From: Michael Wang <wangyun@linux.vnet.ibm.com>
To: Denys Fedoryshchenko <denys@visp.net.lb>
Cc: tomas.winkler@intel.com, linux-kernel@vger.kernel.org
Subject: Re: Intel management, circular locking warning
Date: Tue, 20 Nov 2012 14:26:44 +0800 [thread overview]
Message-ID: <50AB22A4.2000400@linux.vnet.ibm.com> (raw)
In-Reply-To: <217a860b36a4d202a71a281003af2788@visp.net.lb>
On 11/13/2012 07:03 PM, Denys Fedoryshchenko wrote:
> Hi
>
> Just tried to run latest 3.6.6 32bit kernel on my server farm and got
> circular locking warning:
> Please let me know if you need more information.
Hi, Denys
You can take a try on below patch, it should do some help,
please let me know the result.
Regards,
Michael Wang
diff --git a/drivers/misc/mei/init.c b/drivers/misc/mei/init.c
index 98f1430..d9a74e6 100644
--- a/drivers/misc/mei/init.c
+++ b/drivers/misc/mei/init.c
@@ -120,6 +120,8 @@ static void mei_reset_iamthif_params(struct mei_device *dev)
dev->iamthif_timer = 0;
}
+static struct lock_class_key mei_device_key;
+
/**
* init_mei_device - allocates and initializes the mei device structure
*
@@ -140,6 +142,7 @@ struct mei_device *mei_device_init(struct pci_dev *pdev)
INIT_LIST_HEAD(&dev->wd_cl.link);
INIT_LIST_HEAD(&dev->iamthif_cl.link);
mutex_init(&dev->device_lock);
+ lockdep_set_class(&dev->device_lock, &mei_device_key);
init_waitqueue_head(&dev->wait_recvd_msg);
init_waitqueue_head(&dev->wait_stop_wd);
dev->dev_state = MEI_DEV_INITIALIZING;
>
> [ 4.359176]
> [ 4.359316] ======================================================
> [ 4.359461] [ INFO: possible circular locking dependency detected ]
> [ 4.359612] 3.6.6-build-0063 #21 Not tainted
> [ 4.359763] -------------------------------------------------------
> [ 4.359916] watchdog/1375 is trying to acquire lock:
> [ 4.360060] (&dev->device_lock){+.+.+.}, at: [<f8790075>]
> mei_wd_ops_start+0x2d/0x75 [mei]
> [ 4.360530]
> [ 4.360530] but task is already holding lock:
> [ 4.360770] (&wdd->lock){+.+...}, at: [<c02c9661>]
> watchdog_start+0x19/0x53
> [ 4.361113]
> [ 4.361113] which lock already depends on the new lock.
> [ 4.361113]
> [ 4.361454]
> [ 4.361454] the existing dependency chain (in reverse order) is:
> [ 4.361701]
> [ 4.361701] -> #2 (&wdd->lock){+.+...}:
> [ 4.361841] [<c01608b4>] lock_acquire+0x71/0x85
> [ 4.361845] [<c0360eb8>] __mutex_lock_common+0x44/0x2e2
> [ 4.361848] [<c03611d8>] mutex_lock_nested+0x20/0x22
> [ 4.361850] [<c02c9661>] watchdog_start+0x19/0x53
> [ 4.361851] [<c02c9808>] watchdog_open+0x5c/0xa1
> [ 4.361853] [<c0294000>] misc_open+0xf5/0x14f
> [ 4.361855] [<c01ad941>] chrdev_open+0x106/0x124
> [ 4.361857] [<c01a93fd>] do_dentry_open.clone.16+0x12a/0x1c6
> [ 4.361859] [<c01a94b1>] finish_open+0x18/0x22
> [ 4.361860] [<c01b4d34>] do_last.clone.35+0x6fb/0x865
> [ 4.361862] [<c01b4f37>] path_openat+0x99/0x2c3
> [ 4.361864] [<c01b5380>] do_filp_open+0x26/0x67
> [ 4.361865] [<c01a9efb>] do_sys_open+0x5b/0xe6
> [ 4.361867] [<c01a9fac>] sys_open+0x26/0x2c
> [ 4.361868] [<c0362e31>] syscall_call+0x7/0xb
> [ 4.361870]
> [ 4.361870] -> #1 (misc_mtx){+.+.+.}:
> [ 4.361871] [<c01608b4>] lock_acquire+0x71/0x85
> [ 4.361873] [<c0360eb8>] __mutex_lock_common+0x44/0x2e2
> [ 4.361876] [<c03611d8>] mutex_lock_nested+0x20/0x22
> [ 4.361877] [<c02941aa>] misc_register+0x1f/0xfd
> [ 4.361879] [<c02c9ba7>] watchdog_dev_register+0x22/0xef
> [ 4.361880] [<c02c952f>] watchdog_register_device+0xa0/0x165
> [ 4.361883] [<f87904ff>] mei_watchdog_register+0x13/0x41 [mei]
> [ 4.361885] [<f878d169>]
> mei_interrupt_thread_handler+0x2fd/0x12b8 [mei]
> [ 4.361887] [<c016c553>] irq_thread_fn+0x13/0x25
> [ 4.361888] [<c016c2f8>] irq_thread+0x9e/0x138
> [ 4.361891] [<c0140725>] kthread+0x59/0x5e
> [ 4.361892] [<c0363aba>] kernel_thread_helper+0x6/0xd
> [ 4.361894]
> [ 4.361894] -> #0 (&dev->device_lock){+.+.+.}:
> [ 4.361895] [<c016026e>] __lock_acquire+0x9a3/0xc27
> [ 4.361897] [<c01608b4>] lock_acquire+0x71/0x85
> [ 4.361898] [<c0360eb8>] __mutex_lock_common+0x44/0x2e2
> [ 4.361900] [<c03611d8>] mutex_lock_nested+0x20/0x22
> [ 4.361902] [<f8790075>] mei_wd_ops_start+0x2d/0x75 [mei]
> [ 4.361904] [<c02c967f>] watchdog_start+0x37/0x53
> [ 4.361905] [<c02c9808>] watchdog_open+0x5c/0xa1
> [ 4.361907] [<c0294000>] misc_open+0xf5/0x14f
> [ 4.361908] [<c01ad941>] chrdev_open+0x106/0x124
> [ 4.361909] [<c01a93fd>] do_dentry_open.clone.16+0x12a/0x1c6
> [ 4.361910] [<c01a94b1>] finish_open+0x18/0x22
> [ 4.361912] [<c01b4d34>] do_last.clone.35+0x6fb/0x865
> [ 4.361914] [<c01b4f37>] path_openat+0x99/0x2c3
> [ 4.361915] [<c01b5380>] do_filp_open+0x26/0x67
> [ 4.361916] [<c01a9efb>] do_sys_open+0x5b/0xe6
> [ 4.361918] [<c01a9fac>] sys_open+0x26/0x2c
> [ 4.361919] [<c0362e31>] syscall_call+0x7/0xb
> [ 4.361919]
> [ 4.361919] other info that might help us debug this:
> [ 4.361919]
> [ 4.361921] Chain exists of:
> [ 4.361921] &dev->device_lock --> misc_mtx --> &wdd->lock
> [ 4.361921]
> [ 4.361921] Possible unsafe locking scenario:
> [ 4.361921]
> [ 4.361922] CPU0 CPU1
> [ 4.361922] ---- ----
> [ 4.361923] lock(&wdd->lock);
> [ 4.361924] lock(misc_mtx);
> [ 4.361924] lock(&wdd->lock);
> [ 4.361925] lock(&dev->device_lock);
> [ 4.361926]
> [ 4.361926] *** DEADLOCK ***
> [ 4.361926]
> [ 4.361926] 2 locks held by watchdog/1375:
> [ 4.361929] #0: (misc_mtx){+.+.+.}, at: [<c0293f28>]
> misc_open+0x1d/0x14f
> [ 4.361931] #1: (&wdd->lock){+.+...}, at: [<c02c9661>]
> watchdog_start+0x19/0x53
> [ 4.361932]
> [ 4.361932] stack backtrace:
> [ 4.361933] Pid: 1375, comm: watchdog Not tainted 3.6.6-build-0063 #21
> [ 4.361933] Call Trace:
> [ 4.361936] [<c015ec57>] print_circular_bug+0x1ac/0x1b6
> [ 4.361937] [<c016026e>] __lock_acquire+0x9a3/0xc27
> [ 4.361940] [<c015f736>] ? mark_lock+0x26/0x1bb
> [ 4.361941] [<c01608b4>] lock_acquire+0x71/0x85
> [ 4.361943] [<f8790075>] ? mei_wd_ops_start+0x2d/0x75 [mei]
> [ 4.361945] [<c0360eb8>] __mutex_lock_common+0x44/0x2e2
> [ 4.361947] [<f8790075>] ? mei_wd_ops_start+0x2d/0x75 [mei]
> [ 4.361949] [<c036114c>] ? __mutex_lock_common+0x2d8/0x2e2
> [ 4.361951] [<c0160c8d>] ? trace_hardirqs_on_caller+0x10e/0x13f
> [ 4.361953] [<c03611d8>] mutex_lock_nested+0x20/0x22
> [ 4.361955] [<f8790075>] ? mei_wd_ops_start+0x2d/0x75 [mei]
> [ 4.361957] [<f8790075>] mei_wd_ops_start+0x2d/0x75 [mei]
> [ 4.361959] [<c02c967f>] watchdog_start+0x37/0x53
> [ 4.361960] [<c02c9808>] watchdog_open+0x5c/0xa1
> [ 4.361962] [<c0294000>] misc_open+0xf5/0x14f
> [ 4.361963] [<c01ad941>] chrdev_open+0x106/0x124
> [ 4.361964] [<c01ad83b>] ? cdev_put+0x1a/0x1a
> [ 4.361966] [<c01a93fd>] do_dentry_open.clone.16+0x12a/0x1c6
> [ 4.361967] [<c01a94b1>] finish_open+0x18/0x22
> [ 4.361969] [<c01b4d34>] do_last.clone.35+0x6fb/0x865
> [ 4.361970] [<c01b2f73>] ? inode_permission+0x3f/0x41
> [ 4.361972] [<c01b4f37>] path_openat+0x99/0x2c3
> [ 4.361974] [<c01b5380>] do_filp_open+0x26/0x67
> [ 4.361977] [<c01be06c>] ? alloc_fd+0xb7/0xc2
> [ 4.361979] [<c01a9efb>] do_sys_open+0x5b/0xe6
> [ 4.361980] [<c01a9fac>] sys_open+0x26/0x2c
> [ 4.361981] [<c0362e31>] syscall_call+0x7/0xb
>
>
>
> ---
> Denys Fedoryshchenko, Network Engineer, Virtual ISP S.A.L.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
prev parent reply other threads:[~2012-11-20 6:26 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-13 11:03 Intel management, circular locking warning Denys Fedoryshchenko
2012-11-20 6:26 ` Michael Wang [this message]
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=50AB22A4.2000400@linux.vnet.ibm.com \
--to=wangyun@linux.vnet.ibm.com \
--cc=denys@visp.net.lb \
--cc=linux-kernel@vger.kernel.org \
--cc=tomas.winkler@intel.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 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.