[MITgcm-support] Re: MITgcm-support Digest, Vol 25, Issue 3

[eliyahu biton]

mitgcm-support-request at mitgcm.org wrote:

>Send MITgcm-support mailing list submissions to
>	mitgcm-support at mitgcm.org
>
>To subscribe or unsubscribe via the World Wide Web, visit
>	http://mitgcm.org/mailman/listinfo/mitgcm-support
>or, via email, send a message with subject or body 'help' to
>	mitgcm-support-request at mitgcm.org
>
>You can reach the person managing the list at
>	mitgcm-support-owner at mitgcm.org
>
>When replying, please edit your Subject line so it is more specific
>than "Re: Contents of MITgcm-support digest..."
>
>
>Today's Topics:
>
>   1. Re: mpi (Ed Hill)
>
>
>----------------------------------------------------------------------
>
>Message: 1
>Date: Mon, 04 Jul 2005 00:33:55 -0400
>From: Ed Hill <ed at eh3.com>
>Subject: Re: [MITgcm-support] mpi
>To: MITgcm-support <mitgcm-support at mitgcm.org>
>Message-ID: <1120451635.9523.303.camel at ernie>
>Content-Type: text/plain
>
>On Sun, 2005-07-03 at 16:33 +0300, eliyahu biton wrote:
>  
>
>>Dear supporting list.
>>I have downloaded the daily snapshot  'MITgcm_ss_20050629' .
>>I managed to compile my configuration on one processor without any 
>>difficulties, but trying  to compile with mpi fail due to the following 
>>error massage:
>>
>>-L/opt/mpi/lib/alpha/ -lfmpi -lmpi -lkmp_osfp10 -pthread
>>ld:
>>Unresolved:
>>timenow_
>>f77: Severe: Failed while trying to link.
>>gmake: *** [mitgcmuv] Error 1
>>
>>running the  global with_exf verification with mpi  leads to the same 
>>error massage at the compilation step.
>>more information that could help is that I work on  'alpha machine' 
>>running with 'Tru64' operating system.
>>any ideas ?
>>    
>>
>
>
>Hi Eli,
>
>Compiling with MPI is very similar to compiling without it -- you just
>need to set two defines (the easy way is with the genmake2 "-mpi"
>option) and make sure that the right paths are specified for the MPI
>headers and libraries.  Thats it.
>
>The error you're seeing above is a Fortran/C linking problem and
>probably has nothing to do with MPI.  Please add the following to your
>"optfile" and try again:
>
>  FC_NAMEMANGLE="#define FC_NAMEMANGLE(X) X ## _"
>
>Ed
>
>  
>
Thanks Ed it works.
Eli

-- 
Eli Biton
Dept. of Environmental Sciences,
Weizmann Institute of Sciences,
Rehovot, 76100
Israel
Tel: +972-8-9344924
Fax: +972-8-9344124