[PATCH 09/46] mm/rmap: handle device-exclusive entries in folio_referenced_one()

[Lorenzo Stoakes (ARM)]

On Wed, Oct 28, 2026 at 09:08:04AM +0100, David Hildenbrand (Arm) wrote:
> >> -		if (lru_gen_enabled() && !lru_gen_switching() && pvmw.pte) {
> >> +		if (lru_gen_enabled() && !lru_gen_switching() && pvmw.pte &&
> >> +		    pte_present(pteval)) {
> >>  			if (lru_gen_look_around(&pvmw, nr))
> >>  				referenced++;
> >>  		} else if (pvmw.pte) {
> >> -			if (clear_flush_young_ptes_notify(vma, address, pvmw.pte, nr))
> >> +			bool young = false;
> >> +
> >> +			/*
> >> +			 * PFN swap PTEs, such as device-exclusive ones, that
> >
> > You mean "PFN softleaf PTEs"? Better expressed as "softleaf PTEs embedding a PFN" ...
>
> Lorenzo raised this off-list as well; I'm using the existing text we use
> elsewhere for consistency.
>
> This needs a cleanup that adjusts all of them :)

*like a proud dad of his softleaf name :P*

Yeah, agreed, not a big deal, and easier for me to just do a bigger pass
finishing off what I started :P

--
Cheers, Lorenzo