From owner-freebsd-current@FreeBSD.ORG Thu Feb 5 07:44:18 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 09BEE16A4CE; Thu, 5 Feb 2004 07:44:18 -0800 (PST) Received: from postal1.es.net (proxy.es.net [198.128.3.205]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6FD6043D39; Thu, 5 Feb 2004 07:44:16 -0800 (PST) (envelope-from oberman@es.net) Received: from ptavv.es.net ([198.128.4.29]) by postal1.es.net (Postal Node 1) with ESMTP (SSL) id IBA74465; Thu, 05 Feb 2004 07:44:15 -0800 Received: from ptavv (localhost [127.0.0.1]) by ptavv.es.net (Tachyon Server) with ESMTP id CF7D65D07; Thu, 5 Feb 2004 07:44:13 -0800 (PST) To: Joe Marcus Clarke In-Reply-To: Message from Joe Marcus Clarke of "Wed, 04 Feb 2004 18:03:15 EST." <1075935794.761.79.camel@gyros> Date: Thu, 05 Feb 2004 07:44:13 -0800 From: "Kevin Oberman" Message-Id: <20040205154413.CF7D65D07@ptavv.es.net> cc: ports@FreeBSD.org cc: current@FreeBSD.org Subject: Re: HEADS UP: MAJOR changes to the ports system X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 05 Feb 2004 15:44:18 -0000 > From: Joe Marcus Clarke > Date: Wed, 04 Feb 2004 18:03:15 -0500 > > > --=-hTZZlYlR23+XSEWaXjJV > Content-Type: text/plain > Content-Transfer-Encoding: quoted-printable > > On Wed, 2004-02-04 at 18:00, Kevin Oberman wrote: > > > From: Joe Marcus Clarke > > > Date: Wed, 04 Feb 2004 02:49:15 -0500 > > > Sender: owner-freebsd-current@freebsd.org > > > > > > Sorry for the wide distribution, but it is critical that -CURRENT users > > > hear this. Two _MAJOR_ changes just went in to the ports tree that are > > > bound to cause some pain to -CURRENT users. > > > > > > The first is the change in default threading libraries. The ports > > > system now uses -lpthread when linking ports instead of -lc_r. Binaries > > > that wind up with both libc_r and libpthread in them will cause > > > problems. In that case, it is imperative that you report this to the > > > port's maintainer ASAP. You can use pkg_info -W /path/to/binary to > > > determine which port installed it. > > > > > > Second, Perl 5.8.2 is now the default version of Perl in -CURRENT since > > > the 5.2 split. This replaces Perl 5.6.1 as the default. Note: Perl > > > 5.00503 is still the default version of Perl in the 4.X base OS. > > > > > > If you experience problems with any of these changes (or with the other > > > recent changes listed in /usr/ports/CHANGES), please report them on the > > > ports@ list and to the appropriate maintainers as soon as possible. > > > Thanks. > > > > OK. After last night's massive adjustment of PORTREVISION, I set u to > > rebuild most everything in ports. I did the standard 'portupgrade -Ra' > > and away it went. Soon enough of the low-level libs were re-built that > > nothing much would run, but I figured that it would be better as soon as > > everything finished building. > > > > Then, after installing the updated librep, ruby dumped core. Now I can't > > run and of the portupgrade tools except portsdb. I have re-built ruby > > and the things built on it for portupgrade, but I still fail with the > > error: > > ruby in malloc(): error: allocation failed > > Abort (core dumped) > > > > Any idea what might be causing this and what I can do to fix it? Getting > > everything re-built in the correct order without portupgrade will be very > > unpleasant with over 180 ports left to re-build! > > Have you rebuild world and kernel today? Also, check the ruby binary > with ldd to see if it's linked to both libc_r and libpthread. If it is, > report that to knu. ruby does not use threading at all, so that is not involved. I found the problem was a corrupt pkgdb.db file. After rebuilding it, everything started working again. I have saved the old file and I hope to figure out what went wrong. In any case, it seem unrelated to the threading issue. Unfortunately, the reinstallation of threaded ports on both of my current systems died in gtk with the same error: /bin/sh ../../mkinstalldirs /usr/X11R6/etc/gtk-2.0 ../../gtk/gtk-query-immodules-2.0 > /usr/X11R6/etc/gtk-2.0/gtkimmodules Fatal error 'Spinlock called when not threaded.' at line 83 in file /usr/src/lib/libpthread/thread/thr_spinlock.c (errno = 0) Abort trap (core dumped) *** Error code 134 I am rebuilding the system now. I'll see if that helps. Thanks! -- R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: oberman@es.net Phone: +1 510 486-8634