Re: [NTLK] NIE and the TCP stack

From: Martin Joseph <>
Date: Fri Nov 16 2007 - 13:38:21 EST

On Nov 16, 2007, at 10:25 AM, Ben Sanford wrote:

> I'm currently working on finishing up Eckhart's GCC compiler (going to
> finish the rom jump table for the MP2x00 and eventually make one for
> the eMate - this is so I can target the MP2x00 or the eMate
> specifically for the best compatibility) and I'm looking ahead at
> making a port of PuTTY SSH. While I can implement most, if not all,
> the basic functions that handle encryption and whatnot my big problem
> is with the communications part. I could use the Newton's own built
> in handlers (eg. the nasty NIE stack) or I could attempt to port one.
> I found a project called uIP
> ( which is a very
> small, low ram TCP stack for embedded microcontrollers - the code is
> complete but the only thing required to actually port is to make a
> device driver. In my case the driver would be a bridge between the
> NIE TCP stack code and the TCP stack itself - but I'm wondering if
> this is feasible? If so, we could have an actual TCP stack although
> it will have to be compiled into each application individually.
> I'm looking for some feedback on this. I'm going to be a very bored
> grad student starting next semester and wouldn't mind playing around
> with this and giving the Newton a true TCP stack would give it a major
> boost.

Wow! That is exciting. I would suggest if you haven't already done
so, check through the archives for messages from Paul Guyot? He is
the master of modifying/improving Newton's software architecture
beyond where others would consider it feasible.

I love the idea of a different/newer TCP stack that worked with the
older Newtons, but don't have any brilliant ideas for you, other then
the above.


The NewtonTalk Mailing List -
The Official Newton FAQ -
The Newton Glossary -
WikiWikiNewt -
Received on Fri Nov 16 13:38:24 2007

This archive was generated by hypermail 2.1.8 : Fri Nov 16 2007 - 14:30:00 EST