Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT sources DOT redhat DOT com Delivered-To: mailing list cygwin AT sources DOT redhat DOT com From: "Markus Hoenicka" MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15007.50216.75000.569222@gargle.gargle.HOWL> Date: Fri, 2 Mar 2001 16:02:48 +0000 (CT0) To: cygwin AT cygwin DOT com Subject: strace question X-Mailer: VM 6.75 under Emacs 20.6.1 Hi, I hope this is not a completely dumb question, but here we go. I'm trying to debug a dynamically linked C++ app that runs just fine on other platforms including native Win32. As previously reported, the statically linked version runs fine on Cygwin, whereas the dynamically linked version creates all expected output, but then dies with a SIGSEV and causes an error popup ("The instruction at 0xsomething referenced memory at 0xwhatever. The memory could not be read") instead of just exiting. I tried to get a hint from strace as to where the program dies. I ran both the statically linked and the dynamically linked version and get exactly the same strace output. There is no such error popup in this case for the dynamically linked version and no hint in the strace output that the prog receives a SIGSEV signal. Does strace magically protect applications from crashing? Is this expected behaviour? If not, does that tell me anything about what is wrong with the dynamic linking? BTW debugging this directly with gdb fails as any attempt to gain information at the crash point (bt and the like) just show "Cannot access memory at address 0x2000000". Any help is appreciated. Markus -- Markus Hoenicka, PhD UT Houston Medical School Dept. of Integrative Biology and Pharmacology 6431 Fannin MSB4.114 Houston, TX 77030 (713) 500-6313, -7477 (713) 500-7444 (fax) Markus DOT Hoenicka AT uth DOT tmc DOT edu http://ourworld.compuserve.com/homepages/hoenicka_markus/ -- Want to unsubscribe from this list? Check out: http://cygwin.com/ml/#unsubscribe-simple