From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 28651 invoked by alias); 8 Mar 2007 15:47:58 -0000 Received: (qmail 28626 invoked by uid 22791); 8 Mar 2007 15:47:55 -0000 X-Spam-Check-By: sourceware.org Received: from relay1.psi.neteu.net (HELO relay1.psi.neteu.net) (154.15.246.50) by sourceware.org (qpsmtpd/0.31) with ESMTP; Thu, 08 Mar 2007 15:47:52 +0000 Received: from [212.222.53.82] (helo=gg.pruftechnik.com) by relay1.psi.neteu.net with esmtp (Exim 4.65) (envelope-from ) id 1HPKqD-00028n-Ls; Thu, 08 Mar 2007 16:47:41 +0100 Received: from gg.pruftechnik.com (localhost [127.0.0.1]) by gg.pruftechnik.com (8.13.8/8.13.8) with ESMTP id l28FleNr023371; Thu, 8 Mar 2007 16:47:40 +0100 (CET) Received: (from localhost) by gg.pruftechnik.com (MSCAN) id 1/gg.pruftechnik.com/smtp-gw/mscan; Thu Mar 8 16:47:40 2007 In-Reply-To: <20070308151547.GA4632@caradoc.them.org> To: Daniel Jacobowitz Cc: gdb@sourceware.org Subject: Re: gdb 6.6: incomplete backtraces ? MIME-Version: 1.0 X-Mailer: Lotus Notes Release 6.5.4 March 27, 2005 Message-ID: From: Markus.Grunwald@pruftechnik.com Date: Thu, 08 Mar 2007 15:47:00 -0000 Content-Type: text/plain; charset="US-ASCII" X-IsSubscribed: yes Mailing-List: contact gdb-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sourceware.org X-SW-Source: 2007-03/txt/msg00132.txt.bz2 Hello, > -------------------------------------------------------------------------------- > > gru@CMDevLin2 >/home/gru/gdb-6.6/bin/gdb dafit_x86.bin > > GNU gdb 6.6 > > (gdb) catch fork > > Catchpoint 1 (fork) > > (gdb) run > > [...] > > Catchpoint 1 (forked process 26023), 0xb7fa1410 in ?? () > > (gdb) bt > > #0 0xb7fa1410 in ?? () > > #1 0x00000001 in ?? () > > #2 0x00000000 in ?? () > > (gdb) show backtrace limit > > An upper bound on the number of backtrace levels is unlimited. > > > -------------------------------------------------------------------------------- > > > > Hey, that's quite short ! No matter which catchpoint is hit, I get only 3 > > frames ... > > What does info shared say is at that address? Anything? Catchpoint 1 (forked process 27093), 0xb7f10410 in ?? () (gdb) bt #0 0xb7f10410 in ?? () #1 0x00000001 in ?? () #2 0x00000000 in ?? () (gdb) info shared >From To Syms Read Shared Object Library 0xb7f11790 0xb7f222cf Yes /lib/ld-linux.so.2 0xb7eeb0e0 0xb7ef7a74 Yes /lib/libbz2.so.1.0 0xb7700f40 0xb7bf0880 Yes /opt/qt/x86/qt3/lib/libqt-mt.so.3 [...] Hmm, seems we didn't hit anything... This is for gdb 6.3: Catchpoint 1 (forked process 27776), 0xb7f77410 in ?? () (gdb) bt #0 0xb7f77410 in ?? () #1 0x00000001 in ?? () #2 0x00000000 in ?? () #3 0xb728761e in strtold_l () from /lib/tls/i686/cmov/libc.so.6 [...] (gdb) info shared >From To Syms Read Shared Object Library 0xb7f520e0 0xb7f5ea74 Yes /lib/libbz2.so.1.0 0xb7767f40 0xb7c57880 Yes /opt/qt/x86/qt3/lib/libqt-mt.so.3 0xb74f1660 0xb74fb174 Yes /usr/lib/libXext.so.6 0xb7417c40 0xb74a54f4 Yes /usr/lib/libX11.so.6 [...] > It usually helps to install the Debian libc6-dbg package and configure > your GDB with --prefix=/usr. That will improve backtraces through the > system libraries. Sounds interresing... But I am currently testing a lot of gdb stuff here and don't want to interfere with my colleagues who work on the same machine, so I don't want to use --prefix=/usr ... cu Markus Grunwald