Reimplementing OCaml on LLVM

Reimplementing OCaml on LLVM

[Jon Harrop]

Many people including myself have shown considerable interest in using the 
LLVM project from OCaml and, in particular, in implementing an entire FPL 
using this approach.

I'm just wondering if any of the OCaml team would be interested in doing or 
helping with a port themselves? I doubt it could count as "work" but I think 
there are plenty of opportunities for fun tinkering here. :-)

Even if the result were only a toy implementation of a language I am sure 
examples would help non-experts enormously in getting to grips with how such 
a compiler might be written.

This also presents an awesome opportunity to address some of the shortcomings 
of OCaml's current implementation, such as polymorphism, FFI, unboxing and so 
forth and to test alternative approaches.

-- 
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com/products/?e

Re: [Caml-list] Reimplementing OCaml on LLVM

[Francois Rouaix]

[-- Attachment #1: Type: text/plain, Size: 1594 bytes --]

As a quick feasability check, you might start by checking if the backend can
actually support OCaml's run-time concepts. ocamlc has two undocumented
options -dlamba and -dinstr which could be used to generate intermediate
representations. Alternatively, JIT from the bytecode might be a quick hack
as well.--f

On Jan 12, 2008 6:55 AM, Jon Harrop <jon@ffconsultancy.com> wrote:

>
> Many people including myself have shown considerable interest in using the
> LLVM project from OCaml and, in particular, in implementing an entire FPL
> using this approach.
>
> I'm just wondering if any of the OCaml team would be interested in doing
> or
> helping with a port themselves? I doubt it could count as "work" but I
> think
> there are plenty of opportunities for fun tinkering here. :-)
>
> Even if the result were only a toy implementation of a language I am sure
> examples would help non-experts enormously in getting to grips with how
> such
> a compiler might be written.
>
> This also presents an awesome opportunity to address some of the
> shortcomings
> of OCaml's current implementation, such as polymorphism, FFI, unboxing and
> so
> forth and to test alternative approaches.
>
> --
> Dr Jon D Harrop, Flying Frog Consultancy Ltd.
> http://www.ffconsultancy.com/products/?e
>
> _______________________________________________
> Caml-list mailing list. Subscription management:
> http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
> Archives: http://caml.inria.fr
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs
>

[-- Attachment #2: Type: text/html, Size: 2229 bytes --]