Re: [PATCH v2 08/20] mm/mshare: flush all TLBs when updating PTEs in an mshare range

[Anthony Yznaga <anthony.yznaga@oracle.com>]

On 5/30/25 7:41 AM, Jann Horn wrote:
> On Fri, Apr 4, 2025 at 4:18 AM Anthony Yznaga <anthony.yznaga@oracle.com> wrote:
>> Unlike the mm of a task, an mshare host mm is not updated on context
>> switch. In particular this means that mm_cpumask is never updated
>> which results in TLB flushes for updates to mshare PTEs only being
>> done on the local CPU. To ensure entries are flushed for non-local
>> TLBs, set up an mmu notifier on the mshare mm and use the
>> .arch_invalidate_secondary_tlbs callback to flush all TLBs.
>> arch_invalidate_secondary_tlbs guarantees that TLB entries will be
>> flushed before pages are freed when unmapping pages in an mshare region.
> 
> Thanks for working on this, I think this is a really nice feature.
> 
> An issue that I think this series doesn't address is:
> There could be mmu_notifiers (for things like KVM or SVA IOMMU) that
> want to be notified on changes to an mshare VMA; if those are not
> invoked, we could get UAF of page contents. So either we propagate MMU
> notifier invocations in the host mm into the mshare regions that use
> it, or we'd have to somehow prevent a process from using MMU notifiers
> and mshare at the same time.

Thanks, Jann. I've noted this as an issue. Ultimately I think the 
notifiers calls will need to be propagated. It's going to be tricky, but 
I have some ideas.