On Tue, 2025-02-25 at 16:52 +0000, David Hildenbrand wrote:> On 21.02.25 17:07, Patrick Roy wrote:
>> Make secretmem set AS_NO_DIRECT_MAP on its struct address_space, to drop
>> all the vma_is_secretmem()/secretmem_mapping() checks that are based on
>> checking explicitly for the secretmem ops structures.
>>
>> This drops a optimization in gup_fast_folio_allowed() where
>> secretmem_mapping() was only called if CONFIG_SECRETMEM=y. secretmem is
>> enabled by default since commit b758fe6df50d ("mm/secretmem: make it on
>> by default"), so the secretmem check did not actually end up elided in
>> most cases anymore anyway.
>>
>> Signed-off-by: Patrick Roy <roypat@xxxxxxxxxxxx>
>> ---
>
> Ah, there it is. Can both patches somehow be squashed?
Yeah, I'm happy to squash them. I separated them out based on feedback
on the v2, but checking back I realized that I actually just
misunderstood/misremembered what you and Mike were telling me. Sorry
about that, they'll be squashed together in the next posting :)
> --
> Cheers,
>
> David / dhildenb
>
Best,
Patrick
