Messages in this thread | | | Date | Sat, 18 Apr 1998 15:57:49 -0500 | From | David Fries <> | Subject | 2.1.97 oops |
| |
In 2.1.97 I was running bonnie on my harddrive at the time when it locked up, SysRQ was worning, but since the lockup I didn't see any harddrive access, it did not appear to sync or remount the file system as readonly I did write down some of the current and registers.
EIP: 0010:[<c0126b2f>] EFLAGS: 00000202 EAX: 00001027 EBX: c3062a20 ECX: c01f13ac EDX: c3062a20 ESI: 00001a26 EDI: 00000002 EBP: bffffd90 DS: 0018 ES: 0018
I listed a few of them on the computer and only EIP, EFLAGS and EAX were changing, I only wrote down the EIP. Only the last three values were changing. EIP: 0010:[<c0126b2f>] aa0 b2f aa3 a96 b31 b2f
It appeared to come up with only those values. In the System.map, c0126108 T brw_page c0126478 T mark_buffer_uptodate c01264cc T generic_readpage c0126558 t grow_buffers c0126644 T try_to_free_buffer c012685c T show_buffers c0126938 T buffer_init c0126a04 T wakeup_bdflush c0126a60 T sync_old_buffers c0126ba0 T sys_bdflush c0126c38 T bdflush c0126e50 T lookup_vfsmnt c0126e98 t add_vfsmnt c0126fcc t remove_vfsmnt
I rebooted with SysRQ and ran 2.1.97 again, in e2fsck it gave an oops, from the ksymoops output it looks like I must have copied an address wrong, sorry, if the kernel was running right I wouldn't have had to copy by hand :)
Using `/mnt/hda1/2.1.97-map' to map addresses to symbols.
>>EIP: c73dc2c6 cannot be resolved Trace: c0106000 <this_must_match_init_task> Trace: c0107f9c <this_must_match_init_task+1f9c/2000> Trace: c0107f9c <this_must_match_init_task+1f9c/2000> Trace: c0107fe4 <this_must_match_init_task+1fe4/2000> Trace: c0109880 <system_call+38/3c> Trace: c0107fe4 <this_must_match_init_task+1fe4/2000> Trace: c0108073 <cpu_idle+7/18> Trace: c010025b <L6> Code: Code: 1b 00 sbbl (%eax),%eax Code: 00 04 00 addb %al,(%eax,%eax,1) Code: 00 13 addb %dl,(%ebx) Code: 08 13 orb %dl,(%ebx) Code: 08 02 orb %al,(%edx) Code: 00 36 addb %dh,(%esi) Code: 00 60 c2 addb %ah,0xffffffc2(%eax) Code: 3d c7 09 00 00 cmpl $0x9c7,%eax Code: 90 nop Code: 90 nop Code: 90 nop Cannot read eip address from EIP: line. Is this a valid oops file?
date: 04-18-1998,15:15 kernel 2.1.97 Unable to handle kernel NULL pointer dereference at virtual address 00000000 current->tss.cr3 = 00101000, %cr3 = 00101000 *pde = 00000000 Oops: 0000 CPU: 0 EIP: 0010:[<c73dc2c6>] EFLAGS: 00010246 eax: 00000000 ebx: 00000000 ecx: c0106000 edx: 00000023 esi: 00000000 edi: 0001c7fc ebp: c7fcbf90 esp: c0107f7c ds: 0018 es: 0018 ss: 0018 Process swapper (pid: 0, procss nr: 0, stackpage=c0107000) Stack: c0106000 00000000 c7fcbf90 c0107f9c 00000000 00000023 c0106000 00000000 c59da960 c0107fe4 c0109880 00000f00 c0108084 00000000 c0107fe4 c0106000 00009000 00000070 00000018 00000018 00000070 c0108073 00000010 00000246 Call Trac: [<c0106000>] [<c0107f9c>] [<c0106000>] [<c0107fe4>] [<c0109880>] [<c0106000>] [<c0108073>] [<c010025b>] Code: 1b 00 00 04 00 00 13 08 13 08 02 00 36 00 60 c2 3d c7 09 00 Kernel panic: Attempted to kill the idle task! In swapper task - not syncing SysRq: Emergency Sync Syncing device 08:13 ... OK Done.
AMD K5 Tyan 1653D, Tomcat III Intel 82439HX, 82371SB chipset 128 megs memory, NCR53C810 using BSD ported driver gcc 2.7.2.3
-- +---------------------------------+ | David Fries | | dfries@mail.win.org | +---------------------------------+ [unhandled content-type:application/pgp-signature] | |