Google Groups Home Help | Sign in
Message from discussion new release of the MLton Standard ML compiler
The group you are posting to is a Usenet group. Messages posted to this group will make your email address visible to anyone on the Internet.
Your reply message has not been sent.
Your post was successful
Matthew Fluet  
View profile
 More options Aug 30 2007, 2:53 pm
Newsgroups: comp.lang.functional
From: Matthew Fluet <Matthew.Fl...@gmail.com>
Date: Thu, 30 Aug 2007 18:53:24 -0000
Local: Thurs, Aug 30 2007 2:53 pm
Subject: new release of the MLton Standard ML compiler
We are pleased to announce a new release of MLton, the whole-program
optimizing compiler for Standard ML, available at http://mlton.org/.
MLton version 20070826 is the first public release since 20051202.

The major improvements are:

 + Support for 64-bit platforms.
 + Native amd64 codegen.
 + New platforms: AMD64/Linux, AMD64/FreeBSD, HPPA/HPUX,
    PowerPC/AIX, X86/Darwin.
 + Support for specifying default Int.int, Word.word, and Real.real
    types at compile-time.

MLton has the following features.

 + Portability.
  Runs on the following platforms.
  o AMD64: Linux, FreeBSD.
  o HPPA: HPUX (11.00 and above), Linux (Debian).
  o PowerPC: AIX (5.1 and above), Darwin (Mac OS X), Linux (Debian,
Fedora).
  o Sparc: Linux (Debian), Solaris (8 and above).
  o X86: Cygwin/Windows, Darwin (Mac OS X), FreeBSD, Linux, MinGW/
Windows,
     NetBSD, OpenBSD.
 + Robustness.
  o Supports the full SML 97 language as given in The Definition of
     Standard ML.
  o A complete implementation of the Basis Library.
  o Generates standalone executables.
  o Compiles large programs (hundreds of thousands of lines).
  o Support for large amounts of memory (up to 4G on 32-bit systems,
     more on 64-bit systems).
  o Support for large arrays (up to 2G elements).
  o Support for large files (using 64-bit file positions).
 + Performance.
  o Executables with excellent running times.
  o Native integers, reals, and words (untagged and unboxed).
  o Unboxed native arrays.
  o Multiple garbage collection strategies.
  o Fast arbitrary-precision arithmetic based on the GnuMP.
 + Tools.
  o Source-level profiling of both time and allocation.
  o ML-Lex lexer generator.
  o ML-Yacc parser generator.
  o ML-NLFFIGEN.
 + Extensions.
  o Simple and fast C FFI for calling from SML to C and from C to SML.
  o The ML Basis system for programming in the very large.
  o Libraries for C pointers, continuations, interval timers, random
     numbers, resource limits, resource usage, signal handlers,
     system logging, threads, and heap save and restore.

For more information, go to the MLton home page.

  http://mlton.org/

Enjoy!

-- The MLton Team


    Reply to author    Forward  
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.

Create a group - Google Groups - Google Home - Terms of Service - Privacy Policy
©2008 Google