Path: utzoo!attcan!uunet!nih-csl!lhc!adm!cmcl2!kramden.acf.nyu.edu!brnstnd
From: brnstnd@kramden.acf.nyu.edu (Dan Bernstein)
Newsgroups: comp.lang.misc
Subject: Re: Aggressive optimization
Message-ID: <6238:Oct1922:30:0690@kramden.acf.nyu.edu>
Date: 19 Oct 90 22:30:06 GMT
References: <1990Oct18.212844.14728@murdoch.acc.Virginia.EDU> <25336:Oct1823:13:3990@kramden.acf.nyu.edu> <1559@exodus.Eng.Sun.COM>
Organization: IR
Lines: 14

In article <1559@exodus.Eng.Sun.COM> chased@rbbb.Eng.Sun.COM (David Chase) writes:
> And, whether you agree with the choice of dialects or not, it's got
> nothing to do with "buggy optimizations" -- either global variables
> are volatile (i.e, programs execute in serial order), or they aren't.
> I believe that is what were discussing in the first place.  There's no
> way to "debug" the optimization of references to global variables if
> those references are intended to be treated as volatile.

Fair enough. I'll stop calling the assumption of nonvolatility a bug,
though I still consider it a huge mistake.

There are lots of other examples of optimizer bugs, so the point stands.

---Dan