Using kernel 2.4.20 on ix86 and an HP IEEE1394 DVD+RW drive, I
encountered this BUG/oops when attempting to mount a CD-ROM. I at
first could not mount the disc, so I attempted to mount it as HFS:
kernel BUG at buffer.c:2516!
invalid operand: 0000
CPU: 0
EIP: 0010:[<c013747e>] Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010206
eax: 000007ff ebx: 0000000b ecx: 00000800 edx: cfed4280
esi: 00000001 edi: 00000b01 ebp: 00000002 esp: c5f71db4
ds: 0018 es: 0018 ss: 0018
Process mount (pid: 15589, stackpage=c5f71000)
Stack: 00000000 00000b01 00000200 00000002 00002328 c013559a 00000b01 00000002
00000200 cff14c80 c06fc800 00000002 00000001 c01357c8 00000b01 00000002
00000200 c5f71e08 d4c5b703 00000b01 00000002 00000200 cff14c80 c06fc800
Call Trace: [<c013559a>] [<c01357c8>] [<d4c5b703>] [<d4c5a2c9>] [<d4c5b568>]
[<c01d8bfd>] [<c01386fc>] [<d4c5f204>] [<d4c5f204>] [<c01388db>] [<d4c5f204>]
[<c01488f9>] [<c0148bc2>] [<c0148a1d>] [<c0148f44>] [<c0106d27>]
Code: 0f 0b d4 09 00 b6 25 c0 8b 44 24 20 05 00 fe ff ff 3d 00 0e
>>EIP; c013747e <grow_buffers+3e/110> <=====
>>edx; cfed4280 <_end+fba14a8/1250c288>
>>esp; c5f71db4 <_end+5c3efdc/1250c288>
Trace; c013559a <getblk+2a/50>
Trace; c01357c8 <bread+18/70>
Trace; d4c5b703 <[hfs]hfs_buffer_get+23/80>
Trace; d4c5a2c9 <[hfs]hfs_mdb_get+a9/3c0>
Trace; d4c5b568 <[hfs]hfs_read_super+68/1a0>
Trace; c01d8bfd <media_changed+3d/70>
Trace; c01386fc <get_sb_bdev+1dc/250>
Trace; d4c5f204 <[hfs]hfs_fs+0/1c>
Trace; d4c5f204 <[hfs]hfs_fs+0/1c>
Trace; c01388db <do_kern_mount+5b/110>
Trace; d4c5f204 <[hfs]hfs_fs+0/1c>
Trace; c01488f9 <do_add_mount+69/140>
Trace; c0148bc2 <do_mount+152/170>
Trace; c0148a1d <copy_mount_options+4d/a0>
Trace; c0148f44 <sys_mount+84/d0>
Trace; c0106d27 <system_call+33/38>
Code; c013747e <grow_buffers+3e/110>
00000000 <_EIP>:
Code; c013747e <grow_buffers+3e/110> <=====
0: 0f 0b ud2a <=====
Code; c0137480 <grow_buffers+40/110>
2: d4 09 aam $0x9
Code; c0137482 <grow_buffers+42/110>
4: 00 b6 25 c0 8b 44 add %dh,0x448bc025(%esi)
Code; c0137488 <grow_buffers+48/110>
a: 24 20 and $0x20,%al
Code; c013748a <grow_buffers+4a/110>
c: 05 00 fe ff ff add $0xfffffe00,%eax
Code; c013748f <grow_buffers+4f/110>
11: 3d 00 0e 00 00 cmp $0xe00,%eax
Corruptedly yours,
Jeffrey (please CC replies)
Jeffrey Baker <[email protected]> wrote:
> Using kernel 2.4.20 on ix86 and an HP IEEE1394 DVD+RW drive, I
> encountered this BUG/oops when attempting to mount a CD-ROM. I at
> first could not mount the disc, so I attempted to mount it as HFS:
HFS wants 512 blocks while sr sets hardsect size to 2048. You can
work around it by reading it via loopback (losetup or mount -o loop).
--
Debian GNU/Linux 3.0 is out! ( http://www.debian.org/ )
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Mon, Dec 30, 2002 at 02:22:35PM +1100, Herbert Xu wrote:
> Jeffrey Baker <[email protected]> wrote:
> > Using kernel 2.4.20 on ix86 and an HP IEEE1394 DVD+RW drive, I
> > encountered this BUG/oops when attempting to mount a CD-ROM. I at
> > first could not mount the disc, so I attempted to mount it as HFS:
>
> HFS wants 512 blocks while sr sets hardsect size to 2048. You can
> work around it by reading it via loopback (losetup or mount -o loop).
Do you mean make an image of the CD with dd and mount it
with loopback? That sounds like it would work fine.
On Sun, Dec 29, 2002 at 08:04:02PM -0800, Jeffrey Baker wrote:
>
> > HFS wants 512 blocks while sr sets hardsect size to 2048. You can
> > work around it by reading it via loopback (losetup or mount -o loop).
>
> Do you mean make an image of the CD with dd and mount it
> with loopback? That sounds like it would work fine.
It's easier that.
losetup /dev/loop0 /dev/cdrom
mount -rt hfs /dev/loop0 /mnt
--
Debian GNU/Linux 3.0 is out! ( http://www.debian.org/ )
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt