> as i understood, the code is written in such a way that it relies on > assembling to the same adress multiple times. Yes that, and in addition some of the "ORG" statements in the original code have a calculated expression. If I move that information to segment specifications in a linker configuration, I lose (a small bit of) information. > probably quite some work to convert something like that. (but imho worth it, > this is really a bad way to do things) Well, this is code from 1979, They didn't have fancy linkers back then ... And the trick they do to check if the different parts don't overlap is quite neat. Those guys were quite good at working around the limitation of their toolset. The code was supposidly compiled with the "Microtec" cross-assembler back then -- anybody heard of that? I am trying to do the minimal amount of changes to the code to get it to assemble with cc65. This is just a labour of love for me, to get this old code back in a somewhat usable, inspectable state. It is computer archaeology, if you will... I started from a paper version, did the scanning and OCR'ing, and a LOT of proofreading.... I will publish the result on the web in a few weeks. Apart from some very basic search/replace actions, I'm down to about 20 places where "manual intervention" is needed to get it to work. It all works very very nicely, except for this one issue. Best regards, Sidney ---------------------------------------------------------------------- To unsubscribe from the list send mail to majordomo@musoftware.de with the string "unsubscribe cc65" in the body(!) of the mail.Received on Thu Mar 31 14:00:30 2011
This archive was generated by hypermail 2.1.8 : 2011-03-31 14:00:32 CEST