:
>>On Tue, Sep 22, 2026 at 11:36 AM Zhu Yanjun <yanjun.zhu@linux.dev> wrote:
>>>在 2026/9/18 13:06, David Matlack 写道:
>>>>Future Work
>>>>-----------
>>>>
>>>>Following this series, we expect to make further improvements to the PCI
>>>>core support for Live Update:
>>>>
>>>>    - Allow P2P across Live Update by avoiding resizing or moving
>>>>      preserved device BARs and preserving all upstream bridge windows.
>>>>
>>>>    - Support preserving Virtual Functions by preserving SR-IOV
>>>>      configuration on PFs and enumerating VFs after Live Update.
>>>Preserving the PCIe topology, bus numbers, ACS, and Bus Mastering across
>>>kexec is a foundational step for minimizing downtime.
>>>
>>>As we look toward complete end-to-end support for DMA preservation
>>>across Live Update—especially for VFIO device passthrough and dma-buf
>>>sharing scenarios—IOMMU table/domain preservation becomes crucial to
>>>prevent IOMMU page faults when devices continue performing DMA during kexec.
>>>
>>>I would like to ask about the current status and roadmap regarding IOMMU
>>>Live Update / KHO (Kexec Handover) support:
>>>
>>>Is there an ongoing effort or RFC series for IOMMU handover / page-table
>>>preservation currently in development or under discussion?
>>>
>>>How is the coordination between the PCI core Live Update mechanisms and
>>>the IOMMU subsystem being envisioned for preserving IOVA mappings (e.g.,
>>>restoring domains or handing over root tables)?
>>>
>>>Any pointers to active discussion threads, RFCs, or future plans
>>>regarding IOMMU participation in Live Update would be greatly appreciated.
>>The first IOMMU series to support Live update, can be found here:
>>
>>   https://lore.kernel.org/linux-iommu/20260921004834.2601285-1-skhawaja@google.com/
>
>After cloning the source code from samikhawaja/linux, branch 
>iommu/phase1-v5, I tested the VFIO live update functionality in my 
>local KVM environment and encountered the following issue.
>
>When I run the following commands to verify that VFIO live update 
>works correctly, I get the output shown below.
>
>I am currently testing with VFIO v5 and PCI core v9, and I am not sure 
>whether this behavior is expected.
>
>Interestingly, when I revert the following commit:
>
>vfio/pci: Preserve the iommufd state of the vfio cdev
>
>the issue disappears.
>
>Could you please let me know whether this is expected behavior, or 
>whether it indicates a problem with the current VFIO live update 
>implementation?
>
>"
>
>echo "vfio-pci" > /sys/bus/pci/devices/0000:00:04.0/driver_override
>
>echo "0000:00:04.0" > /sys/bus/pci/drivers/vfio-pci/bind
>Preserving device in session
>vfio_pci_liveupdate_kexec_test.c:122: Assertion Failure
>
>  Expression: ret == 0
>  Observed: 0xffffffffffffffea == 0
>  [errno: 22 - Invalid argument]
>"

Hi Yanjun,

Thanks for testing this.

This is expected. With "vfio/pci: Preserve the iommufd state of the vfio
cdev", preserving a vfio cdev depends on the preservation of its
associated iommufd. Since the vfio_pci_liveupdate_kexec_test does not
preserve the associated iommufd, it fails with -EINVAL.

Note that the IOMMU persistence series does not update the
vfio_pci_liveupdate_kexec_test, with those extra bits, as it would
regress after_kexec() part of the test because the retrieval of iommufd
is not supported in phase1.

Please use tools/testing/selftests/iommu/iommufd_liveupdate_kexec_test
added in the last patch of the series, it sets up the iommufd and HWPT
preservation before preserving the vfio cdev.

The wip phase2 can be found here if you are interested.
https://github.com/samikhawaja/linux/tree/iommu/phase2-v1-wip

Thanks,
Sami

>
>Thanks a lot.
>
>Yanjun Zhu
>
>>
>>The IOMMU series builds on top of the PCI core support (this series)
>>and the VFIO series linked further up in the cover letter.
>
>-- 
>Best Regards,
>Yanjun.Zhu
>