Re: [PATCH v3 2/3] mm: mprotect: avoid unnecessary struct page accessing if pte_protnone()

[Kefeng Wang <wangkefeng.wang@huawei.com>]

On 2025/10/16 1:45, David Hildenbrand wrote:
> On 15.10.25 17:43, Lorenzo Stoakes wrote:
>> On Wed, Oct 15, 2025 at 08:35:15PM +0800, Kefeng Wang wrote:
>>> If the pte_protnone() is true, we could avoid unnecessary struct page
>>> accessing and reduce cache footprint when scanning page tables for prot
>>> numa, the performance test of pmbench memory accessing benchmark
>>> should be benifit, see more commit a818f5363a0e ("autonuma: reduce cache
>>> footprint when scanning page tables").
>>
>> Hm should benefit? But you've not tested it? Seems like rather than
>> guessing you should actually give data. Otherwise I don't know why we're
>> referencing a benchmark.
> 

Yes, I don't test it, just find the above commit which does same thing 
before and it shows benefit in a benchmark.

> I guess it might be harder to quantify, but theoretically it makes sense 
> to me to just skip in any case.
> 
> ... and it enables patch #3, which is nice :)
> 

My initial goal was to skip unnecessary steps for pmd folio when prot numa.