From owner-freebsd-gnome@FreeBSD.ORG Thu Mar 25 23:58:48 2004 Return-Path: Delivered-To: freebsd-gnome@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 921F016A4CE for ; Thu, 25 Mar 2004 23:58:48 -0800 (PST) Received: from smtp3.jp.viruscheck.net (smtp3.jp.viruscheck.net [154.33.69.54]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1640643D2D for ; Thu, 25 Mar 2004 23:58:48 -0800 (PST) (envelope-from bland@freebsd.org) Received: from scan2.jp.viruscheck.net ([154.33.69.37] helo=mail3.jp.viruscheck.net) by smtp3.jp.viruscheck.net with esmtp (Exim 3.36 #1) id 1B6mEm-0002hl-00; Fri, 26 Mar 2004 16:58:44 +0900 Received: from [219.167.14.54] (helo=noc.orchid) by mail3.jp.viruscheck.net with esmtp (Exim 3.36 #3) id 1B6mEm-0005Rx-00; Fri, 26 Mar 2004 16:58:44 +0900 Received: from FreeBSD.org (horse.orchid [89.60.10.11]) by noc.orchid (8.12.11/8.12.11) with ESMTP id i2Q7wgpX043609; Fri, 26 Mar 2004 16:58:43 +0900 (JST) (envelope-from bland@FreeBSD.org) Message-ID: <4063E2B2.5030306@FreeBSD.org> Date: Fri, 26 Mar 2004 16:58:42 +0900 From: Alexander Nedotsukov User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7b) Gecko/20040316 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Joe Marcus Clarke References: <4063B4C0.9050001@FreeBSD.org> <1080276619.16291.11.camel@shumai.marcuscom.com> In-Reply-To: <1080276619.16291.11.camel@shumai.marcuscom.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-gnome@FreeBSD.org Subject: Re: Time to put ${PTHREAD_LIBS} in the x11/xscreensaver-gnome? X-BeenThere: freebsd-gnome@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GNOME for FreeBSD -- porting and maintaining List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 26 Mar 2004 07:58:48 -0000 Joe Marcus Clarke wrote: >On Thu, 2004-03-25 at 23:42, Alexander Nedotsukov wrote: > > >>Jeremy Messenger wrote: >> >> >> >>>On Thu, 25 Mar 2004 23:46:28 +0900, Alexander Nedotsukov >>> wrote: >>> >>> >>> >>> >>> >>>>Well to be constructive can people (not only once who currently >>>>expirince a problem with xscreensaver) try the patch attached and report >>>>two things is: it works and what is your GL library Nvidia/Mesa? >>>> >>>> >>>I tried this patch and it doesn't fix the crash. Here using Nvidia >>>driver. >>> >>> >>Jeremy, I looked into your ldd output and it same to mine. There is no >>any single object linked against libc_r, libthr, libkse or libpthread. >>Can this be world + kernel problem? >>About the patch. Its purpose not link GL hacks against libpthread. They >>don't use any pthreads api so why we need this? >> >> > >The problem isn't with the -demo binary, but with the hacks that the >binary loads. On my -CURRENT system, libGL is linked to libpthread, and >so are my GL-based hacks: > > /me confused and feels like missed something important... Wich crash we talking about than? xscreensaver-demo use fork() & exec() for hack preview. GL hack crash should not affect xscreensaver-demo. If I remove libmapping for GL hacks they start coredump but still have no problem with selecting/previewing/using another non-GL hack. I was always thinking about this: hilton>$ xscreensaver-demo xscreensaver-demo: 12:32:45: error closing "/usr/home/stephenh/.xscreensav er": Bad file descriptor xscreensaver-demo: too early for dialog? Fatal error 'Unable to read from thread kernel pipe' at line 1100 in file /usr/src/lib/libc_r/uthread/uthread_kern.c (errno = 0) Abort trap (core dumped) hilton>$ Jeremy, can you try let's say /usr/X11R6/bin/xscreensaver-hacks/glmatrix from cli? >ldd /usr/X11R6/bin/xscreensaver-hacks/glplanet >glplanet: > > ... > libintl.so.6 => /usr/local/lib/libintl.so.6 (0x2857b000) > libstdc++.so.4 => /usr/lib/libstdc++.so.4 (0x28584000) > >I bet if you rebuilt xscreensaver without GL support, it would work >fine. I think adding PTHREAD_LIBS to the build may be the obly way to >fix this for libpthread users. > Ideally if libmap works correct they (libc_r users) should not see any difference. My problem is I still do not understand what we going to fix. I feel like something but not xscreensaver problem. > >Of course, as a test, you might try libmapping libthread to libc_r, and >see if you can reproduce this, Alexander. I can try the same thing. I >know I can. > > This is what I doing all the time to let GL hacks play. I will try to switch back to stock nv driver and MesaGL library later today the only combination I did not test. All the best, Alexander. >Joe > > >