VBCC: building a TextEdit syntax plugin for OS3.2

[boemann]

Uhm the plugins doesn't have to be reentrant but the highlighter hook has to as it will be called from several texteditor.gadgets. The current state is given as part of the highlighter msg. The plugin must not store anything else regarding current state. Also the highlighter is executed in the input task so no dos calls either.

[bebbo]

Originally Posted by phx

...


I agree this is generally a good idea for reentrant programs. Maybe even for normal ones. But it usually forces you to write compiler-specific code when doing OS-calls, which depend on a static library base.

some compilers like sc always refer to global variables for *bases.


Using macros, like gcc does, local variables can be used for *bases too. Maybe one can build something simliar for sc to replace the pragma libcall stuff...

[boemann]

Originally Posted by bebbo

some compilers like sc always refer to global variables for *bases.


Using macros, like gcc does, local variables can be used for *bases too. Maybe one can build something simliar for sc to replace the pragma libcall stuff...

SAS/C can refer to local bases too while still using the pragma stuff

[boemann]

So do you guys prefer stack or registers arguments ?

Also Guy: how is it going. Any progress?

[guy lateur]

Originally Posted by boemann

Also Guy: how is it going. Any progress?

I haven't done any more work on this, I'm afraid, but if I do I'll be sure to post about it here. I've kinda gotten used to cross-assembling in the meantime, tbh..

[phx]

Originally Posted by boemann

So do you guys prefer stack or registers arguments ?

The question is in which situation? For your own, private projects: yes, you can do that, as long as it gives you an advantage (depends on the number of registers allocated for it). For public code I would refrain from it, because it is not a standard m68k ABI and may be implemented differently in C compilers, if at all.

I know SAS/C can do it, but I'm not sure which ABI it follows there. Does it use only scratch registers for arguments? I also remember that function names become

@name

instead of

_name

, when it takes register arguments.

For vbccm68k we have the option -fastcall (and also the

__regargs

attribute for individual functions), which uses d0/d1/a0/a1 for register arguments (a0/a1 for pointers only) and pushes the rest onto the stack. Not sure how compatible that would be. I guess you can always run into situations which are different.

[boemann]

Thanks - I meant for the plugins, and if i were to require passing by registers i would obviously specify which regs using __reg and the like

[phx]

Ah, ok. Thats' fine!