Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sdd.hp.com!spool.mu.edu!cs.umn.edu!wytten
From: wytten@cs.umn.edu (Dale Wyttenbach)
Newsgroups: comp.sys.ibm.pc.programmer
Subject: strange borland 2.0 behavior
Message-ID: <1991Apr22.144727.7170@cs.umn.edu>
Date: 22 Apr 91 14:47:27 GMT
Article-I.D.: cs.1991Apr22.144727.7170
Organization: University of Minnesota, Minneapolis, CSci dept.
Lines: 23

The Borland-C2.0 compiler on our Northgate-486 has developed the annoying
habit of writing bad executable files.  I have a couple programs that
execute correctly when run from within the Borland enviroment, but when
I write an executable and try to run it, it's like a random entry point
is chosen.  One program (a large one) appears to go into graphics mode
and then exits with a divide error.  The reason I think it is not finding
the right entry point is because I put a printf as the first line in
main() and it is apprently not executed.  Another program, which is a
small test program that initializes graphics and outtexts("hello, world")
also runs within the enviroment, but from DOS it exits with no effect.
Both programs are compiled using the medium memory model.  

I turned on duplicate symbol checking and got the message that _setjmp
and _longjmp are defined twice in module SETJMP.  Could this have
anything to do with it?

Thanks in advance,

dale
-- 
 Dale Wyttenbach     | We all shine on,
 wytten@cs.umn.edu   | like the moon, the stars and the sun.
                     |                                      --John Lennon