Re: [Patch v2] mm/huge_memory: fix NULL pointer deference when splitting folio

[Wei Yang <richard.weiyang@gmail.com>]

On Wed, Nov 19, 2025 at 11:53:02PM +0000, Wei Yang wrote:
>Commit c010d47f107f ("mm: thp: split huge page to any lower order
>pages") introduced an early check on the folio's order via
>mapping->flags before proceeding with the split work.
>
>This check introduced a bug: for shmem folios in the swap cache and
>truncated folios, the mapping pointer can be NULL. Accessing
>mapping->flags in this state leads directly to a NULL pointer
>dereference.
>
>This commit fixes the issue by moving the check for mapping != NULL
>before any attempt to access mapping->flags.
>
>Fixes: c010d47f107f ("mm: thp: split huge page to any lower order pages")
>Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
>Cc: Zi Yan <ziy@nvidia.com>
>Cc: "David Hildenbrand (Red Hat)" <david@kernel.org>
>Cc: <stable@vger.kernel.org>
>
>---
>This patch is based on current mm-new, latest commit:
>
>    febb34c02328 dt-bindings: riscv: Add Svrsw60t59b extension description
>
>v2:
>  * just move folio->mapping ahead
>---
> mm/huge_memory.c | 22 ++++++++++------------
> 1 file changed, 10 insertions(+), 12 deletions(-)
>
>diff --git a/mm/huge_memory.c b/mm/huge_memory.c
>index efea42d68157..4e9e920f306d 100644
>--- a/mm/huge_memory.c
>+++ b/mm/huge_memory.c
>@@ -3929,6 +3929,16 @@ static int __folio_split(struct folio *folio, unsigned int new_order,
> 	if (folio != page_folio(split_at) || folio != page_folio(lock_at))
> 		return -EINVAL;
> 
>+	/*
>+	 * Folios that just got truncated cannot get split. Signal to the
>+	 * caller that there was a race.
>+	 *
>+	 * TODO: this will also currently refuse shmem folios that are in the
>+	 * swapcache.
>+	 */
>+	if (!is_anon && !folio->mapping)
>+		return -EBUSY;
>+

This one would have a conflict on direct cherry-pick to current master and
mm-stable.

But if I move this code before (folio != page_folio(split_at) ...), it could
be apply to mm-new and master/mm-stable smoothly.

Not sure whether this could make Andrew's life easier.

> 	if (new_order >= old_order)
> 		return -EINVAL;
> 
>@@ -3965,18 +3975,6 @@ static int __folio_split(struct folio *folio, unsigned int new_order,
> 		gfp_t gfp;
> 
> 		mapping = folio->mapping;
>-
>-		/* Truncated ? */
>-		/*
>-		 * TODO: add support for large shmem folio in swap cache.
>-		 * When shmem is in swap cache, mapping is NULL and
>-		 * folio_test_swapcache() is true.
>-		 */
>-		if (!mapping) {
>-			ret = -EBUSY;
>-			goto out;
>-		}
>-
> 		min_order = mapping_min_folio_order(folio->mapping);
> 		if (new_order < min_order) {
> 			ret = -EINVAL;
>-- 
>2.34.1

-- 
Wei Yang
Help you, Help me