Re: [PATCH v6 3/3] mm/gup: disallow FOLL_LONGTERM GUP-fast writing to file-backed mappings

[David Hildenbrand <david@redhat.com>]

On 02.05.23 18:19, Jason Gunthorpe wrote:
> On Tue, May 02, 2023 at 06:12:39PM +0200, David Hildenbrand wrote:
> 
>>> It missses the general architectural point why we have all these
>>> shootdown mechanims in other places - plares are not supposed to make
>>> these kinds of assumptions. When the userspace unplugs the memory from
>>> KVM or unmaps it from VFIO it is not still being accessed by the
>>> kernel.
>>
>> Yes. Like having memory in a vfio iommu v1 and doing the same (mremap,
>> munmap, MADV_DONTNEED, ...). Which is why we disable MADV_DONTNEED (e.g.,
>> virtio-balloon) in QEMU with vfio.
> 
> That is different, VFIO has it's own contract how it consumes the
> memory from the MM and VFIO breaks all this stuff.
> 
> But when you tell VFIO to unmap the memory it doesn't keep accessing
> it in the background like this does.

To me, this is similar to when QEMU (user space) triggers 
KVM_S390_ZPCIOP_DEREG_AEN, to tell KVM to disable AIF and stop using the 
page (1) When triggered by the guest explicitly (2) when resetting the 
VM (3) when resetting the virtual PCI device / configuration.

Interrupt gets unregistered from HW (which stops using the page), the 
pages get unpinned. Pages get no longer used.

I guess I am still missing (a) how this is fundamentally different (b) 
how it could be done differently.

I'd really be happy to learn how a better approach would look like that 
does not use longterm pinnings.

I don't see an easy way to not use longterm pinnings. When using mmu 
notifiers and getting notified about unmapping of a page (for whatever 
reason ... migration, swapout, unmap), you'd have to disable aif. But 
when to reenable it (maybe there would be a way)? Also, I'm not sure if 
this could even be visible by the guest, if it's suddenly no longer enabled.

Something for the s390x people to explore ... if HW would be providing a 
way to deal with that somehow.

-- 
Thanks,

David / dhildenb