Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
Date: Thu, 23 Sep 2004 15:43:32 -0500
From: Brian Ford <ford AT vss DOT fsi DOT com>
Reply-To: cygwin AT cygwin DOT com
To: cygwin AT cygwin DOT com
Subject: gcc >= 3.3.3-3 and binutils .rdata alignment
Message-ID: <Pine.CYG.4.58.0409231530530.1264@fordpc.vss.fsi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-IsSubscribed: yes

This is just a heads up for anyone compiling with gcc >= 3.3.3-3 and
<= current Cygwin binutils (2.15.91 20040725).  The issue fixed in the
reference below may also affect constant long long or double derrived data
structures since -malign-double is gcc's default on Cygwin:

http://sources.redhat.com/ml/binutils/2004-09/msg00230.html

Note that so far I have only experienced this problem with a custom
compiled gcc 3.4.3 in combination with -mfpmath=sse, so this is somewhat
speculative.  But, if gcc 3.3.3-3 really does put all constants in .rdata,
it seems likely that those data types described above may get misaligned
when sections from differing objects are linked together.

If you experience this mis-match bug, you will see a SIGSEGV due to a load
from a misaligned address.

HTH somebody...

-- 
Brian Ford
Senior Realtime Software Engineer
VITAL - Visual Simulation Systems
FlightSafety International
the best safety device in any aircraft is a well-trained pilot...

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/