Promise SuperTrak100 Oops/Kernel Panic

Promise SuperTrak100 Oops/Kernel Panic

[Rob Lake]

Hi,
    I've been trying to get a Promise SuperTrak 100 working.  I'm using 
linux-2.4.13-ac8 (patched from linux-2.4.13 with ac8 patch, if it 
matters).  Each time I load the i2o_block module, there is a a kernel 
panic.  Running the output through ksymoops produced the what's below - 
hopefully the ???? are supposed to be there.  At first I tried with i2o 
compiled into kernel; It had detected the raid card, but panic'd soon after.

If you're requiring any more information, let me know.

Thanks for any help,
Rob Lake
rlake@colabnet.com

------------
ksymoops 2.4.1 on i686 2.4.13-ac8.  Options used
     -V (default)
     -k ./ksyms.3 (specified)
     -l /proc/modules (default)
     -o /lib/modules/2.4.13-ac8/ (default)
     -m /boot/System.map-2.4.13-ac8 (default)

No modules in ksyms, skipping objects
Oops:   0000
CPU:    0
EIP:    0010:[<e094af5e>]  Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 000102006
eax: e0956044   ebx: 0009e2be     ecx: 0000000a       edx: 0009e2be
esi: 04000001   edi: c18a0d80     edp: c0267fac       esp: c0267f54
ds:  0018       es:  0018     ss: 0018
Process swapper (pid:0, stackpage=c0266000)
Stack:  dd559280 04000001 0000000a e089e10e c18a0d80 c0108229 0000000a 
c18a0280
        c0267fac c0267fac 0000000a c029ab80 88559280 c0108398 0000000a 
c0267fac
        dd559280 c01051a0 c0250de0 c01551a0 000ae000 c0108254 c01051a0 
00000032
Call Trace: [<e089e10c>] [<c0108229>] [<c0108398>] [<c01051a0>] [<e01051a0>]
[<c010a254>] [<c01051a0>] [<c01051bf>] [<c0105232>] [<c0105000>]
Code: 8b 4a 08 89 c8 83 e0 3f 8b 04 85 00 ff 94 e0 85 c0 74 10 8b

 >>EIP; e094af5e <END_OF_CODE+20679fc2/????>   <=====
Trace; e089e10c <END_OF_CODE+205cd170/????>
Trace; c0108229 <handle_IRQ_event+39/60>
Trace; c0108398 <do_IRQ+68/b0>
Trace; c01051a0 <default_idle+0/30>
Trace; e01051a0 <END_OF_CODE+1fe34204/????>
Trace; c010a254 <call_do_IRQ+5/d>
Trace; c01051a0 <default_idle+0/30>
Trace; c01051bf <default_idle+1f/30>
Trace; c0105232 <cpu_idle+42/60>
Trace; c0105000 <_stext+0/0>
Code;  e094af5e <END_OF_CODE+20679fc2/????>
00000000 <_EIP>:
Code;  e094af5e <END_OF_CODE+20679fc2/????>   <=====
   0:   8b 4a 08                  mov    0x8(%edx),%ecx   <=====
Code;  e094af61 <END_OF_CODE+20679fc5/????>
   3:   89 c8                     mov    %ecx,%eax
Code;  e094af63 <END_OF_CODE+20679fc7/????>
   5:   83 e0 3f                  and    $0x3f,%eax
Code;  e094af66 <END_OF_CODE+20679fca/????>
   8:   8b 04 85 00 ff 94 e0      mov    0xe094ff00(,%eax,4),%eax
Code;  e094af6d <END_OF_CODE+20679fd1/????>
   f:   85 c0                     test   %eax,%eax
Code;  e094af6f <END_OF_CODE+20679fd3/????>
  11:   74 10                     je     23 <_EIP+0x23> e094af81 
<END_OF_CODE+20679fe5/????>
Code;  e094af71 <END_OF_CODE+20679fd5/????>
  13:   8b 00                     mov    (%eax),%eax

Re: Promise SuperTrak100 Oops/Kernel Panic

[Alan Cox]

> linux-2.4.13-ac8 (patched from linux-2.4.13 with ac8 patch, if it 
> matters).  Each time I load the i2o_block module, there is a a kernel 
> panic.  Running the output through ksymoops produced the what's below - 
> hopefully the ???? are supposed to be there.  At first I tried with i2o 
> compiled into kernel; It had detected the raid card, but panic'd soon after.

Use either 2.4.9 Red Hat kernel trees or 2.4.18-rc1 and all should be well.

Re: Promise SuperTrak100 Oops/Kernel Panic

[Rob Lake]

It was suggested to try RedHat 2.4.9 trees or 2.4.18-rc1. Both attempts
leave the i2o_block module initializing (no oops/panic, just boredom).

2.4.18-rc1 says this in /var/log/messages:
Feb 13 21:52:08 sphere878 kernel: Linux I2O PCI support (c) 1999 Red Hat
Software.
Feb 13 21:52:08 sphere878 kernel: I2O Core - (C) Copyright 1999 Red Hat
Software
Feb 13 21:52:08 sphere878 kernel: i2o: Checking for PCI I2O
controllers...
Feb 13 21:52:08 sphere878 kernel: PCI: Found IRQ 10 for device 00:0b.1
Feb 13 21:52:08 sphere878 kernel: i2o: I2O controller on bus 0 at 89.
Feb 13 21:52:08 sphere878 kernel: i2o: PCI I2O controller at 0xE9800000
size=4194304
Feb 13 21:52:08 sphere878 kernel: I2O: Promise workarounds activated.
Feb 13 21:52:08 sphere878 kernel: i2o/iop0: Installed at IRQ10
Feb 13 21:52:08 sphere878 kernel: i2o: 1 I2O controller found and
installed.
Feb 13 21:52:08 sphere878 kernel: I2O: Event thread created as pid 1496
Feb 13 21:52:08 sphere878 kernel: Activating I2O controllers...
Feb 13 21:52:08 sphere878 kernel: This may take a few minutes if there
are many devices
Feb 13 21:52:10 sphere878 kernel: i2o/iop0: Reset rejected, trying to
clear
Feb 13 21:52:15 sphere878 kernel: i2o/iop0: LCT has 4 entries.
Feb 13 21:52:15 sphere878 kernel: I2O Block Storage OSM v0.9
Feb 13 21:52:15 sphere878 kernel:    (c) Copyright 1999-2001 Red Hat
Software.
Feb 13 21:52:15 sphere878 kernel: i2o_block: registered device at major
80
Feb 13 21:52:15 sphere878 kernel: i2o_block: Checking for Boot device...
Feb 13 21:52:15 sphere878 kernel: i2o_block: Checking for I2O Block
devices...
Feb 13 21:52:15 sphere878 kernel: i2ob: Installing tid 14 device at unit
0
Feb 13 21:52:15 sphere878 kernel: i2o/hda: Max segments 8, queue depth
4, byte limit 6144.
Feb 13 21:52:15 sphere878 kernel: i2o/hda: Disk Storage: 305274MB, 512
byte sectors.
Feb 13 21:52:15 sphere878 kernel: i2o/hda: Maximum sectors/read set to
32.
Feb 13 21:52:17 sphere878 kernel:  i2o/hda:r 3
Feb 13 21:52:17 sphere878 kernel: I2O: Spurious reply to handler 3
Feb 13 21:52:17 sphere878 last message repeated 454 times

Would a corrupt partition table do this to me?  Because it is corrupt. 
I'm trying this to fdisk it.

Thanks again, I appreciate any insight,
Rob

Re: Promise SuperTrak100 Oops/Kernel Panic

[Alan Cox]

> Feb 13 21:52:15 sphere878 kernel: i2o/hda: Disk Storage: 305274MB, 512
> byte sectors.
> Feb 13 21:52:15 sphere878 kernel: i2o/hda: Maximum sectors/read set to
> 32.

It found the block array happily.

> Feb 13 21:52:17 sphere878 kernel:  i2o/hda:r 3
> Feb 13 21:52:17 sphere878 kernel: I2O: Spurious reply to handler 3
> Feb 13 21:52:17 sphere878 last message repeated 454 times

Then things seem to go a little mad, and its getting bogus messages to
a drivert that has been unloaded

Re: Promise SuperTrak100 Oops/Kernel Panic

[Rob Lake]

On Thu, 2002-02-14 at 10:03, Alan Cox wrote:

> > Feb 13 21:52:17 sphere878 kernel:  i2o/hda:r 3
> > Feb 13 21:52:17 sphere878 kernel: I2O: Spurious reply to handler 3
> > Feb 13 21:52:17 sphere878 last message repeated 454 times
> 
> Then things seem to go a little mad, and its getting bogus messages to
> a drivert that has been unloaded

Is there a driver I should be loading pre i2o_block?
Rob

Re: Promise SuperTrak100 Oops/Kernel Panic

[Alan Cox]

> > > Feb 13 21:52:17 sphere878 kernel:  i2o/hda:r 3
> > > Feb 13 21:52:17 sphere878 kernel: I2O: Spurious reply to handler 3
> > > Feb 13 21:52:17 sphere878 last message repeated 454 times
> > 
> > Then things seem to go a little mad, and its getting bogus messages to
> > a drivert that has been unloaded
> 
> Is there a driver I should be loading pre i2o_block?

i2o_block will load i2o_pci and i2o_core which is all that are needed.
So far I've not been able to duplicate your problem. 

Re: Promise SuperTrak100 Oops/Kernel Panic

[Rob Lake]

No need to try any longer.  I upgraded the firmware and can mount device
now.  Thanks for you help and time.

Rob

On Thu, 2002-02-14 at 15:54, Alan Cox wrote:
> > > > Feb 13 21:52:17 sphere878 kernel:  i2o/hda:r 3
> > > > Feb 13 21:52:17 sphere878 kernel: I2O: Spurious reply to handler 3
> > > > Feb 13 21:52:17 sphere878 last message repeated 454 times
> > > 
> > > Then things seem to go a little mad, and its getting bogus messages to
> > > a drivert that has been unloaded
> > 
> > Is there a driver I should be loading pre i2o_block?
> 
> i2o_block will load i2o_pci and i2o_core which is all that are needed.
> So far I've not been able to duplicate your problem. 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>