Re: Deadlock in kernel on USB shutdown

public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed

* Re: Deadlock in kernel on USB shutdown
@ 2002-01-03  0:13 Roger Leblanc
  2002-01-03  0:18 ` Greg KH
  0 siblings, 1 reply; 14+ messages in thread
From: Roger Leblanc @ 2002-01-03  0:13 UTC (permalink / raw)
  To: Greg KH; +Cc: linux-kernel

----- Original Message -----
From: "Greg KH" <greg@kroah.com>
To: "Roger Leblanc" <r_leblanc@videotron.ca>
Cc: <linux-kernel@vger.kernel.org>
Sent: Wednesday, January 02, 2002 9:40 AM
Subject: Re: Deadlock in kernel on USB shutdown



>> On Tue, Jan 01, 2002 at 04:48:41PM -0500, Roger Leblanc wrote:
>
>>> > Hi,
>>> >
>>> > I just compiled version 2.4.17 of the Linux kernel for my Pentium III.
>>> > It is compiled with modular USB support so I can run my USB scanner (an
>>> > Epson Perfection 1200U).
>>> >
>>> > The scanner works fine but the system freeses when I shut it down. I
>>> > investigated a bit and found that in the file:
>>> > <kernel_root>/drivers/usb/usb.c
>>> > in function:
>>> > usb_disconnect(struct usb_device **pdev)
>>> >
>>> > there is a call to function:
>>> > usbdevfs_remove_device(dev)
>>> > at line 2423.
>>> >
>>> > That is the exact point where it freeses. If I comment out that line,
>>> > everything goes fine. I know! This is not the proper way to fix it! But
>>> > at least, it fixes my problem. Since I'm not a kernel expert, I will
>>> > leave it to you to find the right way to fix it.
>>
>>
>> Does the system lock up when you unload the usbcore module by hand
>> without shutting the system down?
>>
>> Are your init scripts unmounting the usbdevfs filesystem properly before
>> trying to unload the usbcore module?
>
It doesn't get that far. The first thing my init script (or Mandrake 8.1 
script) does at shutdown is to run modprobe -r on modules usb-ohci, 
usb-uhci and uhci. The system freeses when it gets to usb-uhci. It does 
it also if I run these commands on the command line.

Thanks

Roger




^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  0:13 Deadlock in kernel on USB shutdown Roger Leblanc
@ 2002-01-03  0:18 ` Greg KH
  2002-01-03  0:25   ` Roger Leblanc
  0 siblings, 1 reply; 14+ messages in thread
From: Greg KH @ 2002-01-03  0:18 UTC (permalink / raw)
  To: Roger Leblanc; +Cc: linux-kernel

On Wed, Jan 02, 2002 at 07:13:35PM -0500, Roger Leblanc wrote:
> It doesn't get that far. The first thing my init script (or Mandrake 8.1 
> script) does at shutdown is to run modprobe -r on modules usb-ohci, 
> usb-uhci and uhci. The system freeses when it gets to usb-uhci. It does 
> it also if I run these commands on the command line.

Have you unloaded your scanner module before unloading the usb-uhci
module?

thanks,

greg k-h

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  0:18 ` Greg KH
@ 2002-01-03  0:25   ` Roger Leblanc
  2002-01-03  0:28     ` Greg KH
  0 siblings, 1 reply; 14+ messages in thread
From: Roger Leblanc @ 2002-01-03  0:25 UTC (permalink / raw)
  To: Greg KH; +Cc: linux-kernel

Hi Greg,

Greg KH wrote:

>On Wed, Jan 02, 2002 at 07:13:35PM -0500, Roger Leblanc wrote:
>
>>It doesn't get that far. The first thing my init script (or Mandrake 8.1 
>>script) does at shutdown is to run modprobe -r on modules usb-ohci, 
>>usb-uhci and uhci. The system freeses when it gets to usb-uhci. It does 
>>it also if I run these commands on the command line.
>>
>
>Have you unloaded your scanner module before unloading the usb-uhci
>module?
>
No. Actually, I don't even know how it's called. How can I find out?

Thanks

Roger

>
>
>thanks,
>
>greg k-h
>



^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  0:25   ` Roger Leblanc
@ 2002-01-03  0:28     ` Greg KH
  2002-01-03  1:09       ` Roger Leblanc
  0 siblings, 1 reply; 14+ messages in thread
From: Greg KH @ 2002-01-03  0:28 UTC (permalink / raw)
  To: Roger Leblanc; +Cc: linux-kernel

On Wed, Jan 02, 2002 at 07:25:16PM -0500, Roger Leblanc wrote:
> Greg KH wrote:
> >
> >Have you unloaded your scanner module before unloading the usb-uhci
> >module?
> >
> No. Actually, I don't even know how it's called. How can I find out?

I think it's called "scanner".  Just look for any module that the
usbcore module shows as "Used by".

greg k-h

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  0:28     ` Greg KH
@ 2002-01-03  1:09       ` Roger Leblanc
  2002-01-03  1:32         ` Greg KH
  0 siblings, 1 reply; 14+ messages in thread
From: Roger Leblanc @ 2002-01-03  1:09 UTC (permalink / raw)
  To: Greg KH; +Cc: linux-kernel

Hi,

>>Greg KH wrote:
>>
>>>Have you unloaded your scanner module before unloading the usb-uhci
>>>module?
>>>
>>No. Actually, I don't even know how it's called. How can I find out?
>>
>
>I think it's called "scanner".  Just look for any module that the
>usbcore module shows as "Used by".
>
Mmmm, I should have guessed that one. Scanner is quite a good name for a 
scanner module ;-). Anyway, I moved things around so "scanner" and all 
the other device specific modules are unloaded before usb-uhci but yet, 
it doesn't help. It still freeses when unload usb-uhci. Any idea?

Thanks

Roger



^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  1:09       ` Roger Leblanc
@ 2002-01-03  1:32         ` Greg KH
  2002-01-03  2:10           ` Roger Leblanc
  0 siblings, 1 reply; 14+ messages in thread
From: Greg KH @ 2002-01-03  1:32 UTC (permalink / raw)
  To: Roger Leblanc; +Cc: linux-kernel

On Wed, Jan 02, 2002 at 08:09:35PM -0500, Roger Leblanc wrote:
> Mmmm, I should have guessed that one. Scanner is quite a good name for a 
> scanner module ;-). Anyway, I moved things around so "scanner" and all 
> the other device specific modules are unloaded before usb-uhci but yet, 
> it doesn't help. It still freeses when unload usb-uhci. Any idea?

Hm, if you umount usbdevfs before unloading usb-uhci, does that work?

greg k-h

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  1:32         ` Greg KH
@ 2002-01-03  2:10           ` Roger Leblanc
  2002-01-03  3:03             ` Greg KH
  0 siblings, 1 reply; 14+ messages in thread
From: Roger Leblanc @ 2002-01-03  2:10 UTC (permalink / raw)
  To: Greg KH; +Cc: linux-kernel



Greg KH wrote:

>On Wed, Jan 02, 2002 at 08:09:35PM -0500, Roger Leblanc wrote:
>
>>Mmmm, I should have guessed that one. Scanner is quite a good name for a 
>>scanner module ;-). Anyway, I moved things around so "scanner" and all 
>>the other device specific modules are unloaded before usb-uhci but yet, 
>>it doesn't help. It still freeses when unload usb-uhci. Any idea?
>>
>
>Hm, if you umount usbdevfs before unloading usb-uhci, does that work?
>
umount seg-faults while unmounting /proc/bus/usb! Then modprobe also 
seg-faults while unloading usb-uhci!!! But the system stays up.
After that, I cannot make my scanner to work even if I run 
".../init.d/usb start".
B.T.W. Did I tell you that I can bring the kernel down just by turning 
the scanner off and then back again?

Confused Roger



^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  2:10           ` Roger Leblanc
@ 2002-01-03  3:03             ` Greg KH
  2002-01-03  3:26               ` Roger Leblanc
  0 siblings, 1 reply; 14+ messages in thread
From: Greg KH @ 2002-01-03  3:03 UTC (permalink / raw)
  To: Roger Leblanc; +Cc: linux-kernel

On Wed, Jan 02, 2002 at 09:10:16PM -0500, Roger Leblanc wrote:
> umount seg-faults while unmounting /proc/bus/usb! Then modprobe also 
> seg-faults while unloading usb-uhci!!! But the system stays up.
> After that, I cannot make my scanner to work even if I run 
> ".../init.d/usb start".

Can you run that first oops through ksymoops and send it.  Actually do
it for both oopses.

> B.T.W. Did I tell you that I can bring the kernel down just by turning 
> the scanner off and then back again?

Hm, sounds like the scanner driver has some problems :)
You might email the author with this info and see if he has any ideas.

greg k-h

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  3:03             ` Greg KH
@ 2002-01-03  3:26               ` Roger Leblanc
  2002-01-03 18:57                 ` Greg KH
  0 siblings, 1 reply; 14+ messages in thread
From: Roger Leblanc @ 2002-01-03  3:26 UTC (permalink / raw)
  To: Greg KH; +Cc: linux-kernel

Greg KH wrote:

>On Wed, Jan 02, 2002 at 09:10:16PM -0500, Roger Leblanc wrote:
>
>>umount seg-faults while unmounting /proc/bus/usb! Then modprobe also 
>>seg-faults while unloading usb-uhci!!! But the system stays up.
>>After that, I cannot make my scanner to work even if I run 
>>".../init.d/usb start".
>>
>
>Can you run that first oops through ksymoops and send it.  Actually do
>it for both oopses.
>
Oops? I know about cores but I don't know about oopses. I'm only a C++, 
hight level, OO developer. Sorry ;-). Can you explain please?

>>B.T.W. Did I tell you that I can bring the kernel down just by turning 
>>the scanner off and then back again?
>>
>
>Hm, sounds like the scanner driver has some problems :)
>You might email the author with this info and see if he has any ideas.
>
I'll do that

Thanks

Roger



^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03  3:26               ` Roger Leblanc
@ 2002-01-03 18:57                 ` Greg KH
  2002-01-04 21:46                   ` Roger Leblanc
  0 siblings, 1 reply; 14+ messages in thread
From: Greg KH @ 2002-01-03 18:57 UTC (permalink / raw)
  To: Roger Leblanc; +Cc: linux-kernel

On Wed, Jan 02, 2002 at 10:26:41PM -0500, Roger Leblanc wrote:
> Oops? I know about cores but I don't know about oopses. I'm only a C++, 
> hight level, OO developer. Sorry ;-). Can you explain please?

When the seg-fault happens, I am guessing that an oops message is sent
to the kernel log.  Look at 'dmesg'.  Run that oops through ksymoops.
See Documentation/oops-tracing.txt for more info on oopses and how to
get info from them.

thanks,

greg k-h

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-03 18:57                 ` Greg KH
@ 2002-01-04 21:46                   ` Roger Leblanc
  0 siblings, 0 replies; 14+ messages in thread
From: Roger Leblanc @ 2002-01-04 21:46 UTC (permalink / raw)
  To: Greg KH; +Cc: linux-kernel

[-- Attachment #1: Type: text/plain, Size: 567 bytes --]

Hi Greg,

Greg KH wrote:

>
>When the seg-fault happens, I am guessing that an oops message is sent
>to the kernel log.  Look at 'dmesg'.  Run that oops through ksymoops.
>See Documentation/oops-tracing.txt for more info on oopses and how to
>get info from them.
>
>thanks,
>
>greg k-h
>
Sorry for the delay.

I found a lot of oopses and I managed to separate them in two files 
(attached). Each of the files correspond approx. to the time when I had 
the 2 seg-faults.

Thanks a lot for all your help Greg. It is really appreciated.

Happy New Year!

Roger Leblanc


[-- Attachment #2: ksymoops.out.1 --]
[-- Type: text/plain, Size: 5181 bytes --]

ksymoops 2.4.3 on i686 2.4.17.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.17/ (default)
     -m /usr/src/linux/System.map (default)

Warning: You did not tell me where to find symbol information.  I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc.  ksymoops -h explains the options.

Jan  2 21:00:23 roger kernel:  <1>Unable to handle kernel paging request at virtual address a1ae3ae0
Jan  2 21:00:23 roger kernel: e614b46e
Jan  2 21:00:23 roger kernel: Oops: 0000
Jan  2 21:00:23 roger kernel: CPU:    0
Jan  2 21:00:23 roger kernel: EIP:    0010:[scanner:__insmod_scanner_O/lib/modules/2.4.17/kernel/drivers/usb/sc+-41200530/96]    Not tainted
Jan  2 21:00:23 roger kernel: EIP:    0010:[<e614b46e>]    Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
Jan  2 21:00:23 roger kernel: EFLAGS: 00010282
Jan  2 21:00:23 roger kernel: eax: c02b1466   ebx: dcd755a0   ecx: c031ade8   edx: 00000000
Jan  2 21:00:23 roger kernel: esi: e7ff91e0   edi: dcd75480   ebp: e252dba0   esp: e08fbf3c
Jan  2 21:00:23 roger kernel: ds: 0018   es: 0018   ss: 0018
Jan  2 21:00:23 roger kernel: Process mozilla-bin (pid: 1986, stackpage=e08fb000)
Jan  2 21:00:23 roger kernel: Stack: c01ff4b0 dcd755a0 dcd75480 c01ffa40 dcd755a0 ffffffff e517f8c0 00000000 
Jan  2 21:00:23 roger kernel:        e517f8c0 c01344ce dcd75480 e517f8c0 000080d4 00000014 09266be0 bf5ff4b4 
Jan  2 21:00:23 roger kernel:        e517f8c0 e517f8c0 00000000 bf5ff404 c013332d e517f8c0 e2745580 00000000 
Jan  2 21:00:23 roger kernel: Call Trace: [sock_release+16/96] [sock_close+48/64] [fput+78/240] [filp_close+141/160] [sys_close+91/112] 
Jan  2 21:00:23 roger kernel: Call Trace: [<c01ff4b0>] [<c01ffa40>] [<c01344ce>] [<c013332d>] [<c013339b>] 
Jan  2 21:00:23 roger kernel:    [<c0106f1b>] 
Jan  2 21:00:23 roger kernel: Code: 33 3c 7e b8 33 3c 7e b8 33 3c 0a 00 00 00 00 10 00 00 00 00 

>>EIP; e614b46e <_end+25e04bfa/2851978c>   <=====
Trace; c01ff4b0 <sock_release+10/60>
Trace; c01ffa40 <sock_close+30/40>
Trace; c01344ce <fput+4e/f0>
Trace; c013332c <filp_close+8c/a0>
Trace; c013339a <sys_close+5a/70>
Trace; c0106f1a <system_call+32/38>
Code;  e614b46e <_end+25e04bfa/2851978c>
00000000 <_EIP>:
Code;  e614b46e <_end+25e04bfa/2851978c>   <=====
   0:   33 3c 7e                  xor    (%esi,%edi,2),%edi   <=====
Code;  e614b470 <_end+25e04bfc/2851978c>
   3:   b8 33 3c 7e b8            mov    $0xb87e3c33,%eax
Code;  e614b476 <_end+25e04c02/2851978c>
   8:   33 3c 0a                  xor    (%edx,%ecx,1),%edi
Code;  e614b478 <_end+25e04c04/2851978c>
   b:   00 00                     add    %al,(%eax)
Code;  e614b47a <_end+25e04c06/2851978c>
   d:   00 00                     add    %al,(%eax)
Code;  e614b47c <_end+25e04c08/2851978c>
   f:   10 00                     adc    %al,(%eax)

Jan  2 21:00:23 roger kernel:  <1>Unable to handle kernel paging request at virtual address a8b4e5e0
Jan  2 21:00:23 roger kernel: e614b46e
Jan  2 21:00:23 roger kernel: Oops: 0000
Jan  2 21:00:23 roger kernel: CPU:    0
Jan  2 21:00:23 roger kernel: EIP:    0010:[scanner:__insmod_scanner_O/lib/modules/2.4.17/kernel/drivers/usb/sc+-41200530/96]    Not tainted
Jan  2 21:00:23 roger kernel: EIP:    0010:[<e614b46e>]    Not tainted
Jan  2 21:00:23 roger kernel: EFLAGS: 00010293
Jan  2 21:00:23 roger kernel: eax: c02b1466   ebx: e05aab20   ecx: c031ade8   edx: 00000000
Jan  2 21:00:23 roger kernel: esi: e7ff91e0   edi: e05aaa00   ebp: e06d28a0   esp: e17d7e48
Jan  2 21:00:23 roger kernel: ds: 0018   es: 0018   ss: 0018
Jan  2 21:00:23 roger kernel: Process mozilla-bin (pid: 1966, stackpage=e17d7000)
Jan  2 21:00:23 roger kernel: Stack: c01ff4b0 e05aab20 e05aaa00 c01ffa40 e05aab20 ffffffff e04b5ce0 00000000 
Jan  2 21:00:23 roger kernel:        e04b5ce0 c01344ce e05aaa00 e04b5ce0 e17d4bfc e3679960 e7ffbce4 e3679960 
Jan  2 21:00:23 roger kernel:        00007fff e04b5ce0 00000000 00000011 c013332d e04b5ce0 e2745580 00000000 
Jan  2 21:00:23 roger kernel: Call Trace: [sock_release+16/96] [sock_close+48/64] [fput+78/240] [filp_close+141/160] [put_files_struct+77/208] 
Jan  2 21:00:23 roger kernel: Call Trace: [<c01ff4b0>] [<c01ffa40>] [<c01344ce>] [<c013332d>] [<c011881d>] 
Jan  2 21:00:23 roger kernel:    [<c011901f>] [<c010ac98>] [<c0106dc4>] [<c013bf1a>] [<c01133c9>] [<c0106f54>] 
Warning (Oops_read): Code line not seen, dumping what data is available

>>EIP; e614b46e <_end+25e04bfa/2851978c>   <=====
Trace; c01ff4b0 <sock_release+10/60>
Trace; c01ffa40 <sock_close+30/40>
Trace; c01344ce <fput+4e/f0>
Trace; c013332c <filp_close+8c/a0>
Trace; c011881c <put_files_struct+4c/d0>
Trace; c011901e <do_exit+10e/240>
Trace; c010ac98 <call_do_IRQ+6/e>
Trace; c0106dc4 <do_signal+244/2b0>
Trace; c013bf1a <pipe_write+20a/270>
Trace; c01133c8 <schedule+458/510>
Trace; c0106f54 <signal_return+14/18>


2 warnings issued.  Results may not be reliable.

[-- Attachment #3: ksymoops.out.2 --]
[-- Type: text/plain, Size: 12177 bytes --]

ksymoops 2.4.3 on i686 2.4.17.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.17/ (default)
     -m /usr/src/linux/System.map (default)

Warning: You did not tell me where to find symbol information.  I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc.  ksymoops -h explains the options.

Jan  2 21:00:38 roger kernel:  <1>Unable to handle kernel paging request at virtual address aeed88e0
Jan  2 21:00:38 roger kernel: e614b46e
Jan  2 21:00:38 roger kernel: Oops: 0000
Jan  2 21:00:38 roger kernel: CPU:    0
Jan  2 21:00:38 roger kernel: EIP:    0010:[scanner:__insmod_scanner_O/lib/modules/2.4.17/kernel/drivers/usb/sc+-41200530/96]    Not tainted
Jan  2 21:00:38 roger kernel: EIP:    0010:[<e614b46e>]    Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
Jan  2 21:00:38 roger kernel: EFLAGS: 00013297
Jan  2 21:00:38 roger kernel: eax: c02b1466   ebx: e376fca0   ecx: c031ade8   edx: 00000000
Jan  2 21:00:38 roger kernel: esi: e7ff91e0   edi: e376fb80   ebp: e3729440   esp: e5099f3c
Jan  2 21:00:38 roger kernel: ds: 0018   es: 0018   ss: 0018
Jan  2 21:00:38 roger kernel: Process X (pid: 1628, stackpage=e5099000)
Jan  2 21:00:38 roger kernel: Stack: c01ff4b0 e376fca0 e376fb80 c01ffa40 e376fca0 ffffffff e517f440 00000000 
Jan  2 21:00:38 roger kernel:        e517f440 c01344ce e376fb80 e517f440 0000000c e376fca0 e376fca0 0000000a 
Jan  2 21:00:38 roger kernel:        e517f440 e517f440 00000000 bffff758 c013332d e517f440 e6de76c0 00000000 
Jan  2 21:00:38 roger kernel: Call Trace: [sock_release+16/96] [sock_close+48/64] [fput+78/240] [filp_close+141/160] [sys_close+91/112] 
Jan  2 21:00:38 roger kernel: Call Trace: [<c01ff4b0>] [<c01ffa40>] [<c01344ce>] [<c013332d>] [<c013339b>] 
Jan  2 21:00:38 roger kernel:    [<c0106f1b>] 
Jan  2 21:00:38 roger kernel: Code: 33 3c 7e b8 33 3c 7e b8 33 3c 0a 00 00 00 00 10 00 00 00 00 

>>EIP; e614b46e <_end+25e04bfa/2851978c>   <=====
Trace; c01ff4b0 <sock_release+10/60>
Trace; c01ffa40 <sock_close+30/40>
Trace; c01344ce <fput+4e/f0>
Trace; c013332c <filp_close+8c/a0>
Trace; c013339a <sys_close+5a/70>
Trace; c0106f1a <system_call+32/38>
Code;  e614b46e <_end+25e04bfa/2851978c>
00000000 <_EIP>:
Code;  e614b46e <_end+25e04bfa/2851978c>   <=====
   0:   33 3c 7e                  xor    (%esi,%edi,2),%edi   <=====
Code;  e614b470 <_end+25e04bfc/2851978c>
   3:   b8 33 3c 7e b8            mov    $0xb87e3c33,%eax
Code;  e614b476 <_end+25e04c02/2851978c>
   8:   33 3c 0a                  xor    (%edx,%ecx,1),%edi
Code;  e614b478 <_end+25e04c04/2851978c>
   b:   00 00                     add    %al,(%eax)
Code;  e614b47a <_end+25e04c06/2851978c>
   d:   00 00                     add    %al,(%eax)
Code;  e614b47c <_end+25e04c08/2851978c>
   f:   10 00                     adc    %al,(%eax)

Jan  2 21:00:38 roger kernel:  <1>Unable to handle kernel paging request at virtual address b21b72a0
Jan  2 21:00:38 roger kernel: e614b46e
Jan  2 21:00:38 roger kernel: Oops: 0000
Jan  2 21:00:38 roger kernel: CPU:    0
Jan  2 21:00:38 roger kernel: EIP:    0010:[scanner:__insmod_scanner_O/lib/modules/2.4.17/kernel/drivers/usb/sc+-41200530/96]    Not tainted
Jan  2 21:00:38 roger kernel: EIP:    0010:[<e614b46e>]    Not tainted
Jan  2 21:00:38 roger kernel: EFLAGS: 00013287
Jan  2 21:00:38 roger kernel: eax: c02b1466   ebx: e50df180   ecx: c031ade8   edx: 00000000
Jan  2 21:00:38 roger kernel: esi: e7ff91e0   edi: e50df060   ebp: e50d0760   esp: e5099d70
Jan  2 21:00:38 roger kernel: ds: 0018   es: 0018   ss: 0018
Jan  2 21:00:38 roger kernel: Process X (pid: 1628, stackpage=e5099000)
Jan  2 21:00:38 roger kernel: Stack: c01ff4b0 e50df180 e50df060 c01ffa40 e50df180 ffffffff e4e7ff60 00000000 
Jan  2 21:00:38 roger kernel:        e4e7ff60 c01344ce e50df060 e4e7ff60 e4e22bfc e6ed1bc0 e7ffbce4 e6ed1bc0 
Jan  2 21:00:38 roger kernel:        03ffff7f e4e7ff60 00000000 00000003 c013332d e4e7ff60 e6de76c0 00000000 
Jan  2 21:00:38 roger kernel: Call Trace: [sock_release+16/96] [sock_close+48/64] [fput+78/240] [filp_close+141/160] [put_files_struct+77/208] 
Jan  2 21:00:38 roger kernel: Call Trace: [<c01ff4b0>] [<c01ffa40>] [<c01344ce>] [<c013332d>] [<c011881d>] 
Jan  2 21:00:38 roger kernel:    [<c011901f>] [<c01122a0>] [<c0107464>] [<c0112698>] [<c01ff56c>] [<c01133c9>] 
Jan  2 21:00:38 roger kernel:    [<c0190530>] [<c01122e0>] [<c010700c>] [<c01ff4b0>] [<c01ffa40>] [<c01344ce>] 
Jan  2 21:00:38 roger kernel:    [<c013332d>] [<c013339b>] [<c0106f1b>] 
Jan  2 21:00:38 roger kernel: Code: 33 3c 7e b8 33 3c 7e b8 33 3c 0a 00 00 00 00 10 00 00 00 00 

>>EIP; e614b46e <_end+25e04bfa/2851978c>   <=====
Trace; c01ff4b0 <sock_release+10/60>
Trace; c01ffa40 <sock_close+30/40>
Trace; c01344ce <fput+4e/f0>
Trace; c013332c <filp_close+8c/a0>
Trace; c011881c <put_files_struct+4c/d0>
Trace; c011901e <do_exit+10e/240>
Trace; c01122a0 <bust_spinlocks+50/60>
Trace; c0107464 <die+54/70>
Trace; c0112698 <do_page_fault+3b8/500>
Trace; c01ff56c <sock_sendmsg+6c/90>
Trace; c01133c8 <schedule+458/510>
Trace; c0190530 <normal_poll+110/130>
Trace; c01122e0 <do_page_fault+0/500>
Trace; c010700c <error_code+34/3c>
Trace; c01ff4b0 <sock_release+10/60>
Trace; c01ffa40 <sock_close+30/40>
Trace; c01344ce <fput+4e/f0>
Trace; c013332c <filp_close+8c/a0>
Trace; c013339a <sys_close+5a/70>
Trace; c0106f1a <system_call+32/38>
Code;  e614b46e <_end+25e04bfa/2851978c>
00000000 <_EIP>:
Code;  e614b46e <_end+25e04bfa/2851978c>   <=====
   0:   33 3c 7e                  xor    (%esi,%edi,2),%edi   <=====
Code;  e614b470 <_end+25e04bfc/2851978c>
   3:   b8 33 3c 7e b8            mov    $0xb87e3c33,%eax
Code;  e614b476 <_end+25e04c02/2851978c>
   8:   33 3c 0a                  xor    (%edx,%ecx,1),%edi
Code;  e614b478 <_end+25e04c04/2851978c>
   b:   00 00                     add    %al,(%eax)
Code;  e614b47a <_end+25e04c06/2851978c>
   d:   00 00                     add    %al,(%eax)
Code;  e614b47c <_end+25e04c08/2851978c>
   f:   10 00                     adc    %al,(%eax)

Jan  2 21:00:38 roger kernel:  <1>Unable to handle kernel paging request at virtual address ab7068e0
Jan  2 21:00:38 roger kernel: e614b46e
Jan  2 21:00:38 roger kernel: Oops: 0000
Jan  2 21:00:38 roger kernel: CPU:    0
Jan  2 21:00:38 roger kernel: EIP:    0010:[scanner:__insmod_scanner_O/lib/modules/2.4.17/kernel/drivers/usb/sc+-41200530/96]    Not tainted
Jan  2 21:00:38 roger kernel: EIP:    0010:[<e614b46e>]    Not tainted
Jan  2 21:00:38 roger kernel: EFLAGS: 00010286
Jan  2 21:00:38 roger kernel: eax: c02b1466   ebx: e1b86ca0   ecx: c031ade8   edx: 00000000
Jan  2 21:00:38 roger kernel: esi: e7ff91e0   edi: e1b86b80   ebp: e1bac1a0   esp: e33a9f3c
Jan  2 21:00:38 roger kernel: ds: 0018   es: 0018   ss: 0018
Jan  2 21:00:38 roger kernel: Process ksmserver (pid: 1964, stackpage=e33a9000)
Jan  2 21:00:38 roger kernel: Stack: c01ff4b0 e1b86ca0 e1b86b80 c01ffa40 e1b86ca0 ffffffff e1c685e0 00000000 
Jan  2 21:00:38 roger kernel:        e1c685e0 c01344ce e1b86b80 e1c685e0 e1bcd2e0 c0141932 e1bcd2e0 00000004 
Jan  2 21:00:38 roger kernel:        e1c685e0 e1c685e0 00000000 bfffef58 c013332d e1c685e0 e2745a60 00000000 
Jan  2 21:00:38 roger kernel: Call Trace: [sock_release+16/96] [sock_close+48/64] [fput+78/240] [sys_select+1138/1152] [filp_close+141/160] 
Jan  2 21:00:38 roger kernel: Call Trace: [<c01ff4b0>] [<c01ffa40>] [<c01344ce>] [<c0141932>] [<c013332d>] 
Jan  2 21:00:38 roger kernel:    [<c013339b>] [<c0106f1b>] 
Jan  2 21:00:38 roger kernel: Code: 33 3c 7e b8 33 3c 7e b8 33 3c 0a 00 00 00 00 10 00 00 00 00 

>>EIP; e614b46e <_end+25e04bfa/2851978c>   <=====
Trace; c01ff4b0 <sock_release+10/60>
Trace; c01ffa40 <sock_close+30/40>
Trace; c01344ce <fput+4e/f0>
Trace; c0141932 <sys_select+472/480>
Trace; c013332c <filp_close+8c/a0>
Trace; c013339a <sys_close+5a/70>
Trace; c0106f1a <system_call+32/38>
Code;  e614b46e <_end+25e04bfa/2851978c>
00000000 <_EIP>:
Code;  e614b46e <_end+25e04bfa/2851978c>   <=====
   0:   33 3c 7e                  xor    (%esi,%edi,2),%edi   <=====
Code;  e614b470 <_end+25e04bfc/2851978c>
   3:   b8 33 3c 7e b8            mov    $0xb87e3c33,%eax
Code;  e614b476 <_end+25e04c02/2851978c>
   8:   33 3c 0a                  xor    (%edx,%ecx,1),%edi
Code;  e614b478 <_end+25e04c04/2851978c>
   b:   00 00                     add    %al,(%eax)
Code;  e614b47a <_end+25e04c06/2851978c>
   d:   00 00                     add    %al,(%eax)
Code;  e614b47c <_end+25e04c08/2851978c>
   f:   10 00                     adc    %al,(%eax)

Jan  2 21:00:38 roger kernel:  <1>Unable to handle kernel paging request at virtual address aba84920
Jan  2 21:00:38 roger kernel: e614b46e
Jan  2 21:00:38 roger kernel: Oops: 0000
Jan  2 21:00:38 roger kernel: CPU:    0
Jan  2 21:00:38 roger kernel: EIP:    0010:[scanner:__insmod_scanner_O/lib/modules/2.4.17/kernel/drivers/usb/sc+-41200530/96]    Not tainted
Jan  2 21:00:38 roger kernel: EIP:    0010:[<e614b46e>]    Not tainted
Jan  2 21:00:38 roger kernel: EFLAGS: 00010213
Jan  2 21:00:38 roger kernel: eax: c02b1466   ebx: e1d45cc0   ecx: c031ade8   edx: 00000000
Jan  2 21:00:38 roger kernel: esi: e7ff91e0   edi: e1d45ba0   ebp: e1d1e0a0   esp: e33a9d70
Jan  2 21:00:38 roger kernel: ds: 0018   es: 0018   ss: 0018
Jan  2 21:00:38 roger kernel: Process ksmserver (pid: 1964, stackpage=e33a9000)
Jan  2 21:00:38 roger kernel: Stack: c01ff4b0 e1d45cc0 e1d45ba0 c01ffa40 e1d45cc0 ffffffff e61a8120 00000000 
Jan  2 21:00:38 roger kernel:        e61a8120 c01344ce e1d45ba0 e61a8120 e2dacbfc e3679b40 e7ffbce4 e3679b40 
Jan  2 21:00:38 roger kernel:        00007fef e61a8120 00000000 00000003 c013332d e61a8120 e2745a60 00000000 
Jan  2 21:00:38 roger kernel: Call Trace: [sock_release+16/96] [sock_close+48/64] [fput+78/240] [filp_close+141/160] [put_files_struct+77/208] 
Jan  2 21:00:38 roger kernel: Call Trace: [<c01ff4b0>] [<c01ffa40>] [<c01344ce>] [<c013332d>] [<c011881d>] 
Jan  2 21:00:38 roger kernel:    [<c011901f>] [<c01122a0>] [<c0107464>] [<c0112698>] [<c010657a>] [<c01066d4>] 
Jan  2 21:00:38 roger kernel:    [<c01122e0>] [<c010700c>] [<c01ff4b0>] [<c01ffa40>] [<c01344ce>] [<c0141932>] 
Jan  2 21:00:38 roger kernel:    [<c013332d>] [<c013339b>] [<c0106f1b>] 
Jan  2 21:00:38 roger kernel: Code: 33 3c 7e b8 33 3c 7e b8 33 3c 0a 00 00 00 00 10 00 00 00 00 

>>EIP; e614b46e <_end+25e04bfa/2851978c>   <=====
Trace; c01ff4b0 <sock_release+10/60>
Trace; c01ffa40 <sock_close+30/40>
Trace; c01344ce <fput+4e/f0>
Trace; c013332c <filp_close+8c/a0>
Trace; c011881c <put_files_struct+4c/d0>
Trace; c011901e <do_exit+10e/240>
Trace; c01122a0 <bust_spinlocks+50/60>
Trace; c0107464 <die+54/70>
Trace; c0112698 <do_page_fault+3b8/500>
Trace; c010657a <setup_sigcontext+da/120>
Trace; c01066d4 <setup_frame+114/210>
Trace; c01122e0 <do_page_fault+0/500>
Trace; c010700c <error_code+34/3c>
Trace; c01ff4b0 <sock_release+10/60>
Trace; c01ffa40 <sock_close+30/40>
Trace; c01344ce <fput+4e/f0>
Trace; c0141932 <sys_select+472/480>
Trace; c013332c <filp_close+8c/a0>
Trace; c013339a <sys_close+5a/70>
Trace; c0106f1a <system_call+32/38>
Code;  e614b46e <_end+25e04bfa/2851978c>
00000000 <_EIP>:
Code;  e614b46e <_end+25e04bfa/2851978c>   <=====
   0:   33 3c 7e                  xor    (%esi,%edi,2),%edi   <=====
Code;  e614b470 <_end+25e04bfc/2851978c>
   3:   b8 33 3c 7e b8            mov    $0xb87e3c33,%eax
Code;  e614b476 <_end+25e04c02/2851978c>
   8:   33 3c 0a                  xor    (%edx,%ecx,1),%edi
Code;  e614b478 <_end+25e04c04/2851978c>
   b:   00 00                     add    %al,(%eax)
Code;  e614b47a <_end+25e04c06/2851978c>
   d:   00 00                     add    %al,(%eax)
Code;  e614b47c <_end+25e04c08/2851978c>
   f:   10 00                     adc    %al,(%eax)

Jan  2 21:00:38 roger kernel:  <1>Unable to handle kernel paging request at virtual address a92eece0
Jan  2 21:00:38 roger kernel: e614b46e

1 warning issued.  Results may not be reliable.

^ permalink raw reply	[flat|nested] 14+ messages in thread

[parent not found: <000701c193f6$4ca368a0$6800000a@brownell.org>]

* Re: Deadlock in kernel on USB shutdown
       [not found] <000701c193f6$4ca368a0$6800000a@brownell.org>
@ 2002-01-03  2:25 ` Roger Leblanc
  0 siblings, 0 replies; 14+ messages in thread
From: Roger Leblanc @ 2002-01-03  2:25 UTC (permalink / raw)
  To: David Brownell; +Cc: linux-kernel

David Brownell wrote:

>
>The classic question is whether the same thing happens when you
>make your system use the "uhci" driver instead of "usb-uhci".  If the
>symptom is hang on rmmod, then rmmodding a different module
>may well make a difference.
>
It works great with uhci! Thanks a lot! I guess it points to a serious 
bug in usb-uhci?

Thanks to everyone

Roger



^ permalink raw reply	[flat|nested] 14+ messages in thread

* Deadlock in kernel on USB shutdown
@ 2002-01-01 21:48 Roger Leblanc
  2002-01-02 14:40 ` Greg KH
  0 siblings, 1 reply; 14+ messages in thread
From: Roger Leblanc @ 2002-01-01 21:48 UTC (permalink / raw)
  To: linux-kernel

Hi,

I just compiled version 2.4.17 of the Linux kernel for my Pentium III. 
It is compiled with modular USB support so I can run my USB scanner (an 
Epson Perfection 1200U).

The scanner works fine but the system freeses when I shut it down. I 
investigated a bit and found that in the file:
<kernel_root>/drivers/usb/usb.c
in function:
usb_disconnect(struct usb_device **pdev)

there is a call to function:
usbdevfs_remove_device(dev)
at line 2423.

That is the exact point where it freeses. If I comment out that line, 
everything goes fine. I know! This is not the proper way to fix it! But 
at least, it fixes my problem. Since I'm not a kernel expert, I will 
leave it to you to find the right way to fix it.

If you want me to try something or to give you more info, do not 
hesitate to contact me.

Thanks

Roger

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Deadlock in kernel on USB shutdown
  2002-01-01 21:48 Roger Leblanc
@ 2002-01-02 14:40 ` Greg KH
  0 siblings, 0 replies; 14+ messages in thread
From: Greg KH @ 2002-01-02 14:40 UTC (permalink / raw)
  To: Roger Leblanc; +Cc: linux-kernel

On Tue, Jan 01, 2002 at 04:48:41PM -0500, Roger Leblanc wrote:
> Hi,
> 
> I just compiled version 2.4.17 of the Linux kernel for my Pentium III. 
> It is compiled with modular USB support so I can run my USB scanner (an 
> Epson Perfection 1200U).
> 
> The scanner works fine but the system freeses when I shut it down. I 
> investigated a bit and found that in the file:
> <kernel_root>/drivers/usb/usb.c
> in function:
> usb_disconnect(struct usb_device **pdev)
> 
> there is a call to function:
> usbdevfs_remove_device(dev)
> at line 2423.
> 
> That is the exact point where it freeses. If I comment out that line, 
> everything goes fine. I know! This is not the proper way to fix it! But 
> at least, it fixes my problem. Since I'm not a kernel expert, I will 
> leave it to you to find the right way to fix it.

Does the system lock up when you unload the usbcore module by hand
without shutting the system down?

Are your init scripts unmounting the usbdevfs filesystem properly before
trying to unload the usbcore module?

thanks,

greg k-h

^ permalink raw reply	[flat|nested] 14+ messages in thread

end of thread, other threads:[~2002-01-04 21:47 UTC | newest]

Thread overview: 14+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-01-03  0:13 Deadlock in kernel on USB shutdown Roger Leblanc
2002-01-03  0:18 ` Greg KH
2002-01-03  0:25   ` Roger Leblanc
2002-01-03  0:28     ` Greg KH
2002-01-03  1:09       ` Roger Leblanc
2002-01-03  1:32         ` Greg KH
2002-01-03  2:10           ` Roger Leblanc
2002-01-03  3:03             ` Greg KH
2002-01-03  3:26               ` Roger Leblanc
2002-01-03 18:57                 ` Greg KH
2002-01-04 21:46                   ` Roger Leblanc
     [not found] <000701c193f6$4ca368a0$6800000a@brownell.org>
2002-01-03  2:25 ` Roger Leblanc
  -- strict thread matches above, loose matches on Subject: below --
2002-01-01 21:48 Roger Leblanc
2002-01-02 14:40 ` Greg KH

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox