On Mon, Aug 26, 2002 at 12:00:31PM +0200, Xavier Leroy wrote:
> > Alternatively, strip can be fixed to know about ocaml bytecode.
> 
> I wouldn't recommend that -- if I were the maintainer of "strip", I'd
> scream at such a kludge :-)

:)))

That said, maybe strip has a list of things not to strip or something
such ?

> > I was also told that if the ocaml team would have used another
> > label/tag/whatever it is called in the binary code than the one it is
> > using, then strip would know about it.
> 
> Yes, a cleaner solution would be to embed the OCaml bytecode in a
> special ELF section.  But there are two problems with this: 1- the ELF
> file format is quite complex, and 2- this solution would break utterly
> on non-ELF systems, e.g. Windows.

Ok, you know a lot more than me about this, but couldn't it not be
possible to implement this only on ELF systems ?
Anyway, the resulting code is arch/system dependant, since you embed
it with ocamlrun anyway, is it not ?

> The current solution (just stick the bytecode at the end of the native
> executable) is a hack, but it's a hack that works on *every single
> platform* I've tested in the last 12 years...

Sure, a good reason for it i guess, but the result is that strip breaks
ocaml bytecode executables.

Friendly,

Sven Luther
-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners