[MITgcm-support] Correct Options File NetCDF Configuration
Phil Underwood
p.underwood at uea.ac.uk
Wed Jun 23 13:07:02 EDT 2010
Hi Jean-Michel,
Thank you for getting back to me. I did as you suggested, although it isn't immediately clear to me if I've made the correct amendments to the options file. I've attached a copy of the 'linux_ia32_g77' options file I modified based on the one in the tools/build_options directory. My selection of that file as the foundation was based on the platform, instruction set and compiler of the target system I am working with. I have changed the include and lib paths to match those of the NetCDF installation on the target machine, yet I am still receiving the error.
You'll noticed I have not changed the path name in the if comparison lines as I do not know what the comparison is for. Is this where the problem lies?
Cheers,
Phil
P.S. Incase your servers reject attachments I also include a text version of the file:
#
FC=g77
CC=gcc
DEFINES='-D_BYTESWAPIO -DWORDLENGTH=4'
# CPP='gcc -E -traditional -P -'
CPP='cpp -traditional -P'
NOOPTFLAGS='-O0'
EXTENDED_SRC_FLAG='-ffixed-line-length-132'
if test "x$IEEE" = x ; then
# No need for IEEE-754
FFLAGS='-Wimplicit -Wunused -Wuninitialized'
FOPTIM='-O3 -malign-double -funroll-loops'
else
# Try to follow IEEE-754
has_sse2=f
grep flags /proc/cpuinfo | grep sse2 > /dev/null 2>&1 && has_sse2=t
if test "x$has_sse2" = xt ; then
FFLAGS='-Wimplicit -Wunused -mfpmath=sse -msse -msse2'
else
FFLAGS='-Wimplicit -Wunused -ffloat-store'
fi
# echo 'FFLAGS="'$FFLAGS'"'
FOPTIM='-O0 -malign-double'
fi
if test -d /usr/include/netcdf-3 ; then
INCLUDES='-I/cvos/apps/netcdf-4.0.ifc/include'
if test -d /usr/lib/netcdf-3 ; then
if test -f /usr/lib/netcdf-3/libnetcdf_g77.a ; then
LIBS='-L/cvos/apps/netcdf-4.0.ifc/lib -lnetcdf_g77'
else
LIBS='-L/cvos/apps/netcdf-4.0.ifc/lib'
fi
fi
elif test -d /usr/include/netcdf ; then
INCLUDES='-I/cvos/apps/netcdf-4.0.ifc/include'
elif test -d /usr/local/netcdf ; then
INCLUDES='-I/cvos/apps/netcdf-4.0.ifc/include'
LIBS='-L/cvos/apps/netcdf-4.0.ifc/lib'
fi
GSLINC=''
GSLLIB='-lgsl -lgslcblas'
PAPIINC='-I/usr/local/pkg/papi/papi-3.0.8.1/p4/include'
PAPILIB='-L/usr/local/pkg/papi/papi-3.0.8.1/p4/lib -lpapi'
-------------- next part --------------
A non-text attachment was scrubbed...
Name: linux_ia32_g77
Type: application/octet-stream
Size: 1511 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20100623/902223da/attachment.obj>
-------------- next part --------------
On 23 Jun 2010, at 17:00, 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: [MITgcm Manual] section 6.15 NetCDF I/O Integration: MNC
> (Jean-Michel Campin)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 23 Jun 2010 07:52:41 -0400
> From: Jean-Michel Campin <jmc at ocean.mit.edu>
> To: mitgcm-support at mitgcm.org
> Subject: Re: [MITgcm-support] [MITgcm Manual] section 6.15 NetCDF I/O
> Integration: MNC
> Message-ID: <20100623115241.GA11739 at ocean.mit.edu>
> Content-Type: text/plain; charset=us-ascii
>
> Hi Phil,
>
> Just 2 commnets:
> 1) Some of the documentation links you are pointed to are from
> the Release-1 version:
>> http://mitgcm.org/public/sealion/
> Probably better to look at the latest version:
>> http://mitgcm.org/public/r2_manual/latest/
> (more recent but not fully up-to-date neither).
> 2) The "WARNING" you are getting are related to failure in compiling/link
> NetCDF code/library.
> When you run "genmake2", there are several tests to check if we have the
> right things set-up to compile and run with NetCDF. If this step is not
> successful, then ALLOW_MNC is disable and if you try to run with
> useMNC=T (data.pkg) you will get this warning message.
> To fix this, I would recommand to check the optfile you are using for
> this platform (and to speed-up this task, you can probably figure out
> whether this works or not just by running "genmake2").
>
> Cheers,
> Jean-Michel
>
> On Wed, Jun 23, 2010 at 11:55:31AM +0100, Phil Underwood wrote:
>> Comments and questions on page http://mitgcm.org/sealion/online_documents/node229.html
>>
>> Hi there,
>>
>> I am currently working on completion of the biogeochemistry tutorial (http://mitgcm.org/public/r2_manual/latest/online_documents/node155.html) and am amending the path names for the "c62g" release of the MITgcm. I have built a test script in order to submit the job to a Sun Grid Engine running on the parallel computing cluster here at my University. Every time the job is completed I get an output file from the system containing the following text:
>>
>> ***WARNING*** PACKAGES_CHECK: useMNC is TRUE
>> ***WARNING*** but pkg/mnc has not been compiled (#undef ALLOW_MNC)
>> STOP NORMAL END statement executed
>>
>> This is irrespective of whether the MITgcm is compiled with the -enable=mnc or -disable=mnc switches. I have also commented out #mnc in the "packages.conf" file within the "/MITgcm_c62g/verification/tutorial_dic_adjoffline/code_ad" directory.
>>
>> I imagine I would like to use MNC since I am under the impression that using NetCDF is likely to make the process of using the model output quicker. Can you tell me how I can get rid of that warning and exactly what I should in order to get the MITgcm running as I would like.
>>
>> Regards,
>> Phil
>>
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
>
> ------------------------------
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
> End of MITgcm-support Digest, Vol 84, Issue 33
> **********************************************
More information about the MITgcm-support
mailing list