Re: [PATCH] mm/mm_slot: make sure slot is the first element of its wrapper structure

[Kiryl Shutsemau <kirill@shutemov.name>]

On Sun, Sep 14, 2025 at 12:00:26AM +0000, Wei Yang wrote:
> When using mm_slot in ksm/khugepaged, there is code snip like:
> 
> 	slot = mm_slot_lookup(mm_slots_hash, mm);
> 	mm_slot = mm_slot_entry(slot, struct ksm_mm_slot, slot);
> 	if (mm_slot && ..) {
> 	}
> 
> This is only valid when mm_slot is the first element of its wrapper
> structure, otherwise a NULL slot would converted to a mm_slot with
> negative value. And current code thinks it is valid and continue.
> 
> Current code works since mm_slot is the first element, but make sure it
> won't be disturbed.

Seems excessive for this kind of problem.

Why not check slot in mm_slot_entry?

diff --git a/mm/mm_slot.h b/mm/mm_slot.h
index 83f18ed1c4bd..0de24c490b62 100644
--- a/mm/mm_slot.h
+++ b/mm/mm_slot.h
@@ -19,7 +19,7 @@ struct mm_slot {
 };

 #define mm_slot_entry(ptr, type, member) \
-	container_of(ptr, type, member)
+	(ptr ? container_of(ptr, type, member) : NULL)

 static inline void *mm_slot_alloc(struct kmem_cache *cache)
 {
-- 
  Kiryl Shutsemau / Kirill A. Shutemov