From: Louis Rilling <Louis.Rilling@kerlabs.com>
To: Joel.Becker@oracle.com
Cc: Louis.Rilling@kerlabs.com, linux-kernel@vger.kernel.org,
ocfs2-devel@oss.oracle.com
Subject: [RFC][PATCH 3/3] configfs: Silence lockdep when destroying default groups
Date: Tue, 20 May 2008 18:33:23 +0200 [thread overview]
Message-ID: <20080520164318.941457704@kerlabs.com> (raw)
In-Reply-To: 20080520163320.025971210@kerlabs.com
[-- Attachment #1: configfs-silence-lockdep-with-default-group-destruction.patch --]
[-- Type: text/plain, Size: 3852 bytes --]
When destroying a config_group having default groups, lockdep raises a warning
because multiple locks of class I_MUTEX_NORMAL are taken in
configfs_detach_prep() (the first one being in vfs_rmdir()). However this
recursive locking is another instance of I_MUTEX_PARENT -> I_MUTEX_CHILD
dependency, which was already checked correct when creating the config_group,
and which lockdep cannot handle cleanly because of the variable depth.
This patch silences lockdep when destroying default groups by simply hiding the
inode mutex locks from lockdep in configfs_detach_prep(),
configfs_detach_rollback(), and detach_groups().
=============================================
[ INFO: possible recursive locking detected ]
2.6.26-rc3 #5
---------------------------------------------
rmdir/1499 is trying to acquire lock:
(&sb->s_type->i_mutex_key#11){--..}, at: [<ffffffff802d2529>] configfs_detach_prep+0x54/0x89
but task is already holding lock:
(&sb->s_type->i_mutex_key#11){--..}, at: [<ffffffff80296513>] vfs_rmdir+0x49/0xac
other info that might help us debug this:
2 locks held by rmdir/1499:
#0: (&sb->s_type->i_mutex_key#3/1){--..}, at: [<ffffffff80298138>] do_rmdir+0x82/0x108
#1: (&sb->s_type->i_mutex_key#11){--..}, at: [<ffffffff80296513>] vfs_rmdir+0x49/0xac
stack backtrace:
Pid: 1499, comm: rmdir Not tainted 2.6.26-rc3 #5
Call Trace:
[<ffffffff8024afe9>] __lock_acquire+0x8d2/0xc78
[<ffffffff80266399>] filemap_fault+0x1cf/0x332
[<ffffffff802d2529>] configfs_detach_prep+0x54/0x89
[<ffffffff8024b762>] lock_acquire+0x51/0x6c
[<ffffffff802d2529>] configfs_detach_prep+0x54/0x89
[<ffffffff80248331>] debug_mutex_lock_common+0x16/0x23
[<ffffffff805d6244>] mutex_lock_nested+0xcd/0x23b
[<ffffffff802d2529>] configfs_detach_prep+0x54/0x89
[<ffffffff802d3656>] configfs_rmdir+0xb8/0x1c3
[<ffffffff80296535>] vfs_rmdir+0x6b/0xac
[<ffffffff8029816d>] do_rmdir+0xb7/0x108
[<ffffffff8024a2a2>] trace_hardirqs_on+0xef/0x113
[<ffffffff805d7364>] trace_hardirqs_on_thunk+0x35/0x3a
[<ffffffff8020b0cb>] system_call_after_swapgs+0x7b/0x80
Signed-off-by: Louis Rilling <Louis.Rilling@kerlabs.com>
---
fs/configfs/dir.c | 18 +++++++++++++++++-
1 file changed, 17 insertions(+), 1 deletion(-)
Index: b/fs/configfs/dir.c
===================================================================
--- a/fs/configfs/dir.c 2008-05-20 18:19:41.000000000 +0200
+++ b/fs/configfs/dir.c 2008-05-20 18:19:45.000000000 +0200
@@ -352,7 +352,12 @@ static int configfs_detach_prep(struct d
if (sd->s_type & CONFIGFS_NOT_PINNED)
continue;
if (sd->s_type & CONFIGFS_USET_DEFAULT) {
+ /* Hide this mutex from lockdep for the same reasons as in
+ * populate_groups()
+ */
+ lockdep_off();
mutex_lock(&sd->s_dentry->d_inode->i_mutex);
+ lockdep_on();
/* Mark that we've taken i_mutex */
sd->s_type |= CONFIGFS_USET_DROPPING;
@@ -388,7 +393,12 @@ static void configfs_detach_rollback(str
if (sd->s_type & CONFIGFS_USET_DROPPING) {
sd->s_type &= ~CONFIGFS_USET_DROPPING;
+ /* This mutex was hidden from lockdep in
+ * configfs_detach_prep()
+ */
+ lockdep_off();
mutex_unlock(&sd->s_dentry->d_inode->i_mutex);
+ lockdep_on();
}
}
}
@@ -475,8 +485,14 @@ static void detach_groups(struct config_
* has taken our i_mutex for us. Drop it.
* From mkdir/register cleanup, there is no sem held.
*/
- if (sd->s_type & CONFIGFS_USET_DROPPING)
+ if (sd->s_type & CONFIGFS_USET_DROPPING) {
+ /* This mutex was hidden from lockdep in
+ * configfs_detach_prep()
+ */
+ lockdep_off();
mutex_unlock(&child->d_inode->i_mutex);
+ lockdep_on();
+ }
d_delete(child);
dput(child);
--
Dr Louis Rilling Kerlabs
Skype: louis.rilling Batiment Germanium
Phone: (+33|0) 6 80 89 08 23 80 avenue des Buttes de Coesmes
http://www.kerlabs.com/ 35700 Rennes
next prev parent reply other threads:[~2008-05-20 16:51 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-20 16:33 [RFC][PATCH 0/3] configfs: Make nested default groups lockdep-friendly Louis Rilling
2008-05-20 16:33 ` [RFC][PATCH 1/3] configfs: set CONFIGFS_USET_DEFAULT earlier in configfs_attach_group() Louis Rilling
2008-05-20 16:33 ` [RFC][PATCH 2/3] configfs: Silence lockdep when creating nested default groups Louis Rilling
2008-05-20 16:33 ` Louis Rilling [this message]
2008-05-20 16:58 ` [RFC][PATCH 0/3] configfs: Make nested default groups lockdep-friendly Arjan van de Ven
2008-05-20 17:08 ` Louis Rilling
2008-05-20 21:56 ` Joel Becker
2008-05-20 22:13 ` Arjan van de Ven
2008-05-20 22:27 ` Joel Becker
2008-05-20 22:35 ` Arjan van de Ven
2008-05-20 23:51 ` Joel Becker
2008-05-21 9:20 ` Peter Zijlstra
2008-05-21 9:23 ` Peter Zijlstra
2008-05-21 10:25 ` Louis Rilling
2008-05-21 10:59 ` Peter Zijlstra
2008-05-21 12:54 ` Louis Rilling
2008-05-21 22:09 ` Joel Becker
2008-05-21 8:13 ` Louis Rilling
2008-05-20 21:41 ` [Ocfs2-devel] " Joel Becker
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=20080520164318.941457704@kerlabs.com \
--to=louis.rilling@kerlabs.com \
--cc=Joel.Becker@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=ocfs2-devel@oss.oracle.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox