From: Martin Stromberg Message-Id: <200211081254.NAA18624@lws256.lu.erisoft.se> Subject: Re: Weird ceil() and other results To: djgpp-workers AT delorie DOT com Date: Fri, 8 Nov 2002 13:54:42 +0100 (MET) In-Reply-To: <005501c2871e$2a5b9990$0100a8c0@p4> from "Andrew Cottrell" at Nov 08, 2002 10:57:47 PM X-Mailer: ELM [version 2.5 PL3] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: djgpp-workers AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp-workers AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk > > If I comment out the #include from the program below then the > program compiles and links without any errors or warnings. The results are:- > date = 52.680000 > date = 51.680000 > decimalPart = 1072705824.000000 > wholePart = 1077938047.000000 > down = 1077939071.000000 > > If I include the #inclue from the program below then the program > compiles and links without any errors or warnings. The results are:- > date = 52.680000 > date = 51.680000 > decimalPart = 0.680000 > wholePart = 51.000000 > down = 52.000000 > > These results are correct.... > > Does anyone know why I would get wrong results, but no error messages if I > do not include math.h? I would have expected to get error messages > indicating a problem rather than a bad exe. The reason it breaks when it breaks is that undeclared functions in C defaults to returning int. So when something else than int is returned things breaks. Does "-Wall" generate a warning message? Right, MartinS