> That is what I assumed from what had been said.
> At startup it would be easy [tm] to fill the stack with return
> addresses that all went to the same place.

That is possible, am curious how many piclisters can figure out how to=20
do that.

> When max stack depth is to be determined, as long as the stack has
> never overrun, the current depth is known and can be would back to
> zero.

I think you mean: after the program has run, and has returned each call=20
it has done, you could use those 'same addresses' to to determine how=20
many stack entries were never used.

> NOW another pop rolls the stack back to the top and subsequent
> pops produce returns to defined locations as long as the contents have
> not been destroyed.

not pops, those do not exist on 14-bit cores. use RETURN.

 > When you strike the first "used" location the
> return will not go to where you expect.

 > A timer can haul the system
> back into control and note can be made of how many locations returned

clever idea.

But do you realize that the PIC stack is not circular? When you pop a=20
return address, the 'oldest' address in stack is duplicated.

Your scheme could still be used, with a small change, and at the cost of=20
an extra stack entry. The solution is of course left as an exercise to=20
the reader.

--=20

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: www.voti.nl/hvu

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