Re: [RFC] virtio-iommu: Take into account possible aliasing in virtio_iommu_mr()

[Robin Murphy]

On 2023-01-20 15:28, Jean-Philippe Brucker wrote:

For some reason this came through as blank mail with a text attachment, 
so apologies for the lack of quoting, but for reference this is a 
long-standing known issue:

https://lore.kernel.org/linux-iommu/9625faf4-48ef-2dd3-d82f-931d9cf26976@huawei.com/

In summary, yes, hanging {of,acpi}_iommu_configure() off driver probe is 
massively wrong, and pulling that step into iommu_probe_device() in a 
sane and robust manner is one of the next things to start on after 
getting the bus ops out of the way.

Thanks,
Robin.