[PATCH v5 0/7] mm/mprotect: Fix dax puds

[David Hildenbrand]

On 13.11.24 18:56, Peter Xu wrote:
> On Wed, Nov 13, 2024 at 05:42:15PM +0100, David Hildenbrand wrote:
>> On 13.11.24 17:39, Peter Xu wrote:
>>> On Mon, Nov 11, 2024 at 10:20:59PM +0100, Jann Horn wrote:
>>>> On Mon, Aug 12, 2024 at 8:12 PM Peter Xu <peterx at redhat.com> wrote:
>>>>> Dax supports pud pages for a while, but mprotect on puds was missing since
>>>>> the start.  This series tries to fix that by providing pud handling in
>>>>> mprotect().  The goal is to add more types of pud mappings like hugetlb or
>>>>> pfnmaps.  This series paves way for it by fixing known pud entries.
>>>>
>>>> Do people actually use hardware where they can use PUD THP mappings
>>>> for DAX? I thought that was just some esoteric feature that isn't
>>>> actually usable on almost any system.
>>>> Was I wrong about that?
>>>
>>> I did run it with a qemu emulated nvdimm device.  Though in reality I've no
>>> idea on how many people are using it..
>>
>> I wonder if we still have to support it ... or could disable it+rip it out.
> 
> Note that in my previous email, I also mentioned mremap() for PMD on dax
> too.  If that's a real problem, it won't be fixed even if dropping dax PUD
> support.
>

Very true.


> And we definitely want to understand whether there're still users on pud
> dax to consider dropping anything.. it could still be that both mprotect()
> and mremap() are not yet used in the current use cases.

Right, but at least NVDIMMs are getting less important. Just a thought 
if this is really still worth having.

-- 
Cheers,

David / dhildenb