Re: [PATCH v2 1/2] mm: Use a folio in the softleaf_is_device_private path

[Lorenzo Stoakes (ARM)]

On Thu, Jul 23, 2026 at 11:07:51AM +0800, Hongfu Li wrote:
> From: Hongfu Li <lihongfu@xxxxxxxxxx>
>
> Use the folio APIs in the device_private migration path of do_swap_page(),
> replacing four calls to compound_head() with one.
>
> Signed-off-by: Hongfu Li <lihongfu@xxxxxxxxxx>
> Acked-by: David Hildenbrand (Arm) <david@xxxxxxxxxx>

With changes below applied LGTM, so:

Reviewed-by: Lorenzo Stoakes (ARM) <ljs@xxxxxxxxxx>

> ---
>  mm/memory.c | 11 ++++++-----
>  1 file changed, 6 insertions(+), 5 deletions(-)
>
> diff --git a/mm/memory.c b/mm/memory.c
> index d5e87624f692..ae521baa8c84 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -4871,18 +4871,19 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>  				goto unlock;
>

Up here we have:

			vmf->page = softleaf_to_page(entry);
			vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
					vmf->address, &vmf->ptl);
			if (unlikely(!vmf->pte ||
				     !pte_same(ptep_get(vmf->pte),
							vmf->orig_pte)))

Since we're canonicalising to folio why not actually set folio using
softleaf_to_folio()?

So

+			folio = softleaf_to_folio(entry);
-			vmf->page = softleaf_to_page(entry);
+			vmf->page = &folio->page;

This would also make more sense in your 2/2. But more on that in a sec.

>  			/*
> -			 * Get a page reference while we know the page can't be
> +			 * Get a folio reference while we know the folio can't be

NIT: be -> next line for 80 chars.

>  			 * freed.
>  			 */
> -			if (trylock_page(vmf->page)) {
> +			folio = page_folio(vmf->page);
> +			if (folio_trylock(folio)) {
>  				struct dev_pagemap *pgmap;
>
> -				get_page(vmf->page);
> +				folio_get(folio);
>  				pte_unmap_unlock(vmf->pte, vmf->ptl);
>  				pgmap = page_pgmap(vmf->page);
>  				ret = pgmap->ops->migrate_to_ram(vmf);
> -				unlock_page(vmf->page);
> -				put_page(vmf->page);
> +				folio_unlock(folio);
> +				folio_put(folio);
>  			} else {
>  				pte_unmap(vmf->pte);
>  				softleaf_entry_wait_on_locked(entry, vmf->ptl);
> --
> 2.54.0
>

Thanks, Lorenzo