Hi,
I got an oops on 2.5.11 with an software raid 0 setup on 3 scsi disks,
it worked ok on 2.5.8. I get this when booting up and then my /dev/md0
isn't found.. If you need more details/help let me know!
Attached are the dmesg and config, below is the oops.
ksymoops 2.4.5 on i686 2.4.18-wolk3.3. Options used
-v /home/it0/download/kernelstuff/linux-2.5.8/vmlinux (specified)
-K (specified)
-L (specified)
-o /lib/modules/2.5.11/ (specified)
-m /boot/System.map-2.5.11 (specified)
No modules in ksyms, skipping objects
Machine check exception polling timer started.
cpu: 0, clocks: 2672704, slice: 1336352
Unable to handle kernel NULL pointer dereference at virtual address
00000010
c01e13f5
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<c01e13f5>] Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010202
eax: 00000000 ebx: c152fb14 ecx: c0395fc0 edx: 00000000
esi: 00000003 edi: df885000 ebp: c152fb00 esp: df287e08
ds: 0018 es: 0018 ss: 0018
Stack: c152fb14 c023c79a 00000000 c0395fc0 00000002 c152fb00 c152fb00
00000001
c023ca16 c152fb00 df215900 c152fb14 c152fb00 df215900 00000008
000029a0
000029a0 c011ba29 000029a0 00000286 df286000 000029a0 000029a0
00000286
Call Trace: [<c023c79a>] [<c023ca16>] [<c011ba29>] [<c011bcf7>]
[<c011bbd1>]
[<c023ce55>] [<c023d082>] [<c023d217>] [<c023e2e3>] [<c0140809>]
[<c0140900>]
[<c016b978>] [<c013afac>] [<c013aec2>] [<c014286c>] [<c0149a8b>]
[<c0108aaf>]
Code: 0f b7 50 10 b2 00 66 0f b6 40 10 01 c2 89 d0 66 c1 e8 08 0f
>>EIP; c01e13f5 <blk_get_readahead+5/60> <=====
>>ebx; c152fb14 <END_OF_CODE+1195200/????>
>>ecx; c0395fc0 <md_size+0/400>
>>edi; df885000 <END_OF_CODE+1f4ea6ec/????>
>>ebp; c152fb00 <END_OF_CODE+11951ec/????>
>>esp; df287e08 <END_OF_CODE+1eeed4f4/????>
Trace; c023c79a <device_size_calculation+14a/1f0>
Trace; c023ca16 <do_md_run+1d6/350>
Trace; c011ba29 <call_console_drivers+d9/e0>
Trace; c011bcf7 <release_console_sem+b7/c0>
Trace; c011bbd1 <printk+131/160>
Trace; c023ce55 <autorun_array+85/b0>
Trace; c023d082 <autorun_devices+202/230>
Trace; c023d217 <autostart_array+167/1b0>
Trace; c023e2e3 <md_ioctl+323/800>
Trace; c0140809 <deactivate_super+49/a0>
Trace; c0140900 <grab_super+a0/d0>
Trace; c016b978 <devfs_open+b8/170>
Trace; c013afac <dentry_open+dc/180>
Trace; c013aec2 <filp_open+52/60>
Trace; c014286c <blkdev_ioctl+5c/70>
Trace; c0149a8b <sys_ioctl+16b/1c0>
Trace; c0108aaf <syscall_call+7/b>
Code; c01e13f5 <blk_get_readahead+5/60>
00000000 <_EIP>:
Code; c01e13f5 <blk_get_readahead+5/60> <=====
0: 0f b7 50 10 movzwl 0x10(%eax),%edx <=====
Code; c01e13f9 <blk_get_readahead+9/60>
4: b2 00 mov $0x0,%dl
Code; c01e13fb <blk_get_readahead+b/60>
6: 66 0f b6 40 10 movzbw 0x10(%eax),%ax
Code; c01e1400 <blk_get_readahead+10/60>
b: 01 c2 add %eax,%edx
Code; c01e1402 <blk_get_readahead+12/60>
d: 89 d0 mov %edx,%eax
Code; c01e1404 <blk_get_readahead+14/60>
f: 66 c1 e8 08 shr $0x8,%ax
Code; c01e1408 <blk_get_readahead+18/60>
13: 0f 00 00 sldtl (%eax)
Hi,
On 29 Apr 2002, Tommy Faasen wrote:
> I got an oops on 2.5.11 with an software raid 0 setup on 3 scsi disks,
> it worked ok on 2.5.8. I get this when booting up and then my /dev/md0
> isn't found.. If you need more details/help let me know!
The patch below fixes it for me.
rdev doesn't point to a valid raid partition.
bye, Roman
Index: drivers/md/md.c
===================================================================
RCS file: /usr/src/cvsroot/linux-2.5/drivers/md/md.c,v
retrieving revision 1.1.1.8
diff -u -p -r1.1.1.8 md.c
--- drivers/md/md.c 29 Apr 2002 17:35:50 -0000 1.1.1.8
+++ drivers/md/md.c 30 Apr 2002 17:52:04 -0000
@@ -1577,6 +1577,7 @@ static int device_size_calculation(mddev
if (!md_size[mdidx(mddev)])
md_size[mdidx(mddev)] = sb->size * data_disks;
+ rdev = list_entry(mddev->disks.next, mdk_rdev_t, same_set);
readahead = (blk_get_readahead(rdev->bdev) * 512) / PAGE_SIZE;
if (!sb->level || (sb->level == 4) || (sb->level == 5)) {
readahead = (mddev->sb->chunk_size>>PAGE_SHIFT) * 4 * data_disks;
On Tue, 2002-04-30 at 20:58, Roman Zippel wrote:
> Hi,
>
> On 29 Apr 2002, Tommy Faasen wrote:
>
> > I got an oops on 2.5.11 with an software raid 0 setup on 3 scsi disks,
> > it worked ok on 2.5.8. I get this when booting up and then my /dev/md0
> > isn't found.. If you need more details/help let me know!
>
> The patch below fixes it for me.
> rdev doesn't point to a valid raid partition.
>
Thanks, that did it for me!
> bye, Roman
>
> Index: drivers/md/md.c
> ===================================================================
> RCS file: /usr/src/cvsroot/linux-2.5/drivers/md/md.c,v
> retrieving revision 1.1.1.8
> diff -u -p -r1.1.1.8 md.c
> --- drivers/md/md.c 29 Apr 2002 17:35:50 -0000 1.1.1.8
> +++ drivers/md/md.c 30 Apr 2002 17:52:04 -0000
> @@ -1577,6 +1577,7 @@ static int device_size_calculation(mddev
> if (!md_size[mdidx(mddev)])
> md_size[mdidx(mddev)] = sb->size * data_disks;
>
> + rdev = list_entry(mddev->disks.next, mdk_rdev_t, same_set);
> readahead = (blk_get_readahead(rdev->bdev) * 512) / PAGE_SIZE;
> if (!sb->level || (sb->level == 4) || (sb->level == 5)) {
> readahead = (mddev->sb->chunk_size>>PAGE_SHIFT) * 4 * data_disks;
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
>