Re: [PATCH]: Schedule correct qdisc in watchdog.

[Jarek Poplawski <jarkao2@gmail.com>]

On Mon, Aug 18, 2008 at 03:45:29PM +0300, Denys Fedoryshchenko wrote:
> Patch applied, got another warning.

I hope I'll figure this out before evening. Probably it's safer
to stop testing for a while.

Thanks,
Jarek P.

> 
> [  413.646102]
> [  413.646102] =========================
> [  413.646102] [ BUG: held lock freed! ]
> [  413.646102] -------------------------
> [  413.646102] pppd/4105 is freeing memory f6fe6e00-f6fe6eff, with a lock 
> still held there!
> [  413.646102]  (&qdisc_rx_lock){-+..}, at: [<c026e3ab>] 
> shutdown_scheduler_queue+0x1c/0x2e
> [  413.646102] 3 locks held by pppd/4105:
> [  413.646102]  #0:  (all_ppp_mutex){--..}, at: [<f8a671ae>] 
> ppp_shutdown_interface+0x17/0xb0 [ppp_generic]
> [  413.646102]  #1:  (rtnl_mutex){--..}, at: [<c026967a>] rtnl_lock+0xf/0x11
> [  413.647102]  #2:  (&qdisc_rx_lock){-+..}, at: [<c026e3ab>] 
> shutdown_scheduler_queue+0x1c/0x2e
> [  413.647102]
> [  413.647102] stack backtrace:
> [  413.647102] Pid: 4105, comm: pppd Not tainted 
> 2.6.27-rc3-build-0031-12578-g96d2031-dirty #32
> [  413.647102]  [<c02ba6f7>] ? printk+0xf/0x18
> [  413.647102]  [<c013db8f>] debug_check_no_locks_freed+0xce/0xff
> [  413.647102]  [<c01643ce>] kfree+0x78/0xc8
> [  413.647102]  [<c026e38b>] ? qdisc_destroy+0x8c/0x90
> [  413.647102]  [<c026e38b>] qdisc_destroy+0x8c/0x90
> [  413.647102]  [<c026e3b2>] shutdown_scheduler_queue+0x23/0x2e
> [  413.647102]  [<c026e3fa>] dev_shutdown+0x3d/0x59
> [  413.647102]  [<c0262240>] rollback_registered+0x128/0x1b1
> [  413.647102]  [<c02622fb>] unregister_netdevice+0x32/0x6e
> [  413.647102]  [<c0262349>] unregister_netdev+0x12/0x1a
> [  413.647102]  [<f8a671f3>] ppp_shutdown_interface+0x5c/0xb0 [ppp_generic]
> [  413.647102]  [<f8a6742c>] ppp_release+0x2c/0x5a [ppp_generic]
> [  413.647102]  [<c0167db8>] __fput+0x99/0x136
> [  413.647102]  [<c0167e6c>] fput+0x17/0x19
> [  413.647102]  [<c016593b>] filp_close+0x47/0x51
> [  413.647102]  [<c01659b8>] sys_close+0x73/0xad
> [  413.647102]  [<c0103861>] sysenter_do_call+0x12/0x35
> [  413.647102]  =======================
> 
> On Monday 18 August 2008, Jarek Poplawski wrote: