!!!OLD!!! ZCPU Documentation

[Black Phoenix]

That's going to be a bitch to maintain when you're working on code and you're using any of the jump relative commands, but all right, if that's how that opcode is supposed to be handled by the compiler...

You're not supposed to use them that often, they are more like service opcodes rather than ones you use everyday.

[sukasa]

Ah. Well, I look at them and see a great way to make location-independant code without having to mess with using CS or LS or any sort of 'you are here' register. If everything's relative, then it makes maintaining code just that much easier.

[Black Phoenix]

Ah. Well, I look at them and see a great way to make location-independant code without having to mess with using CS or LS or any sort of 'you are here' register. If everything's relative, then it makes maintaining code just that much easier.

You should mess with CS/LS registers, they are "you are here" registers. That's why they exist.

This code will properly run independantly on it's location:

myloop:
  inc CS:#nice_variable; //offset relative to current code execution base
jmp myloop; //offset relative to current code execution base

alloc nice_variable;

To call this code from address 123456 in RAM you do this:

callf 0,123456;

To return from that code you do:

retf

[Nikita]

Correct me if I'm wrong, it appears that CPU has no support for integers? So if I wanted to divide 7 by 2 and get 3, I'd have to make hidden rounding? And if I wanted an 8-bit char, overflowing would have to be faked too? This is for my compiler, I want to have all kinds of variables there...

I see there is a binary precision and processor precision variables, but what do they do?

[Black Phoenix]

Correct me if I'm wrong, it appears that CPU has no support for integers? So if I wanted to divide 7 by 2 and get 3, I'd have to make hidden rounding? And if I wanted an 8-bit char, overflowing would have to be faked too? This is for my compiler, I want to have all kinds of variables there...

I see there is a binary precision and processor precision variables, but what do they do?

They are supposed to emulate integer-only arithmetics, but they don't do that yet. They should be doing that in next update though, but that's not yet here