🚀 go-pugleaf

In article <richard.726756238@dutepp6>, Richard.Kooijman@dnpap.et.tudelft.nl (Richard Kooijman) writes:
> 
> Hi,
> 
> 
> I have a rather large key definition file for Kermit. Alas, Kermit
> just doesn't make it to the last 4 definitions.
> 
> So, I picked up the MS-Kermit distribution from watsun and
> enlarged the key definition buffers. But again, alas, Kermit seems
> to be filled to the last byte. It is compiled in small memory model
> and compiling it in large memory model would solve the problem, but
> the TCP software is relying on the small mm.
> 
> Has anybody succeeded it doing this? It appears to me that there must
> be other people with large key definitions. 
> Any help greatly appreciated!
> Richard.
> 
> --
> -- ir Richard Kooijman ------------------- Email: R.Kooijman@ET.TUDelft.NL --
> -- Intelligent Network Management Project (INEMA) ---------------------------
> -- CARDIT, Delft University of Technology ------------ Tel: (31)-15-786209 --
> -- P.O.Box 5031, 2600 GA Delft, The Netherlands ------ Fax: (31)-15-784898 --
------------
	There is some hope of space in the next release, but all things have
limits. The compromise in the current releases is between what you would like
and the memory consumption experienced by others, and it's not easy saying
"enough" to anyone. I presume you are already loading an appropriate Code
Page so many key mappings are not necessary.
	Btw, the assembly language code (the vast majority of MS-DOS Kermit)
has no memory model. It has several code and data segments, two stacks, and
allocates free memory from DOS. It exploits near references whereever possible 
to save space and time. The Telnet code in C is the small memory model, for
the same reasons. Far references are expensive! Changing the Telnet C code
won't help one bit (er, byte).
	Joe D.
	Joe D.