[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <66c5a94a-4e6d-7719-7b03-de77a3f18c1b@linux.alibaba.com>
Date: Thu, 11 Aug 2022 11:52:12 +0800
From: haoxin <xhao@...ux.alibaba.com>
To: Hyeonggon Yoo <42.hyeyoo@...il.com>
Cc: cl@...ux.com, penberg@...nel.org, rientjes@...gle.com,
iamjoonsoo.kim@....com, akpm@...ux-foundation.org, vbabka@...e.cz,
roman.gushchin@...ux.dev, linux-mm@...ck.org,
linux-kernel@...r.kernel.org, xhao@...ux.alibaba.com
Subject: Re: [PATCH] mm/slub: release kobject if kobject_init_and_add failed
in sysfs_slab_add
在 2022/8/11 上午11:31, Hyeonggon Yoo 写道:
> On Thu, Aug 11, 2022 at 10:52:58AM +0800, Xin Hao wrote:
>> In kobject_init_and_add() function, the refcount is setted by calling
>> kobject_init() function, regardless of whether the return value is zero
>> or not, therefore, we must call kobject_del(&s->kobj) to prevent memory
> Hello and thanks!
>
> Should kobject_del() be called when kobject_add() failed?
Yes, I'm afraid think so, you can see the comment about kobject_add()
* Return: If this function returns an error, kobject_put() must be
* called to properly clean up the memory associated with the
* object. Under no instance should the kobject that is passed
* to this function be directly freed with a call to kfree(),
* that can leak memory.
*
* If this function returns success, kobject_put() must also be
called
* in order to properly clean up the memory associated with the
object.
*
* In short, once this function is called, kobject_put() MUST
be called
* when the use of the object is finished in order to properly free
* everything.
*/
int kobject_add(struct kobject *kobj, struct kobject *parent,
const char *fmt, ...)
>
> its comments says:
>
> 597 * kobject_del() - Unlink kobject from hierarchy.
> 598 * @kobj: object.
> 599 *
> 600 * This is the function that should be called to delete an object
> 601 * successfully added via kobject_add().
>
> AFAIK kobject_put() is proper function to call when
> kobject_init_and_add() failed as stated in its comment:
>
> 417 /**
> 418 * kobject_init_and_add() - Initialize a kobject structure and add it to
> 419 * the kobject hierarchy.
> 420 * @kobj: pointer to the kobject to initialize
> 421 * @ktype: pointer to the ktype for this kobject.
> 422 * @parent: pointer to the parent of this kobject.
> 423 * @fmt: the name of the kobject.
> 424 *
> 425 * This function combines the call to kobject_init() and kobject_add().
> 426 *
> 427 * If this function returns an error, kobject_put() must be called to
> 428 * properly clean up the memory associated with the object. This is the
> 429 * same type of error handling after a call to kobject_add() and kobject
> 430 * lifetime rules are the same here.
>
>> of s->kobj is leaked.
>>
>> Signed-off-by: Xin Hao <xhao@...ux.alibaba.com>
>> ---
>> mm/slub.c | 7 +++----
>> 1 file changed, 3 insertions(+), 4 deletions(-)
>>
>> diff --git a/mm/slub.c b/mm/slub.c
>> index b1281b8654bd..63b0a8a3a71f 100644
>> --- a/mm/slub.c
>> +++ b/mm/slub.c
>> @@ -5981,19 +5981,18 @@ static int sysfs_slab_add(struct kmem_cache *s)
>>
>> err = sysfs_create_group(&s->kobj, &slab_attr_group);
>> if (err)
>> - goto out_del_kobj;
>> + goto out;
>>
>> if (!unmergeable) {
>> /* Setup first alias */
>> sysfs_slab_alias(s, s->name);
>> }
>> + return err;
>> out:
>> if (!unmergeable)
>> kfree(name);
>> - return err;
>> -out_del_kobj:
>> kobject_del(&s->kobj);
Maybe there use kobject_put will be better.
>> - goto out;
>> + return err;
>> }
>>
>> void sysfs_slab_unlink(struct kmem_cache *s)
>> --
>> 2.31.0
Powered by blists - more mailing lists