Re: exception handling bug in b18 g++

["Wei Ku" <weiku at UTKUX1 dot UTK dot EDU>]

 Can g++ treat 'exception handling' correctly ? I thought it can not even in
UNIX machine.

Sincerely,
Wei Ku

***************************************
Department of Physics and Astronomy
The University of Tennessee
1408 Circle Drive
Knoxville, Tennessee 37996-1200
weiku@utkux.utcc.utk.edu
---------------------------------------
Solid State Division
Oak Ridge National Laboratory
P.O.Box 2008
Oak Ridge, TN 37831-6032
Phone: (423) 574-5795
Fax: (423) 574-4143
weiku@solid.ssd.ornl.gov
***************************************

-----Original Message-----
From: Benjamin Sean Scarlet <scarlet@mit.edu>
To: gnu-win32 Mailing List <gnu-win32@cygnus.com>
Date: Tuesday, July 29, 1997 8:34 AM
Subject: exception handling bug in b18 g++



>I seem to have found some odd bug in exception handling: after
>a catch block in a member function, the this pointer is corrupted:
>
>--------------------------test.C------------------------------------
>#include <iostream.h>
>
>class X
>{
>  int x;
>public:
>  void foo();
>};
>
>void printit(void *p)
>{
>  cerr << p << endl;
>}
>
>void X::foo()
>{
>  printit(this);
>  try
>

>      printit(this);
>      throw 1;
>    }
>  catch(int)
>    {
>      printit(this);
>    }
>  printit(this);
>}
>
>main()
>{
>  X x;
>  x.foo();
>}
>----------------------------------------------------------------------
>When I try this, I get:
>
>bash$ g++ -fexceptions test.C -o test
>bash$ ./test
>0x241f464
>0x241f464
>0x241f464
>0x40c0f8
>
>As a side note, a possible quick fix is to turn on optimization:
>bash$ g++ -O -fexceptions test.C -o test
>bash$ ./test
>0x241f464
>0x241f464
>0x241f464
>0x241f464
>
> Benjamin Scarlet
> scarlet@mit.edu
>
>-----------------------------------------------------------------------
>Configuration particulars:
>Pentium II, Windows NT 4.0 (service pack 3)
>
>An even further aside:  I don't know how extensive the Cygnus changes to
>the g++ exception handling code are, so I figured it wouldn't hurt to
>try this
>program on a Linux box with vanilla gcc-2.7.2.1:  There I get no
>problem.
>-
>For help on using this list (especially unsubscribing), send a message to
>"gnu-win32-request@cygnus.com" with one line of text: "help".
>

-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".