From owner-nwchem-users@emsl.pnl.gov Tue May 20 11:49:18 2008 Received: from odyssey.emsl.pnl.gov (localhost.pnl.gov [127.0.0.1]) by odyssey.emsl.pnl.gov (8.14.1/8.14.1) with ESMTP id m4KInHdL002865 for ; Tue, 20 May 2008 11:49:18 -0700 (PDT) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.14.1/8.14.1/Submit) id m4KInHAs002864 for nwchem-users-outgoing-0915; Tue, 20 May 2008 11:49:17 -0700 (PDT) X-Authentication-Warning: odyssey.emsl.pnl.gov: majordom set sender to owner-nwchem-users@emsl.pnl.gov using -f X-Ironport-SG: OK_Domains X-Ironport-SBRS: 5.7 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AqcAAMC5MkiA/H/InWdsb2JhbACSMQEBAQEBCA0HnEQB X-IronPort-AV: E=Sophos;i="4.27,516,1204531200"; d="scan'208";a="76410419" Date: Tue, 20 May 2008 13:49:20 -0500 (CDT) From: Lev Gelb X-X-Sender: gelb@localhost.localdomain To: nwchem-users@emsl.pnl.gov Subject: [NWCHEM] Problem with plane-wave calculation in 5.1, (openMPI compile) Message-ID: User-Agent: Alpine 1.10 (LFD 962 2008-03-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk (Apologies if this is a repost - some odd mailer behavior here.) Dear All, I'm getting a strange error near completion of a simple plane-wave calculation (below; basically a single lithium atom in a largish simulation cell). I can reproduce this with separate builds of NWchem 5.1 on two different systems - 1. Intel quad-core, FC9 x86_64, standard BLAS, openMPI 1.2.4 2. Opteron quad-core, Centos 5.0 x86_64, AMD BLAS, openMPI 1.2.5rc1 The error looks like: output psi filename:/tmp//pysim_nwchem.movecs ------------------------------------------------------------------------ D1dB_end:freeing heap memory 0 ------------------------------------------------------------------------ ------------------------------------------------------------------------ current input line : 35: task pspw energy 0:0:D1dB_end:freeing heap memory:: 0 0:0:D1dB_end:freeing heap memory:: 0 Last System Error Message from Task 0:: Inappropriate ioctl for device [druid:29529] MPI_ABORT invoked on rank 0 in communicator MPI_COMM_WORLD with errorcode 0 which occurs after completion of the SCF iteration, but before completion of the program. It occurs whether I use one or two processors. Also, it doesn't matter what I set the "total memory" to. This error does *not* occur with my 5.0 build. Likewise, 'bigger' input files, with 30+ lithium atoms in the cell, finish correctly. Here's the input file: #-------------------------------------------- title "test.inp" scratch_dir /tmp/ permanent_dir /tmp/ memory 400 mb start nwchem_test geometry units angstrom noautosym noautoz print system crystal lat_a 9.0 lat_b 9.0 lat_c 9.0 alpha 90.0 beta 90.0 gamma 90.0 end Li 5.0 5.0 5.0 end nwpw pseudopotentials Li library library1 end simulation_cell ngrid 16 16 16 end unrestricted xc vosko ewald_ncut 3 end set nwpw:psi_nolattice .true. set nwpw:lcao_skip .true. set nwpw:mimimizer 2 task pspw energy #------------------------------------------------ (I know this isn't the most compact way to write this input.) Anyway, any suggestions you might have would be welcome. Best regards, Lev -------------------------------------------------------------------- Lev Gelb email: gelb@wustl.edu Associate Professor phone: (314)935-5026 Department of Chemistry, fax: (314)935-4481 Washington University in St. Louis, St. Louis, MO 63130 USA www.chemistry.wustl.edu/~gelb --------------------------------------------------------------------