2.5.48 modules problem

2.5.48 modules problem

[Douglas Gilbert]

Can't see other reports on this in the lists **.
When I upgraded to lk 2.5.48, my kernel can't load
modules. modprobe and friends all fail with:
   QM_MODULES: Function not implemented

RH8.0 base system. Suggestions?


** Tried Adam J. Richter patch from lkml entitled:
"Patch(2.5.48): Restore module device ID tables" but it
didn't help.

Re: 2.5.48 modules problem

[J.E.J. Bottomley]

dougg@torque.net said:
> Can't see other reports on this in the lists **. When I upgraded to lk
> 2.5.48, my kernel can't load modules. modprobe and friends all fail
> with:
>    QM_MODULES: Function not implemented 

The entire method the kernel uses to construct and load modules changed in 
2.5.48.  You need Rusty's new module loader routines:

http://www.kernel.org/pub/linux/kernel/people/rusty/module-init-tools-0.7.tar.g
z

For the record, I still can't get quite a few modules to load even with this.

James

Re: 2.5.48 modules problem

[Douglas Gilbert]

J.E.J. Bottomley wrote:
> dougg@torque.net said:
> 
>>Can't see other reports on this in the lists **. When I upgraded to lk
>>2.5.48, my kernel can't load modules. modprobe and friends all fail
>>with:
>>   QM_MODULES: Function not implemented 
> 
> 
> The entire method the kernel uses to construct and load modules changed in 
> 2.5.48.  You need Rusty's new module loader routines:
> 
> http://www.kernel.org/pub/linux/kernel/people/rusty/module-init-tools-0.7.tar.g
> z

James,
Thanks. I made some progress. Hope Rusty is going to write
some documentation, man pages, update the Changes file, etc.

The new modprobe doesn't seem to pass parameters (options)
properly, at least that is what I'm seeing with scsi_debug.

Enough other things are broken with 2.5.48 to force me back
to lk 2.4.18 .

Doug Gilbert