From: Ming Lei <tom.leiming@gmail.com>
To: Alan Stern <stern@rowland.harvard.edu>,
Greg Kroah-Hartman <gregkh@suse.de>,
Jiri Kosina <jkosina@suse.cz>
Cc: linux-usb@vger.kernel.org, linux-input@vger.kernel.org,
Ming Lei <tom.leiming@gmail.com>,
Ming Lei <ming.lei@canonical.com>
Subject: [PATCH] HID: usbhid: fix dead lock between open and disconect
Date: Thu, 12 Jan 2012 17:42:22 +0800 [thread overview]
Message-ID: <1326361342-4141-1-git-send-email-tom.leiming@gmail.com> (raw)
There is no reason to hold hiddev->existancelock before
calling usb_deregister_dev, so move it out of the lock.
The patch fixes the lockdep warning below [1].
[1], lockdep warning
[ 5733.386271] ======================================================
[ 5733.386274] [ INFO: possible circular locking dependency detected ]
[ 5733.386278] 3.2.0-custom-next-20120111+ #1 Not tainted
[ 5733.386281] -------------------------------------------------------
[ 5733.386284] khubd/186 is trying to acquire lock:
[ 5733.386288] (minor_rwsem){++++.+}, at: [<ffffffffa0011a04>] usb_deregister_dev+0x37/0x9e [usbcore]
[ 5733.386311]
[ 5733.386312] but task is already holding lock:
[ 5733.386315] (&hiddev->existancelock){+.+...}, at: [<ffffffffa0094d17>] hiddev_disconnect+0x26/0x87 [usbhid]
[ 5733.386328]
[ 5733.386329] which lock already depends on the new lock.
[ 5733.386330]
[ 5733.386333]
[ 5733.386334] the existing dependency chain (in reverse order) is:
[ 5733.386336]
[ 5733.386337] -> #1 (&hiddev->existancelock){+.+...}:
[ 5733.386346] [<ffffffff81082d26>] lock_acquire+0xcb/0x10e
[ 5733.386357] [<ffffffff813df961>] __mutex_lock_common+0x60/0x465
[ 5733.386366] [<ffffffff813dfe4d>] mutex_lock_nested+0x36/0x3b
[ 5733.386371] [<ffffffffa0094ad6>] hiddev_open+0x113/0x193 [usbhid]
[ 5733.386378] [<ffffffffa0011971>] usb_open+0x66/0xc2 [usbcore]
[ 5733.386390] [<ffffffff8111a8b5>] chrdev_open+0x12b/0x154
[ 5733.386402] [<ffffffff811159a8>] __dentry_open.isra.16+0x20b/0x355
[ 5733.386408] [<ffffffff811165dc>] nameidata_to_filp+0x43/0x4a
[ 5733.386413] [<ffffffff81122ed5>] do_last+0x536/0x570
[ 5733.386419] [<ffffffff8112300b>] path_openat+0xce/0x301
[ 5733.386423] [<ffffffff81123327>] do_filp_open+0x33/0x81
[ 5733.386427] [<ffffffff8111664d>] do_sys_open+0x6a/0xfc
[ 5733.386431] [<ffffffff811166fb>] sys_open+0x1c/0x1e
[ 5733.386434] [<ffffffff813e7c79>] system_call_fastpath+0x16/0x1b
[ 5733.386441]
[ 5733.386441] -> #0 (minor_rwsem){++++.+}:
[ 5733.386448] [<ffffffff8108255d>] __lock_acquire+0xa80/0xd74
[ 5733.386454] [<ffffffff81082d26>] lock_acquire+0xcb/0x10e
[ 5733.386458] [<ffffffff813e01f5>] down_write+0x44/0x77
[ 5733.386464] [<ffffffffa0011a04>] usb_deregister_dev+0x37/0x9e [usbcore]
[ 5733.386475] [<ffffffffa0094d2d>] hiddev_disconnect+0x3c/0x87 [usbhid]
[ 5733.386483] [<ffffffff8132df51>] hid_disconnect+0x3f/0x54
[ 5733.386491] [<ffffffff8132dfb4>] hid_device_remove+0x4e/0x7a
[ 5733.386496] [<ffffffff812c0957>] __device_release_driver+0x81/0xcd
[ 5733.386502] [<ffffffff812c09c3>] device_release_driver+0x20/0x2d
[ 5733.386507] [<ffffffff812c0564>] bus_remove_device+0x114/0x128
[ 5733.386512] [<ffffffff812bdd6f>] device_del+0x131/0x183
[ 5733.386519] [<ffffffff8132def3>] hid_destroy_device+0x1e/0x3d
[ 5733.386525] [<ffffffffa00916b0>] usbhid_disconnect+0x36/0x42 [usbhid]
[ 5733.386530] [<ffffffffa000fb60>] usb_unbind_interface+0x57/0x11f [usbcore]
[ 5733.386542] [<ffffffff812c0957>] __device_release_driver+0x81/0xcd
[ 5733.386547] [<ffffffff812c09c3>] device_release_driver+0x20/0x2d
[ 5733.386552] [<ffffffff812c0564>] bus_remove_device+0x114/0x128
[ 5733.386557] [<ffffffff812bdd6f>] device_del+0x131/0x183
[ 5733.386562] [<ffffffffa000de61>] usb_disable_device+0xa8/0x1d8 [usbcore]
[ 5733.386573] [<ffffffffa0006bd2>] usb_disconnect+0xab/0x11f [usbcore]
[ 5733.386583] [<ffffffffa0008aa0>] hub_thread+0x73b/0x1157 [usbcore]
[ 5733.386593] [<ffffffff8105dc0f>] kthread+0x95/0x9d
[ 5733.386601] [<ffffffff813e90b4>] kernel_thread_helper+0x4/0x10
[ 5733.386607]
[ 5733.386608] other info that might help us debug this:
[ 5733.386609]
[ 5733.386612] Possible unsafe locking scenario:
[ 5733.386613]
[ 5733.386615] CPU0 CPU1
[ 5733.386618] ---- ----
[ 5733.386620] lock(&hiddev->existancelock);
[ 5733.386625] lock(minor_rwsem);
[ 5733.386630] lock(&hiddev->existancelock);
[ 5733.386635] lock(minor_rwsem);
[ 5733.386639]
[ 5733.386640] *** DEADLOCK ***
[ 5733.386641]
[ 5733.386644] 6 locks held by khubd/186:
[ 5733.386646] #0: (&__lockdep_no_validate__){......}, at: [<ffffffffa00084af>] hub_thread+0x14a/0x1157 [usbcore]
[ 5733.386661] #1: (&__lockdep_no_validate__){......}, at: [<ffffffffa0006b77>] usb_disconnect+0x50/0x11f [usbcore]
[ 5733.386677] #2: (hcd->bandwidth_mutex){+.+.+.}, at: [<ffffffffa0006bc8>] usb_disconnect+0xa1/0x11f [usbcore]
[ 5733.386693] #3: (&__lockdep_no_validate__){......}, at: [<ffffffff812c09bb>] device_release_driver+0x18/0x2d
[ 5733.386704] #4: (&__lockdep_no_validate__){......}, at: [<ffffffff812c09bb>] device_release_driver+0x18/0x2d
[ 5733.386714] #5: (&hiddev->existancelock){+.+...}, at: [<ffffffffa0094d17>] hiddev_disconnect+0x26/0x87 [usbhid]
[ 5733.386727]
[ 5733.386727] stack backtrace:
[ 5733.386731] Pid: 186, comm: khubd Not tainted 3.2.0-custom-next-20120111+ #1
[ 5733.386734] Call Trace:
[ 5733.386741] [<ffffffff81062881>] ? up+0x34/0x3b
[ 5733.386747] [<ffffffff813d9ef3>] print_circular_bug+0x1f8/0x209
[ 5733.386752] [<ffffffff8108255d>] __lock_acquire+0xa80/0xd74
[ 5733.386756] [<ffffffff810808b4>] ? trace_hardirqs_on_caller+0x15d/0x1a3
[ 5733.386763] [<ffffffff81043a3f>] ? vprintk+0x3f4/0x419
[ 5733.386774] [<ffffffffa0011a04>] ? usb_deregister_dev+0x37/0x9e [usbcore]
[ 5733.386779] [<ffffffff81082d26>] lock_acquire+0xcb/0x10e
[ 5733.386789] [<ffffffffa0011a04>] ? usb_deregister_dev+0x37/0x9e [usbcore]
[ 5733.386797] [<ffffffff813e01f5>] down_write+0x44/0x77
[ 5733.386807] [<ffffffffa0011a04>] ? usb_deregister_dev+0x37/0x9e [usbcore]
[ 5733.386818] [<ffffffffa0011a04>] usb_deregister_dev+0x37/0x9e [usbcore]
[ 5733.386825] [<ffffffffa0094d2d>] hiddev_disconnect+0x3c/0x87 [usbhid]
[ 5733.386830] [<ffffffff8132df51>] hid_disconnect+0x3f/0x54
[ 5733.386834] [<ffffffff8132dfb4>] hid_device_remove+0x4e/0x7a
[ 5733.386839] [<ffffffff812c0957>] __device_release_driver+0x81/0xcd
[ 5733.386844] [<ffffffff812c09c3>] device_release_driver+0x20/0x2d
[ 5733.386848] [<ffffffff812c0564>] bus_remove_device+0x114/0x128
[ 5733.386854] [<ffffffff812bdd6f>] device_del+0x131/0x183
[ 5733.386859] [<ffffffff8132def3>] hid_destroy_device+0x1e/0x3d
[ 5733.386865] [<ffffffffa00916b0>] usbhid_disconnect+0x36/0x42 [usbhid]
[ 5733.386876] [<ffffffffa000fb60>] usb_unbind_interface+0x57/0x11f [usbcore]
[ 5733.386882] [<ffffffff812c0957>] __device_release_driver+0x81/0xcd
[ 5733.386886] [<ffffffff812c09c3>] device_release_driver+0x20/0x2d
[ 5733.386890] [<ffffffff812c0564>] bus_remove_device+0x114/0x128
[ 5733.386895] [<ffffffff812bdd6f>] device_del+0x131/0x183
[ 5733.386905] [<ffffffffa000de61>] usb_disable_device+0xa8/0x1d8 [usbcore]
[ 5733.386916] [<ffffffffa0006bd2>] usb_disconnect+0xab/0x11f [usbcore]
[ 5733.386921] [<ffffffff813dff82>] ? __mutex_unlock_slowpath+0x130/0x141
[ 5733.386929] [<ffffffffa0008aa0>] hub_thread+0x73b/0x1157 [usbcore]
[ 5733.386935] [<ffffffff8106a51d>] ? finish_task_switch+0x78/0x150
[ 5733.386941] [<ffffffff8105e396>] ? __init_waitqueue_head+0x4c/0x4c
[ 5733.386950] [<ffffffffa0008365>] ? usb_remote_wakeup+0x56/0x56 [usbcore]
[ 5733.386955] [<ffffffff8105dc0f>] kthread+0x95/0x9d
[ 5733.386961] [<ffffffff813e90b4>] kernel_thread_helper+0x4/0x10
[ 5733.386966] [<ffffffff813e24b8>] ? retint_restore_args+0x13/0x13
[ 5733.386970] [<ffffffff8105db7a>] ? __init_kthread_worker+0x55/0x55
[ 5733.386974] [<ffffffff813e90b0>] ? gs_change+0x13/0x13
Signed-off-by: Ming Lei <ming.lei@canonical.com>
---
drivers/hid/usbhid/hiddev.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/hid/usbhid/hiddev.c b/drivers/hid/usbhid/hiddev.c
index 7c297d3..b1ec0e2 100644
--- a/drivers/hid/usbhid/hiddev.c
+++ b/drivers/hid/usbhid/hiddev.c
@@ -922,11 +922,11 @@ void hiddev_disconnect(struct hid_device *hid)
struct hiddev *hiddev = hid->hiddev;
struct usbhid_device *usbhid = hid->driver_data;
+ usb_deregister_dev(usbhid->intf, &hiddev_class);
+
mutex_lock(&hiddev->existancelock);
hiddev->exist = 0;
- usb_deregister_dev(usbhid->intf, &hiddev_class);
-
if (hiddev->open) {
mutex_unlock(&hiddev->existancelock);
usbhid_close(hiddev->hid);
--
1.7.5.4
next reply other threads:[~2012-01-12 9:42 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-12 9:42 Ming Lei [this message]
2012-01-12 9:54 ` [PATCH] HID: usbhid: fix dead lock between open and disconect Oliver Neukum
2012-01-12 10:33 ` Ming Lei
[not found] ` <CACVXFVP5V88_cTcpDYnsTy-M9aCaZ5Yneo=yVfB4+2_YjjWFFQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2012-02-02 4:05 ` Ming Lei
[not found] ` <1326361342-4141-1-git-send-email-tom.leiming-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2012-02-02 8:50 ` Jiri Kosina
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=1326361342-4141-1-git-send-email-tom.leiming@gmail.com \
--to=tom.leiming@gmail.com \
--cc=gregkh@suse.de \
--cc=jkosina@suse.cz \
--cc=linux-input@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=ming.lei@canonical.com \
--cc=stern@rowland.harvard.edu \
/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;
as well as URLs for NNTP newsgroup(s).