X-Recipient: archive-cygwin AT delorie DOT com X-SWARE-Spam-Status: No, hits=-0.9 required=5.0 tests=AWL,BAYES_00,SPF_NEUTRAL X-Spam-Check-By: sourceware.org Message-ID: <4E695104.3030104@cs.utoronto.ca> Date: Thu, 08 Sep 2011 19:34:28 -0400 From: Ryan Johnson User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:5.0) Gecko/20110624 Thunderbird/5.0 MIME-Version: 1.0 To: cygwin AT cygwin DOT com Subject: Re: [ANNOUNCEMENT] Updated: stable compiler package gcc4-4.5.3-2 References:

<4E68AA10 DOT 4070205 AT cs DOT utoronto DOT ca> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-IsSubscribed: yes Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com Delivered-To: mailing list cygwin AT cygwin DOT com On 08/09/2011 4:17 PM, Frédéric Bron wrote: >> In my experience, these sorts of errors arise when there really is something >> wrong with the code, usually involving sizes of integers (e.g. 'long' in >> struct's definition and 'int' in another, on a 64-bit machine). However, the >> linker is poorly-equipped to detect such errors unless the resulting objects >> have different sizes. >> >> So, while I wouldn't rule out miscompilation, I would first check for >> silently conflicting definitions in different compilation units. > Does the linker message gives any hint on where to look? Sort of... the error message mentions Test.o and the section `.rdata$_ZTISs' You also presumably have the linker command line, which should tell you what other .o and .a are involved. Invoke `objdump -h' and pass as args all those .o/.a and examine the output. Most likely you'll see several files containing the offending section, and among those there are probably two distinct sizes reported. From there you have to work backwards to figure out what was different about #defines and #includes in the two source files that might have caused the discrepancy. Top suspects: custom typedefs or #defines which confuse system headers, or including Windows headers which make superficially equivalent declarations (but which have completely different and incompatible implementations). Given that mingw doesn't complain my money's on the latter. Ryan -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple