>Number:         25094
>Category:       bin
>Synopsis:       gcc generates wrong debugging symbols per default
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    bin-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Apr 07 16:15:00 UTC 2004
>Closed-Date:
>Last-Modified:
>Originator:     Jan Schaumann
>Release:        NetBSD 1.6ZK
>Organization:
	
>Environment:
	
	
System: NetBSD dogfish-head.cs.stevens-tech.edu 1.6ZK NetBSD 1.6ZK (BOCK.current) #0: Tue Mar 9 11:41:58 EST 2004 jschauma@doppelbock.hpcf.cs.stevens-tech.edu:/share/src/sys/arch/i386/compile/obj/BOCK.current i386
Architecture: i386
Machine: i386
>Description:

Under i386, it appears that gcc if passed '-g' generates the wrong
debugging symbols per default:

#include <stdio.h>

struct astruct {
  unsigned char c;
  double d;
};

int main() {

  struct astruct x;

  x.c = 'a';
  x.d = 0.184;
  printf("%f\n", x.d);
  return 0;
}

When compiled with '-g', running it through gdb it will correctly print
0.184000, but if you break and inspect x, it will give:

(gdb) p x
$1 = {c = 97 'a', d = 5.2867121265460283e-315}
(gdb)

Now using '-gstabs' or '-gstabs+' or '-gstabs3', it will behave
correctly.

This is not obvious to the end user and somewhat embarrasing if the system
ships with a compiler that produces output that the systems debugger can't
use.  I'd say that shipping 2.0 with this would be a Bad Thing.

Also see http://mail-index.netbsd.org/current-users/2004/04/05/0018.html,
Nathan Williams' response to my mail.  I'm not sure if this is port-specific.

>How-To-Repeat:
See above.
>Fix:
Dunno.  Tell gcc to use stabs per default?
>Release-Note:
>Audit-Trail:
>Unformatted: