linux-kernel - Re: [PATCH] slub: fix failure when we delete and create a slab cache

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <20180615155809.77862e1f6376d5779da9d991@linux-foundation.org>
Date:   Fri, 15 Jun 2018 15:58:09 -0700
From:   Andrew Morton <akpm@...ux-foundation.org>
To:     Mikulas Patocka <mpatocka@...hat.com>
Cc:     Christoph Lameter <cl@...ux.com>,
        Pekka Enberg <penberg@...nel.org>,
        David Rientjes <rientjes@...gle.com>,
        Joonsoo Kim <iamjoonsoo.kim@....com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] slub: fix failure when we delete and create a slab
 cache

On Fri, 15 Jun 2018 18:25:29 -0400 (EDT) Mikulas Patocka <mpatocka@...hat.com> wrote:

> In the kernel 4.17 I removed some code from dm-bufio that did slab cache
> merging (21bb13276768) - both slab and slub support merging caches with
> identical attributes, so dm-bufio now just calls kmem_cache_create and
> relies on implicit merging.
> 
> This uncovered a bug in the slub subsystem - if we delete a cache and
> immediatelly create another cache with the same attributes, it fails
> because of duplicate filename in /sys/kernel/slab/. The slub subsystem
> offloads freeing the cache to a workqueue - and if we create the new cache
> before the workqueue runs, it complains because of duplicate filename in
> sysfs.

Huh.  Surprised that such an obvious blooper survived this long.  I
guess a rapid del+add is uncommon.

> This patch fixes the bug by moving the call of kobject_del from 
> sysfs_slab_remove_workfn to shutdown_cache. kobject_del must be called 
> while we hold slab_mutex - so that the sysfs entry is deleted before a 
> cache with the same attributes could be created.
> 
> 
> Running device-mapper-test-suite with:

Nice changelog, btw.

> --- linux-2.6.orig/include/linux/slub_def.h
> +++ linux-2.6/include/linux/slub_def.h
> @@ -156,8 +156,12 @@ struct kmem_cache {
>  
>  #ifdef CONFIG_SYSFS
>  #define SLAB_SUPPORTS_SYSFS
> +void sysfs_slab_unlink(struct kmem_cache *);
>  void sysfs_slab_release(struct kmem_cache *);
>  #else
> +static inline void sysfs_slab_unlink(struct kmem_cache *s)
> +{
> +}
>  static inline void sysfs_slab_release(struct kmem_cache *s)
>  {
>  }

hm, that's pretty old-school.  We could replace SLAB_SUPPORTS_SYSFS
with CONFIG_SLAB_SUPPORTS_SYSFS, move the above logic into slab.h and..

> --- linux-2.6.orig/mm/slab_common.c
> +++ linux-2.6/mm/slab_common.c
> @@ -566,10 +566,14 @@ static int shutdown_cache(struct kmem_ca
>  	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);
>  	} else {
>  #ifdef SLAB_SUPPORTS_SYSFS
> +		sysfs_slab_unlink(s);
>  		sysfs_slab_release(s);
>  #else
>  		slab_kmem_cache_release(s);

remove a bunch of ifdefs.  But that would be a separate thing.