Many apologies, it's late and I ran the default ksymoops (0.7c) instead of
./ksymoops
However, after running the real ksymoops against it, it gives nothing
new; I suppose since 'ls' is dying and the stack is from 'ls', that's why.
Is there anything else I should do before booting back to a 2.2.17 kernel?
As proof that I ran the right ksymoops this time:
ksymoops 2.3.5 on i686 2.4.0-test10. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.4.0-test10/ (default)
-m /boot/System.map-2.4.0-test10 (specified)
Nov 8 00:11:42 chewbacca kernel: Oops: 0000
Nov 8 00:11:42 chewbacca kernel: CPU: 0
Nov 8 00:11:42 chewbacca kernel: EIP: 0010:[usb_stor_exit+12588154/-1072693296]
Nov 8 00:11:42 chewbacca kernel: EFLAGS: 00010206
Nov 8 00:11:42 chewbacca kernel: eax: 00c014aa ebx: cf0a1fa4 ecx: da155940 edx: c4a70440
Nov 8 00:11:42 chewbacca kernel: esi: 00000000 edi: bffff020 ebp: bffff00c esp: cf0a1f94
Nov 8 00:11:42 chewbacca kernel: ds: 0018 es: 0018 ss: 0018
Nov 8 00:11:42 chewbacca kernel: Process ls (pid: 9036, stackpage=cf0a1000)
Nov 8 00:11:42 chewbacca kernel: Stack: c0135dd7 c4a70440 cf0a0000 bffff020 c4a70440 eff83540 00000003 bfffe830
Nov 8 00:11:42 chewbacca kernel: bfffe81c 00000008 00000001 c010a4fb bffff020 080549ac 4010fd60 bffff020
Nov 8 00:11:42 chewbacca kernel: bffff020 bffff00c 000000c4 0000002b 0000002b 000000c4 400c4e1c 00000023
Nov 8 00:11:42 chewbacca kernel: Call Trace: [sys_lstat64+55/112] [system_call+51/56]
Nov 8 00:11:42 chewbacca kernel: Code: Bad EIP value.
Using defaults from ksymoops -t elf32-i386 -a i386
- Matt
-----------------------------------------------------------------------
Matthew Hanselman QUIQ Inc.
(608) 230-7205 (voice) 25 Kessel Ct #201
(608) 230-7299 (fax) Madison, WI 53711
On Wed, 8 Nov 2000, Keith Owens wrote:
> On Wed, 8 Nov 2000 00:38:33 -0600 (CST),
> Matthew Hanselman <mjhans@quiq.com> wrote:
> >I can poke around tomorrow morning without a reboot, but then I'll have to
> >reboot (so please respond via email if I can do anything). I tried
> >grinding the message through ksymoops-2.3.5, and it complains with this
> >error:
> >
> >Warning (Oops_code): trailing garbage ignored on Code: line
> > Text: 'Code: Bad EIP value. '
> > Garbage: 'IP value. '
> >Using defaults from ksymoops -t elf32-i386 -a i386
> >Error (Oops_code_values): invalid value 0xBad in Code line, must be 2, 4, 8 or 1
> >Error (Oops_code_values): invalid value 0xE in Code line, must be 2, 4, 8 or 16
>
> Are you sure that is ksymoops 2.3.5? My version of 2.3.5 handles
> 'Code: Bad EIP value. '. ksymoops >= 2.3.4 should handle it correctly.
>
> >Nov 8 00:11:42 chewbacca kernel: Unable to handle kernel paging request at virtual address 00c014aa
> >Nov 8 00:11:42 chewbacca kernel: printing eip:
> >Nov 8 00:11:42 chewbacca kernel: 00c014aa
> >Nov 8 00:11:42 chewbacca kernel: *pde = 00000000
> >Nov 8 00:11:42 chewbacca kernel: Oops: 0000
> >Nov 8 00:11:42 chewbacca kernel: CPU: 0
> >Nov 8 00:11:42 chewbacca kernel: EIP: 0010:[usb_stor_exit+12588154/-1072693296]
> >Nov 8 00:11:42 chewbacca kernel: EFLAGS: 00010206
> >Nov 8 00:11:42 chewbacca kernel: eax: 00c014aa ebx: cf0a1fa4 ecx: da155940 edx: c4a70440
> >Nov 8 00:11:42 chewbacca kernel: esi: 00000000 edi: bffff020 ebp: bffff00c esp: cf0a1f94
> >Nov 8 00:11:42 chewbacca kernel: ds: 0018 es: 0018 ss: 0018
> >Nov 8 00:11:42 chewbacca kernel: Process ls (pid: 9036, stackpage=cf0a1000)
> >Nov 8 00:11:42 chewbacca kernel: Stack: c0135dd7 c4a70440 cf0a0000 bffff020 c4a70440 eff83540 00000003 bfffe830
> >Nov 8 00:11:42 chewbacca kernel: bfffe81c 00000008 00000001 c010a4fb bffff020 080549ac 4010fd60 bffff020
> >Nov 8 00:11:42 chewbacca kernel: bffff020 bffff00c 000000c4 0000002b 0000002b 000000c4 400c4e1c 00000023
> >Nov 8 00:11:42 chewbacca kernel: Call Trace: [sys_lstat64+55/112] [system_call+51/56]
> >Nov 8 00:11:42 chewbacca kernel: Code: Bad EIP value.
>
> sys_lstat64+55 has branched to the value in eax and blown up. It looks like
> the data in eax is a kernel address shifted right one byte.
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Nov 15 2000 - 21:00:11 EST