From owner-freebsd-questions Wed Sep 6 10:52:52 2000 Delivered-To: freebsd-questions@freebsd.org Received: from ipamzlx.physik.uni-mainz.de (ipamzlx.Physik.Uni-Mainz.DE [134.93.180.54]) by hub.freebsd.org (Postfix) with ESMTP id A05F437B422 for ; Wed, 6 Sep 2000 10:52:47 -0700 (PDT) Received: from ipamzlx.Physik.Uni-Mainz.DE (ipamzlx.Physik.Uni-Mainz.DE [134.93.180.54]) by ipamzlx.physik.uni-mainz.de (8.11.0/8.9.3) with ESMTP id e86HtGT01009; Wed, 6 Sep 2000 19:55:16 +0200 (CEST) (envelope-from ohartman@ipamzlx.physik.uni-mainz.de) Date: Wed, 6 Sep 2000 19:55:16 +0200 (CEST) From: "O. Hartmann" To: Alfred Perlstein Cc: freebsd-questions@FreeBSD.ORG Subject: Re: Lahey Fortran95 and FreeBSD/Linux Emulation In-Reply-To: <20000906094056.B18862@fw.wintelcom.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Wed, 6 Sep 2000, Alfred Perlstein wrote: Dear Alfred Perlstein. Well, I'm glad to tell you: it works! My mistake was obviously "branding" the shared objects. What I did is: I installed linux_base (I tried linux_devtools also, but it works with both!). Then I installed Lahey/Fujitsu Fortran 95 Compiler and I followed the suggestions on what kind of glibc was installed (I think it is 2.1 with the recent port). I installed the stuff in /usr/local/lf9555. Then I went into /usr/local/lf9555/bin and "brandelf'ed" all the stuff herein "brandelf -t Linux *". That worked. Then I walked into /compat/linux/etc/ and added the line "/usr/local/lf955/lib" in ld.so.conf. Then I put a symlink into /compat/linux/usr/local/ pointing to /usr/local/lf9555 (doing "ln -s /usr/local/lf9555 /compat/linux/usr/local/lf9555"). Well, because I'm a little bit stupid and foolish on how to work with Linux, I do not know how to rehash the shared object cache, so I rebooted the machine. Then I went into the source tree of a standard application, a numerical model called ddscat. It compiled very well and after starting the binary from the right place and with the right option, I produced well known and correct output. Next time I'll see whether this works with all of our numerical stuff at our department, we have some projects on fortran based software and all the oncoming numerical calculations would be done on a new FBSD SMP server - and I hope that will work! Thanks a lot for your really helpfull assistance! Best wishes, O. Hartmann :>* O. Hartmann [000906 05:22] wrote: :>> On Tue, 5 Sep 2000, Alfred Perlstein wrote: :>> Well ... :>> Here I am again, a new issue of the old problem. :>> :>> I'm soory, but acting like a foolish newbie isn't of mine and I :>> do not want myself to be one, but I ran into new problems. After :>> I realized, that binaries has to be "branded" for usage with Linux :>> and after I realized, that compat/linux is the "root tree" for all :>> lib-searches, after installing the linux_devtools port out of the :>> ports-collection, I got rid of many previous problems. :>> :>> Now I made some links of the lib-path of the installation of :>> Lahey Fujitsu Fortran 96 compiler into /compat/linux/usr/local/lib :>> and I added the specific row into ld.so.conf. Fortran compiler :>> now runs perfectly through the code, but after compiling, I try to :>> start the binary - and I get this error message: :> :>> :>> ddscat: error in loading shared libraries: libfj9i6.so.1: ELF file OS ABI invalid. :>> :>> What is that? I branded both the lib and the binary to be Linux :>> type ... and now it says, that I have some OS ABI problems? :> :>Ack, dude, you _must_ fix your mailer to wrap lines at 70 characters. :) :> :>I think you weren't supposed to brandelf the library :(, do you have :>a backup copy of it? :> :>maybe you can run the binary like so: :> :> truss executable :> :>and put the output up on some url? :> :>I'm sending an email to Lahey to ask for an evaluation copy of thier :>software so I can see if I can get this working. :> :>-- :>-Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] :>"I have the heart of a child; I keep it in a jar on my desk." :> Gruss O. Hartmann ------------------------------------------------------------------- ohartman@ipamzlx.physik.uni-mainz.de Klimadatenserver des IPA, Universitaet Mainz Netzwerk- und Systembetreuung To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message