Re: Re: [cc65] HEAP from here to there

From: Thomas Giesel <>
Date: 2009-06-05 07:32:10
Thanks for your responses.

> Since segments have fixed sizes,
> your suggestion is not satisfactory.  A fixed-size heap segment is either
> too small or too large for > 99% of all programs. :-) 

I didn't realize the fixed size. So currently one could use __HEAP_START__ but still there has to be a dependency from __RAM_END__ to get the maximal possible heap size. This dependency is created based on SP at the moment (because after running the program SP is near __RAM_END__). Correct me if I'm wrong.

(The symbol names may be wrong because I'm in a hurry and can't look them up now)

It's also totally intelligible that both, heap and stack are in HEAP.

A way to have both, the maximal size of the HEAP segment and the freedom to put it into another memory area could be this: There could be a new attribute for a linker segment "maximize = yes". This shall mean that the segment goes up to the end of the current area (and must be the last one in this area). With this feature one could put the HEAP segment in any area and both, heap and stack can be calculated from HEAP_START and HEAP_END,

Possibly I still didn't understand the cross-platform issues and that's why my writing may be nonsense. I'm sorry in this case.

> Uz is not as old as he thinks that he is. :-)
Oh, I also have some half-finished changes in my code I forgot why I've started them. And I feel young nevertheless ;-)



To unsubscribe from the list send mail to with
the string "unsubscribe cc65" in the body(!) of the mail.
Received on Fri Jun 5 07:32:19 2009

This archive was generated by hypermail 2.1.8 : 2009-06-05 07:32:22 CEST