Re: [cc65] Banking out Kernal

From: Oliver Schmidt <ol.sc1web.de>
Date: 2013-02-16 14:47:59
Hi,

> I would prefer a header file with two forward declarations

Prefer over what? I'd help if you were somewhat more specific ;-)

> void switch_rom_off (void) and
> void switch_rom_on (void) (or even macros)
> so that the programmer can decide by himself when to switch the memory configuration.

That's so far part of one of my proposals.

> This way one could design the program so, that all functions which call kernal routines are side by side eachother and one needs only to switch memory configuration before all these calls and after.

How is a C programmer supposed to know which function calls Kernal
routines? How is the fact that a function calls Kernal routines to
stay fixed over cc65 releases - otherwise programs which used to work
start to break when built with later releases? How is Uz supposed to
handle the questions regarding this magic?

> Anything else would bloat the library unnecessary.

1. How do you define "bloat" ?
2. Why is size increase of the library in order to gain access to the
full 64kB RAM unnecessary?

BTW:

1.  Presuming that's you'd like to have interrupts working while the
"ROM is off" you'd need some part of the "bloat" nevertheless.

2. How do you want to make use of the RAM behind the Kernal? For
dynamically generated data? Then you're into double buffering in case
it needs to be accessed by the Kernal. For code? Then the code needs
to be moved there so you'd some more of the "bloat".

Regards,
Oliver
----------------------------------------------------------------------
To unsubscribe from the list send mail to majordomo@musoftware.de with
the string "unsubscribe cc65" in the body(!) of the mail.
Received on Sat Feb 16 14:48:08 2013

This archive was generated by hypermail 2.1.8 : 2013-02-16 14:48:11 CET