From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 6093 invoked by alias); 11 Nov 2002 03:37:56 -0000 Mailing-List: contact gdb-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sources.redhat.com Received: (qmail 6084 invoked from network); 11 Nov 2002 03:37:54 -0000 Received: from unknown (HELO tbdnetworks.com) (63.209.25.99) by sources.redhat.com with SMTP; 11 Nov 2002 03:37:54 -0000 Received: from defiant (defiant.tbdnetworks.com [192.168.4.126]) by tbdnetworks.com (8.12.2/8.12.2) with ESMTP id gAB3brii018687; Sun, 10 Nov 2002 19:37:53 -0800 (PST) Received: from nkiesel by defiant with local (Exim 3.36 #1 (Debian)) id 18B5Oc-000715-00; Sun, 10 Nov 2002 19:37:54 -0800 Date: Sun, 10 Nov 2002 19:37:00 -0000 From: nkiesel@tbdnetworks.com To: gdb@sources.redhat.com Cc: nkiesel@tbdnetworks.com Subject: gdb-5.2.1/g++-3.1 on Solaris5.7 severely broken?! Message-ID: <20021111033754.GA26897@defiant> Reply-To: nkiesel@tbdnetworks.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4i X-SW-Source: 2002-11/txt/msg00106.txt.bz2 Hi, GDB was my truly friend ever since I submitted the patch for "set print null-stop" about 10 years ago, but now I have severe problems with gdb-5.2.1 while trying to debug C++ programs compiled with g++-3.1 under Solaris 5.7. I compile my programs with "-gstabs" because with "-ggdb" gdb will core dump as soon as I try to walk the call stack (e.g. running "bt"). Even with "-gstabs", I always get the following message on startup: dbxread.c:1717: gdb-internal-error: sect_index_data not initialized The situation gets worse with multi-threaded programs (using pthreads): gdb ignores breakpoints and the call stack seems to be wrong (see below for an slightly edited typescript of a gdb session). I also tried the latest weekly snapshot from the branch, but this just resulted in more gdb internal error messages (sorry I lost these messages, but can reproduce them if needed). One additional information: my program is dynamically linked against some 3rd party C++ libs (e.g. Xerces and STLport) which are compiled with "-g" instead of "-gstabs". Would recompiling them with "-gstabs" make any difference? Would upgrading to g++-3.2 make any difference (not that this is really an option right now about one week before deadline)? I would be very thankful to get any hints (debugging using "cout" really sucks :-). so long Norbert P.S.: Please CC to nkiesel@tbdnetworks.com as I'm not subscribed to this mailing list. --nk Script started on Sun Nov 10 19:15:22 2002 $ gdb ./CAGTarget GNU gdb 5.2.1 Copyright 2002 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 "sparc-sun-solaris2.7"... (gdb) b main Breakpoint 1 at 0x54ad18: file CAGMain.cpp, line 182. (gdb) r Starting program: /tmp/TBD_79.26487/TBD/agent/CAGTarget dbxread.c:1717: gdb-internal-error: sect_index_data not initialized An internal GDB error was detected. This may make further debugging unreliable. Quit this debugging session? (y or n) n Create a core file containing the current state of GDB? (y or n) n [New LWP 1] [New LWP 2] [New LWP 3] *** Current System Time: 2002-11-11T03:16:01Z *** CAG local IP Address is: 192.168.4.99 Setti[New LWP 4] ng global MONPERIOD to 0 seconds Setting global COLLECTMONDATA to true Setting global AUDITINGPERIODMIN to 0 seconds Setting global AUDITINGPERIODMAX to 0 seconds Listening on port 6600 Sending notifications to port 7700 N[New LWP 5] otify::startup: entering wait state.. [New LWP 6] ^C Program received signal SIGINT, Interrupt. 0xfea98044 in _poll () from /usr/lib/libc.so.1 (gdb) bt #0 0xfea98044 in _poll () from /usr/lib/libc.so.1 #1 0xfea4f8d0 in select () from /usr/lib/libc.so.1 #2 0x00340d4c in vtable for SchemaGrammar () #3 0x000b1dac in AttrMapImpl::cloneAttrMap(NodeImpl*) () at ../../../../libstdc++-v3/libsupc++/tinfo2.cc:61 (gdb) q The program is running. Exit anyway? (y or n) y script done on Sun Nov 10 19:16:27 2002