All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jiri Slaby <jslaby@suse.cz>
To: Aristeu Rozanski <aris@redhat.com>
Cc: linux-kernel@vger.kernel.org, Dave Jones <davej@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Tejun Heo <tj@kernel.org>, Li Zefan <lizefan@huawei.com>,
	James Morris <jmorris@namei.org>,
	Pavel Emelyanov <xemul@openvz.org>,
	Serge Hallyn <serge.hallyn@canonical.com>,
	cgroups@vger.kernel.org
Subject: Re: [PATCH 1/4] cgroup: fix invalid rcu dereference
Date: Tue, 23 Oct 2012 14:50:48 +0200	[thread overview]
Message-ID: <508692A8.6090706@suse.cz> (raw)
In-Reply-To: <20121022134536.543579196@napanee.usersys.redhat.com>

On 10/22/2012 03:45 PM, Aristeu Rozanski wrote:
> From: Jiri Slaby <jslaby@suse.cz>

No, I'm *not* the author of the patch... Or at least I don't remember
:P. Where did you take it from?

> Commit "device_cgroup: convert device_cgroup internally to policy +
> exceptions" removed rcu locks which are needed in task_devcgroup
> called in this chain: devcgroup_inode_mknod OR
> __devcgroup_inode_permission -> __devcgroup_inode_permission ->
> task_devcgroup -> task_subsys_state -> task_subsys_state_check.
> 
> Change the code so that task_devcgroup is safely called with rcu read
> lock held.
> 
> ===============================
> [ INFO: suspicious RCU usage. ]
> 3.6.0-rc5-next-20120913+ #42 Not tainted
> -------------------------------
> include/linux/cgroup.h:553 suspicious rcu_dereference_check() usage!
> 
> other info that might help us debug this:
> 
> rcu_scheduler_active = 1, debug_locks = 0
> 2 locks held by kdevtmpfs/23:
>  #0:  (sb_writers){.+.+.+}, at: [<ffffffff8116873f>]
> mnt_want_write+0x1f/0x50
>  #1:  (&sb->s_type->i_mutex_key#3/1){+.+.+.}, at: [<ffffffff811558af>]
> kern_path_create+0x7f/0x170
> 
> stack backtrace:
> Pid: 23, comm: kdevtmpfs Not tainted 3.6.0-rc5-next-20120913+ #42
> Call Trace:
>  [<ffffffff810c638d>] lockdep_rcu_suspicious+0xfd/0x130
>  [<ffffffff8121541d>] devcgroup_inode_mknod+0x19d/0x240
>  [<ffffffff8107bf54>] ? ns_capable+0x44/0x80
>  [<ffffffff81156b21>] vfs_mknod+0x71/0xf0
>  [<ffffffff813a8332>] handle_create.isra.2+0x72/0x200
>  [<ffffffff813a85d4>] devtmpfsd+0x114/0x140
>  [<ffffffff813a84c0>] ? handle_create.isra.2+0x200/0x200
>  [<ffffffff81093ad6>] kthread+0xd6/0xe0
>  [<ffffffff81654f24>] kernel_thread_helper+0x4/0x10
>  [<ffffffff8165369d>] ? retint_restore_args+0xe/0xe
>  [<ffffffff81093a00>] ? kthread_create_on_node+0x140/0x140
>  [<ffffffff81654f20>] ? gs_change+0xb/0xb
> 
> Cc: Dave Jones <davej@redhat.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Tejun Heo <tj@kernel.org>
> Cc: Li Zefan <lizefan@huawei.com>
> Cc: James Morris <jmorris@namei.org>
> Cc: Pavel Emelyanov <xemul@openvz.org>
> Cc: Serge Hallyn <serge.hallyn@canonical.com>
> Signed-off-by: Jiri Slaby <jslaby@suse.cz>
> 
> ---
> 
> And this should fix it.
> 
>  security/device_cgroup.c |   15 +++++++--------
>  1 file changed, 7 insertions(+), 8 deletions(-)
> 
> --- github.orig/security/device_cgroup.c	2012-10-17 11:11:08.514793906 -0400
> +++ github/security/device_cgroup.c	2012-10-19 16:35:37.936804289 -0400
> @@ -533,10 +533,10 @@ struct cgroup_subsys devices_subsys = {
>   *
>   * returns 0 on success, -EPERM case the operation is not permitted
>   */
> -static int __devcgroup_check_permission(struct dev_cgroup *dev_cgroup,
> -					short type, u32 major, u32 minor,
> +static int __devcgroup_check_permission(short type, u32 major, u32 minor,
>  				        short access)
>  {
> +	struct dev_cgroup *dev_cgroup;
>  	struct dev_exception_item ex;
>  	int rc;
>  
> @@ -547,6 +547,7 @@ 	memset(&ex, 0, sizeof(ex));
>  	ex.access = access;
>  
>  	rcu_read_lock();
> +	dev_cgroup = task_devcgroup(current);
>  	rc = may_access(dev_cgroup, &ex);
>  	rcu_read_unlock();
>  
> @@ -558,7 +559,6 @@ 	return 0;
>  
>  int __devcgroup_inode_permission(struct inode *inode, int mask)
>  {
> -	struct dev_cgroup *dev_cgroup = task_devcgroup(current);
>  	short type, access = 0;
>  
>  	if (S_ISBLK(inode->i_mode))
> @@ -570,13 +570,12 @@ 	short type, access = 0;
>  	if (mask & MAY_READ)
>  		access |= ACC_READ;
>  
> -	return __devcgroup_check_permission(dev_cgroup, type, imajor(inode),
> -					    iminor(inode), access);
> +	return __devcgroup_check_permission(type, imajor(inode), iminor(inode),
> +			access);
>  }
>  
>  int devcgroup_inode_mknod(int mode, dev_t dev)
>  {
> -	struct dev_cgroup *dev_cgroup = task_devcgroup(current);
>  	short type;
>  
>  	if (!S_ISBLK(mode) && !S_ISCHR(mode))
> @@ -587,7 +586,7 @@ 		return 0;
>  	else
>  		type = DEV_CHAR;
>  
> -	return __devcgroup_check_permission(dev_cgroup, type, MAJOR(dev),
> -					    MINOR(dev), ACC_MKNOD);
> +	return __devcgroup_check_permission(type, MAJOR(dev), MINOR(dev),
> +			ACC_MKNOD);
>  
>  }
> 


-- 
js
suse labs

WARNING: multiple messages have this Message-ID (diff)
From: Jiri Slaby <jslaby-AlSwsSmVLrQ@public.gmane.org>
To: Aristeu Rozanski <aris-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Cc: linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Dave Jones <davej-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Andrew Morton
	<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
	Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
	Li Zefan <lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>,
	James Morris <jmorris-gx6/JNMH7DfYtjvyW6yDsg@public.gmane.org>,
	Pavel Emelyanov <xemul-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org>,
	Serge Hallyn
	<serge.hallyn-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org>,
	cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH 1/4] cgroup: fix invalid rcu dereference
Date: Tue, 23 Oct 2012 14:50:48 +0200	[thread overview]
Message-ID: <508692A8.6090706@suse.cz> (raw)
In-Reply-To: <20121022134536.543579196-cd6kKtb6gxi3M6m420IelR/sF2h8X+2i0E9HWUfgJXw@public.gmane.org>

On 10/22/2012 03:45 PM, Aristeu Rozanski wrote:
> From: Jiri Slaby <jslaby-AlSwsSmVLrQ@public.gmane.org>

No, I'm *not* the author of the patch... Or at least I don't remember
:P. Where did you take it from?

> Commit "device_cgroup: convert device_cgroup internally to policy +
> exceptions" removed rcu locks which are needed in task_devcgroup
> called in this chain: devcgroup_inode_mknod OR
> __devcgroup_inode_permission -> __devcgroup_inode_permission ->
> task_devcgroup -> task_subsys_state -> task_subsys_state_check.
> 
> Change the code so that task_devcgroup is safely called with rcu read
> lock held.
> 
> ===============================
> [ INFO: suspicious RCU usage. ]
> 3.6.0-rc5-next-20120913+ #42 Not tainted
> -------------------------------
> include/linux/cgroup.h:553 suspicious rcu_dereference_check() usage!
> 
> other info that might help us debug this:
> 
> rcu_scheduler_active = 1, debug_locks = 0
> 2 locks held by kdevtmpfs/23:
>  #0:  (sb_writers){.+.+.+}, at: [<ffffffff8116873f>]
> mnt_want_write+0x1f/0x50
>  #1:  (&sb->s_type->i_mutex_key#3/1){+.+.+.}, at: [<ffffffff811558af>]
> kern_path_create+0x7f/0x170
> 
> stack backtrace:
> Pid: 23, comm: kdevtmpfs Not tainted 3.6.0-rc5-next-20120913+ #42
> Call Trace:
>  [<ffffffff810c638d>] lockdep_rcu_suspicious+0xfd/0x130
>  [<ffffffff8121541d>] devcgroup_inode_mknod+0x19d/0x240
>  [<ffffffff8107bf54>] ? ns_capable+0x44/0x80
>  [<ffffffff81156b21>] vfs_mknod+0x71/0xf0
>  [<ffffffff813a8332>] handle_create.isra.2+0x72/0x200
>  [<ffffffff813a85d4>] devtmpfsd+0x114/0x140
>  [<ffffffff813a84c0>] ? handle_create.isra.2+0x200/0x200
>  [<ffffffff81093ad6>] kthread+0xd6/0xe0
>  [<ffffffff81654f24>] kernel_thread_helper+0x4/0x10
>  [<ffffffff8165369d>] ? retint_restore_args+0xe/0xe
>  [<ffffffff81093a00>] ? kthread_create_on_node+0x140/0x140
>  [<ffffffff81654f20>] ? gs_change+0xb/0xb
> 
> Cc: Dave Jones <davej-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
> Cc: Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
> Cc: Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
> Cc: Li Zefan <lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
> Cc: James Morris <jmorris-gx6/JNMH7DfYtjvyW6yDsg@public.gmane.org>
> Cc: Pavel Emelyanov <xemul-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org>
> Cc: Serge Hallyn <serge.hallyn-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org>
> Signed-off-by: Jiri Slaby <jslaby-AlSwsSmVLrQ@public.gmane.org>
> 
> ---
> 
> And this should fix it.
> 
>  security/device_cgroup.c |   15 +++++++--------
>  1 file changed, 7 insertions(+), 8 deletions(-)
> 
> --- github.orig/security/device_cgroup.c	2012-10-17 11:11:08.514793906 -0400
> +++ github/security/device_cgroup.c	2012-10-19 16:35:37.936804289 -0400
> @@ -533,10 +533,10 @@ struct cgroup_subsys devices_subsys = {
>   *
>   * returns 0 on success, -EPERM case the operation is not permitted
>   */
> -static int __devcgroup_check_permission(struct dev_cgroup *dev_cgroup,
> -					short type, u32 major, u32 minor,
> +static int __devcgroup_check_permission(short type, u32 major, u32 minor,
>  				        short access)
>  {
> +	struct dev_cgroup *dev_cgroup;
>  	struct dev_exception_item ex;
>  	int rc;
>  
> @@ -547,6 +547,7 @@ 	memset(&ex, 0, sizeof(ex));
>  	ex.access = access;
>  
>  	rcu_read_lock();
> +	dev_cgroup = task_devcgroup(current);
>  	rc = may_access(dev_cgroup, &ex);
>  	rcu_read_unlock();
>  
> @@ -558,7 +559,6 @@ 	return 0;
>  
>  int __devcgroup_inode_permission(struct inode *inode, int mask)
>  {
> -	struct dev_cgroup *dev_cgroup = task_devcgroup(current);
>  	short type, access = 0;
>  
>  	if (S_ISBLK(inode->i_mode))
> @@ -570,13 +570,12 @@ 	short type, access = 0;
>  	if (mask & MAY_READ)
>  		access |= ACC_READ;
>  
> -	return __devcgroup_check_permission(dev_cgroup, type, imajor(inode),
> -					    iminor(inode), access);
> +	return __devcgroup_check_permission(type, imajor(inode), iminor(inode),
> +			access);
>  }
>  
>  int devcgroup_inode_mknod(int mode, dev_t dev)
>  {
> -	struct dev_cgroup *dev_cgroup = task_devcgroup(current);
>  	short type;
>  
>  	if (!S_ISBLK(mode) && !S_ISCHR(mode))
> @@ -587,7 +586,7 @@ 		return 0;
>  	else
>  		type = DEV_CHAR;
>  
> -	return __devcgroup_check_permission(dev_cgroup, type, MAJOR(dev),
> -					    MINOR(dev), ACC_MKNOD);
> +	return __devcgroup_check_permission(type, MAJOR(dev), MINOR(dev),
> +			ACC_MKNOD);
>  
>  }
> 


-- 
js
suse labs

  parent reply	other threads:[~2012-10-23 12:50 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-22 13:45 [PATCH 0/4] Rebase device_cgroup v2 patchset Aristeu Rozanski
2012-10-22 13:45 ` [PATCH 1/4] cgroup: fix invalid rcu dereference Aristeu Rozanski
2012-10-22 16:11   ` Serge Hallyn
2012-10-22 16:11     ` Serge Hallyn
2012-10-23 12:50   ` Jiri Slaby [this message]
2012-10-23 12:50     ` Jiri Slaby
2012-10-23 13:17     ` Aristeu Rozanski
2012-10-23 13:17       ` Aristeu Rozanski
2012-10-23 13:53       ` Jiri Slaby
2012-10-23 13:53         ` Jiri Slaby
2012-10-22 13:45 ` [PATCH 2/4] device_cgroup: rename deny_all to behavior Aristeu Rozanski
2012-10-22 16:12   ` Serge Hallyn
2012-10-22 13:45 ` [PATCH 3/4] device_cgroup: stop using simple_strtoul() Aristeu Rozanski
2012-10-22 16:14   ` Serge Hallyn
2012-10-22 16:14     ` Serge Hallyn
2012-10-22 13:45 ` [PATCH 4/4] device_cgroup: add proper checking when changing default behavior Aristeu Rozanski
2012-10-22 16:16   ` Serge Hallyn
2012-10-22 16:16     ` Serge Hallyn
2012-10-22 19:58 ` [PATCH 0/4] Rebase device_cgroup v2 patchset Andrew Morton
2012-10-22 19:58   ` Andrew Morton
2012-10-22 20:14   ` Aristeu Rozanski
2012-10-22 20:14     ` Aristeu Rozanski
2013-05-14 15:05 ` Serge Hallyn
2013-05-14 15:51   ` Aristeu Rozanski
2013-05-14 15:51     ` Aristeu Rozanski
2013-05-14 16:22     ` Serge Hallyn
2013-05-14 16:22       ` Serge Hallyn
2013-05-14 21:02       ` Eric W. Biederman
2013-05-14 21:02         ` Eric W. Biederman
2013-05-16  1:14         ` Serge E. Hallyn
2013-05-16  1:14           ` Serge E. Hallyn
2013-05-16  1:23           ` Serge E. Hallyn
2013-05-16  1:23             ` Serge E. Hallyn

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=508692A8.6090706@suse.cz \
    --to=jslaby@suse.cz \
    --cc=akpm@linux-foundation.org \
    --cc=aris@redhat.com \
    --cc=cgroups@vger.kernel.org \
    --cc=davej@redhat.com \
    --cc=jmorris@namei.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lizefan@huawei.com \
    --cc=serge.hallyn@canonical.com \
    --cc=tj@kernel.org \
    --cc=xemul@openvz.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.