On Mon, 28 Dec 1998, Julian Munoz Dominguez wrote:
> Tomi, thank you very much for the explanation.
> But what I don't understand is: Why the "conf.modules" is not well
> configured from the begining, by the distribuitor (RedHat for example).
Ok, that was a point that I was going to make in my mailing but it started
to become too long...
I too don't understand why the distribution makers don't try to compile
even a semi-complete conf.modules "skeleton". At least something to start
with, possibly with a warning comment that this isn't necessarily 100%
correct/complete.
> So "someone" has decided that kiss id is "5".
> What is failing is the assotiation between "5" and the name of the
> module (wich is part of the kernel, it comes with the kernel).
>
> In the Kernel sources, the "conf.modules" is not documented (in
> Documentation directory), so that's maybe the source of the problem.
The association is missing I believe because of flexibility. In the
general case the module/driver doesn't have to be a part of the kernel or
come with it. Having special cases for the drivers that come with the
kernel would probably be painful to maintain, even though I think there
are a few such special cases in the kernel...
But I do agree that in the Documentation directory there should be a file
that would list all the drivers needing aliases in conf.modules. Last time
I would have needed that I ended up reading kernel ppp driver sources to
find out what those ppp-compress-xx modules are that my kernel tries to
load. (A hint if anyone needs: ppp-compress-1 and 2 are Predictor 1 and 2
and linux kernel doesn't support them so you can alias them to "off" if
your ISP happens to offer them and you are annoyed by the warnings in
syslog.)
--
--... Tomi Manninen / [EMAIL PROTECTED] / OH2BNS @ OH2RBI.FIN.EU ...--
