Ooops in FDDI defxx driver

Marc MERLIN (marc_news@merlins.org)
6 Jul 1999 03:18:41 GMT


I contacted the driver author, Larry Stefani, but didn't hear back. Maybe
someone else has an idea of what's happening:

Here's a little info about my machine:
Pentium 133 VX chipset (single processor machine)
3 ethernet cards (2 ISA 10MBit and 1 PCI 100Mbit) and a PCI DEFPA FDDI card.

Since I upgraded from 2.0.34 to the 2.2.x series, your driver will randomly
oops at ifconfig time (the machine will boot fine 4 times out of 5). The
version the FDDI driver in 2.2.6 is 1.04 09/16/96, so some interface could
have changed in the 2.2.x series without the driver ever being updated.

Unable to handle kernel NULL pointer dereference at virtual address 00000080
current->tss.cr3 = 02dd9000, %cr3 = 02dd9000
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<c01aa551>]
EFLAGS: 00010246
eax: 00000000 ebx: 00000000 ecx: c0092000 edx: 00000080
esi: 00000004 edi: 00000068 ebp: 00000360 esp: c2d8fe78
ds: 0018 es: 0018 ss: 0018
Process ifconfig (pid: 88, process nr: 10, stackpage=c2d8f000)
Stack: 00000000 00000080 c0092000 c01fb2e0 c008d7a8 00001043 0000006c c01aa284
c0092000 c0092000 c01a93b2 c0092000 c0092000 c00921b4 c01a949f c0092000
00001002 c01fb2e0 00000000 c0153e7d c01fb2e0 00001002 c01fb2e0 00000000
Call Trace: [<c01aa284>] [<c01a93b2>] [<c01a949f>] [<c0153e7d>] [<c015492a>] [<c0174440>] [<c015516d>]
[<c0176253>] [<c014ff8d>] [<c012b488>] [<c014ff6c>] [<c0107a40>]
Code: 8b 8b 80 00 00 00 8d 44 0a ff f7 da 21 d0 29 c8 01 c1 89 8b

>>EIP: c01aa551 <unregister_netdev+14ad/27a4>
Trace: c01aa284 <unregister_netdev+11e0/27a4>
Trace: c01a93b2 <unregister_netdev+30e/27a4>
Trace: c01a949f <unregister_netdev+3fb/27a4>
Trace: c0153e7d <dev_open+25/80>
Trace: c015492a <dev_set_allmulti+82/5f0>
Trace: c0174440 <devinet_ioctl+250/d84>
Trace: c015516d <dev_ioctl+2d5/3d0>
Trace: c0176253 <inet_poll+387/99c>
Code: c01aa551 <unregister_netdev+14ad/27a4> 00000000 <_EIP>: <===
Code: c01aa551 <unregister_netdev+14ad/27a4> 0: 8b 8b 80 00 00 movl 0x80(%ebx),%ecx <===
Code: c01aa556 <unregister_netdev+14b2/27a4> 5: 00
Code: c01aa557 <unregister_netdev+14b3/27a4> 6: 8d 44 0a ff leal 0xffffffff(%edx,%ecx,1),%eax
Code: c01aa55b <unregister_netdev+14b7/27a4> a: f7 da negl %edx
Code: c01aa55d <unregister_netdev+14b9/27a4> c: 21 d0 andl %edx,%eax
Code: c01aa55f <unregister_netdev+14bb/27a4> e: 29 c8 subl %ecx,%eax
Code: c01aa561 <unregister_netdev+14bd/27a4> 10: 01 c1 addl %eax,%ecx
Code: c01aa563 <unregister_netdev+14bf/27a4> 12: 89 8b 00 00 00 movl %ecx,0x0(%ebx)
Code: c01aa568 <unregister_netdev+14c4/27a4> 17: 00

If the kernel doesn't oops at boot time (during the ifconfig command) then
everything works fine. I have another machine (K6/2 350 with another DEC
card (cddi this time) running 2.2.6 also and haven't experienced any such
problems yet..

While I can't unfortunately reproduce this bug reliably, I'll still be happy
to assist in debugging this if I can.

Best,
Marc

-- 
Microsoft is to software what McDonalds is to gourmet cooking
 
Home page: http://marc.merlins.org/ (friendly to non IE browsers)
Finger marc_f@merlins.org for PGP key and other contact information

- 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/