Re: [PATCH v3 02/11] KVM: guest_memfd: Handle final folio_put() of guest_memfd pages

[David Hildenbrand <david@redhat.com>]

>>> +#endif
>>> +
>>>   #endif
>>> diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
>>> index 6dc2494bd002..734afda268ab 100644
>>> --- a/include/linux/page-flags.h
>>> +++ b/include/linux/page-flags.h
>>> @@ -933,6 +933,17 @@ enum pagetype {
>>>        PGTY_slab       = 0xf5,
>>>        PGTY_zsmalloc   = 0xf6,
>>>        PGTY_unaccepted = 0xf7,
>>> +     /*
>>> +      * guestmem folios are used to back VM memory as managed by guest_memfd.
>>> +      * Once the last reference is put, instead of freeing these folios back
>>> +      * to the page allocator, they are returned to guest_memfd.
>>> +      *
>>> +      * For now, guestmem will only be set on these folios as long as they
>>> +      * cannot be mapped to user space ("private state"), with the plan of
>>
>> Might be a bit misleading as I don't think it's set yet as of this series.
>> But I guess we can keep it to avoid another update later.
> 
> You're right, it's not in this series. But as you said, the idea is to
> have the least amount of churn in the core mm code.

I proposed adding the latter, to make it clearer that the purpose of 
these things is to be mapped to user space long-term (-> be a proper 
folio, like hugetlb folios will have to be), compared to long-term not 
being a folio (slab, offline, unaccepted, ...).

Might become relevant for people working on the memdesc work, to know 
what to do here.

(or at least have less questions :) )

-- 
Cheers,

David / dhildenb