Hi,

The hci_sysfs uses work queue to finish the sysfs add/del fuction.

But when the same device connection failed, if another connection of

same device come in before the delete work finish, sysfs will warn

about duplicate filename creating.
Sep 19 12:30:27 darkstar kernel: sysfs: duplicate filename

'acl00194FDB6C71' can not be created

Sep 19 12:30:27 darkstar kernel: WARNING: at fs/sysfs/dir.c:433 sysfs_add_one()

Sep 19 12:30:27 darkstar kernel:  [<c01c3ad0>] sysfs_add_one+0xa0/0xe0

Sep 19 12:30:27 darkstar kernel:  [<c01c3e38>] create_dir+0x48/0xb0

Sep 19 12:30:27 darkstar kernel:  [<c0178df5>] __slab_alloc+0x205/0x250

Sep 19 12:30:27 darkstar kernel:  [<c01c3ee9>] sysfs_create_dir+0x29/0x50

Sep 19 12:30:27 darkstar kernel:  [<c024cf5b>] create_dir+0x1b/0x50

Sep 19 12:30:27 darkstar kernel:  [<c024d226>] kobject_add+0x46/0x150

Sep 19 12:30:27 darkstar kernel:  [<c024d404>] kobject_set_name+0x84/0xf0

Sep 19 12:30:27 darkstar kernel:  [<c02b7035>] device_add+0x95/0x350

Sep 19 12:30:27 darkstar kernel:  [<f88aec60>] add_conn+0x0/0x90 [bluetooth]

Sep 19 12:30:27 darkstar kernel:  [<f88aec6f>] add_conn+0xf/0x90 [bluetooth]

Sep 19 12:30:27 darkstar kernel:  [<c01663d0>] vmstat_update+0x0/0x30

Sep 19 12:30:27 darkstar kernel:  [<c013bdfe>] run_workqueue+0x5e/0x110

Sep 19 12:30:27 darkstar kernel:  [<c013bf5c>] worker_thread+0xac/0x100

Sep 19 12:30:27 darkstar kernel:  [<c0140340>] autoremove_wake_function+0x0/0x50

Sep 19 12:30:27 darkstar kernel:  [<c0140340>] autoremove_wake_function+0x0/0x50

Sep 19 12:30:27 darkstar kernel:  [<c013beb0>] worker_thread+0x0/0x100

Sep 19 12:30:27 darkstar kernel:  [<c013fd49>] kthread+0x59/0xa0

Sep 19 12:30:27 darkstar kernel:  [<c013fcf0>] kthread+0x0/0xa0

Sep 19 12:30:27 darkstar kernel:  [<c0104fb3>] kernel_thread_helper+0x7/0x14

Sep 19 12:30:27 darkstar kernel:  =======================

Sep 19 12:30:27 darkstar kernel: kobject_add failed for...
[ message continues ]

which kernel are we talking about. This problem should not exist

anymore. However the usage of a work queue is unavoidable.
Regards
Marcel
-

Recently, I encounted this issue again:

Oct 26 16:57:00 darkstar kernel: sysfs: duplicate filename

'acl00194FA4E517' can not be created

Oct 26 16:57:00 darkstar kernel: WARNING: at fs/sysfs/dir.c:424 sysfs_add_one()

Oct 26 16:57:00 darkstar kernel:  [<c01c0120>] sysfs_add_one+0xa0/0xe0

Oct 26 16:57:00 darkstar kernel:  [<c01c0d7b>] sysfs_create_link+0x9b/0x140

Oct 26 16:57:00 darkstar kernel:  [<c01c1671>] create_files+0x31/0x60

Oct 26 16:57:00 darkstar kernel:  [<c02b537b>] bus_add_device+0x5b/0xf0

Oct 26 16:57:00 darkstar kernel:  [<c02b391c>] device_add+0x11c/0x350

Oct 26 16:57:00 darkstar kernel:  [<f8879410>] add_conn+0x0/0x90 [bluetooth]

Oct 26 16:57:00 darkstar kernel:  [<f887941f>] add_conn+0xf/0x90 [bluetooth]

Oct 26 16:57:00 darkstar kernel:  [<c013baee>] run_workqueue+0x5e/0x110

Oct 26 16:57:00 darkstar kernel:  [<c013bc4c>] worker_thread+0xac/0x100

Oct 26 16:57:00 darkstar kernel:  [<c0140000>] autoremove_wake_function+0x0/0x50

Oct 26 16:57:00 darkstar kernel:  [<c0140000>] autoremove_wake_function+0x0/0x50

Oct 26 16:57:00 darkstar kernel:  [<c013bba0>] worker_thread+0x0/0x100

Oct 26 16:57:00 darkstar kernel:  [<c013fa39>] kthread+0x59/0xa0

Oct 26 16:57:00 darkstar kernel:  [<c013f9e0>] kthread+0x0/0xa0

Oct 26 16:57:00 darkstar kernel:  [<c0104f83>] kernel_thread_helper+0x7/0x14

Oct 26 16:57:00 darkstar kernel:  =======================

Oct 26 16:57:00 darkstar kernel: add_conn: Failed to register connection device
Seems same problem as before. This time, I just used  several rfcomm

connections.

I think if we can add an atomic variable 'going_away' like this:

in hci_conn_del_sysfs:

before INIT_WORK;

atomic_set(&going_away, 1)
in del_conn:

after deleting completed

atomic_set(&going_away, 0)
in add_conn:

busy waiting the going_away

while (atomic_read(&going_away))

        cpu_relax()
Regards

-