[MITgcm-support] make errors for contrib model

[Jody Klymak]

Hi David, 

Its in PARAMS.h (model/inc/PARAMS.h), at least in newer versions (c61t has it, which isn't particularly new).  I don't have "icefront" in my verification directory, so its odd you'd have an out-of-date PARAMS.h that you are linking to...  If I do "grep inAdMode gad_calc_rhs.f " in my build directory, it certainly finds inAdMode's defn.

You have a linking problem.  I'd delete everything in the "build" directory, and rerun makedepends.  

Cheers, Jody





On Jan 3, 2013, at  12:37 PM, David Porter <dporter at ldeo.columbia.edu> wrote:

> Michael, thanks for the quick help.  Unfortunately, I'm still having the same issue after removing "-fimplicit-none" in my -optfile.  Also, I can't find where "inadmode" is defined in PARMS.h.  I've tried grepping for "inAdMode" in the icefront directory and only get a few "IF ( inAdMode ) THEN" like results. 
> 
> -> make
> Update AD_CONFIG.h and make ./mitgcmuvmake -f Makefile ./mitgcmuv
> make[1]: Entering directory `/local/data/gravity/users/dporter/code/MITgcm/verification/icefront/2D_example/build'
> gfortran -fconvert=big-endian -fPIC  -O3 -funroll-loops -c gad_calc_rhs.f
> gad_calc_rhs.f:3471.23:
> 
>          IF ( inAdMode ) THEN                                          
>                       1
> Error: Symbol 'inadmode' at (1) has no IMPLICIT type
> make[1]: *** [gad_calc_rhs.o] Error 1
> make[1]: Leaving directory `/local/data/gravity/users/dporter/code/MITgcm/verification/icefront/2D_example/build'
> make: *** [fwd_exe_target] Error 2
> 
> Maybe I'm misunderstanding how this contributed code works.  Anyway, thanks again.
> Dave
> 
> 
> On Jan 3, 2013, at 1:33 PM, michael schaferkotter <schaferk at bellsouth.net> wrote:
> 
>> 
>>>      IF ( inAdMode ) THEN                                          
>>>                   1
>>> Error: Symbol 'inadmode' at (1) has no IMPLICIT type
>> 
>> 
>>> gfortran -fconvert=big-endian -fimplicit-none -fPIC  -O3 -funroll-loops -c gad_calc_rhs.f
>> 
>>                                                       ^^^^^^^^^^^^^
>> 
>> remove the implicit none compiler directive.
>> 
>> inAdMode is defined in model/inc/PARAMS.h and PARAMS.h, is in general, an important file.
>> 
>> seems that PARAMS.h would be 'included'  in the source file.
>> 
>> michael
>> 
>> On Jan 3, 2013, at 10:04 AM, David Porter wrote:
>> 
>>> Hi,
>>> 
>>> I have had success building and running many of the verification examples included with the main code.  I am now trying to build/run the contributed content "icefront/2D_example)".  When trying to build this contributed code it in a similar way, the code fails to compile.  Any ideas?
>>> 
>>> $ make
>>> Update AD_CONFIG.h and make ./mitgcmuv
>>> make -f Makefile ./mitgcmuv
>>> make[1]: Entering directory `/local/data/gravity/users/dporter/code/MITgcm/verification/icefront/2D_example/build'
>>> cat gad_calc_rhs.F |  cpp -traditional -P -DWORDLENGTH=4 -DNML_TERMINATOR -DHAVE_SYSTEM -DHAVE_FDATE -DHAVE_CLOC -DHAVE_SETRLSTK -DHAVE_SIGREG -DHAVE_STAT -DHAVE_NETCDF -DHAVE_FLUSH  -I/local/data/gravity/users/dporter/code/netcdf-3.6.3/include | ../../../../tools/set64bitConst.sh  > gad_calc_rhs.f
>>> gfortran -fconvert=big-endian -fimplicit-none -fPIC  -O3 -funroll-loops -c gad_calc_rhs.f
>>> gad_calc_rhs.f:3471.23:
>>> 
>>>      IF ( inAdMode ) THEN                                          
>>>                   1
>>> Error: Symbol 'inadmode' at (1) has no IMPLICIT type
>>> make[1]: *** [gad_calc_rhs.o] Error 1
>>> make[1]: Leaving directory `/local/data/gravity/users/dporter/code/MITgcm/verification/icefront/2D_example/build'
>>> make: *** [fwd_exe_target] Error 2
>>> 
>>> ?
>>> 
>>> Thanks for any help you can provide,
>>> Dave
>>> 
>>> 
>>> 
>>> _______________________________________________
>>> 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
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support

--
Jody Klymak    
http://web.uvic.ca/~jklymak/