FDTI-SIO: kernel-oops after resume

FDTI-SIO: kernel-oops after resume

[Martin Zwickel]

[-- Attachment #1: Type: text/plain, Size: 3187 bytes --]

Hi all,

I got an oops after resuming from suspend S3 state.

I had the ttyUSB0 opened with minicom and went to suspend S3.
While in suspend I removed the physical serial device converter.
While resuming I got this kernel oops:

[output]
BUG: unable to handle kernel NULL pointer dereference at 00000044
IP: [<c012ba40>] del_timer+0xc/0x4c
*pde = 00000000 
Oops: 0000 [#1] PREEMPT SMP 
last sysfs file: /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq
Modules linked in: iptable_filter ip_tables x_tables i915 fb drm
i2c_algo_bit cfbcopyarea i2c_core cfbimgblt cfbfillrect ipv6
snd_hda_codec_conexant snd_hda_intel ftdi_sio iwlagn usbserial iwlcore
snd_hda_codec pcspkr evdev

Pid: 16129, comm: minicom Not tainted (2.6.29.4 #3) ESPRIMO Mobile V5505           
   
EIP: 0060:[<c012ba40>] EFLAGS: 00210286 CPU: 0
EIP is at del_timer+0xc/0x4c
EAX: 00000044 EBX: 00000044 ECX: f2224b00 EDX: 00000044
ESI: 00000000 EDI: 00000000 EBP: f6489df0 ESP: f6489de4
 DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
Process minicom (pid: 16129, ti=f6488000 task=f1b54000 task.ti=f6488000)
Stack:
 00000000 f1a31c00 00000034 f6489e38 c01311a3 00000044 00200246 f6489e20
 00200246 00000000 c0609e00 00000002 00000000 f6489e44 00200246 f6489e30
 c014e2a6 c1aa7f80 f1a31c00 00000400 00000000 f6489e40 c01312c3 f6489e58
Call Trace:
 [<c01311a3>] ? __cancel_work_timer+0x20/0x135
 [<c014e2a6>] ? __pagevec_free+0x17/0x1e
 [<c01312c3>] ? cancel_delayed_work_sync+0xb/0xd
 [<f8103692>] ? ftdi_close+0xc3/0xe2 [ftdi_sio]
 [<c0175977>] ? fasync_helper+0xb3/0xbc
 [<f820d722>] ? serial_close+0x7b/0x113 [usbserial]
 [<c02a165c>] ? tty_release_dev+0x15b/0x392
 [<c014e2a6>] ? __pagevec_free+0x17/0x1e
 [<c0167c27>] ? add_partial+0x44/0x49
 [<c017d860>] ? mntput_no_expire+0x18/0xe7
 [<c0178b67>] ? dput+0x1d/0xfc
 [<c02a18a5>] ? tty_release+0x12/0x1c
 [<c016d193>] ? __fput+0xaf/0x155
 [<c016d45d>] ? fput+0x17/0x19
 [<c016ae36>] ? filp_close+0x51/0x5b
 [<c0125870>] ? put_files_struct+0x68/0xaa
 [<c01258e9>] ? exit_files+0x37/0x3c
 [<c012681d>] ? do_exit+0x150/0x62e
 [<c0126d5a>] ? do_group_exit+0x5f/0x88
 [<c0126d92>] ? sys_exit_group+0xf/0x13
 [<c0102df1>] ? sysenter_do_call+0x12/0x25
Code: 04 0f 0b eb fe 39 50 08 75 0a 83 38 00 b9 01 00 00 00 75 07 e8 10 fe ff ff 89 c1 5d 89 c8 c3 55 89 e5 56 31 f6 53 89 c3 83 ec 04 <83> 38 00 74 34 8d 55 f4 e8 4c fd ff ff 83 3b 00 89 c1 74 1b 8b 
EIP: [<c012ba40>] del_timer+0xc/0x4c SS:ESP 0068:f6489de4
---[ end trace 679defbdbe4aa1e4 ]---
[/output]

Minicom said it lost the device, but was unable to close it.

The attached device:
[output]
USB Serial support registered for FTDI USB Serial Device
ftdi_sio 6-1:1.0: FTDI USB Serial Device converter detected
usb 6-1: Detected FT8U232AM
usb 6-1: FTDI USB Serial Device converter now attached to ttyUSB0
usbcore: registered new interface driver ftdi_sio
ftdi_sio: v1.4.3:USB FTDI Serial Converters Driver
[/output]


My kernel runs with SMP and HIGHMEM if it matters:
Linux tt-business 2.6.29.4 #3 SMP PREEMPT Fri Jun 5 00:03:39 WEST 2009 i686 Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz GenuineIntel GNU/Linux


Best regards,
   Martin

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re: FDTI-SIO: kernel-oops after resume

[Andrew Morton]

On Tue, 9 Jun 2009 08:39:48 +0200 Martin Zwickel <martin.zwickel@novabasedigitaltv.de> wrote:

> Hi all,
> 
> I got an oops after resuming from suspend S3 state.
> 
> I had the ttyUSB0 opened with minicom and went to suspend S3.
> While in suspend I removed the physical serial device converter.
> While resuming I got this kernel oops:
> 
> [output]
> BUG: unable to handle kernel NULL pointer dereference at 00000044
> IP: [<c012ba40>] del_timer+0xc/0x4c
> *pde = 00000000 
> Oops: 0000 [#1] PREEMPT SMP 
> last sysfs file: /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq
> Modules linked in: iptable_filter ip_tables x_tables i915 fb drm
> i2c_algo_bit cfbcopyarea i2c_core cfbimgblt cfbfillrect ipv6
> snd_hda_codec_conexant snd_hda_intel ftdi_sio iwlagn usbserial iwlcore
> snd_hda_codec pcspkr evdev
> 
> Pid: 16129, comm: minicom Not tainted (2.6.29.4 #3) ESPRIMO Mobile V5505           
>    
> EIP: 0060:[<c012ba40>] EFLAGS: 00210286 CPU: 0
> EIP is at del_timer+0xc/0x4c
> EAX: 00000044 EBX: 00000044 ECX: f2224b00 EDX: 00000044
> ESI: 00000000 EDI: 00000000 EBP: f6489df0 ESP: f6489de4
>  DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
> Process minicom (pid: 16129, ti=f6488000 task=f1b54000 task.ti=f6488000)
> Stack:
>  00000000 f1a31c00 00000034 f6489e38 c01311a3 00000044 00200246 f6489e20
>  00200246 00000000 c0609e00 00000002 00000000 f6489e44 00200246 f6489e30
>  c014e2a6 c1aa7f80 f1a31c00 00000400 00000000 f6489e40 c01312c3 f6489e58
> Call Trace:
>  [<c01311a3>] ? __cancel_work_timer+0x20/0x135
>  [<c014e2a6>] ? __pagevec_free+0x17/0x1e
>  [<c01312c3>] ? cancel_delayed_work_sync+0xb/0xd
>  [<f8103692>] ? ftdi_close+0xc3/0xe2 [ftdi_sio]
>  [<c0175977>] ? fasync_helper+0xb3/0xbc
>  [<f820d722>] ? serial_close+0x7b/0x113 [usbserial]
>  [<c02a165c>] ? tty_release_dev+0x15b/0x392
>  [<c014e2a6>] ? __pagevec_free+0x17/0x1e
>  [<c0167c27>] ? add_partial+0x44/0x49
>  [<c017d860>] ? mntput_no_expire+0x18/0xe7
>  [<c0178b67>] ? dput+0x1d/0xfc
>  [<c02a18a5>] ? tty_release+0x12/0x1c
>  [<c016d193>] ? __fput+0xaf/0x155
>  [<c016d45d>] ? fput+0x17/0x19
>  [<c016ae36>] ? filp_close+0x51/0x5b
>  [<c0125870>] ? put_files_struct+0x68/0xaa
>  [<c01258e9>] ? exit_files+0x37/0x3c
>  [<c012681d>] ? do_exit+0x150/0x62e
>  [<c0126d5a>] ? do_group_exit+0x5f/0x88
>  [<c0126d92>] ? sys_exit_group+0xf/0x13
>  [<c0102df1>] ? sysenter_do_call+0x12/0x25
> Code: 04 0f 0b eb fe 39 50 08 75 0a 83 38 00 b9 01 00 00 00 75 07 e8 10 fe ff ff 89 c1 5d 89 c8 c3 55 89 e5 56 31 f6 53 89 c3 83 ec 04 <83> 38 00 74 34 8d 55 f4 e8 4c fd ff ff 83 3b 00 89 c1 74 1b 8b 
> EIP: [<c012ba40>] del_timer+0xc/0x4c SS:ESP 0068:f6489de4
> ---[ end trace 679defbdbe4aa1e4 ]---
> [/output]
> 
> Minicom said it lost the device, but was unable to close it.
> 
> The attached device:
> [output]
> USB Serial support registered for FTDI USB Serial Device
> ftdi_sio 6-1:1.0: FTDI USB Serial Device converter detected
> usb 6-1: Detected FT8U232AM
> usb 6-1: FTDI USB Serial Device converter now attached to ttyUSB0
> usbcore: registered new interface driver ftdi_sio
> ftdi_sio: v1.4.3:USB FTDI Serial Converters Driver
> [/output]
> 
> 
> My kernel runs with SMP and HIGHMEM if it matters:
> Linux tt-business 2.6.29.4 #3 SMP PREEMPT Fri Jun 5 00:03:39 WEST 2009 i686 Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz GenuineIntel GNU/Linux
> 

The oops is very very similar to the one which Daniel Mack reported,
and which was fixed by

commit c16199fcd341f10147eb2e29a60aed41ceca55b4
Author: Alan Stern <stern@rowland.harvard.edu>
Date:   Thu Apr 30 10:06:19 2009 -0400

    usb-serial: ftdi_sio: fix reference counting of ftdi_private

but your 2.6.29.4 contained that fix, so it seems that we still have a
problem here?

Re: FDTI-SIO: kernel-oops after resume

[Martin Zwickel]

[-- Attachment #1: Type: text/plain, Size: 1018 bytes --]

On Wed, Jun 10, 2009 at 08:50:18AM +0100, <Andrew Morton> bubbled:
> On Tue, 9 Jun 2009 08:39:48 +0200 Martin Zwickel <martin.zwickel@novabasedigitaltv.de> wrote:
> 
> > My kernel runs with SMP and HIGHMEM if it matters:
> > Linux tt-business 2.6.29.4 #3 SMP PREEMPT Fri Jun 5 00:03:39 WEST 2009 i686 Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz GenuineIntel GNU/Linux
> > 
> 
> The oops is very very similar to the one which Daniel Mack reported,
> and which was fixed by
> 
> commit c16199fcd341f10147eb2e29a60aed41ceca55b4
> Author: Alan Stern <stern@rowland.harvard.edu>
> Date:   Thu Apr 30 10:06:19 2009 -0400
> 
>     usb-serial: ftdi_sio: fix reference counting of ftdi_private
> 
> but your 2.6.29.4 contained that fix, so it seems that we still have a
> problem here?

Could be, or I'm doing something wrong.

I will clean the kernel, rebuild 2.6.29.4, reboot, try to crash it and
let you know.
If I should try some patches, please let me know.

Best regards and thanks,
     Martin

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re: FDTI-SIO: kernel-oops after resume

[Martin Zwickel]

[-- Attachment #1: Type: text/plain, Size: 1535 bytes --]

On Wed, Jun 10, 2009 at 09:24:17AM +0100, <Martin Zwickel> bubbled:
> On Wed, Jun 10, 2009 at 08:50:18AM +0100, <Andrew Morton> bubbled:
> > On Tue, 9 Jun 2009 08:39:48 +0200 Martin Zwickel <martin.zwickel@novabasedigitaltv.de> wrote:
> >
> > > My kernel runs with SMP and HIGHMEM if it matters:
> > > Linux tt-business 2.6.29.4 #3 SMP PREEMPT Fri Jun 5 00:03:39 WEST 2009 i686 Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz GenuineIntel GNU/Linux
> > >
> >
> > The oops is very very similar to the one which Daniel Mack reported,
> > and which was fixed by
> >
> > commit c16199fcd341f10147eb2e29a60aed41ceca55b4
> > Author: Alan Stern <stern@rowland.harvard.edu>
> > Date:   Thu Apr 30 10:06:19 2009 -0400
> >
> >     usb-serial: ftdi_sio: fix reference counting of ftdi_private
> >
> > but your 2.6.29.4 contained that fix, so it seems that we still have a
> > problem here?
> 
> Could be, or I'm doing something wrong.
> 
> I will clean the kernel, rebuild 2.6.29.4, reboot, try to crash it and
> let you know.
> If I should try some patches, please let me know.

I have seen that 2.6.30 is out. I'm now running with 2.6.30 and
tried to crash the ftdi_sio driver, I had no luck till now.

So it seems to be fixed at least in 2.6.30, and I believe it was this
patch:

commit 80193195f87ebca6d7417516d6edeb3969631c15
Author: David Woodhouse <dwmw2@infradead.org>
Date:   Mon May 18 13:07:35 2009 +0100

    Fix oops on close of hot-unplugged FTDI serial converter


Best regards,
     Martin



[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]