PCI daughter card with USB on a 405GP

Laurent Mohin laurent.mohin at acterna.com
Tue Feb 24 04:01:08 EST 2004


Hello,

The pci_assign_bus_resource function checks that your PCI board is able to answer to Memory space and
IO space cycles (PCI:pci_assign_resource: type_mask must match<3>).
Because this check doesn't match, it is unable to map any memory ressource
for your PCI board and thus generate an exception when the ohci driver is
launched.

In your case, I would suspect an hardware problem with your PCI adaptors
on the Walnut board (I used to have a problem with a graphic daughter card
on it and it was due to the lack of one of PCI power supply).

You can check this by deactivating the usb driver in Linux and let the
board booting without any exception. At the prompt try the command : lspci
-xx and check wether the configuration registers of your board are OK or
not.

Hope this helps,

Laurent





v.sudeep at exgate.tek.com
Sent by: owner-linuxppc-embedded at lists.linuxppc.org
23/02/04 17:11


        To:     linuxppc-embedded at lists.linuxppc.org
        cc:
        Subject:        PCI daughter card with USB on a 405GP


Hi Everyone !!!

I am finding it a bit difficult to overcome a problem related to the PCI
daughter card with USB on my 405GP IBM Walnut board. If anyone can help
me,
I would be greatful to them.

My problem is as follows....
I want to install a USB printer on my 405GP board. Unfortunately the 405GP
does not have a USB port. So I am trying to use a PCI daughter card with
USB. I tried using two different cards  - OPTi 82c861 and ALi M5273. In
both
cases it seems to have some problems in booting up. i.e when the kernel is
probing the pci bus, some errors seem to happen. I snooped the kernel code
and identified as to when the error is being thrown out but am not sure
what
exactly is wrong and how I can set it right.

This error occurs in :
                 file            - ./drivers/pci/setup-res.c
                 function - pci_assign_bus_resource
                 line            - if ((res->flags ^ r->flags) &
type_mask)
                                                 continue;

This results in a kernel panic.  I am not sure of how I could go around
this
problem. Any suggestions on how to get the USB card up will be greatly
appreciated.

And thanks in advance.


Note:

1.               Our setup over here is as follows:
                                 Host Machine : Intel x86 having Red Hat 9
installed
                                 Target Board : IBM Walnut board - 405GP
with PowerPC
processor
                                 Cross-Compiler: MonteVista Hard Hat Linux
2.0  for the
ibm-walnut board

2.               Also attaching a file containing all the messages that
the kernel
pattered out.

 <<booting-up-messages>>

- S u d e e p    V
Don't change horses in mid stream.




 <<booting-up.messages>>


-------------- next part --------------
A non-text attachment was scrubbed...
Name: booting-up-messages
Type: application/octet-stream
Size: 2852 bytes
Desc: not available
Url : http://ozlabs.org/pipermail/linuxppc-embedded/attachments/20040223/2165be94/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: booting-up.messages
Type: application/octet-stream
Size: 2852 bytes
Desc: not available
Url : http://ozlabs.org/pipermail/linuxppc-embedded/attachments/20040223/2165be94/attachment-0001.obj 


More information about the Linuxppc-embedded mailing list