From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-0.3 required=5.0 tests=BAYES_00,
	REPLYTO_WITHOUT_TO_CC,XPRIO autolearn=no autolearn_force=no
	version=3.4.4
X-Google-Thread: 103376,afb4d45672b1e262,start
X-Google-Attributes: gid103376,public
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news1.google.com!news2.google.com!news3.google.com!border1.nntp.dca.giganews.com!nntp.giganews.com!local01.nntp.dca.giganews.com!nntp.pipex.net!news.pipex.net.POSTED!not-for-mail
NNTP-Posting-Date: Sat, 01 Apr 2006 07:47:53 -0600
Reply-To: "Doobs" <doobs@doobs.com>
From: "Doobs" <doobs@doobs.com>
Newsgroups: comp.lang.ada
Subject: Any way of persuading GNAT/GCC to implement a true overlay and not a
 pointer?
Date: Sat, 1 Apr 2006 14:47:59 +0100
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2900.2527
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2527
X-RFC2646: Format=Flowed; Original
Message-ID: <LIednWDbPNgXGbPZRVnysQ@pipex.net>
NNTP-Posting-Host: 85.210.60.99
X-Trace: 
 sv3-DUbt6HM8rx4h3JAi5DB4BZF3lsUoxE8ReomYV50t1/tCYj4SZaQYlTIuOjJ5/domLHq2X4una3fWkHi!rX5QNiQ+4A6XD80riA/1AKAjUB7j01KwP65wRaPgZUAvZ8hgz49+EBZ7uI+mT9WxAGY4U28Cyy52!S9kIC6Md0Q==
X-Complaints-To: abuse@dsl.pipex.net
X-DMCA-Complaints-To: abuse@dsl.pipex.net
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint
 properly
X-Postfilter: 1.3.32
Xref: g2news1.google.com comp.lang.ada:3694
Date: 2006-04-01T14:47:59+01:00
List-Id: <comp.lang.ada>

I was under the impression that code of the following form :

X : <Some Type>;
Y : <Some Type>;
for Y'Address use X'Address;

would result in an overlay in the resulting code.   However on a project 
where external test equipment showed that the contents of  X were NOT the 
same as Y I investigated further.  I checked on both a Windows host version 
of GCC (3.4.2) and a target cross compiler (GCC 3.4.4) and on both the 
construct above is actually implemented using indirection (a pointer).  The 
map file and symbol table show that irrespective of the data type and size 
of Y it is always implemented as a pointer to X.  Although this appears 
semantially identical to an overlay as far as the progam is concerned it is 
NOT identical as far as an external observer is concerned.    I have a bit 
packed record which I wanted to return as an array of longwords to pretty 
dumb test equipment. The easiest way *would* be to overlay a longword 
array - however this would not appear to be possible.....