Re: Current work in compiler/language design.

hwloidl@risc.uni-linz.ac.at (Hans Wolfgang Loidl)
Tue, 12 Nov 1991 06:30:19 GMT

          From comp.compilers

Related articles
Current work in compiler/language design. hackeron@Athena.MIT.EDU (Harris L. Gilliam - MIT Project Athena) (1991-11-10)
Re: Current work in compiler/language design. preston@dawn.cs.rice.edu (1991-11-11)
Re: Current work in compiler/language design. hwloidl@risc.uni-linz.ac.at (1991-11-12)
Current work in compiler/language design. objsys@netcom.com (1991-11-14)
Re: Current work in compiler/language design. preston@dawn.cs.rice.edu (1991-11-16)
Re: Current work in compiler/language design. martens@laurel.cis.ohio-state.edu (1991-11-17)
Re: Current work in compiler/language design. objsys@netcom.com (Bob Hathaway) (1991-11-18)
Re: Current work in compiler/language design. carlton@husc8.harvard.edu (1991-11-19)
Re: Current work in compiler/language design. chambers@cs.washington.edu (1991-11-18)
[10 later articles]
| List of all articles for this month |
Newsgroups: comp.compilers
From: hwloidl@risc.uni-linz.ac.at (Hans Wolfgang Loidl)
Keywords: design
Organization: RISC, J.K. University of Linz, Austria
References: 91-11-030
Date: Tue, 12 Nov 1991 06:30:19 GMT

In article 91-11-030, hackeron@Athena.MIT.EDU (Harris L. Gilliam) writes:
> What are the current hot spots in compiler/language design ?


>From my experience in the area of parallel computation I can say that
currently really much work goes into the direction of a kind of
automatically parallelizing compilers.


Many users of really big FORTRAN programs (most of these users are
non-computer scientists of course, who else would program in FORTRAN :-)
are dreaming of compilers that automatically distribute the work over
several processors and speed up computation without having to change
anything in the source code, which can't be understood by any human being
anyways. But despite really much efforts into this direction there hasn't
been a breakthrough yet even when one restricts oneself to FORTRAN not to
speak of C.


However, another direction is to look for other languages that are more
suitable for automatic parallelization as they themselves contain more
inherent parallelism than the usual programming languages. Among these
languages there are for example functional and logic languages. As I
myself am currently working on an automatically parallelizing compiler for
a functional language I just *have* to think that this approach is quite
promising.


I hope this information helps a bit.


                                                                            Hans Wolfgang Loidl
                                                                      hwloidl@risc.uni-linz.ac.at
--


Post a followup to this message

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