Re: [PATCH v2 2/7] btrfs: do not take the uuid_mutex in btrfs_rm_device

[Anand Jain <anand.jain@oracle.com>]

On 02/09/2021 01:10, Josef Bacik wrote:
> On 9/1/21 8:01 AM, Anand Jain wrote:
>> On 28/07/2021 05:01, Josef Bacik wrote:
>>> We got the following lockdep splat while running xfstests (specifically
>>> btrfs/003 and btrfs/020 in a row) with the new rc.  This was uncovered
>>> by 87579e9b7d8d ("loop: use worker per cgroup instead of kworker") which
>>> converted loop to using workqueues, which comes with lockdep
>>> annotations that don't exist with kworkers.  The lockdep splat is as
>>> follows
>>>
>>> ======================================================
>>> WARNING: possible circular locking dependency detected
>>> 5.14.0-rc2-custom+ #34 Not tainted
>>> ------------------------------------------------------
>>> losetup/156417 is trying to acquire lock:
>>> ffff9c7645b02d38 ((wq_completion)loop0){+.+.}-{0:0}, at: 
>>> flush_workqueue+0x84/0x600
>>>
>>> but task is already holding lock:
>>> ffff9c7647395468 (&lo->lo_mutex){+.+.}-{3:3}, at: 
>>> __loop_clr_fd+0x41/0x650 [loop]
>>>
>>> which lock already depends on the new lock.
>>>
>>> the existing dependency chain (in reverse order) is:
>>>
>>> -> #5 (&lo->lo_mutex){+.+.}-{3:3}:
>>>         __mutex_lock+0xba/0x7c0
>>>         lo_open+0x28/0x60 [loop]
>>>         blkdev_get_whole+0x28/0xf0
>>>         blkdev_get_by_dev.part.0+0x168/0x3c0
>>>         blkdev_open+0xd2/0xe0
>>>         do_dentry_open+0x163/0x3a0
>>>         path_openat+0x74d/0xa40
>>>         do_filp_open+0x9c/0x140
>>>         do_sys_openat2+0xb1/0x170
>>>         __x64_sys_openat+0x54/0x90
>>>         do_syscall_64+0x3b/0x90
>>>         entry_SYSCALL_64_after_hwframe+0x44/0xae
>>>
>>> -> #4 (&disk->open_mutex){+.+.}-{3:3}:
>>>         __mutex_lock+0xba/0x7c0
>>>         blkdev_get_by_dev.part.0+0xd1/0x3c0
>>>         blkdev_get_by_path+0xc0/0xd0
>>>         btrfs_scan_one_device+0x52/0x1f0 [btrfs]
>>>         btrfs_control_ioctl+0xac/0x170 [btrfs]
>>>         __x64_sys_ioctl+0x83/0xb0
>>>         do_syscall_64+0x3b/0x90
>>>         entry_SYSCALL_64_after_hwframe+0x44/0xae
>>>
>>> -> #3 (uuid_mutex){+.+.}-{3:3}:
>>>         __mutex_lock+0xba/0x7c0
>>>         btrfs_rm_device+0x48/0x6a0 [btrfs]
>>>         btrfs_ioctl+0x2d1c/0x3110 [btrfs]
>>>         __x64_sys_ioctl+0x83/0xb0
>>>         do_syscall_64+0x3b/0x90
>>>         entry_SYSCALL_64_after_hwframe+0x44/0xae
>>>
>>> -> #2 (sb_writers#11){.+.+}-{0:0}:
>>>         lo_write_bvec+0x112/0x290 [loop]
>>>         loop_process_work+0x25f/0xcb0 [loop]
>>>         process_one_work+0x28f/0x5d0
>>>         worker_thread+0x55/0x3c0
>>>         kthread+0x140/0x170
>>>         ret_from_fork+0x22/0x30
>>>
>>> -> #1 ((work_completion)(&lo->rootcg_work)){+.+.}-{0:0}:
>>>         process_one_work+0x266/0x5d0
>>>         worker_thread+0x55/0x3c0
>>>         kthread+0x140/0x170
>>>         ret_from_fork+0x22/0x30
>>>
>>> -> #0 ((wq_completion)loop0){+.+.}-{0:0}:
>>>         __lock_acquire+0x1130/0x1dc0
>>>         lock_acquire+0xf5/0x320
>>>         flush_workqueue+0xae/0x600
>>>         drain_workqueue+0xa0/0x110
>>>         destroy_workqueue+0x36/0x250
>>>         __loop_clr_fd+0x9a/0x650 [loop]
>>>         lo_ioctl+0x29d/0x780 [loop]
>>>         block_ioctl+0x3f/0x50
>>>         __x64_sys_ioctl+0x83/0xb0
>>>         do_syscall_64+0x3b/0x90
>>>         entry_SYSCALL_64_after_hwframe+0x44/0xae
>>>
>>> other info that might help us debug this:
>>> Chain exists of:
>>>    (wq_completion)loop0 --> &disk->open_mutex --> &lo->lo_mutex
>>>   Possible unsafe locking scenario:
>>>         CPU0                    CPU1
>>>         ----                    ----
>>>    lock(&lo->lo_mutex);
>>>                                 lock(&disk->open_mutex);
>>>                                 lock(&lo->lo_mutex);
>>>    lock((wq_completion)loop0);
>>>
>>>   *** DEADLOCK ***
>>> 1 lock held by losetup/156417:
>>>   #0: ffff9c7647395468 (&lo->lo_mutex){+.+.}-{3:3}, at: 
>>> __loop_clr_fd+0x41/0x650 [loop]
>>>
>>> stack backtrace:
>>> CPU: 8 PID: 156417 Comm: losetup Not tainted 5.14.0-rc2-custom+ #34
>>> Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 
>>> 02/06/2015
>>> Call Trace:
>>>   dump_stack_lvl+0x57/0x72
>>>   check_noncircular+0x10a/0x120
>>>   __lock_acquire+0x1130/0x1dc0
>>>   lock_acquire+0xf5/0x320
>>>   ? flush_workqueue+0x84/0x600
>>>   flush_workqueue+0xae/0x600
>>>   ? flush_workqueue+0x84/0x600
>>>   drain_workqueue+0xa0/0x110
>>>   destroy_workqueue+0x36/0x250
>>>   __loop_clr_fd+0x9a/0x650 [loop]
>>>   lo_ioctl+0x29d/0x780 [loop]
>>>   ? __lock_acquire+0x3a0/0x1dc0
>>>   ? update_dl_rq_load_avg+0x152/0x360
>>>   ? lock_is_held_type+0xa5/0x120
>>>   ? find_held_lock.constprop.0+0x2b/0x80
>>>   block_ioctl+0x3f/0x50
>>>   __x64_sys_ioctl+0x83/0xb0
>>>   do_syscall_64+0x3b/0x90
>>>   entry_SYSCALL_64_after_hwframe+0x44/0xae
>>> RIP: 0033:0x7f645884de6b
>>>
>>> Usually the uuid_mutex exists to protect the fs_devices that map
>>> together all of the devices that match a specific uuid.  In rm_device
>>> we're messing with the uuid of a device, so it makes sense to protect
>>> that here.
>>>
>>> However in doing that it pulls in a whole host of lockdep dependencies,
>>> as we call mnt_may_write() on the sb before we grab the uuid_mutex, thus
>>> we end up with the dependency chain under the uuid_mutex being added
>>> under the normal sb write dependency chain, which causes problems with
>>> loop devices.
>>>
>>> We don't need the uuid mutex here however.  If we call
>>> btrfs_scan_one_device() before we scratch the super block we will find
>>> the fs_devices and not find the device itself and return EBUSY because
>>> the fs_devices is open.  If we call it after the scratch happens it will
>>> not appear to be a valid btrfs file system.
>>>
>>> We do not need to worry about other fs_devices modifying operations here
>>> because we're protected by the exclusive operations locking.
>>>
>>> So drop the uuid_mutex here in order to fix the lockdep splat.
>>
>>
>> I think uuid_mutex should stay. Here is why.
>>
>>   While thread A takes %device at line 816 and deref at line 880.
>>   Thread B can completely remove and free that %device.
>>   As of now these threads are mutual exclusive using uuid_mutex.
>>
>> Thread A
>>
>> btrfs_control_ioctl()
>>    mutex_lock(&uuid_mutex);
>>      btrfs_scan_one_device()
>>        device_list_add()
>>        {
>>   815                 mutex_lock(&fs_devices->device_list_mutex);
>>
>>   816                 device = btrfs_find_device(fs_devices, devid,
>>   817                                 disk_super->dev_item.uuid, NULL);
>>
>>   880         } else if (!device->name || strcmp(device->name->str, 
>> path)) {
>>
>>   933                         if (device->bdev->bd_dev != path_dev) {
>>
>>   982         mutex_unlock(&fs_devices->device_list_mutex);
>>         }
>>
>>
>> Thread B
>>
>> btrfs_rm_device()
>>
>> 2069         mutex_lock(&uuid_mutex);  <-- proposed to remove
>>
>> 2150         mutex_lock(&fs_devices->device_list_mutex);
    2151         list_del_rcu(&device->dev_list);   <----
>>
>> 2172         mutex_unlock(&fs_devices->device_list_mutex);
>>
>> 2180                 btrfs_scratch_superblocks(fs_info, device->bdev,
>> 2181                                           device->name->str);
>>
>> 2183         btrfs_close_bdev(device);
>> 2184         synchronize_rcu();
>> 2185         btrfs_free_device(device);
>>
>> 2194         mutex_unlock(&uuid_mutex);  <-- proposed to remove
>>
>>
> 
> This is fine, we're protected by the fs_devices->device_list_mutex here. 

>   We'll remove our device from the list before dropping the 
> device_list_mutex,

You are right. I missed that point.
Changes looks good to me.

Reviewed-by: Anand Jain <anand.jain@oracle.com>

Thanks.

> so we won't be able to find the old device if we're 
> removing it.  Thanks,





> 
> Josef