Re: Analysis and optimization frameworks related work?

cliffc@risc.sps.mot.com (Cliff Click)
15 Oct 1996 09:39:31 -0400

          From comp.compilers

Related articles
Analysis and optimization frameworks related work? jdean@stash.pa.dec.com (1996-10-12)
Re: Analysis and optimization frameworks related work? creusil@cri.ensmp.fr (1996-10-15)
Re: Analysis and optimization frameworks related work? cliffc@risc.sps.mot.com (1996-10-15)
Re: Analysis and optimization frameworks related work? jyates@gis.net (1996-10-16)
Re: Analysis and optimization frameworks related work? assmann@ipd.info.uni-karlsruhe.de (1996-10-18)
Re: Analysis and optimization frameworks related work? uday@cs.unipune.ernet.in (Uday P Khedker) (1996-10-20)
| List of all articles for this month |
From: cliffc@risc.sps.mot.com (Cliff Click)
Newsgroups: comp.compilers
Date: 15 Oct 1996 09:39:31 -0400
Organization: none
References: 96-10-044
Keywords: tools

jdean@stash.pa.dec.com (Jeff Dean) writes:


      This is a request for information about related work on frameworks
      that make it easier to design and implement compiler optimizations.


      I'd also appreciate any philosophical comments or anecdotes that
      people might have about such frameworks..


Try my IR'95 paper or any of the VDG papers for a different kind of
framework. I do things completely sparse, no bitvectors please.




          o Multiple analyses & transformations can easily be combined
      together to run in parallel when there are synergistic effects from
      doing so (Click & Cooper have developed a theoretical framework that
      specifies when combining multiple optimizations together is desirable;
      our framework makes combining such optimizations easy to do in
      practice).


Wow! I always wondered if this sank into the deep, or if
somebody read it.


Cliff


--
Cliff Click, Ph.D. Compiler Researcher & Designer
RISC Software, Motorola PowerPC Compilers
cliffc@risc.sps.mot.com (512) 891-7240
http://members.aol.com/mjclick1
--


Post a followup to this message

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