another oops message

Tuukka Toivonen (tuukkat@stekt.oulu.fi)
Fri, 7 Nov 1997 14:59:40 +0200 (EET)


It seems that I can reproduce realiably either a kernel (2.0.31) crash or
complete hang in my system under heavy load, althought I tried to compile
as stable kernel as possible (ie. using standard gcc 2.7.2, not any extra
patches, and all unnecessary features off or as unused modules)

The kernel complete crash happened when I had very low free memory (4 megs
free swap or so, 64 megs total swap, 24 megs main RAM). First X crashed,
then I tried to restore textmode from serial terminal, but then kernel just
suddenly crashed... only big reset button helped. No oopses, nothing.

The oops message happened also under quite heavy load. When I got it, I
started reading /usr/src/linux/Documentation/oops-tracing.txt. At first it
suggested running gdb /usr/linux/vmlinux. Unfortunately one of my processes
was "make clean" which had deleted vmlinux so that didn't work :(

Now the question is: at which point I should report the oopses or other
kernel bug-messages on this list? I know that there's high volume of
message in this list and I definitely don't want to make the Kernel
Hackers' life any harder. If I just get some oops, is it worth of posting?
How much I should gather/give extra information? Should I stay away from
this list? ;-)

Anyway, there's my last oops message... hope you don't mind :)

------ I was running kernel 'make clean' among other things:
rm -f core `find . -type f -name 'core' -print`
Unable to handle kernel paging request at virtual address c48c0016
current->tss.cr3 = 01446000, %cr3 = 01446000
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<00110614>]
EFLAGS: 00010206
eax: 0000ffee ebx: 048c0016 ecx: 00d46538 edx: 007006f4
esi: 0018f950 edi: 00d46534 ebp: 0146eee4 esp: 0146eed8
ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018
Process find (pid: 810, process nr: 36, stackpage=0146e000)
Stack: 00d464ec 0018f950 00d464ec 0019b088 00122b05 00d46538 00d580fc 0146ef74
00000004 002c6000 00151f9d 00151fd0 0019b088 00012730 00000001 00d580fc
0146ef74 0146ef74 00000004 00d580fc 00012730 00129946 00d580fc 002c6000
Call Trace: [<00122b05>] [<00151f9d>] [<00151fd0>] [<00129946>] [<00129b5f>] [<00129c18>] [<00127dd9>]
[<0010a5e5>]
Code: 8b 13 8b 5b 04 85 d2 74 76 8b 02 83 f8 02 74 07 8b 02 83 f8
rm -f vmlinux System.map
----------
After this message I ran top:
----------
8:37pm up 31 min, 11 users, load average: 7.10, 7.16, 5.53
58 processes: 52 sleeping, 6 running, 0 zombie, 0 stopped
CPU states: 59.9% user, 27.7% system, 7.1% nice, 12.7% idle
Mem: 23112K av, 20420K used, 2692K free, 11692K shrd, 192K buff
Swap: 65480K av, 25052K used, 40428K free 3392K cached

Oh, I've Slackware 3.1 system with many parts upgraded (libc, binutils
etc), Pentium 120MHz x86 system, Triton chipset... chipset support was NOT
compiled into kernel, IDE hard disks.

| Tuukka Toivonen <tuukkat@stekt.oulu.fi>
| Homepage: http://stekt.oulu.fi/~tuukkat/
| Try also finger -l tuukkat@stekt.oulu.fi
| Studying information engineering at the University of Oulu
+-----------------------------------------------------------