Re: Queue machine compiler

George Neuner <>
11 Jun 2004 14:11:38 -0400

          From comp.compilers

Related articles
Queue machine compiler (Ben) (2000-09-08)
Queue machine compiler (Benjamin Ylvisaker) (2004-06-09)
Re: Queue machine compiler (George Neuner) (2004-06-11)
Re: Queue machine compiler (Benjamin Ylvisaker) (2004-06-21)
Re: Queue machine compiler (George Neuner) (2004-07-13)
| List of all articles for this month |

From: George Neuner <>
Newsgroups: comp.compilers
Date: 11 Jun 2004 14:11:38 -0400
Organization: Compilers Central
References: 04-06-030
Keywords: architecture
Posted-Date: 11 Jun 2004 14:11:38 EDT

On 9 Jun 2004 00:27:48 -0400, Benjamin Ylvisaker
<> wrote:

>Does anyone know of a compiler that can generate code for a machine
>based on an operand queue? I would be particularly interested in an
>open source/research compiler.

The description is vague, but it sounds like "queue machine" might be
refering to a "data flow" architecture. Unfortunately, data flow has
no particular hardware implementation - there are a number of
reasonable ways to do it and they are all topics of research. No
available compiler is likely to do what you want without major

The most popular technique for data flow is combinator graph
reduction. A number of functional languages use this approach because
it can be simulated fairly easily on stock hardware. CG compilers
normally target either a virtual machine or an abstract one simulated
by their runtime library. Starting with one of these FPL compilers is
probably your best bet.

There is also a less popular approach which uses associative memory
and a resolution engine as the execution model. It's primarily used
by some constraint logic and expert system languages, and by a few
parallel programming "coordination" languages. If you have real
hardware available, some of these might be worth looking into.

Send real email to GNEUNER2 at COMCAST o NET

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.