[MITgcm-support] Error: mnc_cw_model_attr.f
Martin Losch
Martin.Losch at awi.de
Thu Aug 6 10:52:08 EDT 2015
Well,
the normal procedure to build the executable is just that (as described in the manual, “Getting Started”)
create the makefile (make makefile is just a shortcut that’s available if you have already created a makefile earlier)
make CLEAN (remove everything but the makefile)
make depend (checks for the dependencies in the code and appends to the makefile. For this all source code is linked into the build directory. Without “make depend” the makefile does not work properly)
make (compile the code)
M.
> On 06 Aug 2015, at 16:08, WangYang <effwang at 126.com> wrote:
>
> Dear Martin,
> Thank you very much for your advice. I checked the folder MITgcm_c61z and can only find
>
> #ifndef MNC_OPTIONS_H
> #define MNC_OPTIONS_H
> #include "PACKAGES_CONFIG.h"
> #ifdef ALLOW_MNC
> #include "CPP_OPTIONS.h"
> #include "MNC_ID_HEADER.h"
>
> in 'MNC_OPTIONS.h'
>
> and then I followed your suggestion 'make makefile && make CLEAN && make depend && make'
> There was no error anymore !
> Could you please tell me why these command useful ?
>
> Sincerely yours,
> Wang Yang
>
>
>
>
> At 2015-08-06 20:15:31, "Martin Losch" <Martin.Losch at awi.de> wrote:
> >Dear WangYang,
> >
> >MNC_TAG_ID is a CPP-FLAG that should have been replaced by a string during the “cpp” step (when the lower case “f” file is created); so at compile time this name should not appear in “mnc_cw_model_attr.f”. Check, if you have a MNC_ID_HEADER.h that contains a like much like this
> >#define MNC_TAG_ID '1.2165 2015/06/19’
> >If that’s not the case then I would recommend redoing the make procedure startiing at the makefile:
> >
> >make makefile && make CLEAN && make depend && make
> >
> >(the “make depend” is very important, because it creates links that you need during “make”).
> >
> >otherwise something goes wrong at the “cpp” step (cat mnc_cw_model_attr.F | cpp -traditional -P -DWORDLENGTH=4 -DHAVE_SYSTEM -DHAVE_FDATE -DHAVE_ETIME -DHAVE_CLOC -DHAVE_SETRLSTK -DHAVE_SIGREG -DHAVE_STAT -DHAVE_NETCDF -DHAVE_FLUSH -I/public/soft/netcdf362-intel101/include | ../../../tools/set64bitConst.sh > mnc_cw_model_attr.f) I but really can’t what should go wrong there ...
> >
> >Martin
> >
> >
> >> On 06 Aug 2015, at 13:55, WangYang <effwang at 126.com> wrote:
> >>
> >> Dear all,
> >> I was trying to run the example internal_wave, I modified the SIZE.h file as follows:
> >> & sNx = 300,
> >> & sNy = 1,
> >> & OLx = 2,
> >> & OLy = 2,
> >> & nSx = 2,
> >> & nSy = 1,
> >> & nPx = 1,
> >> & nPy = 1,
> >> & Nx = sNx*nSx*nPx,
> >> & Ny = sNy*nSy*nPy,
> >> & Nr = 20)
> >> because I have changed the horizontal resolution from 60grids to 600grids;
> >> I used the ifort compiler. When doing make,I got the error:
> >> ifort -w95 -W0 -WB -convert big_endian -assume byterecl -fPIC -O2 -align -xW -ip -c mnc_cw_missingvals.f
> >> cat mnc_cw_model_attr.F | cpp -traditional -P -DWORDLENGTH=4 -DHAVE_SYSTEM -DHAVE_FDATE -DHAVE_ETIME -DHAVE_CLOC -DHAVE_SETRLSTK -DHAVE_SIGREG -DHAVE_STAT -DHAVE_NETCDF -DHAVE_FLUSH -I/public/soft/netcdf362-intel101/include | ../../../tools/set64bitConst.sh > mnc_cw_model_attr.f
> >> ifort -w95 -W0 -WB -convert big_endian -assume byterecl -fPIC -O2 -align -xW -ip -c mnc_cw_model_attr.f
> >> fortcom: Error: mnc_cw_model_attr.f, line 1795: This name does not have a type, and must have an explicit type. [MNC_TAG_ID]
> >> & MNC_TAG_ID, myThid )
> >> -----------^
> >> compilation aborted for mnc_cw_model_attr.f (code 1)
> >> make[1]: *** [mnc_cw_model_attr.o] Error 1
> >> make[1]: Leaving directory `/public/home/wang_yang/MITgcm01/verification/internal_wave/build'
> >> make: *** [fwd_exe_target] Error 2
> >>
> >> My MITgcm version is MITgcm_c61z,
> >> anybody can help me to solve the problem?
> >> thanks a lot!
> >> WangYang
> >>
> >>
> >> _______________________________________________
> >> 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
More information about the MITgcm-support
mailing list