Re: [PATCH v3 1/1] iommu/sva: Invalidate KVA range on kernel TLB flush

[Dave Hansen]

On 8/26/25 07:33, Matthew Wilcox wrote:
...
> One memdesc type already assigned is for page tables.  Maybe iommu page
> tables are the same

One bit of context: This is an IOMMU problem but the IOMMU is walking
the normal x86 page tables in this case, specifically kernel page
tables. There are separate IOMMU page tables, of course, but those
aren't at issue here.