Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!sdd.hp.com!zaphod.mps.ohio-state.edu!think!yale!cmcl2!stealth.acf.nyu.edu!brnstnd
From: brnstnd@stealth.acf.nyu.edu
Newsgroups: comp.lang.misc
Subject: Re: JLG's flogging of horses
Message-ID: <7149:Apr1102:11:2590@stealth.acf.nyu.edu>
Date: 11 Apr 90 02:11:25 GMT
References: <20026@megaron.cs.arizona.edu> <14319@lambda.UUCP>
Reply-To: brnstnd@stealth.acf.nyu.edu (Dan Bernstein)
Distribution: usa
Organization: IR
Lines: 13
X-Original-Subject: Re: Relationship between C and C++)

In article <14319@lambda.UUCP> jlg@lambda.UUCP (Jim Giles) writes:
> From article <20026@megaron.cs.arizona.edu>, by gudeman@cs.arizona.edu (David Gudeman):
> > but a good enough compiler can detect all possible aliases by global
> > control flow analysis, even with seperate compilation.
> That's just the point.  The _compiler_ can't do any such thing!  That's
> the _definition_ separate compilation - the different program units
> are unknown to each other at compile time.

That isn't true under ANSI C. Everything must work *as if* the program
units were separately compiled; but optimizations don't have to respect
separate compilation.

---Dan