From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sergey Senozhatsky Subject: Re: [PATCH] Battery: sysfs_remove_battery(): possible circular locking Date: Fri, 5 Aug 2011 10:20:44 +0300 Message-ID: <20110805072044.GB8311@swordfish> References: <20110805003322.GA8311@swordfish> <1312521008.2096.173.camel@lantianyu-ws> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mail-pz0-f42.google.com ([209.85.210.42]:44514 "EHLO mail-pz0-f42.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753845Ab1HEHV4 (ORCPT ); Fri, 5 Aug 2011 03:21:56 -0400 Content-Disposition: inline In-Reply-To: <1312521008.2096.173.camel@lantianyu-ws> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: "lan,Tianyu" Cc: Len Brown , "linux-acpi@vger.kernel.org" , "linux-kernel@vger.kernel.org" On (08/05/11 13:10), lan,Tianyu wrote: > I think changing 'the marker' to 'battery->bat.name' will introduce > problem. > In the sysfs_add_battery(), when the 'battery->bat.name' is assigned, > the power_supply_register() and device_create_file() have not been > invoked. In this time, maybe sysfs_remove_battery() will be invoked and > cause device_remove_file() and power_supply_unregister() invoked without > device file created and power supply registered. > Hm, good point! It was ~3.30AM when I was writing the patch, so I obviously didn't think carefully enought. Here the whole trace (I'll get back to my laptop within 2 hours): [14818.477168] [14818.477170] ======================================================= [14818.477200] [ INFO: possible circular locking dependency detected ] [14818.477221] 3.1.0-dbg-07865-g1280ea8-dirty #668 [14818.477236] ------------------------------------------------------- [14818.477257] s2ram/1599 is trying to acquire lock: [14818.477276] (s_active#8){++++.+}, at: [] sysfs_addrm_finish+0x31/0x5a [14818.477323] [14818.477325] but task is already holding lock: [14818.477350] (&battery->lock){+.+.+.}, at: [] sysfs_remove_battery+0x10/0x4b [battery] [14818.477395] [14818.477397] which lock already depends on the new lock. [14818.477399] [14818.477433] [14818.477435] the existing dependency chain (in reverse order) is: [14818.477461] [14818.477463] -> #1 (&battery->lock){+.+.+.}: [14818.477497] [] lock_acquire+0x138/0x1ac [14818.477527] [] mutex_lock_nested+0x5e/0x325 [14818.477558] [] acpi_battery_get_state+0x6a/0x16f [battery] [14818.477592] [] acpi_battery_get_property+0x2d/0x1cf [battery] [14818.477627] [] power_supply_show_property+0x59/0x147 [14818.477662] [] power_supply_uevent+0x93/0x17f [14818.477692] [] dev_uevent+0x12f/0x155 [14818.477722] [] show_uevent+0xa9/0xf4 [14818.477750] [] dev_attr_show+0x22/0x49 [14818.477779] [] sysfs_read_file+0xb4/0x139 [14818.477809] [] vfs_read+0xaa/0x13a [14818.477840] [] sys_read+0x45/0x6c [14818.477867] [] system_call_fastpath+0x16/0x1b [14818.477899] [14818.477900] -> #0 (s_active#8){++++.+}: [14818.477938] [] __lock_acquire+0xfe4/0x164b [14818.477968] [] lock_acquire+0x138/0x1ac [14818.477997] [] sysfs_deactivate+0x9b/0xec [14818.478026] [] sysfs_addrm_finish+0x31/0x5a [14818.478055] [] sysfs_hash_and_remove+0x54/0x77 [14818.478086] [] sysfs_remove_file+0x12/0x14 [14818.478115] [] device_remove_file+0x12/0x14 [14818.478145] [] device_del+0x119/0x17c [14818.478172] [] device_unregister+0xe/0x1a [14818.478201] [] power_supply_unregister+0x23/0x27 [14818.478233] [] sysfs_remove_battery+0x34/0x4b [battery] [14818.478267] [] battery_notify+0x2c/0x3a [battery] [14818.478299] [] notifier_call_chain+0x74/0xa1 [14818.478330] [] __blocking_notifier_call_chain+0x6c/0x89 [14818.478363] [] blocking_notifier_call_chain+0xf/0x11 [14818.478394] [] pm_notifier_call_chain+0x15/0x27 [14818.478426] [] enter_state+0xa7/0xd5 [14818.478454] [] state_store+0xaa/0xc0 [14818.478482] [] kobj_attr_store+0x17/0x19 [14818.478511] [] sysfs_write_file+0x103/0x13f [14818.478541] [] vfs_write+0xad/0x13d [14818.478568] [] sys_write+0x45/0x6c [14818.478595] [] system_call_fastpath+0x16/0x1b [14818.478626] [14818.478628] other info that might help us debug this: [14818.478630] [14818.478665] Possible unsafe locking scenario: [14818.478668] [14818.478694] CPU0 CPU1 [14818.478713] ---- ---- [14818.478731] lock(&battery->lock); [14818.478754] lock(s_active); [14818.478782] lock(&battery->lock); [14818.478810] lock(s_active); [14818.478831] [14818.478832] *** DEADLOCK *** [14818.478835] [14818.478868] 5 locks held by s2ram/1599: [14818.478885] #0: (&buffer->mutex){+.+.+.}, at: [] sysfs_write_file+0x37/0x13f [14818.478932] #1: (s_active#106){.+.+.+}, at: [] sysfs_write_file+0xe2/0x13f [14818.478980] #2: (pm_mutex){+.+...}, at: [] enter_state+0x2a/0xd5 [14818.479021] #3: ((pm_chain_head).rwsem){++++..}, at: [] __blocking_notifier_call_chain+0x57/0x89 [14818.479070] #4: (&battery->lock){+.+.+.}, at: [] sysfs_remove_battery+0x10/0x4b [battery] [14818.479119] [14818.479121] stack backtrace: [14818.479148] Pid: 1599, comm: s2ram Not tainted 3.1.0-dbg-07865-g1280ea8-dirty #668 [14818.479175] Call Trace: [14818.479198] [] print_circular_bug+0x293/0x2a4 [14818.479228] [] __lock_acquire+0xfe4/0x164b [14818.479260] [] ? sysfs_addrm_finish+0x31/0x5a [14818.479288] [] lock_acquire+0x138/0x1ac [14818.479316] [] ? sysfs_addrm_finish+0x31/0x5a [14818.479345] [] sysfs_deactivate+0x9b/0xec [14818.479373] [] ? sysfs_addrm_finish+0x31/0x5a [14818.479405] [] sysfs_addrm_finish+0x31/0x5a [14818.479433] [] sysfs_hash_and_remove+0x54/0x77 [14818.479461] [] sysfs_remove_file+0x12/0x14 [14818.479488] [] device_remove_file+0x12/0x14 [14818.479516] [] device_del+0x119/0x17c [14818.479542] [] device_unregister+0xe/0x1a [14818.479570] [] power_supply_unregister+0x23/0x27 [14818.479601] [] sysfs_remove_battery+0x34/0x4b [battery] [14818.479632] [] battery_notify+0x2c/0x3a [battery] [14818.479662] [] notifier_call_chain+0x74/0xa1 [14818.479692] [] __blocking_notifier_call_chain+0x6c/0x89 [14818.479722] [] blocking_notifier_call_chain+0xf/0x11 [14818.479751] [] pm_notifier_call_chain+0x15/0x27 [14818.479770] [] enter_state+0xa7/0xd5 [14818.479782] [] state_store+0xaa/0xc0 [14818.479795] [] ? pm_async_store+0x45/0x45 [14818.479807] [] kobj_attr_store+0x17/0x19 [14818.479820] [] sysfs_write_file+0x103/0x13f [14818.479834] [] vfs_write+0xad/0x13d [14818.479847] [] sys_write+0x45/0x6c [14818.479860] [] system_call_fastpath+0x16/0x1b Sergey > sysfs_remove_battery() will be invoked in the battery_notify(), > acpi_battery_refresh() and sysfs_remove_battery() which causes the > situation. This is also the cause of bug 35642. > > > I've changed `the marker' from `battery->bat.dev' to `battery->bat.name', so > > the basic idea should remain the same, now we just can release battery->lock > > more quicker, before device_remove_file() call. > > > > Signed-off-by: Sergey Senozhatsky > > > > --- > > > > drivers/acpi/battery.c | 6 ++++-- > > 1 files changed, 4 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c > > index 87c0a8d..398cbfb 100644 > > --- a/drivers/acpi/battery.c > > +++ b/drivers/acpi/battery.c > > @@ -574,15 +574,17 @@ static int sysfs_add_battery(struct acpi_battery *battery) > > static void sysfs_remove_battery(struct acpi_battery *battery) > > { > > mutex_lock(&battery->lock); > > - if (!battery->bat.dev) { > > + if (!battery->bat.name) { > > mutex_unlock(&battery->lock); > > return; > > } > > > > + battery->bat.name = NULL; > > + mutex_unlock(&battery->lock); > > + > > device_remove_file(battery->bat.dev, &alarm_attr); > > power_supply_unregister(&battery->bat); > > battery->bat.dev = NULL; > > - mutex_unlock(&battery->lock); > > } > > > > /* > > > >