Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!news.glorb.com!usenet.stanford.edu!news.iecc.com!.POSTED!nerds-end
From: christian.bau@cbau.wanadoo.co.uk
Newsgroups: comp.compilers
Subject: Re: Compiler optimization and floating point operations
Date: Mon, 28 Oct 2013 15:50:07 -0700 (PDT)
Organization: Compilers Central
Lines: 8
Sender: johnl@iecc.com
Approved: comp.compilers@iecc.com
Message-ID: <13-10-037@comp.compilers>
References: <13-10-026@comp.compilers> <13-10-029@comp.compilers>
NNTP-Posting-Host: news.iecc.com
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
X-Trace: leila.iecc.com 1383000713 74383 64.57.183.58 (28 Oct 2013 22:51:53 GMT)
X-Complaints-To: abuse@iecc.com
NNTP-Posting-Date: Mon, 28 Oct 2013 22:51:53 +0000 (UTC)
Keywords: arithmetic, optimize
Posted-Date: 28 Oct 2013 18:51:53 EDT
X-submission-address: compilers@iecc.com
X-moderator-address: compilers-request@iecc.com
X-FAQ-and-archives: http://compilers.iecc.com
Xref: csiph.com comp.compilers:1020

> [I think you mean they're not associative.  I don't know any situations
> where a+b != b+a, but lots where a+(b+c) != (a+b)+c -John]

Take fused multiply-add, and the expression a*b + c*d.
It can be evaluated either as fma (a, b, c*d) or fma (c, d, a*b).

It would seem natural that the code for a*b + c*d and for c*d + a*b
would be different, with different results.