Re: Stack-based IR vs. Register-based IR

Patrick Logan <patrick_d_logan@ccm.hf.intel.com>
23 Jul 1996 23:36:29 -0400

          From comp.compilers

Related articles
Stack-based IR vs. Register-based IR evan@top.cis.syr.edu (Evan Cheng) (1996-07-20)
Stack-based IR vs. Register-based IR gough@dstc.qut.edu.au (John Gough) (1996-07-22)
Re: Stack-based IR vs. Register-based IR patrick_d_logan@ccm.hf.intel.com (Patrick Logan) (1996-07-23)
Re: Stack-based IR vs. Register-based IR anton@a0.complang.tuwien.ac.at (1996-07-26)
| List of all articles for this month |

From: Patrick Logan <patrick_d_logan@ccm.hf.intel.com>
Newsgroups: comp.compilers
Date: 23 Jul 1996 23:36:29 -0400
Organization: Intel
References: 96-07-137
Keywords: optimize

Evan Cheng wrote:
>
> I am hoping to induce some discussion on the merits (or lack of)
> of stack-based intermediate representations and register-based ones.
> Pointers to papers would be much appreciated as well.


See Richard Kelsey's thesis at Yale (ca. 1988-89?) and the subsequent
work done by Appel, et al. on the SML/NJ compiler.


These works indicate that a continuation-passing/closure-passing
style of IR is better than either stack or register based IR for
a variety of reasons.


--
mailto:Patrick_D_Logan@ccm.hf.intel.com


"Analysis means taking things apart for examination, while
design is to work out the structure or form of something."
-Seamless Object-Oriented Software Architecture, Walden & Nerson
--


Post a followup to this message

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