From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 28239 invoked by alias); 30 Aug 2008 00:44:19 -0000 Received: (qmail 28231 invoked by uid 22791); 30 Aug 2008 00:44:19 -0000 X-Spam-Check-By: sourceware.org Received: from smtp-out.google.com (HELO smtp-out.google.com) (216.239.33.17) by sourceware.org (qpsmtpd/0.31) with ESMTP; Sat, 30 Aug 2008 00:43:46 +0000 Received: from spaceape23.eur.corp.google.com (spaceape23.eur.corp.google.com [172.28.16.75]) by smtp-out.google.com with ESMTP id m7U0heFk027880 for ; Sat, 30 Aug 2008 01:43:40 +0100 Received: from wa-out-1112.google.com (wahj4.prod.google.com [10.114.236.4]) by spaceape23.eur.corp.google.com with ESMTP id m7U0hcPH008369 for ; Fri, 29 Aug 2008 17:43:39 -0700 Received: by wa-out-1112.google.com with SMTP id j4so976655wah.19 for ; Fri, 29 Aug 2008 17:43:38 -0700 (PDT) Received: by 10.115.111.1 with SMTP id o1mr3267628wam.114.1220057018206; Fri, 29 Aug 2008 17:43:38 -0700 (PDT) Received: by 10.114.78.12 with HTTP; Fri, 29 Aug 2008 17:43:38 -0700 (PDT) Message-ID: <8ac60eac0808291743i67ed9872wca32b21389e48dd3@mail.gmail.com> Date: Sun, 31 Aug 2008 05:05:00 -0000 From: "Paul Pluzhnikov" To: "Laurent Carlier" Subject: Re: Cannot find new threads: generic error Cc: gdb@sourceware.org In-Reply-To: <200808300213.57239.lordheavym@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <200808281553.37056.lordheavym@gmail.com> <200808300213.57239.lordheavym@gmail.com> 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: 2008-08/txt/msg00294.txt.bz2 On Fri, Aug 29, 2008 at 5:13 PM, Laurent Carlier wrote: > I've find a workaround, i've link the debugging app with libpthread and now gdb > feels happy, and i can run it with gdb ! Are you dlopen()ing a library which is linked against libpthread.so into a program which isn't? Very few OSes support processes "suddenly" becoming multithreaded (I know only of Solaris 10 and AIX 5). If you do that on Linux, you should expect all kinds of trouble, because e.g. glibc very early on decides whether pthreads are present, and permanently switches to thread-unsafe behavior if they are not. Cheers, -- Paul Pluzhnikov