Re: [RFC/PATCH v3 04/16] s390/mm: add gmap PMD invalidation notification

[Janosch Frank <frankja@linux.vnet.ibm.com>]

[-- Attachment #1.1: Type: text/plain, Size: 1188 bytes --]

On 14.02.2018 16:15, David Hildenbrand wrote:
>>> So what can happen is (please correct me if I'm wrong)
>>>
>>> a) PMD is split. SSKE writes storage key with _PAGE_CHANGED, ends up in
>>> PGSTE. The real storage key doesn't match the requested storage key.
>>> b) Split PMD is replaced, triggers a removal of the split PMD ->
>>> gmap_pmd_split_free(pmdp). The requested storage key is partially lost
>>> (pgste removed).
>>> c) PMD is mapped in again. If the guest reads the storage key now, the
>>> value is wrong.
>>
>> Yes, we loose GR and GC.
>> Is there a case when the VM is running, where this would happen?
> 
> It should already happen when migrating storage keys. The fake PGSTE are
> not considered in get_guest_storage_key().

Yes, that should break it reliably :)

> 
> For the other parts, the original user space PMD would have to be
> changed. A simply mprotect() should achieve that. Or dirty tracking. But
> not sure how that applies to huge pages at all.

I've only seen that 'till now when the VM is already dead and the kernel
cleans up. Anyhow, you already proposed a solution, so there's no sense
in defending against corner cases.



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]