Re: [MiNT] MiNTNet "land" bug fixed

[Mario Becroft <mb@tos.pl.net>]

On Tue, 11 May 1999, Guido Flohr wrote:

> As you've said yourself, something like a hack (and btw, in time-critical
> if-and-statements you should always place the least probable condition
> first).

Yes, I suppose it would be beneficial to change the order of the
conditions; the way I did it was just how I had thought of it in my mind,
and was not based on any performance considerations.

> Isn't it possible to find the real reason for the freeze and fix that
> instead?

Probably, but this would require much more complicated looking into the
internals of the TCP implementation which I didn't want to do. Some people
complained about MiNTNet being frozen by this land program, and this does
fix that, that's all I was worried about.

> Torsten Lang (attending this list) is the current maintainer.  Sorry, I've
> forgotten the address but it should be somewhere in the MiNTNet sources.
> He is also actively working on MiNTNet so you should send your
> patche to too so that they can get integrated into the next distribution.

Well if someone else is already actively working on MiNTNet then there may
be no place for my hack which is, as you point out, not optimal. I had got
the impression that MiNTNet development was quite dead lately. Anyhow,
presumably Torsten Lang is reading this mailing list and will be able to
integrate my hack into the distribution if he wants to.

Torsten, if you are reading this, maybe you could write a little about
what you are doing with MiNTNet, so that I can determine whether it
affects my implementation of IP masquerading.

-- 
+-----------------------------------------------------------------------+
| Mario Becroft                * Tariland, New Zealand Atari User Group |
| mb@tos.pl.net                * Atari Hardware Developments            |
| http://www.ak.planet.gen.nz/~mario/ * Atari Serial Mouse Interface    |
+-----------------------------------------------------------------------+