Re: [PATCH v2] mm/slab_common: Deleting kobject in kmem_cache_destroy() without holding slab_mutex/cpu_hotplug_lock

[Waiman Long <longman@redhat.com>]

On 8/10/22 14:10, Roman Gushchin wrote:
> On Wed, Aug 10, 2022 at 12:49:46PM -0400, Waiman Long wrote:
>> A circular locking problem is reported by lockdep due to the following
>> circular locking dependency.
>>
>>    +--> cpu_hotplug_lock --> slab_mutex --> kn->active --+
>>    |                                                     |
>>    +-----------------------------------------------------+
>>
>> The forward cpu_hotplug_lock ==> slab_mutex ==> kn->active dependency
>> happens in
>>
>>    kmem_cache_destroy():	cpus_read_lock(); mutex_lock(&slab_mutex);
>>    ==> sysfs_slab_unlink()
>>        ==> kobject_del()
>>            ==> kernfs_remove()
>> 	      ==> __kernfs_remove()
>> 	          ==> kernfs_drain(): rwsem_acquire(&kn->dep_map, ...);
>>
>> The backward kn->active ==> cpu_hotplug_lock dependency happens in
>>
>>    kernfs_fop_write_iter(): kernfs_get_active();
>>    ==> slab_attr_store()
>>        ==> cpu_partial_store()
>>            ==> flush_all(): cpus_read_lock()
>>
>> One way to break this circular locking chain is to avoid holding
>> cpu_hotplug_lock and slab_mutex while deleting the kobject in
>> sysfs_slab_unlink() which should be equivalent to doing a write_lock
>> and write_unlock pair of the kn->active virtual lock.
>>
>> Since the kobject structures are not protected by slab_mutex or the
>> cpu_hotplug_lock, we can certainly release those locks before doing
>> the delete operation.
>>
>> Move sysfs_slab_unlink() and sysfs_slab_release() to the newly
>> created kmem_cache_release() and call it outside the slab_mutex &
>> cpu_hotplug_lock critical sections.
>>
>> Signed-off-by: Waiman Long <longman@redhat.com>
>> ---
>>   [v2] Break kmem_cache_release() helper into 2 separate ones.
>>
>>   mm/slab_common.c | 54 +++++++++++++++++++++++++++++++++---------------
>>   1 file changed, 37 insertions(+), 17 deletions(-)
>>
>> diff --git a/mm/slab_common.c b/mm/slab_common.c
>> index 17996649cfe3..7742d0446d8b 100644
>> --- a/mm/slab_common.c
>> +++ b/mm/slab_common.c
>> @@ -392,6 +392,36 @@ kmem_cache_create(const char *name, unsigned int size, unsigned int align,
>>   }
>>   EXPORT_SYMBOL(kmem_cache_create);
>>   
>> +#ifdef SLAB_SUPPORTS_SYSFS
>> +static void kmem_cache_workfn_release(struct kmem_cache *s)
>> +{
>> +	sysfs_slab_release(s);
>> +}
>> +#else
>> +static void kmem_cache_workfn_release(struct kmem_cache *s)
>> +{
>> +	slab_kmem_cache_release(s);
>> +}
>> +#endif
>> +
>> +/*
>> + * For a given kmem_cache, kmem_cache_destroy() should only be called
>> + * once or there will be a use-after-free problem. The actual deletion
>> + * and release of the kobject does not need slab_mutex or cpu_hotplug_lock
>> + * protection. So they are now done without holding those locks.
>> + */
>> +static void kmem_cache_release(struct kmem_cache *s)
>> +{
>> +#ifdef SLAB_SUPPORTS_SYSFS
>> +	sysfs_slab_unlink(s);
>> +#endif
>> +
>> +	if (s->flags & SLAB_TYPESAFE_BY_RCU)
>> +		schedule_work(&slab_caches_to_rcu_destroy_work);
>> +	else
>> +		kmem_cache_workfn_release(s);
>> +}
>> +
>>   static void slab_caches_to_rcu_destroy_workfn(struct work_struct *work)
>>   {
>>   	LIST_HEAD(to_destroy);
>> @@ -418,11 +448,7 @@ static void slab_caches_to_rcu_destroy_workfn(struct work_struct *work)
>>   	list_for_each_entry_safe(s, s2, &to_destroy, list) {
>>   		debugfs_slab_release(s);
>>   		kfence_shutdown_cache(s);
>> -#ifdef SLAB_SUPPORTS_SYSFS
>> -		sysfs_slab_release(s);
>> -#else
>> -		slab_kmem_cache_release(s);
>> -#endif
>> +		kmem_cache_workfn_release(s);
>>   	}
>>   }
>>   
>> @@ -437,20 +463,10 @@ static int shutdown_cache(struct kmem_cache *s)
>>   	list_del(&s->list);
>>   
>>   	if (s->flags & SLAB_TYPESAFE_BY_RCU) {
>> -#ifdef SLAB_SUPPORTS_SYSFS
>> -		sysfs_slab_unlink(s);
>> -#endif
>>   		list_add_tail(&s->list, &slab_caches_to_rcu_destroy);
>> -		schedule_work(&slab_caches_to_rcu_destroy_work);
> Hi Waiman!
>
> This version is much more readable, thank you!
>
> But can we, please, leave this schedule_work(&slab_caches_to_rcu_destroy_work)
> call here? I don't see a good reason to move it, do I miss something?
> It's nice to have list_add_tail() and schedule_work() calls nearby, so
> it's obvious we can't miss the latter.

The reason that I need to move out schedule_work() as well is to make 
sure that sysfs_slab_unlink() is called before sysfs_slab_release(). I 
can't guarantee that if I do schedule_work() first. On the other hand, 
moving sysfs_slab_unlink() into kmem_cache_workfn_release() introduces 
unknown delay of when the sysfs file will be removed. I can add some 
comment to make it more clear.

Please let me know if you have a better idea of dealing with this issue.

Thanks,
Longman