xfs
[Top] [All Lists]

Re: Oops with 2.4.16

To: Pascal Haakmat <a.haakmat@xxxxxxxxx>
Subject: Re: Oops with 2.4.16
From: Steve Lord <lord@xxxxxxx>
Date: 10 Jan 2002 15:29:14 -0600
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <20020110221155.A912@xxxxxxxxxxxxx>
References: <20020110221155.A912@xxxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
Could you do this please:

cd linux
rm fs/xfs/xfs_vfsops.o
make CFLAGS_xfs_vfsops.o=-g vmlinux
s=$(sed -ne '/xfs_syncsub/s/ .*//p' System.map | tr '[a-z]' '[A-Z]')
e=$(echo -e "obase=16\nibase=16\n$s+3112" | bc)
objdump -S --start-address=0x$s --stop-address=0x$e vmlinux

And send me the output, if I got that right it will generate a mixed
source and disassembly listing of the function.

Steve

On Thu, 2002-01-10 at 15:11, Pascal Haakmat wrote:
> Another Oops today.
> 
> Background: these oopses always happen when I am about to write to a file;
> usually when I'm about to record audio.
> 
> Kernel version: 2.4.16
> 
> GCC version: gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)
> 
> Hardware: Dual PIII at 600MHz, 440GX SuperMicro P6DGU mainboard, 2 IDE disks
> mounted as XFS volumes, SCSI CDROM & CDRW on built-in AIC7890 (not in use at
> the time), 512MB of RAM.
> 
> I can reproduce Oopses like this quite reliably. They usually occur after a
> day or so but sometimes they happen earlier. They happen with or without
> low-latency patches, and have also occurred with kernel 2.4.5/XFS release
> 1.0.1, and kernel 2.4.14/XFS release 1.0.2.
> 
> After this Oops has occurred, the machine usually continues to be
> somewhat usable, but data can no longer be written; that is, logging in or
> executing 'sync' or rebooting properly are not possible.
> 
> ksymoops output:
> 
> ksymoops 2.3.4 on i686 2.4.16-xfs-ll.  Options used
>      -V (default)
>      -k /proc/ksyms (default)
>      -l /proc/modules (default)
>      -o /lib/modules/2.4.16-xfs-ll/ (default)
>      -m /usr/src/linux/System.map (default)
> 
> Warning: You did not tell me where to find symbol information.  I will
> assume that the log matches the kernel and modules that are running
> right now and I'll use the default options above for symbol resolution.
> If the current kernel and/or modules do not match the log, you can get
> more accurate output by telling me the kernel version and where to find
> map, modules, ksyms etc.  ksymoops -h explains the options.
> 
> Jan 10 21:41:38 awacs kernel: Unable to handle kernel NULL pointer 
> dereference at virtual address 00000008
> Jan 10 21:41:38 awacs kernel: c01ccb41
> Jan 10 21:41:38 awacs kernel: *pde = 00000000
> Jan 10 21:41:38 awacs kernel: Oops: 0000
> Jan 10 21:41:38 awacs kernel: CPU:    0
> Jan 10 21:41:38 awacs kernel: EIP:    0010:[xfs_syncsub+2393/3112]    Not 
> tainted
> Jan 10 21:41:38 awacs kernel: EFLAGS: 00010246
> Jan 10 21:41:38 awacs kernel: eax: 00000000   ebx: 00000008   ecx: dffdac00   
> edx: d5ff6a24
> Jan 10 21:41:38 awacs kernel: esi: 00000000   edi: d2eaaec0   ebp: c03201e0   
> esp: c1955f30
> Jan 10 21:41:38 awacs kernel: ds: 0018   es: 0018   ss: 0018
> Jan 10 21:41:38 awacs kernel: Process kupdated (pid: 7, stackpage=c1955000)
> Jan 10 21:41:38 awacs kernel: Stack: dffea000 dffea044 c1954660 0008e000 
> 00000001 00000010 00000001 dffdad18 
> Jan 10 21:41:38 awacs kernel:        dffdad18 00000000 00000010 00000001 
> 00000001 00000000 00000008 00000008 
> Jan 10 21:41:38 awacs kernel:        00000040 00000000 00000000 00000000 
> 00000026 ffffff00 c1954000 00000000 
> Jan 10 21:41:38 awacs kernel: Call Trace: [xfs_sync+21/28] 
> [linvfs_write_super+43/48] [sync_supers+220/272] [sync_old_buffers+47/144] 
> [kupdate+297/304] 
> Jan 10 21:41:38 awacs kernel: Code: 39 56 08 74 10 8b 5c 24 38 89 5c 24 3c 85 
> f6 0f 85 fa f7 ff 
> Using defaults from ksymoops -t elf32-i386 -a i386
> 
> Code;  00000000 Before first symbol
> 00000000 <_EIP>:
> Code;  00000000 Before first symbol
>    0:   39 56 08                  cmp    %edx,0x8(%esi)
> Code;  00000003 Before first symbol
>    3:   74 10                     je     15 <_EIP+0x15> 00000015 Before first 
> symbol
> Code;  00000005 Before first symbol
>    5:   8b 5c 24 38               mov    0x38(%esp,1),%ebx
> Code;  00000009 Before first symbol
>    9:   89 5c 24 3c               mov    %ebx,0x3c(%esp,1)
> Code;  0000000d Before first symbol
>    d:   85 f6                     test   %esi,%esi
> Code;  0000000f Before first symbol
>    f:   0f 85 fa f7 ff 00         jne    fff80f <_EIP+0xfff80f> 00fff80f 
> Before first symbol
> 
> 
> 1 warning issued.  Results may not be reliable.
-- 

Steve Lord                                      voice: +1-651-683-3511
Principal Engineer, Filesystem Software         email: lord@xxxxxxx


<Prev in Thread] Current Thread [Next in Thread>