[RFC PATCH v2 07/11] iommufd: Use large IOMMU pages for provider mappings

[Jason Gunthorpe]

On Wed, Oct 14, 2026 at 09:37:29AM +0000, Fred Griffoul wrote:
> From: Fred Griffoul <fgriffo at amazon.co.uk>
> 
> Provider pages are mapped with one PAGE_SIZE entry each. A device that
> uses a large provider block then misses in the IOTLB on every page, and
> the IOMMU needs more page table memory than the block requires. A
> revoke also flushes the IOTLB once per run of pages it unmaps.
> 
> Map each block that the provider reports with large IOMMU pages where
> the IOVA and the frames are aligned alike, and with smaller ones where
> they are not. Only the pages of the block are mapped, as before.
> 
> A large IOMMU page can only be unmapped whole, so a revoke that cuts
> one widens its range to the whole page. It maps back what the provider
> backs now, and a device may also fault on the rest of that page while
> it is remapped. The widening finds the page with iova_to_phys():
> pages that only look like one large page make the revoke remap more
> than it must, which is harmless.

You are going to need to go into generic_pt and implement cut and a
more suitable unmap to make this stuff work properly. We can't over
unmap on conversions.

> iommu_unmap() may not split the range of one map call either, and
> virtio-iommu refuses to. Map each IOMMU page with its own call: the
> widened range ends on IOMMU page boundaries, so it also ends on map
> call boundaries.

Also please don't write commit messages with AI. WTF is this even
trying to say? The subject is "use large pages" and this text says
"map each page with its own call"

Jason