From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261625AbVGNWBF (ORCPT ); Thu, 14 Jul 2005 18:01:05 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262378AbVGNWAv (ORCPT ); Thu, 14 Jul 2005 18:00:51 -0400 Received: from oban.u-picardie.fr ([193.49.184.8]:21096 "EHLO mailx.u-picardie.fr") by vger.kernel.org with ESMTP id S261625AbVGNV6h (ORCPT ); Thu, 14 Jul 2005 17:58:37 -0400 Message-ID: <1121378308.42d6e00473669@webmail.u-picardie.fr> Date: Thu, 14 Jul 2005 23:58:28 +0200 From: FyD To: Alistair John Strachan Cc: linux-kernel@vger.kernel.org Subject: Re: Problem with kernel 2.6.11 References: <1121369685.42d6be556505f@webmail.u-picardie.fr> <200507142106.06682.s0348365@sms.ed.ac.uk> In-Reply-To: <200507142106.06682.s0348365@sms.ed.ac.uk> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT User-Agent: Internet Messaging Program (IMP) 3.2.6 X-Originating-IP: 137.131.252.204 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Quoting Alistair John Strachan : > > I have a problem with a program named Gaussian (http://www.gaussian.com) > > (versions g98 or g03) and FC 4.0 (default kernel 2.6.11): I am used to take > > Gaussian binaries compiled on the RedHat 9.0 version, and used them on FC > > 2.0 or FC 3.0. If I try to do so, on FC 4.0. (with the default kernel) > > Gaussian stops (both g98 and g03 versions) with the following error > > message: > > > > [fyd@localhost ~]$ g98 < toto-g98.com> toto-g98.log > > [fyd@localhost ~]$ g03 < toto-g03.com> toto-g03.log > > Erroneous write during file extend. write 208 instead of 4096 > > Probably out of disk space. > > Write error in NtrExt1: No such file or directory > > > > And obviously, I do _not_ have any problem of space and no NFS server on my > > machine... > > > > Now if I compile and use the kernel 2.6.12, this message dispears and the > > program Gaussian g98 works fine, but I still have problems with the version > > g03 which stops without providing any message... > > > > From my tests, it seems to be a problem relative to the kernel. Being not a > > programmer, it is difficult for me to imagine the problem, and even to > > describe it... > > Install "strace", then run the programs like: > strace g98 < toto-g98.com> toto-g98.log > You'll be able to see what syscalls fail and their decoded return values. > This > is extremely useful for determining how a binary application is interacting > with libc and, inevitably, the kernel. If I do strace g98 < toto-g98.com> toto-g98.log The job finished well and the end of the strace output is: stat64("/usr/local/g98/l1.exe", {st_mode=S_IFREG|0755, st_size=516678, ...}) = 0 write(1, " Entering Gaussian System, Link "..., 137) = 137 rt_sigaction(SIGINT, {SIG_IGN}, {SIG_DFL}, 8) = 0 rt_sigaction(SIGQUIT, {SIG_IGN}, {SIG_DFL}, 8) = 0 rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0 clone(child_stack=0, flags=CLONE_PARENT_SETTID|SIGCHLD, parent_tidptr=0xbf9db204) = 3175 waitpid(3175, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 3175 rt_sigaction(SIGINT, {SIG_DFL}, NULL, 8) = 0 rt_sigaction(SIGQUIT, {SIG_DFL}, NULL, 8) = 0 rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0 --- SIGCHLD (Child exited) @ 0 (0) --- stat64("/home/fyd/0QM_SCR/Gau-3174.inp", 0xbf9db114) = -1 ENOENT (No such file or directory) munmap(0x43a45000, 4096) = 0 exit_group(0) = ? If I do strace g03 < toto-g03.com> toto-g03.log The job finished stop with any warning and the end of the strace output is: stat64("/usr/local/g03/l1.exe", {st_mode=S_IFREG|0750, st_size=460870, ...}) = 0 write(1, " Entering Gaussian System, Link "..., 137) = 137 rt_sigaction(SIGINT, {SIG_IGN}, {SIG_DFL}, 8) = 0 rt_sigaction(SIGQUIT, {SIG_IGN}, {SIG_DFL}, 8) = 0 rt_sigprocmask(SIG_BLOCK, [CHLD], [RTMIN], 8) = 0 fork() = 3237 wait4(3237, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGSEGV}], 0, NULL) = 3237 rt_sigaction(SIGINT, {SIG_DFL}, NULL, 8) = 0 rt_sigaction(SIGQUIT, {SIG_DFL}, NULL, 8) = 0 rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0 --- SIGCHLD (Child exited) @ 0 (0) --- stat64("/home/fyd/0QM_SCR/Gau-3236.inp", {st_mode=S_IFREG|0644, st_size=554, ...}) = 0 unlink("/home/fyd/0QM_SCR/Gau-3236.inp") = 0 _exit(1) = ? There is a difference at the end of each strace log file. But what does it mean ? Thanks, Francois PS Please once again, add my email address in CC since I did not register...