> On the other hand, why not allocate LKM major numbers statically?   If
> you do that, you can have the old standard static /dev and everything
> works fine.   Is it really likely that there will be >2048 (or
> whatever) LKMified device drivers?   Is it likely that there will be
> >4096 drivers in all?

My point in this discussion about this is that it doesn't need to be
done now.

If we want to do 'magic bit' checking at some point in the future, we
can.  It doesn't change the user's view of things at all, and wouldn't
be too hard to implement in the kernel (no harder than it would be to
implement them now).  In the mean time, we can use a bunch of
'lkmentry' entries in the device switch.  If we decide do to a bit
hack, then we can later reclaim the slots -- it's not like LKM's
particularly care, at this point in time, what their numbers are.


We shouldn't make a decision which adds coding and code complexity now
if we don't know that it's necessary, unless we can't do it later.

In this case, we could do it later just as easily, so *punt*.  8-)


chris