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

[David Hildenbrand (Arm)]

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@xxxxxxxxx>
> ---
>  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.

freeze=true really only applies during page migration, where swap entries don't
apply.

I think it's time to clean that up ... that is

a) Expose a helper called split_pmd_to_migration_entries() that is only used by
code that installs migration entries.

b) Hide that "freeze" flag from all other file-external functions

c) Rename the boolean to "use_migration_entries"

Then it's rather clear in this code that this should never happen.



> +		old_pmd = *pmd;
> +		soft_dirty = pmd_swp_soft_dirty(old_pmd);
> +		uffd_wp = pmd_swp_uffd(old_pmd);
> +		anon_exclusive = pmd_swp_exclusive(old_pmd);
>  	} else {
>  		/*
>  		 * Up to this point the pmd is present and huge and userland has
> @@ -3388,6 +3396,25 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>  			VM_WARN_ON(!pte_none(ptep_get(pte + i)));
>  			set_pte_at(mm, addr, pte + i, entry);
>  		}
> +	} else if (pmd_is_swap_entry(old_pmd)) {
> +		softleaf_t sl_entry = softleaf_from_pmd(old_pmd);

No existing code uses "sl_entry". Maybe just call it "pmd_swp_entry"/"swp_entry"
and below "pte_swp_entry".



Apart from that nothing jumped at me :)

-- 
Cheers,

David