2.6.3-bk5 isdn oops
More modprobe/rmmod fun.. (no card present).
Dave
ISDN subsystem Rev: 1.1.2.3/1.1.2.3/1.1.2.2/1.1.2.3/1.1.2.2/1.1.2.2 loaded
ICN-ISDN-driver Rev 1.65.6.8 mem=0x000d0000
icn: (line0) ICN-2B, port 0x320 added
Debug: sleeping function called from invalid context at include/linux/rwsem.h:43in_atomic():0, irqs_disabled():1
Call Trace: [<c0123214>] __might_sleep+0x80/0x8a [<c011df07>] do_page_fault+0x71/0x46c [<c7b0ffd7>] isdn_status_callback+0x8e0/0x968 [isdn] [<c0121447>] schedule+0x5a6/0x631 [<c01213c8>] schedule+0x527/0x631 [<c011de96>] do_page_fault+0x0/0x46c [<c010c165>] error_code+0x2d/0x38 [<c79aba66>] icn_exit+0xb4/0x155 [icn] [<c01214d2>] default_wake_function+0x0/0xc [<c0138926>] kthread_stop+0x83/0xa0 [<c013ab6f>] try_stop_module+0x94/0x9c [<c013ad14>] sys_delete_module+0x11b/0x13d [<c015063c>] unmap_vma_list+0xe/0x17 [<c0150aea>] do_munmap+0x17e/0x18a [<c010b663>] syscall_call+0x7/0xb Unable to handle kernel paging request at virtual address 0000297c printing eip: c79aba66 *pde = 00000000 Oops: 0000 [#1] SMP CPU: 0 EIP: 0060:[<c79aba66>] Not tainted EFLAGS: 00010046
EIP is at icn_exit+0xb4/0x155 [icn]
eax: 00000000 ebx: 00000000 ecx: 00000000 edx: 00000001
esi: 00000000 edi: 00000292 ebp: c0320298 esp: c2323ee4
ds: 007b es: 007b ss: 0068
Process rmmod (pid: 1327, threadinfo=c2322000 task=c4ae87d0)
Stack: c111acc0 c2323f20 00000000 0000010a c01214d2 00100100 00200200 00000000
00000007 00000286 00000000 00000246 c21e1270 c032010c c2323f58 c79ad900
c0138926 c21e1270 c2323f44 c013ab6f c2323f6c c79ad900 00000880 c2323f6c
Call Trace:
[<c01214d2>] default_wake_function+0x0/0xc
[<c0138926>] kthread_stop+0x83/0xa0
[<c013ab6f>] try_stop_module+0x94/0x9c
[<c013ad14>] sys_delete_module+0x11b/0x13d
[<c015063c>] unmap_vma_list+0xe/0x17
[<c0150aea>] do_munmap+0x17e/0x18a
[<c010b663>] syscall_call+0x7/0xb
Code: 81 bb 7c 29 00 00 ad 4e ad de 74 08 0f 0b 5d 00 b3 bb 9a c7Dave Jones <[email protected]> :
[...]
Quoted 4 lines
ISDN subsystem Rev: 1.1.2.3/1.1.2.3/1.1.2.2/1.1.2.3/1.1.2.2/1.1.2.2 loaded
ICN-ISDN-driver Rev 1.65.6.8 mem=0x000d0000
icn: (line0) ICN-2B, port0x320added
Debug: sleeping function called from invalid context at include/linux/rwsem.h:43in_atomic():0, irqs_disabled():1
Quoted 9 lines
Call Trace: [<c0123214>] __might_sleep+0x80/0x8a [<c011df07>] do_page_fault+0x71/0x46c [<c7b0ffd7>] isdn_status_callback+0x8e0/0x968 [isdn] [<c0121447>] schedule+0x5a6/0x631 [<c01213c8>] schedule+0x527/0x631 [<c011de96>] do_page_fault+0x0/0x46c [<c010c165>] error_code+0x2d/0x38 [<c79aba66>] icn_exit+0xb4/0x155 [icn]
drivers/isdn/i4l/isdn_common.c
745 isdn_status_callback(isdn_ctrl *c)
[...]
767 case ISDN_STAT_UNLOAD:
768 rc = fsm_event(&drv->fi, EV_STAT_UNLOAD, c);
[...]
886 put_drv(drv);
^^^ -> let's name this one "catherine"drivers/isdn/i4l/isdn_common.c
630 drv_stat_unload(struct fsm_inst *fi, int pr, void *arg)
[...]
638 put_drv(drv);
^^^ -> let's name this one "therese"One has:
therese = catherine->fi->userdata
Given:
drivers/isdn/i4l/isdn_common.c
896 register_isdn(isdn_if *iif)
[...]
915 drv->fi.userdata = drv;One checks that:
catherine = catherine->fi->userdata
So catherine = therese and there is a double put_drv() on the same pointer. put_drv(drv)
-> drv_destroy(drv)
-> kfree(drv->slots);
Oops.
--
Ueimor