[From nobody Wed Dec 10 01:38:11 2008 Return-Path: <janjaap.poos@wur.nl> X-Original-To: developers@admb-project.org Delivered-To: admb-developers@mercury.nceas.ucsb.edu Received: from net3.wur.nl (net3.wur.NL [137.224.11.3]) by mercury.nceas.ucsb.edu (Postfix) with ESMTP id D12593583FA for <developers@admb-project.org>; Wed, 10 Dec 2008 01:01:11 -0800 (PST) Received: from mppd (localhost.localdomain [127.0.0.1]) by net3.wur.nl (Postfix) with ESMTP id 30678B80E8 for <developers@admb-project.org>; Wed, 10 Dec 2008 10:01:11 +0100 (CET) Received: from scomp0237.wurnet.nl (scomp0237.wurnet.nl [10.91.3.10]) by net3.wur.nl (Postfix) with ESMTP id 3164AB80B4 for <developers@admb-project.org>; Wed, 10 Dec 2008 10:01:08 +0100 (CET) Received: from scomp0040.wurnet.nl ([10.90.3.52]) by scomp0237.wurnet.nl with Microsoft SMTPSVC(6.0.3790.3959); Wed, 10 Dec 2008 10:00:42 +0100 x-mimeole: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Subject: RE: problems with Linux version... Date: Wed, 10 Dec 2008 10:00:42 +0100 Message-ID: <131934CF212D284E8C0C3359B783231FDFEBB0@scomp0040.wurnet.nl> In-Reply-To: <493E9232.2030105@cornell.edu> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: problems with Linux version... Thread-Index: AclapcfgHPGcqkZWToGoBRl5okn3Gg== References: <493E9232.2030105@cornell.edu> From: "Poos, Jan Jaap" <janjaap.Poos@wur.nl> To: <developers@admb-project.org> X-OriginalArrivalTime: 10 Dec 2008 09:00:42.0791 (UTC) FILETIME=[C7F9CB70:01C95AA5] X-CTCH-RefID: str=0001.0A0B0203.493F8556.0094,ss=1,fgs=0 X-Scanned-By: MPP/Commtouch http://www.messagepartners.com X-Scanned-By: This message was scanned by MPP v.3 (www.messagepartners.com) X-CMAE-Analysis: v=1.0 c=1 a=3PszB9wi_l0A:10 a=b01i9L7eAAAA:8 a=QxOOI3OMhqLE7v0Ox_sA:9 a=qix9B3K_F1lLqETQKYcA:7 a=9E0lvCjxkvnX-_sKQi-SkrK-wDMA:4 a=gJcimI5xSWUA:10 X-Scanned-By: MPP/Cloudmark http://www.messagepartners.com X-Spam-Score-W: -2.00 X-Spam-Rules: AWL X-Scanned-By: MPP/Spamassassin http://www.messagepartners.com X-Scanned-By: MPP/Clamd http://www.messagepartners.com X-Scanned-By: MPP/UVScan http://www.messagepartners.com X-NCEAS-MailScanner-Information: Please contact the ISP for more information X-NCEAS-Mercury-MailScanner: Not scanned: please contact your Internet E-Mail Service Provider for details X-NCEAS-MailScanner-MCPCheck: X-NCEAS-Mercury-MailScanner-SpamCheck: not spam, SpamAssassin (not cached, score=-0.002, required 6, autolearn=not spam, SPF_HELO_PASS -0.00, SPF_PASS -0.00) X-MailScanner-From: janjaap.poos@wur.nl =20 -----Original Message----- From: Evan Cooch [mailto:evan.cooch@cornell.edu]=20 Sent: Tuesday, December 09, 2008 4:44 PM To: developers@admb-project.org Subject: problems with Linux version... Greetings - Trying to install ADMB on a multi-Opteron box, that is binary compatible for both 32- and 64-bit apps. Has gcc 4.1.2 installed. Since it doesn't look as though there is a gcc 4.xx version for 64-bits (why?), I downloaded the archive for the 32-bit version. Followed the directions for install, cd'd to ADMB_HOME, ran make, and got a whole slew of errors, starting with In file included from /usr/lib/gcc/x86_64-redhat-linux/4.1.2/../../../../include/c++/4.1.2/bac kward/strstream:51, from //home/egc/Desktop/admb/include/fvar.hpp:446, from ham4.cpp:1: /usr/lib/gcc/x86_64-redhat-linux/4.1.2/../../../../include/c++/4.1.2/bac kward/backward_warning.h:32:2:=20 warning: #warning This file includes at least one deprecated or antiquated header. Please consider using one of the 32 headers found in section 17.4.1.2 of the C++ standard. Examples include substituting the <X> header for the <X.h> header for C++ includes, or <iostream> instead of the deprecated header <iostream.h>. To disable this warning use -Wno-deprecated.... and culminating with: /usr/bin/ld: cannot find -ldf1b2stub collect2: ld returned 1 exit status /bin/sh: ./ham4: No such file or directory make: *** [ham4] Error 127 My guess is that that when ADMB tries to compile using gcc, it isn't smart' enough to tell gcc to compile for 32-bits (the default on this machine is to compile for 64 unless you throw the -m32 flag at it during compilation). Moreover, 32-bit and 64-bit libs are generally both installed on 64-bit machines, and ADMB would need to know where to find what it needs. Suggestions? Since ADMB is now being given away freely, why not consider making the source available, and letting users do a config/make/make install sequence? Would make your lives simpler than trying to make binaries which are unlikely to run on all permutations of Linux. Hi Evan, We were in a similar situation, and Arni Magnusson gave the tip to use the 9rc2 release=20 (http://admb-project.googlecode.com/files/admb-9.0.9rc2-linux64-gcc3.4.z ip) for x86_64 and gcc 3.4. This has indeed worked very well for us (there are some minor issues). If you want to also use the random effects stuff, Arni Magnusson also pointed out that:=20 "I recommend creating a separate script (e.g. mygccr) for random-effects models, where -ldf1b2o=20 is used instead of -ldf1b2stub." This means removing -ldf1b2stub _twice_ from the myggco script, and uncommenting -ldf1b2o . Of course I don't know what issues there are with rc2 compared to the newer versions, but we were very happy to see it work. Now just wait for an official build for x86_64 and gcc 4.x. Cheers, Jan Jaap=20 ]