On 01/06/2011 19:40, Wouter van Ooijen wrote:
> A problem: a stack overflow is harmless. It is the stack*underflow*
> that kills your application. Instrumenting RETURN is probably no
> problem, but instrumenting RETLW instructions might cause problems with
> jump tables.
>
> An if you use a HLL you will need access to the sources of the compiler.
> With anything but Jal that might be a problem:)

Is the source for your original JAL available?

I know the JalV2 Kyle C source is available. As I studied it to try and=20
understand which bit makes the P-code and which bit turns his p-code=20
into PIC Assembler

Hurt my poor head. I decided wasn't *THAT* determined to do an alternate=20
back end (nice though the thought might be) for some other cpu type.

The current JAL compiler of course reports how much stack is used=20
(adding a function call inside Interrupt handler takes the 16F876 from=20
50% to 75% which is not surprising, For more "meaty" projects I think I=20
prefer the 18F family. I bought the pickit2 to play with the 18FxxJxx=20
types, and some of Wouter's handy DIL SMD adaptor boards for the chips=20
till I get some PCBs made.

Not to hard to hand solder with 2.5mm tip if PCB tinned and any bumps=20
removed with solder wick. I just press at the two opposite corners to=20
reflow in position and if aligned then press all the pins in turn.

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.