Re: [PATCH v3 2/2] mm/mm_slot.h: add comments for mm_slot_lookup/insert

[Qi Zheng <qi.zheng@linux.dev>]

On 7/14/26 9:28 AM, xu.xin16@zte.com.cn wrote:
> From: xu xin <xu.xin16@zte.com.cn>
> 
> mm_slot_lookup() and mm_slot_insert() are the only helpers in
> this header that are implemented as macros rather than static inline
> functions. This may look inconsistent without explanation.
> 
> Explain they must be macros because hash_for_each_possible()
> needs the table as an array (for sizeof), not a pointer.
> 
> Signed-off-by: xu xin <xu.xin16@zte.com.cn>
> ---
>   mm/mm_slot.h | 6 ++++++
>   1 file changed, 6 insertions(+)
> 
> diff --git a/mm/mm_slot.h b/mm/mm_slot.h
> index 5de3e91d86b4..9b09b68e5742 100644
> --- a/mm/mm_slot.h
> +++ b/mm/mm_slot.h
> @@ -33,6 +33,12 @@ static inline void mm_slot_free(struct kmem_cache *cache, void *objp)
>   	kmem_cache_free(cache, objp);
>   }
> 
> +/*
> + * Note: mm_slot_lookup and mm_slot_insert cannot be converted to static inline
> + * functions because the hash helpers (hash_for_each_possible and hash_add) rely
> + * on the actual array argument 'hashtable' for sizeof() instead of pointers.
> + */

Right, that's why I wrote them as macros to begin with.

Reviewed-by: Qi Zheng <qi.zheng@linux.dev>

Thanks,
Qi

> +
>   #define mm_slot_lookup(_hashtable, _mm) 				       \
>   ({									       \
>   	struct mm_slot *tmp_slot, *mm_slot = NULL;			       \