oops with 2.0.29

dave (tel1dvw@is.ups.com)
Thu, 27 Feb 1997 19:29:15 -0500 (EST)

Its best that I start a step back or so. I have a P120, 32 megs, two 2.1
gig scsi, one 630 ide, 200 watt power supply, aha2940 with the two scsi's,
and a 1510 with a Pioneer DRM-604x multi-disc changer. I suspected a bad
motherboard so I replaced it. Upon replacing it, I tried to reproduce the
problem (scsi and ide timeout) by running a script as follows:

while [ 1 ]
do
cp /usr/* /mnt/scsi1/mount &
cp /usr/* /mnt/scsi2/mount &
cp /usr/* /mnt/hda1_tmp &
cp /mnt/cdrom0 /mnt/scsi1/mount1 &
sleep 900
done

The problem went away, and figured I got it right by replacing the
motherboard. After letting it run all night last night, I found this
message scrolling on the screen, and the machine hosed:

wait_queue is bad (eip = 00112b11)

So this morning I shut the machine off, and now I am checking
/var/log/messages and I find:

general protection: 0000
CPU: 0
EIP: 0010:[verify_area+61/440]
EFLAGS: 00010286
eax: e80804b2 ebx: bffffa08 ecx: bffffa08 edx: 00000000
esi: bffffa08 edi: 00000040 ebp: bffff98c esp: 0077bf84
ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018
Process cp (pid: 714, process nr: 101, stackpage=0077b000)
Stack: bffffa08 08050328 00000000 bffff98c 00121328 0108d440 00128b7b
00000001
bffffa08 00000040 0033a018 080502f0 0033a018 0010a5f5 08050328
bffffa08
08050328 080502f0 00000000 bffff98c ffffffda 0000002b 0000002b
0000002b
Call Trace: [close_fp+76/92] [sys_newstat+23/88] [system_call+85/128]
Code: 39 70 08 76 0e 89 c2 39 72 04 76 0f 8b 42 14 eb eb 8d 36 8b
Unable to handle kernel paging request at virtual address ca5fa00c
current->tss.cr3 = 00101000, Lr3 = 00101000
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[exit_mmap+62/164]
EFLAGS: 00010202
eax: 0a5fa004 ebx: 00548558 ecx: 00548250 edx: 00548000
esi: 006a3c98 edi: 00000000 ebp: 0077bf48 esp: 0077bed8
ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018
Process cp (pid: 714, process nr: 101, stackpage=0077b000)
Stack: 006a3c98 00000014 0077c000 0011593b 006a3c98 0000002b 00000014
0077c000
0077bf48 0010aba3 0000000b 001b1ae0 00000000 bffffa08 00000040
bffff98c
0000008b 03000000 02800000 00000018 0010afb8 001b1b99 0077bf48
00000000
Call Trace: [do_exit+171/492] [die_if_kernel+695/704] [<03000000>]
[<02800000>]
[do_general_protection+40/84] [do_general_protection+0/84]
[error_code+64/80]
[verify_area+61/440] [close_fp+76/92] [sys_newstat+23/88]
[system_call+85
/128]
Code: 8b 48 08 85 c9 74 10 8b 53 04 8b 43 08 29 d0 50 52 53 ff d1
kfree of non-kmalloced memory: 00909598, next= 003c2000, order=1
1.3-3: restart.
klogd 1.3-3, log source = /proc/kmsg started.
Loaded 3646 symbols from /usr/src/linux/System.map.
Symbols match kernel version 2.0.29.
No module symbols loaded.

Please let me know what I need to do to fix this. Should I go back to
2.0.27, which came with RedHat?