Path: utzoo!yunexus!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!think!snorkelwacker!spdcc!ima!esegue!compilers-sender
From: larus@primost.cs.wisc.edu (James Larus)
Newsgroups: comp.compilers
Subject: Re: No Aliasing Compile Option
Keywords: aliases, optimizing compilers, C
Message-ID: <1990May29.161255.1963@esegue.segue.boston.ma.us>
Date: 29 May 90 16:12:55 GMT
Article-I.D.: esegue.1990May29.161255.1963
References: <54866@microsoft.UUCP> <4886@daffy.cs.wisc.edu> <26547195.F69@tct.uucp> <1990May19.141401.4350@ux1.cso.uiuc.edu> <265861D7.3293@tct.uucp> <1913@cod.NOSC.MIL>
Sender: compilers-sender@esegue.segue.boston.ma.us
Reply-To: larus@primost.cs.wisc.edu (James Larus)
Organization: University of Wisconsin--Madison
Lines: 21
Approved: compilers@esegue.segue.boston.ma.us

Let me stick in a plug for some work that I've done recently.  Univ. of
Wisconsin Computer Science Tech Report #929** contains some measurements of
6 C programs (including gcc).  The measurements were done to estimate the
potential parallelism of the programs, but as a side-effect, I measured the
number and type of loop-carried data dependences in the program.  These
dependences are very frequent and give an indication of the consequences of
blindly ignoring aliases and heavily optimizing a program.

/Jim

** Copies available from Kathy Schultz (kathy@cs.wisc.edu) or

Technical Reports Librarian
Computer Sciences Department
University of Wisconsin
1210 W. Dayton St.
Madison, WI 53706
-- 
Send compilers articles to compilers@esegue.segue.boston.ma.us
{spdcc | ima | lotus}!esegue.  Meta-mail to compilers-request@esegue.
Please send responses to the author of the message, not the poster.