Re: insmod segfault in pci_find_subsys()

[Toralf Lund <toralf@procaptura.com>]

Greg KH wrote:

>On Tue, Mar 29, 2005 at 04:15:37PM +0200, Toralf Lund wrote:
>  
>
>>Greg KH wrote:
>>
>>    
>>
>>>On Fri, Mar 18, 2005 at 10:12:05AM +0100, Toralf Lund wrote:
>>>
>>>
>>>      
>>>
>>>>Am I seeing an issue with the PCI functions here, or is it just that I 
>>>>fail to spot an obvious mistake in the module itself?
>>>>  
>>>>
>>>>        
>>>>
>>>I think it's a problem in your code.  I built and ran the following
>>>example module just fine (based on your example, which wasn't the
>>>smallest or cleanest...), with no oops.  Does this code work for you?
>>>
>>>
>>>      
>>>
>>OK, I've finally been able to test this, and no, it does not work. 
>>insmod segfaults and the system log says
>>
>>kernel: Unable to handle kernel paging request at virtual address 533e3762
>>    
>>
>
>Then I think you have a broken build system or makefile or gcc.  It
>works fine here.
>  
>
Yes. You are right. I actually mentioned this on a different thread: I 
eventually found out that the kernel was compiled with -mregparam=3, and 
the module was not. This option seems to have been added to the default 
config and/or Red Hat's build setup sometime before the current kernel 
release, but after the start of the 2.6 series...

- Toralf