Skip to contentSkip to conversation

Removing USB Bluetooth dongle Oopses 2.6.4

OOPS Silla Rizzoli 2 participants 3 messages Mar 13, 2004 – Mar 14, 2004

lore
Share & tools
lore
#0 lore raw

Hello lkml!

Here comes the oops:

PREEMPT
CPU: 0
EIP: 0060:[<e1921149>] Not tainted
EIP is at urb_unlink+0x31/0x8e [usbcore]
eax: c17c069c ebx: 00000000 ecx: 00000000 edx: dec74000
esi: 00000246 edi: c17c0694 ebp: de5aa1f0 esp: dec75e14
ds: 007b es: 007b ss: 0068
Process hotplug (pid: 6150, threadinfo=dec74000 task=dfa6a100)

Stack: c17c0694 de5aa000 dec75edc c17c0694 dec75edc de5aa000 e19218af c17c0694
       c17c0694 dec74000 c17c0694 e19f5db4 de5aa000 c17c0694 dec75edc dec75edc
       de5aa1d0 de5aa1d0 de5aa000 dec75edc e19f5f72 de5aa000 dec75edc 00701300
Call trace:
 [<e19218af>] usb_hcd_giveback_urb+0x1b/0x39 [usbcore]
 [<e19f5db4>] uhci_finish_completion+0x61/0x9c [uhci_hcd]
 [<e19f5f72>] uhci_irq+0x103/0x165 [uhci_hcd]
 [<e1921903>] usb_hcd_irq+0x36/0x67 [usbcore]
 [<c010ae8a>] handle_IRQ_event+0x3a/0x64
 [<c010b1f7>] do_IRQ+0x94/0x136
 [<c0116eea>] do_page_fault+0x0/0x50c
 [<c0109750>] common_interrupt+0x18/0x20
 [<c0116eea>] do_page_fault+0x0/0x50c
 [<c0116f11>] do_page_fault+0x27/0x50c
 [<c0141c0b>] unmap_vmas+0xdc/0x212
 [<c01451b7>] unmap_vma+0x40/0x7d
 [<c0145210>] unmap_vma_list+0x1c/0x28
 [<c0145625>] do_munmap+0x146/0x183
 [<c0116eea>] do_page_fault+0x0/0x50c
 [<c010980d>] error_code+0x2d/0x38

Code: 89 59 04 89 0b 89 40 04 89 47 08 8b 5f 14 56 9d 8b 42 08 83
 <0>Kernel panic: Fatal exception in interrupt

In interrupt handler - not syncing

Here is the modules list:

Module                  Size  Used by
rfcomm                 33820  0 
l2cap                  21124  3 rfcomm
hci_usb                 9728  1 
bluetooth              43108  3 rfcomm,l2cap,hci_usb
ide_cd                 37636  0 
cdrom                  35872  1 ide_cd
ds                     11524  2 
hid                    23040  0 
intel_agp              15772  1 
agpgart                25896  1 intel_agp
uhci_hcd               29200  0 
ehci_hcd               23428  0 
snd_intel8x0           28804  0 
snd_ac97_codec         58756  1 snd_intel8x0
snd_mpu401_uart         6272  1 snd_intel8x0
snd_rawmidi            20128  1 snd_mpu401_uart
snd_seq_oss            31488  0 
snd_seq_midi_event      6272  1 snd_seq_oss
snd_seq                50192  4 snd_seq_oss,snd_seq_midi_event
snd_seq_device          6536  3 snd_rawmidi,snd_seq_oss,snd_seq
snd_pcm_oss            48164  0 
snd_pcm                84900  2 snd_intel8x0,snd_pcm_oss
snd_page_alloc          9348  2 snd_intel8x0,snd_pcm
snd_timer              21380  2 snd_seq,snd_pcm
snd_mixer_oss          17280  1 snd_pcm_oss
snd                    45796  12 
snd_intel8x0,snd_ac97_codec,snd_mpu401_uart,snd_rawmidi,snd_seq_oss,snd_seq_midi_event,snd_seq,snd_seq_device,snd_pcm_oss,snd_pcm,snd_timer,snd_mixer_oss
soundcore               7392  1 snd
rtc                    10552  0 
usbcore                89820  6 hci_usb,hid,uhci_hcd,ehci_hcd
md5                     3712  1 
ipv6                  228160  11 
fan                     4236  0 
button                  6296  0 
thermal                16400  0 
processor              24496  1 thermal
battery                10636  0 
ac                      5004  0 
e100                   29056  0 
yenta_socket           14336  0 
pcmcia_core            59872  2 ds,yenta_socket
unix                   23472  12 

You'll find the kernel config attached:

Steps to reproduce the problem:

1) Insert the usb bt dongle
2) Start the bluetooth services (rfcomm, hcid...)
3) Stop the bluetooth services
4) Extrach the dongle - Doing rmmod uhci_hcd also causes the same oops.

Notice that SCO BT support is NOT compiled into the kernel, so this is a different oops than similar ones reported before.
After step 3 it is possible to remove rfcomm and l2cap, but hci_usb and bluetooth seem to lock one another, making it impossible to remove them from the running kernel.

Regards,
Silla

#1 lore raw

Hi Silla,

Quoted 33 lines

Here comes the oops:

PREEMPT
CPU: 0
EIP: 0060:[<e1921149>] Not tainted
EIP is at urb_unlink+0x31/0x8e [usbcore]
eax: c17c069c ebx: 00000000 ecx: 00000000 edx: dec74000
esi: 00000246 edi: c17c0694 ebp: de5aa1f0 esp: dec75e14
ds: 007b es: 007b ss: 0068
Process hotplug (pid: 6150, threadinfo=dec74000 task=dfa6a100)

Stack: c17c0694 de5aa000 dec75edc c17c0694 dec75edc de5aa000 e19218af c17c0694
       c17c0694 dec74000 c17c0694 e19f5db4 de5aa000 c17c0694 dec75edc dec75edc
       de5aa1d0 de5aa1d0 de5aa000 dec75edc e19f5f72 de5aa000 dec75edc 00701300
Call trace:
 [<e19218af>] usb_hcd_giveback_urb+0x1b/0x39 [usbcore]
 [<e19f5db4>] uhci_finish_completion+0x61/0x9c [uhci_hcd]
 [<e19f5f72>] uhci_irq+0x103/0x165 [uhci_hcd]
 [<e1921903>] usb_hcd_irq+0x36/0x67 [usbcore]
 [<c010ae8a>] handle_IRQ_event+0x3a/0x64
 [<c010b1f7>] do_IRQ+0x94/0x136
 [<c0116eea>] do_page_fault+0x0/0x50c
 [<c0109750>] common_interrupt+0x18/0x20
 [<c0116eea>] do_page_fault+0x0/0x50c
 [<c0116f11>] do_page_fault+0x27/0x50c
 [<c0141c0b>] unmap_vmas+0xdc/0x212
 [<c01451b7>] unmap_vma+0x40/0x7d
 [<c0145210>] unmap_vma_list+0x1c/0x28
 [<c0145625>] do_munmap+0x146/0x183
 [<c0116eea>] do_page_fault+0x0/0x50c
 [<c010980d>] error_code+0x2d/0x38

Code: 89 59 04 89 0b 89 40 04 89 47 08 8b 5f 14 56 9d 8b 42 08 83
 <0>Kernel panic: Fatal exception in interrupt
Quoted 1 line

In interrupt handler - not syncing

this looks like another unlink-during-submit bug in the uhci-hcd host driver. With the latest 2.6.4-bk2 the ohci-hcd is now free from it and an unplug works again without any oops or freezes. Post your oops to the USB developer mailing list.

Regards

Marcel

#2 lore raw
Quoted 8 lines

this looks like another unlink-during-submit bug in the uhci-hcd host driver. With the latest 2.6.4-bk2 the ohci-hcd is now free from it and an unplug works again without any oops or freezes. Post your oops to the USB developer mailing list.

Regards

Marcel

I've just tried 2.6.4-bk3 but the oops is still there; I'll forward it to the usb mailing list.

Regards, Silla