[bug] killing cardmgr causes oops, 2.3.99-pre6-3, ray_cs driver

From: David Ford (david@kalifornia.com)
Date: Fri Apr 21 2000 - 19:03:17 EST


Starting conditions:

    running system, eth0 (aviator 2.2 wireless, ray_cs incore) up
    cardmgr running, daemons, etc

running "init 1" causes the below:

Apr 21 15:58:55 nifty init: Switching to runlevel: 1
Apr 21 15:58:55 nifty exiting on signal 15
Unable to handle kernel paging request at virtual address cc861001
 printing eip:
c01dd9c8
*pde = 0bfca063
*pte = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<c01dd9c8>]
EFLAGS: 00010212
eax: cc861001 ebx: c0d53d20 ecx: 00000003 edx: c0b1dbf4
esi: cbf12400 edi: cbf2da00 ebp: c0375480 esp: c0b1db8c
ds: 0018 es: 0018 ss: 0018
Process cardmgr (pid: 54, stackpage=c0b1d000)
Stack: cc861001 c028bf0c 0000073c c0d53d20 00000001 00000003 c0375480
c73248e0
       c010bc0d 00000003 cbf2da00 c0b1dbf4 00000000 c0301960 00000003
c0b1dbec
       c010bd72 00000003 c0b1dbf4 c0d53d20 00000246 0000003e cbfe4c00
c0d53d20
Call Trace: [<cc861001>] [<c028bf0c>] [<c010bc0d>] [<c010bd72>]
[<c010adb8>] [<c01cf21c>] [<c01d4b32>]
       [<c01d5f7c>] [<c01d1541>] [<c01d2def>] [<c01dc307>] [<c01db88e>]
[<c01d0bfa>] [<c01d1424>] [<c01e55e9>]
       [<c017c2fa>] [<c017fa96>] [<c017f319>] [<c011491e>] [<c01093ae>]
[<c011491e>] [<c01221dc>] [<c012252b>]
       [<c013b763>] [<c010ad10>]
Code: 8a 00 83 c4 0c 84 c0 0f 8c f9 02 00 00 3c 3f 0f 87 17 02 00

>>EIP; c01dd9c8 <ray_interrupt+3c/354> <=====
Trace; cc861001 <END_OF_CODE+c4dce71/????>
Trace; c028bf0c <exponent+10d8/434e>
Trace; c010bc0d <handle_IRQ_event+35/60>
Trace; c010bd72 <do_IRQ+6e/b4>
Trace; c010adb8 <ret_from_intr+0/20>
Trace; c01cf21c <pci_write_config_word+30/40>
Trace; c01d4b32 <yenta_set_socket+1ea/228>
Trace; c01d5f7c <pci_set_socket+24/30>
Trace; c01d1541 <set_socket+15/1c>
Trace; c01d2def <pcmcia_release_configuration+bf/11c>
Trace; c01dc307 <ray_release+8b/bc>
Trace; c01db88e <ray_detach+5e/d0>
Trace; c01d0bfa <unbind_request+6a/c0>
Trace; c01d1424 <ds_ioctl+500/5cc>
Trace; c01e55e9 <vgacon_cursor+181/188>
Trace; c017c2fa <set_cursor+6e/84>
Trace; c017fa96 <con_flush_chars+12/18>
Trace; c017f319 <do_con_write+5bd/6b0>
Trace; c011491e <schedule+25a/3c4>
Trace; c01093ae <__down+19e/1a8>
Trace; c011491e <schedule+25a/3c4>
Trace; c01221dc <unmap_fixup+70/134>
Trace; c012252b <do_munmap+207/214>
Trace; c013b763 <sys_ioctl+15f/184>
Trace; c010ad10 <system_call+34/38>
Code; c01dd9c8 <ray_interrupt+3c/354>
00000000 <_EIP>:
Code; c01dd9c8 <ray_interrupt+3c/354> <=====
   0: 8a 00 mov (%eax),%al <=====
Code; c01dd9ca <ray_interrupt+3e/354>
   2: 83 c4 0c add $0xc,%esp
Code; c01dd9cd <ray_interrupt+41/354>
   5: 84 c0 test %al,%al
Code; c01dd9cf <ray_interrupt+43/354>
   7: 0f 8c f9 02 00 00 jl 306 <_EIP+0x306> c01ddcce
<ray_interrupt+342/354>
Code; c01dd9d5 <ray_interrupt+49/354>
   d: 3c 3f cmp $0x3f,%al
Code; c01dd9d7 <ray_interrupt+4b/354>
   f: 0f 87 17 02 00 00 ja 22c <_EIP+0x22c> c01ddbf4
<ray_interrupt+268/354>

(repeating oopses for a magic period of time omitted)

However, if I first do:

cardctl suspend
cardctl eject
kill <nn> (pid of cardmgr)

then it won't oops. but i lose my keyboard and mouse when i pull the
card out. effectively hanging the machine. the hardware Fn keys still
work. this is on an NEC versa LX laptop.

-d

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sun Apr 23 2000 - 21:00:19 EST