> people maintaining them. Ftape already does this, and it works. (Although
It works? It doesn't break if something changes in the kernel? Hmmm,
perhaps I should downgrade to ftape from zftape. My zftape-1.02 broke when
I upgraded from 1.3.45 to 1.3.59+m6. But then, I've heard that ftape is
broken with SMP, which I use.
> he is mumbling about integrating it with the kernel...)
Good idea, that makes it easier to have it uptodate when the kernel changes.
Consider a change like this:
x(abc) is being phased out, and must be replaced by calls to xa(),xb(),xc().
Don't know if the happened, but it could happen.
a) ftape is not a kernel module -> the interface change has to be tracked
by the author, because probably Linus doesn't use the driver, so he
doesn't check it.
b) ftape is a kernel module -> the fixes for the interface change are
probably integreted by the proposer of the interface change, because:
*) in some occurances the author just greps for all places to ``upgrade''
because the change breaks much of the code.
*) some changes could even make compiling the kernel without ftape
impossible.
So basically, having a module part of the kernel is a good thing,
especially after the module as such has stabilized (because every patch has
to be submited to Linus).
Andreas
-- Andreas Kostyrka Email: andreas@medman.ag.or.at Fax: +43/1/7070750 Tel: +43/1/7077571, +43/664/3020166 (cellular) Copyright 1996 Andreas Kostyrka. Microsoft Network is prohibited from redistributing this work in any form, in whole or in part.