Re: [Caml-list] Re: (GC issues) Alternative Bytecodes for OCaml

["Basile Starynkevitch [local]" <basile.starynkevitch@inria.fr>]

On Sun, Aug 29, 2004 at 12:12:05PM +0200, Nicolas Cannasse wrote:
> > >     * OCaml GC but with all C allocations made directly into the old
> > > generation (for conservativness)
> >
> > You still has to tell the (exact) minor collector about these
> > allocations (ie something like putting them on the store list) - so
> > you won't gain much in C coding style, and you'll lose in
> > performance. The result of C primitives is usually a temporary value
> > (e.g. because C primitives tend to be wrapped by Ocaml code doing
> > checking and changing C errors into Ocaml exceptions), so you'll
> > probably better put then in the young generation.
> 
> Since old generation collection is mostly conservative (compaction taken
> appart) , allocating *by default* into the old generation and having a C
> stack scanner could be a good compromise. [....]

No, Ocaml's major GC (the garbage collection of the old generation) is
not conservative; it is precise (it knows exactly where the pointers
are), but is an incremental mark&sweep - with occasionnal call to the
compactor whichs compacts the whole heap (including old stuff) by
moving values & updating pointers.

So I don't think Ocaml's GC could be easily patched to have the old GC
work conservatively - hence avoiding the requirement of telling the
runtime what are the pointers (ie CAMLlocal* & CAMLparam* macros).

(I'm not sure that Nicolas Cannasse uses the "conservative" word in
the established sense of the GC experts - perhaps my previous
explanations where not clear enough).

There is no way to conservatively scan the C stack, because the Ocaml
GC may move pointers (and it has to, in order to be performant).

(Damien Doligez would probably explain better than I could)

Regards.

-- 
Basile STARYNKEVITCH -- basile dot starynkevitch at inria dot fr
other email : basile at starynkevitch dot net
Project cristal.inria.fr - temporarily
http://cristal.inria.fr/~starynke --- all opinions are only mine 

-------------------
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