scsiinfo -d kills the kernel

[Ulrich Windl]

As I have been told to supply some details, I crashed my system again 
yesterday. The report is not complete, because I had to write all the 
information to a sheet of paper, and after several lines I simply 
stopped (having in mind that I did no longer have the symbol table 
for that kernel).

The kernel used was 2.0.30 with some pre-patches and the aic7xxx 
4.009. My adapter is a 2940, but I think it does not matter here:

(scsi0:0:0) Data overrun of 16773218 bytes detected in Data-In phase, tag 2; forcing a retry
Have seen Data Phase. Length = 4096, NumSGs=1.
	sg[0] - Addr 0x81000 : Length 4096
#above message (3 lines) repeats again#
SCSI0 channel 0 : resetting for second half of retries.
SCSI bus is being reset for host 0 channel 0.
(scsi0:-1:0) Reset device, active_scb 2
SCSI0: Resetting current channel A
SCSI0: Channel reset, sequencer restarted
Unable to handle kernel paging request at virtual address c5e7024b
current->tss.cr3 = 00101000, %cr3 = 00101000
*pde = 00000000
Oops: 0002
CPU: 0
EIP: 0010:[<0019a303>]
EFLAGS: 00010006
eax: 05e70200  ebx: 00000202  ecx: 0000e000
edx: 00082018  esi: 0008a410  edi: 000826c
ebp: 00000001  esp: 001e3850  ds: 18  es: 18
fs: 2b  gs: 18  ss: 18
Process swapper (pid 0, process nr:0, stackpage=001e19bc)
stack: 0019a4c7 0008a410 ...
...
code: 80 48 4b c0 8b 52 10 85 d2 75  f2 c3 90 8b 44 24 04 8b 4c 24
Aiee, killing interrupt handler
kfree of non-kmalloced memory 001e3a04, next=00000000, order=0
# two more similar messages #
idle task may not sleep
# repeats 5 times #
SCSI host 0 channel 0 reset (pid 867) timed out - trying harder
SCSI bus is being reset for host 0 channel 0
(scsi0:-1:0) Reset device, active_scb 2
(scsi0: (targ -1/chan A) matching scb to (targ 0/chan A)
# above line repeats once #
SCSI0: resetting current channel A
SCSI0: Channel A reset, sequencer restarted
(scsi0:0:0) Aborting scb 2
Unable to handle...
...
SCSI host 0 reset (pid 867) timed out again - probably an unrecoverable SCSI bus or device 
hang.

# here the system is dead #

# Pressing ^ScrollLock shows that bas and scsiinfo only have 4 bytes of free stack; kswapd has 
# 3900

# Pressing C-A-D gives:
SCSI0: Warning!! Interrupt routine called reentrantly!
#last message repeats fast and forever #

I hope I have pointed out that there is some problem with "read defect list" in
the current kernel. Hopefully it can be fixed in a proper way.

Ulrich