savannah-hackers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Savannah-hackers] submission of The ML-- Compiler - savannah.nongnu.org


From: lpadovan
Subject: [Savannah-hackers] submission of The ML-- Compiler - savannah.nongnu.org
Date: Mon, 05 Jan 2004 09:22:21 -0500
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031107 Galeon/1.3.11a (Debian package 1.3.11a-1)

A package was submitted to savannah.nongnu.org
This mail was sent to address@hidden, address@hidden


Luca Padovani <address@hidden> described the package as follows:
License: gpl
Other License: 
Package: The ML--  Compiler
System name: mlmm
Type: non-GNU

Description:
The goal of the ML-- project is to build a compiler for an ML-like
language that works as nicely as possible with the other development
tools of the GNU family (compilers, linkers, debuggers, configuration
tools), with the minimum effort from the developer required.  In a
first stage the ML-- compiler will output C code (using some of the
GCC C extensions) but other possibilities will be taken into account
for the future, in particular: (a) interfacing with GCC backend, 
either by direct coding or through the KSI interface (see
http://gont.pld.org.pl/); (b) generating C-- code
(http://www.cminusminus.org/) which, although less compatible with
code produced by GCC, allows for interesting code optimizations.  In
the (a) scenario the runtime system will use the Boehm garbage
collector, in the (b) scenario an ad-hoc, efficient garbage collector
may be implemented.  There is an existing ML dialect
(http://caml.inria.fr) which is becoming increasingly popular.
However here are some of its shortcomings: (1) it is difficult to
make it work nicely with GNU systems. It has its own library management system, 
its own function calling conventions, its own
autoconfiguration facilities. (2) it is used as a prototype
implementation for research purposes, which means that fancy features 
are added periodically which increase the language complexity and
also the number of different ways for implementing the same thing. Although the 
exact ML dialect that ML-- is meant to implement is
still to be determined, ML-- aims at being simpler than Ocaml, and
by being tightly coupled with GCC it should favor multi-language 
development rather than trying to be a all-in-one language.  Ocaml
will be used in the early stages of the development until the
compiler will be able to bootstrap itself.

Source URL (NOT released, this is a preliminary snapshot of the code
on my hard drive):
http://www.cs.unibo.it/~lpadovan/secret/mlmm-0.0.1.tar.gz

Other Software Required:
Gont/KSI: http://gont.pld.org.pl/
C--: http://www.cminusminus.org/
Ocaml: http://caml.inria.fr
Boehm: http://www.hpl.hp.com/personal/Hans_Boehm/gc/


Other Comments:



_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/







reply via email to

[Prev in Thread] Current Thread [Next in Thread]