Subject: Re: 2.4.0-test11 - get_module_symbol() goes away, how to get mol_interface[]?
From: David Wagner (dhwagner@swbell.net)
Date: Wed Nov 29 2000 - 19:34:18 MST
on 11/24/00 3:55 PM, Samuel Rydh at samuel@ibrium.se wrote:
> On Fri, Nov 24, 2000 at 02:11:32PM -0700, Derrik Pates wrote:
>> Subject says it all. I've been trying to figure out how to get the address
>> of mol_interface using inter_module_get and inter_module_get_request, but
>> it's not working. The get_module_symbol() call is no more. Anything we can
>> do about it? Is there any other way to get the symbol?
>>
>
> Since the 2.2 kernel is never run-time patched, it is not necessary
> to lookup the symbol manually - just use the variable directly (on 2.0
> kernels, the symbol is defined if and only if the kernel is not
> runtime patched).
>
> A fix for this is in the MOL bitkeeper tree
> (bk://waltari.theophys.kth.se:5000).
Samuel,
I am having the same problem with the sources downloaded from your MOL web
site. There is also the bit about list_head not getting properly type-def'd
when mol/emulation/main.c includes linux/module.h , which I mentioned on the
mol mail list and on the linuxppc users list.
If you are not going to update those archives soon, could I please
have read access to the bitkeeper archive?
I am not terribly familiar with bitkeeper, but from reading the help file,
it appears that the command I would need is:
bk clone waltari.theophys.kth.se:5000 mydirectory, or
bk rssync waltari.theophys.kth.se:5000 mydirectory.
Please advise if this is not correct.
p.s. The reason I am going through all this trouble is that the current
binary rpm for the 2.4 kernel will not run with test-11.
Thanks,
David Wagner
This archive was generated by hypermail 2a24 : Wed Nov 29 2000 - 19:38:53 MST