Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!uunet!munnari.oz.au!mel.dit.csiro.au!yarra!melba.bby.oz.au!zvs
From: zvs@bby.oz.au (Zev Sero)
Newsgroups: comp.lang.c
Subject: Re: Portability question for structures.
Message-ID: <1991Jun20.225938.18098@melba.bby.oz.au>
Date: 20 Jun 91 22:59:38 GMT
References: <1991Jun19.194046.17317@axis.dsg.ti.com>
Sender: usenet@melba.bby.oz.au (news READER id)
Organization: Burdett, Buckeridge and Young Ltd.
Lines: 21
In-Reply-To: olson@cuda.aaet.csc.ti.com's message of 19 Jun 91 19:40:46 GMT
Nntp-Posting-Host: leo-gw

DO> What I need to know is, how portable are the list processing
DO> functions.  They are based on the assumption that the first three
DO> fields of the list structures will be void *, void * and char *.
DO> Is this a valid assumption? or can the order that fields are
DO> listed in a structure be changed by the compiler?


  `C compilers are constrained to assign components increasing memory
  addresses in a strict order, with the first component starting at
  the beginning address of the structure itself.'

			- H&S2 page 104

So the answer is, yes, they are portable (assuming that you have
provided for the holes in the struct).
--
				Zev Sero  -  zvs@bby.oz.au
`This constant intervention by government in tasks that belong to the
individual must cease, or ours will become as bureaucratic a government
as that of Russia.'   - Senator Thomas Bayard, speaking on an animal
			inspection bill, 28 Apr 1884.