Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm Sender: cygwin-owner AT sourceware DOT cygnus DOT com Delivered-To: mailing list cygwin AT sourceware DOT cygnus DOT com Message-ID: <19990524175534.4577.rocketmail@web116.yahoomail.com> Date: Mon, 24 May 1999 10:55:34 -0700 (PDT) From: Earnie Boyd Reply-To: earnie_boyd AT yahoo DOT com Subject: Re: gdb + cygwin To: Charles Wilson Cc: cygwin AT sourceware DOT cygnus DOT com MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii --- Charles Wilson wrote: > Earnie Boyd wrote: > > > > Hi Chuck, > > > > I've seen this when I have a pointer reference that isn't being referenced > > correctly. Perhaps a prototype is incorrect or not specified. Perhaps you > > have a pointer becoming an integer or an integer becoming a pointer, either > way > > you're in trouble. > > > > Use the -Wall switch on the compilation and don't ignore the warnings. If > it > > has to do with a pointer, it's an ERROR not a WARNING. > > Okay, I tried this: > gcc -Wall -c smain.c ===> no errors > gcc -Wall -c gaaout.c ===> a few errors having to do with a return (no > args) > in a function that isn't supposed to return void. After I fixed 'em, > > gcc -Wall -c gaaout.c ===> no errors > gcc -o smain smain.o gaaout.o ===> no errors > > execute `smain' ==> core dump. sigh. How about referencing memory location zero due to a NULL pointer reference? This will always SIGSEGV. BTW, Mumit always suggests using Purify or ElectricFence to check the memory allocation. I've often suggested using the -lmmalloc library. === "Earnie Boyd" CYGWIN RELATED HELP: DOCUMENTATION: DLLHELP: ARCHIVE SEARCH: OR _____________________________________________________________ Do You Yahoo!? Free instant messaging and more at http://messenger.yahoo.com -- Want to unsubscribe from this list? Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com