Re: LF: On-calc assembler
[Prev][Next][Index][Thread]
Re: LF: On-calc assembler
On Fri, 13 Dec 1996, Hmmm wrote:
> I remember messages being posted earlier that an on-calc assembler is
> impractical because the calc has to be reset if an error occurs. Therefore,
> you'd have to resend the assembler and everything after resetting the calc.
> Well, the error vectors (illegal instruction, etc) are stored at the
> beginning of RAM, correct? Why not add custom error routines to fargo by
> changing those vectors to point to new routines rather than the ones in ROM?
> Then, when an error occurs, rather than freeze the calc, just give an error
> message and then reenter the OS. If one did that, an on-calc assembler
> would be possible, not to mention it save a lot of headaches finding bugs in
> beta software.
>
> -Mark-
I was working on an oncalc assembler (it's almost ready) and I intended
to do just that; Patch all exception vectors and save stackpointers prior
to running a program. It wouldnt help problems like memory trashing but
it'd propably help a bit.
//Sasq (Jonas Minnberg)
Follow-Ups:
References: