Re: [PATCH v6 08/25] KVM: arm64: iommu: Shadow host stage-2 page table

[Jason Gunthorpe <jgg@ziepe.ca>]

On Tue, May 12, 2026 at 10:42:30AM +0000, Mostafa Saleh wrote:
> > Also io-pgtable arm cannot trigger BBM behaviors, so how do you
> > implement it?
> 
> At the moment, we workaround this by mapping all the memory with PTE
> level, while MMIO remains at block level as they never change ownership
> at the moment.
> 
> This is one of the missing features I plan to add after this series,
> if you look in the cover letter, these are listed under “Future work”

Hrm, I would not be happy to see io-pgtable modified to have bbm
features :\ They are very complex this is what iommupt is for..

> > So that's the real explanation, KVM cannot manage the S2 in the right
> > way so you can't share it. RMM/etc are managing the S2 without
> > pointless page faults so they can share it.
> 
> Well, there is not really a right way, even with a fully populated
> stage-2 page table, you can’t guarantee not getting TLB conflict aborts
> without FEAT_BBML3 (which is quite recent), unless you map everything
> with a leaf level, which then impacts performance.

I mean, there is a right way - you need to rely on the BBM features in
all HW and fully populate the S2 for no faults. So if that isn't
present then you are doing things that will harm performance to work
around the issue. It's fine, but just explain in the commit message.

Jason