Bug #2: char* Exception not propogated to caller

[Dima Volodin]

Use -fno-const-strings or change the catch exception-declatation to

char const* c


Cheers!

Dima

On Thu, 06 Jan 2000 23:14:13 +0800, Monique Kuo wrote:

>Hello,
>
>I have found a bug re: C++ exceptions not being properly propogated to
>the caller.
>
>The same problem exists on both linux and Sun Solaris when I use g++.
>However, when I use Sun's CC the problem does not exist.  gcc was built
>using the --enable-shared configuration option.
>
>The source code to recreate the problem is in the following file.
>
>The bug is that when the exception "thrown" is simply of type char* the
>exception is not propogated outside that method and so the "try ...
>catch" structure does not get notification of the exception and the
>program core dumps.
>
>This exception "try ... catch" works (i.e., no core dump).when Sun's CC
>is used.
>
>(The following output was generated on a linus machine
>#uname -a
>Linux timestock.jpc.com.tw 2.2.5-15CLEsmp #1 SMP Mon May 3 04:48:56 CST
>1999 i686 unknown)
>
>All machine and version info is the same as that specified in a bug I
>also submitted today (exception not propogated from a shared library) so
>I will not include the .ii and other info here.  I hope that this is not
>too inconvenient.  I am sure that you will probably want to reproduce it
>yourselves anyway :)
>
>---------------------------------------------------------
>
>tst.cpp  :
>
>#include <iostream.h>
>void tstMethod () {
>  throw "abc";
>}
>
>int main (int argc, char *argv[])
>{
>  try {
>    tstMethod();
>  }
>  catch (char* c) {
>    cout << c << endl;
>  }
>}
>
># g++ tst.cpp -o tst.exe
>
>------------------------------------------------------------------------
>
>The program output is: "Aborted (core dumped)".
>
>gdb tst.exe core
>GNU gdb 4.17.0.11 with Linux support
>Copyright 1998 Free Software Foundation, Inc.
>GDB is free software, covered by the GNU General Public License, and you
>are
>welcome to change it and/or distribute copies of it under certain
>conditions.
>Type "show copying" to see the conditions.
>There is absolutely no warranty for GDB.  Type "show warranty" for
>details.
>This GDB was configured as "i386-redhat-linux"...
>Core was generated by `tst.exe'.
>Program terminated with signal 6, Aborted.
>Reading symbols from /usr/share/chinese/xa+cv/wrap.so...done.
>Reading symbols from /usr/lib/libstdc++-libc6.1-2.so.3...done.
>Reading symbols from /lib/libm.so.6...done.
>Reading symbols from /lib/libc.so.6...done.
>Reading symbols from /lib/libdl.so.2...done.
>Reading symbols from /lib/ld-linux.so.2...done.
>#0  0x400a2111 in __kill ()
>(gdb) bt
>#0  0x400a2111 in __kill ()
>#1  0x400a1d66 in raise (sig=6) at ../sysdeps/posix/raise.c:27
>#2  0x400a3447 in abort () at ../sysdeps/generic/abort.c:88
>#3  0x40050c5b in __default_terminate ()
>#4  0x40050c7c in __terminate ()
>#5  0x40051574 in throw_helper (eh=0x40066144, pc=0x4009bcb2,
>    my_udata=0xbffff748, offset_p=0xbffff744)
>#6  0x4005172c in __throw ()
>#7  0x8048a46 in main ()
>#8  0x4009bcb3 in __libc_start_main (main=0x8048a34 <main>, argc=1,
>    argv=0xbffff824, init=0x8048820 <_init>, fini=0x8048bf4 <_fini>,
>    rtld_fini=0x4000a350 <_dl_fini>, stack_end=0xbffff81c)
>    at ../sysdeps/generic/libc-start.c:78
>------------------------------------------------------------------------------------------------
>
>Thanks in advance for your attention.  Please let me know if there is
>more that I can help with :)
>
>Sincerely,
>
>Leslie
>
>--
>// Leslie Ruskin
>// EMail: Shanti@TheHelm
>// http://members.tripod.com/~sattva/
>// Jade Pacific Corporation
>// 7F #58, Min-Chuan E. Road, Sec 3, Taipei, Taiwan
>// TEL: (886)(2)2515-0253 x428   FAX: (886)(2)2504-5078
>// U.S.A. Fax (703) 940-5515
>// U.S.A. Voice Mail Box 1(661)420-9511
>