Failure of 1.3.6, Cyclades implicated again.

G.W. Wettstein (
Mon, 3 Jul 1995 17:49:32 CDT

Allan was probably right when he said the new buffer handling code in
the network might be unstable. We had 1.3.6 take out our kernel
development machine within a few minutes after booting the machine up.

Interestingly the machine let go the instant one of my user's tried
accessing the mainframe async server. For those of you following
along at home these are serial sessions serviced by a Cyclades board
that continually hangs when accessed over the network. This may be,
again happenstance, but I find it somewhat unnerving.

The following is typed in by hand from the screen since the kernel was
taken out so fast that klogd/syslogd didn't have a chance to capture
any of this:

Oops: 0000
EIP: 10:00138836
EFLAGS: 00010016

EAX: 00000000 EBX: 0000000d ECX: 00000016 EDX: 006dc000
ESI: 0000003c EDI: 00098098 EBP: 00000c00 ESP: 00199f30
ds: 0018 es: 0018 fs: 002b gs: 0018 ss: 0018

Process swapper (pid: 0, process nr: 0 stackpage = 00198030)

Stack: 0018a070 0000003c 00000001 001a79ac 00000290 00098098 00000216 001a79ac
ffff0290 00000034 00000001 0000000d 00000290 00400d01 00189ac7 001a79ac

Call Trace: 0018a070 00189ac7 00110297 0010ccb5 0010c072 00110018 001096e0
0010a97d 00190018

Code: 83 40 4c 10 83 40 50 10 c3 90 8b 44 24 04 80 78 3d 00 0f 95

Aiee, killing interrupt handler.

kfree of non-kmalloced memory: 0019a028, next = 00000018, order =

task[0] (swapper) killed: unable to recover.

kernel panic: Trying to free up swapper memory space.

In swapper task - not syncing.

Address decomposition:

00138824 T _dev_alloc_skb

Call Trace:
0018a070: 00189ee0 t _ei_receive
00189ac7: 001899b0 T _ei_interrupt
00110297: 0010fe3c t ksyms.o
0010ccb5: 0010cc98 T _do_IRQ
0010c072: 0010c01c t _IRQ5_interrupt
00110018: 0010fe3c t ksyms.o
001096e0: 00109688 T _sys_idle
0010a97d: 0010a924 T _system_call
00190018: 0018fdc4 T _get_address_16

If there is any other additional information that would be useful
please feel free to give me a holler. Take care, have a pleasant
remainder of the week.

As always,
Dr. G.W. Wettstein Oncology Research Div. Computing Facility
Roger Maris Cancer Center INTERNET:
820 4th St. N.
Fargo, ND 58122
Phone: 701-234-7556
`The truest mark of a man's wisdom is his ability to listen to other
men expound their wisdom.' -- GWW