> 
> On Nov 11, 2003, at 11:49 PM, Martin Husemann wrote:
> 
> > This has the additional advantage that it enables LKMs to register and 
> > handler
> > additional ethernet frame types. It would not affect the efficiency of
> > hardwired frame types.
> >
> > Do you think this is worth the effort?
> 
> Yes.  I would actually like to see the entire input path use 
> dynamically-registered handlers, and eliminate that big switch 
> statement entirely.
> 
In that case some performance measurements is required, so that 
an implementation won't give any penalty.

But yes, if it won't hurt performance it is really a good idea.

-- Ragge