Re: Fw: Oops with 2.6.13-rc6-mm2 and USB mouse

Re: Fw: Oops with 2.6.13-rc6-mm2 and USB mouse

[Rogério Brito]

Hi, Andrew.

I just tested the USB mouse with 2.6.13-rc6-mm2 and ACPI disabled
(which, according to Linus, is one of the "usual suspects") and the
problem still occurred.

On the other hand, with kernel 2.6.13-rc5-mm1 (which I am running now),
I didn't have any problems plugging and unplugging the mouse. Here are
the messages I get in dmesg (2.6.13-rc5-mm1) after I plug/unplug the
mouse:

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
usb 1-1.2: new low speed USB device using uhci_hcd and address 4
input: USB HID v1.00 Mouse [USB Wheel Mouse] on usb-0000:00:04.2-1.2
usb 1-1.2: USB disconnect, address 4
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Just thought you might like to know about this. If you want me to test
any other version, please let me know.


Thanks, Rogério Brito.

-- 
Rogério Brito : rbrito@ime.usp.br : http://www.ime.usp.br/~rbrito
Homepage of the algorithms package : http://algorithms.berlios.de
Homepage on freshmeat:  http://freshmeat.net/projects/algorithms/

Re: Fw: Oops with 2.6.13-rc6-mm2 and USB mouse

[Mattia Dongili]

On Sat, Aug 27, 2005 at 05:09:04PM -0300, Rog???rio Brito wrote:
> Hi, Andrew.
> 
> I just tested the USB mouse with 2.6.13-rc6-mm2 and ACPI disabled
> (which, according to Linus, is one of the "usual suspects") and the
> problem still occurred.

see here
http://marc.theaimsgroup.com/?l=linux-kernel&m=112481438512222&w=2

Reverting driver-core-fix-bus_rescan_devices-race.patch and applying the
patch attached to the above message fixed the oops for me.

-- 
mattia
:wq!

Re: Fw: Oops with 2.6.13-rc6-mm2 and USB mouse

[Andrew Morton]

Rogério Brito <rbrito@ime.usp.br> wrote:
>
> Hi, Andrew.
> 
> I just tested the USB mouse with 2.6.13-rc6-mm2 and ACPI disabled
> (which, according to Linus, is one of the "usual suspects") and the
> problem still occurred.
> 
> On the other hand, with kernel 2.6.13-rc5-mm1 (which I am running now),
> I didn't have any problems plugging and unplugging the mouse. Here are
> the messages I get in dmesg (2.6.13-rc5-mm1) after I plug/unplug the
> mouse:
> 
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> usb 1-1.2: new low speed USB device using uhci_hcd and address 4
> input: USB HID v1.00 Mouse [USB Wheel Mouse] on usb-0000:00:04.2-1.2
> usb 1-1.2: USB disconnect, address 4
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> 
> Just thought you might like to know about this. If you want me to test
> any other version, please let me know.
> 

Unfortunately there are 71 USB patches in 2.6.13-rc6-mm2 and I don't know
which ones to suspect.

Hopefully Greg (when he returns) or one of the other USB developers can
identify the buggy patch - we wouldn't want this leaking into Linus's
kernel..

Re: Fw: Oops with 2.6.13-rc6-mm2 and USB mouse

[Mattia Dongili]

On Sat, Aug 27, 2005 at 03:31:57PM -0700, Andrew Morton wrote:
> Rogério Brito <rbrito@ime.usp.br> wrote:
> >
> > Hi, Andrew.
> > 
> > I just tested the USB mouse with 2.6.13-rc6-mm2 and ACPI disabled
> > (which, according to Linus, is one of the "usual suspects") and the
> > problem still occurred.
> > 
> > On the other hand, with kernel 2.6.13-rc5-mm1 (which I am running now),
> > I didn't have any problems plugging and unplugging the mouse. Here are
> > the messages I get in dmesg (2.6.13-rc5-mm1) after I plug/unplug the
> > mouse:
> > 
> > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> > usb 1-1.2: new low speed USB device using uhci_hcd and address 4
> > input: USB HID v1.00 Mouse [USB Wheel Mouse] on usb-0000:00:04.2-1.2
> > usb 1-1.2: USB disconnect, address 4
> > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> > 
> > Just thought you might like to know about this. If you want me to test
> > any other version, please let me know.
> > 
> 
> Unfortunately there are 71 USB patches in 2.6.13-rc6-mm2 and I don't know
> which ones to suspect.

Really, what I mentioned[1] it my previous message fixes the very same
issue with the usb device removal, see below the oops I had here.
Not that I can't wait for Greg to come back but just to have a pointer
for a possible fix :)

[1]: http://marc.theaimsgroup.com/?l=linux-kernel&m=112481438512222&w=2

usb 2-1: new low speed USB device using uhci_hcd and address 2
input: USB HID v1.10 Mouse [Logitech USB Mouse] on usb-0000:00:1d.1-1
usb 2-1: USB disconnect, address 2
BUG: atomic counter underflow at:
 [kref_put+89/160] kref_put+0x59/0xa0
 [klist_dec_and_del+27/32] klist_dec_and_del+0x1b/0x20
 [klist_release+0/80] klist_release+0x0/0x50
 [klist_remove+33/80] klist_remove+0x21/0x50
 [__device_release_driver+86/160] __device_release_driver+0x56/0xa0
 [device_release_driver+28/48] device_release_driver+0x1c/0x30
 [bus_remove_device+104/144] bus_remove_device+0x68/0x90
 [device_del+49/112] device_del+0x31/0x70
 [device_unregister+16/32] device_unregister+0x10/0x20
 [pg0+273833078/1069683712] hub_port_connect_change+0x56/0x450 [usbcore]
 [pg0+273822726/1069683712] clear_port_feature+0x56/0x60 [usbcore]
 [pg0+273834837/1069683712] hub_events+0x2e5/0x490 [usbcore]
 [pg0+273835264/1069683712] hub_thread+0x0/0xe0 [usbcore]
 [pg0+273835285/1069683712] hub_thread+0x15/0xe0 [usbcore]
 [autoremove_wake_function+0/96] autoremove_wake_function+0x0/0x60
 [kthread+182/192] kthread+0xb6/0xc0
 [kthread+0/192] kthread+0x0/0xc0
 [kernel_thread_helper+5/12] kernel_thread_helper+0x5/0xc
Unable to handle kernel NULL pointer dereference at virtual address 00000000
 printing eip:
c02f1477
*pde = 00000000
Oops: 0002 [#1]
PREEMPT 
last sysfs file: /class/usb_device/usbdev2.2/dev
Modules linked in: nfsd exportfs lockd sunrpc ipt_MASQUERADE iptable_nat ipt_state ip_conntrack nfnetlink ip table_filter ip_tables e100 mii yenta_socket rsrc_nonstatic pcmcia_core sd_mod snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mix er_oss snd_pcm snd_timer snd soundcore snd_page_alloc i2c_i801 i2c_core hw_random usb_storage scsi_mod reiser4 zlib_deflate zlib_inflate dm _mod rtc tun crc32 psmouse sonypi speedstep_ich speedstep_lib evdev pcspkr intel_agp agpgart usbhid uhci_hcd usbcore cpufreq_powersave
CPU:    0
EIP:    0060:[wait_for_completion+103/256]    Not tainted VLI
EFLAGS: 00010046   (2.6.13-rc6-mm2-1) 
EIP is at wait_for_completion+0x67/0x100
eax: cf6d449c   ebx: cecf8000   ecx: cecf8000   edx: 00000000
esi: cf6d4498   edi: cecf9ec0   ebp: cecf9ed4   esp: cecf9eb4
ds: 007b   es: 007b   ss: 0068
Process khubd (pid: 1303, threadinfo=cecf8000 task=cecb0a90)
Stack: 00000001 cecb0a90 c0115e50 cf6d449c 00000000 cf6d44c4 cf6d4464 d0919a20 
       cf3c6800 c0251cc6 cf6d4488 c03178ec cf6d4464 cf3c6864 c13d1414 c0251d2c 
       cf6d4464 cf6d4464 c0251458 cf6d4464 cf6d4464 cf6d4464 c02503c1 cf6d4464 
Call Trace:
 [default_wake_function+0/32] default_wake_function+0x0/0x20
 [__device_release_driver+86/160] __device_release_driver+0x56/0xa0
 [device_release_driver+28/48] device_release_driver+0x1c/0x30
 [bus_remove_device+104/144] bus_remove_device+0x68/0x90
 [device_del+49/112] device_del+0x31/0x70
 [device_unregister+16/32] device_unregister+0x10/0x20
 [pg0+273833078/1069683712] hub_port_connect_change+0x56/0x450 [usbcore]
 [pg0+273822726/1069683712] clear_port_feature+0x56/0x60 [usbcore]
 [pg0+273834837/1069683712] hub_events+0x2e5/0x490 [usbcore]
 [pg0+273835264/1069683712] hub_thread+0x0/0xe0 [usbcore]
 [pg0+273835285/1069683712] hub_thread+0x15/0xe0 [usbcore]
 [autoremove_wake_function+0/96] autoremove_wake_function+0x0/0x60
 [kthread+182/192] kthread+0xb6/0xc0
 [kthread+0/192] kthread+0x0/0xc0
 [kernel_thread_helper+5/12] kernel_thread_helper+0x5/0xc
Code: ec 00 00 00 00 c7 45 f0 00 00 00 00 8b 01 c7 45 e8 50 5e 11 c0 c7 45 e0 01 00 00 00 89 45 e4 8d 46 04 8b 50 04 89 45 ec 89 78 04 <89> 3a 89 55 f0 8d 74 26 00 8b 03 c7 00 02 00 00 00 fb 8b 43 08 
 <3>BUG: khubd[1303] exited with nonzero preempt_count 1!

-- 
mattia
:wq!

Re: Fw: Oops with 2.6.13-rc6-mm2 and USB mouse

[Ed Tomlinson]

On Saturday 27 August 2005 16:39, Mattia Dongili wrote:
> On Sat, Aug 27, 2005 at 05:09:04PM -0300, Rog???rio Brito wrote:
> > Hi, Andrew.
> > 
> > I just tested the USB mouse with 2.6.13-rc6-mm2 and ACPI disabled
> > (which, according to Linus, is one of the "usual suspects") and the
> > problem still occurred.
> 
> see here
> http://marc.theaimsgroup.com/?l=linux-kernel&m=112481438512222&w=2
> 
> Reverting driver-core-fix-bus_rescan_devices-race.patch and applying the
> patch attached to the above message fixed the oops for me.

I my case removing a usb device or hub triggers an immediate reboot with mm2.

This is fixed removing the above patch.  All is ok with mm1 too.  

I tried the above process but the second patch does not apply.  

Thanks
Ed Tomlinson