Re: [PATCH] mm/huge_memory: refactor after-split (page) cache code.

[Lorenzo Stoakes <lorenzo.stoakes@oracle.com>]

On Wed, Jul 16, 2025 at 01:11:12PM -0400, Zi Yan wrote:
> Smatch/coverity checkers report NULL mapping referencing issues[1][2][3]
> every time the code is modified, because they do not understand that
> mapping cannot be NULL when a folio is in page cache in the code.
> Refactor the code to make it explicit.
>
> No functional change is intended.
>
> [1]https://lore.kernel.org/linux-mm/2afe3d59-aca5-40f7-82a3-a6d976fb0f4f@stanley.mountain/
> [2]https://lore.kernel.org/oe-kbuild/64b54034-f311-4e7d-b935-c16775dbb642@suswa.mountain/
> [3]https://lore.kernel.org/linux-mm/20250716145804.4836-1-antonio@mandelbit.com/
>
> Suggested-by: David Hildenbrand <david@redhat.com>
> Signed-off-by: Zi Yan <ziy@nvidia.com>

This is fantastic, thanks Zi! There's a nit below but I actually almost
_don't_ want you to address it :P

Therefore:

Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>

> ---
>  mm/huge_memory.c | 43 ++++++++++++++++++++++++++++---------------
>  1 file changed, 28 insertions(+), 15 deletions(-)
>
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 31b5c4e61a57..fe17b0a157cd 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -3804,6 +3804,8 @@ static int __folio_split(struct folio *folio, unsigned int new_order,
>  		 */
>  		for (new_folio = folio_next(folio); new_folio != next_folio;
>  		     new_folio = next) {
> +			unsigned long nr_pages = folio_nr_pages(new_folio);
> +
>  			next = folio_next(new_folio);
>
>  			expected_refs = folio_expected_ref_count(new_folio) + 1;
> @@ -3811,25 +3813,36 @@ static int __folio_split(struct folio *folio, unsigned int new_order,
>
>  			lru_add_split_folio(folio, new_folio, lruvec, list);
>
> -			/* Some pages can be beyond EOF: drop them from cache */
> -			if (new_folio->index >= end) {
> -				if (shmem_mapping(mapping))
> -					nr_shmem_dropped += folio_nr_pages(new_folio);
> -				else if (folio_test_clear_dirty(new_folio))
> -					folio_account_cleaned(
> -						new_folio,
> -						inode_to_wb(mapping->host));
> -				__filemap_remove_folio(new_folio, NULL);
> -				folio_put_refs(new_folio,
> -					       folio_nr_pages(new_folio));
> -			} else if (mapping) {
> -				__xa_store(&mapping->i_pages, new_folio->index,
> -					   new_folio, 0);
> -			} else if (swap_cache) {
> +			/*
> +			 * Anonymous folio with swap cache.
> +			 * NOTE: shmem in swap cache is not supported yet.

Nice added context!

> +			 */
> +			if (swap_cache) {
>  				__xa_store(&swap_cache->i_pages,
>  					   swap_cache_index(new_folio->swap),
>  					   new_folio, 0);
> +				continue;
> +			}
> +
> +			/* Anonymouse folio without swap cache */

I almost don't want to comment here because 'anony-mouse' is really cute :P
but yeah nit I think you have a trailing 'e' here that my cats would be
VERY interested in... ;)

> +			if (!mapping)
> +				continue;
> +
> +			/* Add the new folio to the page cache. */
> +			if (new_folio->index < end) {
> +				__xa_store(&mapping->i_pages, new_folio->index,
> +					   new_folio, 0);
> +				continue;
>  			}
> +
> +			/* Drop folio beyond EOF: ->index >= end */
> +			if (shmem_mapping(mapping))
> +				nr_shmem_dropped += nr_pages;
> +			else if (folio_test_clear_dirty(new_folio))
> +				folio_account_cleaned(
> +					new_folio, inode_to_wb(mapping->host));
> +			__filemap_remove_folio(new_folio, NULL);
> +			folio_put_refs(new_folio, nr_pages);
>  		}
>  		/*
>  		 * Unfreeze @folio only after all page cache entries, which
> --
> 2.47.2
>

Since we no longer need to make new_folio->index >= end work for anon
folios, can we drop the end = -1 in the if (is_anon) { ... } branch?