Re: How to implement lexical closures?

Chris F Clark <cfc@shell01.TheWorld.com>
Mon, 17 May 2010 17:02:34 -0400

          From comp.compilers

Related articles
[7 earlier articles]
Re: How to implement lexical closures? gene.ressler@gmail.com (Gene) (2010-05-12)
Re: How to implement lexical closures? gah@ugcs.caltech.edu (glen herrmannsfeldt) (2010-05-13)
Re: How to implement lexical closures? gneuner2@comcast.net (George Neuner) (2010-05-15)
Re: How to implement lexical closures? gneuner2@comcast.net (George Neuner) (2010-05-15)
Re: How to implement lexical closures? gah@ugcs.caltech.edu (glen herrmannsfeldt) (2010-05-16)
Re: How to implement lexical closures? gneuner2@comcast.net (George Neuner) (2010-05-17)
Re: How to implement lexical closures? cfc@shell01.TheWorld.com (Chris F Clark) (2010-05-17)
Re: How to implement lexical closures? gah@ugcs.caltech.edu (glen herrmannsfeldt) (2010-05-19)
Re: How to implement lexical closures? anton@mips.complang.tuwien.ac.at (2010-05-20)
| List of all articles for this month |

From: Chris F Clark <cfc@shell01.TheWorld.com>
Newsgroups: comp.compilers
Date: Mon, 17 May 2010 17:02:34 -0400
Organization: The World Public Access UNIX, Brookline, MA
References: 10-05-031 10-05-072 10-05-076 10-05-084 10-05-089 10-05-091 10-05-099
Keywords: PL/I, history, comment
Posted-Date: 19 May 2010 00:50:39 EDT

> [That sounds right, but it depends on the version of PL/I. I have
> the impenetrable 1976 PL/I standard and don't see anything about
> threads. -John]


When I worked at Pr1me, we had Freiberghouse's "full" PL/I compiler as
well as the subset G compiler. Neither had the concept of threads to
my knowledge, even the OS didn't. Instead Pr1me had a low cost
process exchange, comparable in cost to some threading context swaps,
IIRC save/restoring 4 to 8 registers could effect a process exchange.
A call instruction save/restored the same set. Thus, a procedure call
and a process exchange were essentially interchangeable.


I never used the IBM PL/I compilers, nor the Burrough's, VAX, or DG
ones. I did use the Multics one and don't remember threads there
either. However, my memory is that PL/I predates threading as a
significant concept. I recall threading being important on the DEC
Alpha, but that was in C.


Hope this helps,
-Chris


******************************************************************************
Chris Clark email: christopher.f.clark@compiler-resources.com
Compiler Resources, Inc. Web Site: http://world.std.com/~compres
23 Bailey Rd voice: (508) 435-5016
Berlin, MA 01503 USA twitter: @intel_chris
------------------------------------------------------------------------------
[PL/I was developed at the same time as OS/360, and the threading in IBM
PL/I was basically the threading in OS/MVT. -John]


Post a followup to this message

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