From owner-freebsd-ports@FreeBSD.ORG Sun Jul 11 13:46:55 2010 Return-Path: Delivered-To: freebsd-ports@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E1E471065674; Sun, 11 Jul 2010 13:46:54 +0000 (UTC) (envelope-from bf1783@googlemail.com) Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by mx1.freebsd.org (Postfix) with ESMTP id 51D458FC15; Sun, 11 Jul 2010 13:46:53 +0000 (UTC) Received: by wyb34 with SMTP id 34so3292413wyb.13 for ; Sun, 11 Jul 2010 06:46:53 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:received:received:reply-to:date :message-id:subject:from:to:cc:content-type; bh=SSgHHtY4/96JYxcfTqUFkPUbAH3F9pXoaqYTKhOoPwI=; b=ZjabeSL1w2RG4wADQ2QLQk1mceEyy6Qeqfe8URnt4qUKhUlfjiJaBu2EK5OGawaolb MVQnsYRlwXNMBYv4WW1mVdp24e/ba5dQ53dX6B231moGVhO7bqahsGT2pmwMTuehQ1KY 1pmdBnsl+hTXERV3pgUMTMfDRD0eR1drCoXW0= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:reply-to:date:message-id:subject:from:to:cc :content-type; b=aba8QdUkDMilNSoJ3y4ERp457O7uejAETHkya1cscExy3sBGp5l4C5veT63CIEfknk 4+TgbRoyhap9zUWqXAXEUZEk7OvzEG6/YZMsvvj4/qAyQR0wlkW07a3RhOKL6bBfW4h9 Z+FOtn+XunbgNMqjvcMkFjI1FbK2IGkm8yk4I= MIME-Version: 1.0 Received: by 10.227.146.142 with SMTP id h14mr8606269wbv.25.1278856013192; Sun, 11 Jul 2010 06:46:53 -0700 (PDT) Received: by 10.216.171.10 with HTTP; Sun, 11 Jul 2010 06:46:53 -0700 (PDT) Date: Sun, 11 Jul 2010 13:46:53 +0000 Message-ID: From: "b. f." To: freebsd-ports@FreeBSD.org Content-Type: text/plain; charset=ISO-8859-1 Cc: Anonymous , Doug Barton Subject: Re: graphics/png does not compile with gcc 4.5.1 X-BeenThere: freebsd-ports@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: bf1783@gmail.com List-Id: Porting software to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 11 Jul 2010 13:46:55 -0000 > Anonymous writes: > > > Doug Barton writes: > > > >> readelf -s /lib/libc.so.7 | fgrep __stack_chk_fail > >> 952: 00028060 26 FUNC GLOBAL DEFAULT 10 __stack_chk_fail@@FBSD_1.0 > >> 1457: 00028060 26 FUNC GLOBAL DEFAULT 10 __stack_chk_fail_local at FBSD_1.0 > > > >> 45: 000ecec0 29 FUNC LOCAL HIDDEN 10 __stack_chk_fail_local > > > > I don't see such HIDDEN symbol on i386/amd64 snapshots from Jun 30 at > > allbsd.org. Can you *install* libc.so.7 from there and try again? > > Did it help? Based on result from above suggestion either my guesses > were incorrect[1] or you have messed up environment/make.conf[2]. > To debug the latter you'd have to bisect your own build configuration. > > [1] the problem is reproducable on libc from snapshot > [2] ... *not* reproducable ... > > > > >> 1542: 00028060 26 FUNC GLOBAL DEFAULT 10 __stack_chk_fail > >> 2047: 00028060 26 FUNC GLOBAL DEFAULT 10 __stack_chk_fail_local at FB I have not attempted to reproduce this particular problem, but some of these symbols in the base system libraries appear similar to others I have seen when parts of the base system are built while remapping the base system compiler libraries to their counterparts in lang/gcc4* ports via libmap.conf(5), as recommended in, for example, http://www.freebsd.org/doc/en_US.ISO8859-1/articles/custom-gcc/configuring-ports-gcc.html Despite claims of 100% backwards-compatibility for these libraries, this remapping will have some consequences, so you may want to check if you have disabled the remapping while building/installing the base system, which I would recommend. And to be on the safe side, I would do this by editing or temporarily moving libmap.conf, rather than by setting LD_LIBMAP_DISABLE in the build/installation environment, to avoid potential problems with sgid/suid binaries. Also, when updating or reinstalling ports, I would first deinstall the old versions (including old versions of the shared libraries that may be in *compat*) before rebuilding the new versions, to avoid any linking problems. (We are still eagerly awaiting this or an equivalent sandbox "safe rebuild" option in ports-mgmt/portmaster. ;) ) Regards, b.