[lwip-users] Re: [lwip] RS485

[Adam Dunkels]

On Thursday 11 October 2001 12:29, you wrote:
> > There is a bug in the uIP checksum code when compiled for big endian 
> > systems.
>
> Nice to know. I just fixed my copy of uIP sources.

I'll put out a new version of uIP once all the portabliity bugs have been 
located and squashed. The new version will also include code for more 
platforms as well.

> As a matter of fact, the main trouble with TCP/IP v4 lies in the
> addressing. IPv6 is supposed to solve this, and many other problems as
> well, but so far it hasn't caught enough to reach a critical mass.

Indeed. The advantage with IPv6 addressing over IPv4 is not only the size of 
the address space, but also the addition of link local addressing, stateless 
autoconfiguration of address, the requirement of multicast operationg among 
others. I believe that IPv6 is ideal for embedded systems because of those 
features. Stateless autoconfiguration means that there is no need for DHCP 
servers and multicast means that systems have a mean to make themselves 
known. 

lwIP has some very basic support for IPv6. (If someone wondered why IP 
addresses in lwIP was repressented by a "struct ip_addr *" instead of a 
simple 32-bit type - this is why.) The IPv6 code will be further developed 
for future releases with the goal of making a lightweight IPv6 stack. The 
idea of using a proxy to lessen some of the stack's burden in this case will 
be investigated more thuroughly as well.

> Best regards, and kudos for your network stack implementation. I also
> enjoyed the MiniWeb demonstration code a lot.

Thanks!

/adam
-- 
Adam Dunkels <address@hidden>
http://www.sics.se/~adam

[This message was sent through the lwip discussion list.]