Related articles |
---|
automatic inlining? dash@hpbs3308.boi.hp.com (1996-02-23) |
Re: automatic inlining? jmattson@hpax.cup.hp.com (Jim Mattson) (1996-02-23) |
Re: automatic inlining? brianmcg@interaccess.com (1996-02-24) |
Re: automatic inlining? snovack@justright.ICS.UCI.EDU (Steven Novack) (1996-02-24) |
Re: automatic inlining? jejones@microware.com (1996-03-08) |
From: | Steven Novack <snovack@justright.ICS.UCI.EDU> |
Newsgroups: | comp.compilers |
Date: | 24 Feb 1996 23:32:43 -0500 |
Organization: | UC Irvine Department of ICS |
References: | 96-02-273 |
Keywords: | optimize, bibliography |
dash@hpbs3308.boi.hp.com (Dan Ash) writes:
>[are there compilers that in-line a routine if it's only called from one
>place?]
You might want to take a look at the FIAT system by Mary Hall et. al.
It not only handles automatic inlining, but also cloning (i.e., don't
inline, but make a copy that will be called from a known, fixed set of
places, thus allowing for better interprocedural analysis and
optimizations).
I've included a reference below (there may be better references ---
this is just one that I happen to have on hand).
Good luck,
Steve
@INPROCEEDINGS{fiat.lcpc93,
author = "Mary W. Hall and John M. Mellor-Brummey
and Alan Carle and Ren\'{e} G. Rodr\'{i}guez",
title = "FIAT: A Framework for Interprocedural Analysis and Transforma
tions",
pages = "522--545",
key = "",
crossref = "lcpc.93",
}
@PROCEEDINGS{lcpc.93,
key = "LCPC93",
editor = "Utpal Banerjee and David Gelernter and Alex Nicolau and
David Padua",
title = "Proc. Sixth Workshop on Languages and Compilers for Parallel
Computing",
booktitle = "Proc. Sixth Workshop on Languages and Compilers for
Parallel Computing",
series = "Lecture Notes in Computer Science",
volume = "768",
month = aug,
year = 1993,
address = "Portland, OR",
publisher = "Springer Verlag",
}
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.