Richard Mulvey wrote :

> > > In one of the books
> > > I'm reading, the author explicitly states that he never uses
> > > relocatable mode, because the linker is too slow, etc.
> >
> > Just silly... Must be a realy old book from the IBM PC-XT
> > timeframe :-)
> >
>
> Actually, it's a new edition of a book by an author who has gotten
> mentioned here in a positive light several times, which is why I
> didn't specify the title.  ;-)

Ops ! What a mistake !
Now *I* might have to "relocate" and get me a hidden home address :-)

> > > Anyhow - is there some inherent advantage of absolute mode that
> > > I'm missing?
> >
> > It *is* easier to learn as a newbie.
> >
> I suppose in some senses it could be, though from my perspective
> as someone who has bought a couple of different PICs to play with,
> it seems much easier to me to just be able to specify "This is code,
> this is data", and not have to remember that I can start at
> 0x20 on one processor, and 0x3c on the next.  ;-)

Yes, you are right in that that is a *better* way of handling
thins issues, even if I still claim that it has a more step
learning curve to *begin* directly with reloc.code.

Actualy, this is an issue I'm a bit concerned about right now.
I'm discussing with our local technical high school to use PIC's as
the base for running some last-term-projects. I'm are currently
specifying what should be included in the setup and how things
should be run. The plan is to use Wouters Dwarf boards
("http://www.voti.nl/dwarf/index.html") as the hardware
platform, and MPLAB for the development.

Now, *currently* I think that we just simply let them
use absolute mode. I don't want this to be too complicated.
And it's far easier to write a few short code examples
i abs code, without having to go into all the bits to get
a relocatable environment up-n-running. And Olins
environment is, I think, just too much in this case.

And also, they will probably just write a single application,
so there is no real reuse of code to benefit from here, and we
don't want them to simple copy others code either...

> Especially when I've also noticed that a lot of example code
> on the web seems to have large number of subroutines that
> never actually get called, but are still being included in the
> absoluete code... where a library and linker would remove the
> extra cruft.

I think that the MPLINK linker *do* includes unused subs,
if they wasn't written as individual "code sections", not ?
I might be wrong here...

Regards,
Jan-Erik.

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu