Re: [PATCH v6 03/12] mm: add PMD swap entry splitting support

[Kiryl Shutsemau <kas@kernel.org>]

On Tue, Aug 18, 2026 at 07:53:48PM +0200, David Hildenbrand (Arm) wrote:
> On 8/18/26 15:09, Usama Arif wrote:
> > Add a swap branch in __split_huge_pmd_locked() that splits a PMD swap
> > entry into 512 PTE swap entries. No folio reference is needed because
> > swap entries point to swap slots rather than pages. Each PTE inherits
> > the correct sub-slot offset and preserves soft_dirty, uffd_wp, and
> > exclusive flags.
> > 
> > The folio_remove_rmap_pmd() gate at the end must inspect old_pmd
> > rather than *pmd: for a present THP split, *pmd has already been
> > cleared by pmdp_invalidate(), and that invalidated bit pattern can
> > decode as a plausible swap entry.
> > 
> > This branch is reached from the explicit __split_huge_pmd() callers
> > that hit a non-present PMD: partial-range mprotect / munmap, the
> > wp_huge_pmd() PMD-COW fallback, and the swap-in / swapoff fallbacks
> > added in later patches when the cached folio is no longer PMD-sized.
> > page_vma_mapped_walk() does not iterate PMD swap entries, so
> > try_to_unmap_one() and try_to_migrate_one() do not reach this branch
> > and freeze=true cannot occur in this branch today.  page and folio
> > are therefore left uninitialized in the swap branch; a
> > VM_WARN_ON_ONCE(freeze) catches any future caller that breaks this
> > invariant before the freeze path dereferences page_to_pfn(page + i)
> > or put_page(page).
> > 
> > Signed-off-by: Usama Arif <usama.arif@linux.dev>
> > ---
> >  mm/huge_memory.c | 29 ++++++++++++++++++++++++++++-
> >  1 file changed, 28 insertions(+), 1 deletion(-)
> > 
> > diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> > index 1b6b0aa2baa3b..a473e85d30f51 100644
> > --- a/mm/huge_memory.c
> > +++ b/mm/huge_memory.c
> > @@ -3252,6 +3252,14 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
> >  			folio_add_anon_rmap_ptes(folio, page, HPAGE_PMD_NR,
> >  						 vma, haddr, rmap_flags);
> >  		}
> > +	} else if (pmd_is_swap_entry(*pmd)) {
> > +		VM_WARN_ON_ONCE(freeze);
> > +		/* Swap entries have no page for the migration freeze path. */
> > +		freeze = false;
> 
> It's odd to VM_WARN_ON_ONCE() and then set freeze=false;
> 
> I'd just add the comment above the VM_WARN_ON_ONCE() and drop the =false.

VM_WARN_ON_ONCE() is compiled out without DEBUG_VM. The freeze = false is
what actually keeps the swap entry out of

      if (freeze || pmd_is_migration_entry(old_pmd)) {
              ...
              make_writable_migration_entry(page_to_pfn(page + i));

and out of the put_page(page) below it, where page is uninitialized in the
swap branch. Dropping it leaves nothing on production builds.

If freeze = false goes, folio and page must be set to NULL there (or
something along the lines).

-- 
  Kiryl Shutsemau / Kirill A. Shutemov