On Wed, Oct 18, 2023 at 06:29:52PM +0800, Liu Shixin wrote:
> diff --git a/mm/kmemleak.c b/mm/kmemleak.c
> index 7c9125c18956..a956b2734324 100644
> --- a/mm/kmemleak.c
> +++ b/mm/kmemleak.c
> @@ -816,16 +816,25 @@ static void delete_object_full(unsigned long ptr)
> */
> static void delete_object_part(unsigned long ptr, size_t size, bool is_phys)
> {
> - struct kmemleak_object *object;
> - unsigned long start, end;
> + struct kmemleak_object *object, *object_l, *object_r;
> + unsigned long start, end, flags;
> +
> + object_l = __alloc_object(GFP_KERNEL);
> + if (!object_l)
> + return;
>
> - object = find_and_remove_object(ptr, 1, is_phys);
> + object_r = __alloc_object(GFP_KERNEL);
> + if (!object_r)
> + goto out;
> +
> + raw_spin_lock_irqsave(&kmemleak_lock, flags);
> + object = __find_and_remove_object(ptr, 1, is_phys);
> if (!object) {
> #ifdef DEBUG
> kmemleak_warn("Partially freeing unknown object at 0x%08lx (size %zu)\n",
> ptr, size);
> #endif
> - return;
> + goto unlock;
> }
>
> /*
> @@ -835,14 +844,25 @@ static void delete_object_part(unsigned long ptr, size_t size, bool is_phys)
> */
> start = object->pointer;
> end = object->pointer + object->size;
> - if (ptr > start)
> - __create_object(start, ptr - start, object->min_count,
> - GFP_KERNEL, is_phys);
> - if (ptr + size < end)
> - __create_object(ptr + size, end - ptr - size, object->min_count,
> - GFP_KERNEL, is_phys);
> + if ((ptr > start) &&
> + !__link_object(object_l, start, ptr - start,
> + object->min_count, is_phys))
> + object_l = NULL;
> + if ((ptr + size < end) &&
> + !__link_object(object_r, ptr + size, end - ptr - size,
> + object->min_count, is_phys))
> + object_r = NULL;
> +
> +unlock:
> + raw_spin_unlock_irqrestore(&kmemleak_lock, flags);
> + if (object)
> + __delete_object(object);
>
> - __delete_object(object);
> +out:
> + if (object_l)
> + mem_pool_free(object_l);
> + if (object_r)
> + mem_pool_free(object_r);
> }
Ah, I see now why __link_object() needs to do further object
initialisation under the lock as prior to find_and_remove_object() we
don't have all the information needed for object_l and object_r.
Maybe in the __create_object splitting patch you can leave
__alloc_object() just do the actual allocation and __link_object() do
the full initialisation. Up to you, you can send a separate patch on top
as I can see Andrew picked them up already.
Otherwise the series looks fine.
Reviewed-by: Catalin Marinas <[email protected]>