Re: [PATCH v2 4/4] mm: remove boolean output parameters from folio_pte_batch_ext()

[Oscar Salvador <osalvador@suse.de>]

On Wed, Jul 02, 2025 at 12:49:26PM +0200, David Hildenbrand wrote:
> Instead, let's just allow for specifying through flags whether we want
> to have bits merged into the original PTE.
> 
> For the madvise() case, simplify by having only a single parameter for
> merging young+dirty. For madvise_cold_or_pageout_pte_range() merging the
> dirty bit is not required, but also not harmful. This code is not that
> performance critical after all to really force all micro-optimizations.

Seems a bit odd to me to have the dirty-young bit being treat as "one".
You mention that this is done because the only user of it doesn't really
care about dirty vs non-dirty and it's not harmful eitherway, so micro-optimizing
this isn't worth at this moment.

But what if we grop another user which wants to make this distinction and
where it matters dirty vs non-dirty/young vs not-young.
Wouldn't be better to have it separated from the start?

And I'm not talking about micro-optimizing, because that's clear that it
doesn't matter, but for clarity purposes.
It seems a lot more organic/natural/obvious to me.



-- 
Oscar Salvador
SUSE Labs