| |
cadaver
Registered: Feb 2002
Posts: 1153 |
Relying on $ae/$af after load
Another slightly odd question .. are you aware of modified C64 Kernals or filebrowser programs that *don't* have the ZP location $ae/$af point to the load end address after load?
Practically, the usual method is to use it as a load pointer and increment after each received byte.
If there are any, would be interesting to know which. |
|
| |
ChristopherJam
Registered: Aug 2004
Posts: 1378 |
Possibly tangential, but I had to add an option to nucrunch for setting 2d/2e after decrunch, as something was relying on those being set to load end address. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
The original DTV2/DTV3 kernal sets $ae/$af correctly when loading from flash, but X/Y is incorrect. |
| |
cadaver
Registered: Feb 2002
Posts: 1153 |
Quote: Possibly tangential, but I had to add an option to nucrunch for setting 2d/2e after decrunch, as something was relying on those being set to load end address.
I was more used to checking $2d/$2e after a load too. But think that's not set unless it's the actual Basic LOAD command getting called, and for example the SD2IEC filebrowser didn't go that route (but still sets $ae/$af as a byproduct of using the Kernal LOAD) |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
A lot of older packers rely on $2d/$2e being set up correctly when being run so if a loader does not set them up there will be compatibility problems. Relying on $ae/$af in the same manner isn't as common from what I have seen. |
