INLET CASE (SWAN-ROM) error when compiling with ifort

[dnavidad]

Hi,


i am trying to compile the inlet (swan-roms coupling) test.

>./build.bash

the process start and look works but after a while i get this error

Code: Select all

ROMS/Bin/cpp_clean /home/dnavidad/ROMS/PROJECTS/Inlet/Build/strings.f90
cd /home/dnavidad/ROMS/PROJECTS/Inlet/Build; /opt/intel/fce/10.0.023/bin/ifort -c -heap-arrays -fp-model precise -g -check uninit -ftrapuv -traceback -I/usr/local/mct/include -free strings.f90
cd /home/dnavidad/ROMS/PROJECTS/Inlet/Build; /opt/intel/fce/10.0.023/bin/ifort -c -heap-arrays -fp-model precise -g -check uninit -ftrapuv -traceback -I/usr/local/mct/include -free mod_netcdf.f90
cd /home/dnavidad/ROMS/PROJECTS/Inlet/Build; /opt/intel/fce/10.0.023/bin/ifort -c -heap-arrays -fp-model precise -g -check uninit -ftrapuv -traceback -I/usr/local/mct/include -free mod_sources.f90
cd /home/dnavidad/ROMS/PROJECTS/Inlet/Build; /opt/intel/fce/10.0.023/bin/ifort -c -heap-arrays -fp-model precise -g -check uninit -ftrapuv -traceback -I/usr/local/mct/include -free analytical.f90
fortcom: Error: analytical.f90, line 573: This is not a field name that is defined in the encompassing structure.   [UBAR_EAST]
          BOUNDARY(ng)%ubar_east(j)=0.0_r8
-----------------------^
fortcom: Error: analytical.f90, line 573: The structure-name is invalid or is missing.   [BOUNDARY]
          BOUNDARY(ng)%ubar_east(j)=0.0_r8
----------^
fortcom: Error: analytical.f90, line 576: This is not a field name that is defined in the encompassing structure.   [VBAR_EAST]
          BOUNDARY(ng)%vbar_east(j)=0.0_r8
-----------------------^
fortcom: Error: analytical.f90, line 576: The structure-name is invalid or is missing.   [BOUNDARY]
          BOUNDARY(ng)%vbar_east(j)=0.0_r8
----------^
fortcom: Error: analytical.f90, line 581: This is not a field name that is defined in the encompassing structure.   [UBAR_WEST]
          BOUNDARY(ng)%ubar_west(j)=0.0_r8
-----------------------^
fortcom: Error: analytical.f90, line 581: The structure-name is invalid or is missing.   [BOUNDARY]
          BOUNDARY(ng)%ubar_west(j)=0.0_r8
----------^
fortcom: Error: analytical.f90, line 584: This is not a field name that is defined in the encompassing structure.   [VBAR_WEST]
          BOUNDARY(ng)%vbar_west(j)=0.0_r8
-----------------------^
fortcom: Error: analytical.f90, line 584: The structure-name is invalid or is missing.   [BOUNDARY]
          BOUNDARY(ng)%vbar_west(j)=0.0_r8
----------^
fortcom: Error: analytical.f90, line 589: This is not a field name that is defined in the encompassing structure.   [UBAR_SOUTH]
          BOUNDARY(ng)%ubar_south(i)=0.0_r8
-----------------------^
fortcom: Error: analytical.f90, line 589: The structure-name is invalid or is missing.   [BOUNDARY]
          BOUNDARY(ng)%ubar_south(i)=0.0_r8
----------^
fortcom: Error: analytical.f90, line 592: This is not a field name that is defined in the encompassing structure.   [VBAR_SOUTH]
          BOUNDARY(ng)%vbar_south(i)=0.0_r8
-----------------------^
fortcom: Error: analytical.f90, line 592: The structure-name is invalid or is missing.   [BOUNDARY]
          BOUNDARY(ng)%vbar_south(i)=0.0_r8
----------^
fortcom: Error: analytical.f90, line 597: This is not a field name that is defined in the encompassing structure.   [UBAR_NORTH]
          BOUNDARY(ng)%ubar_north(i)=0.0_r8
-----------------------^
fortcom: Error: analytical.f90, line 597: The structure-name is invalid or is missing.   [BOUNDARY]
          BOUNDARY(ng)%ubar_north(i)=0.0_r8
----------^
fortcom: Error: analytical.f90, line 600: This is not a field name that is defined in the encompassing structure.   [VBAR_NORTH]
          BOUNDARY(ng)%vbar_north(i)=0.0_r8
-----------------------^
fortcom: Error: analytical.f90, line 600: The structure-name is invalid or is missing.   [BOUNDARY]
          BOUNDARY(ng)%vbar_north(i)=0.0_r8
----------^
fortcom: Error: analytical.f90, line 656: This is not a field name that is defined in the encompassing structure.   [BED]
     &                        OCEAN(ng) % bed,                          &
------------------------------------------^
fortcom: Error: analytical.f90, line 657: This is not a field name that is defined in the encompassing structure.   [BED_FRAC]
     &                        OCEAN(ng) % bed_frac,                     &
------------------------------------------^
fortcom: Error: analytical.f90, line 658: This is not a field name that is defined in the encompassing structure.   [BED_MASS]
     &                        OCEAN(ng) % bed_mass,                     &
------------------------------------------^
fortcom: Error: analytical.f90, line 659: This is not a field name that is defined in the encompassing structure.   [BOTTOM]
     &                        OCEAN(ng) % bottom)
------------------------------------------^
fortcom: Error: analytical.f90, line 656: The shape matching rules of actual arguments and dummy arguments have been violated.   [BED]
     &                        OCEAN(ng) % bed,                          &
------------------------------------------^
fortcom: Error: analytical.f90, line 657: The shape matching rules of actual arguments and dummy arguments have been violated.   [BED_FRAC]
     &                        OCEAN(ng) % bed_frac,                     &
------------------------------------------^
fortcom: Error: analytical.f90, line 658: The shape matching rules of actual arguments and dummy arguments have been violated.   [BED_MASS]
     &                        OCEAN(ng) % bed_mass,                     &
------------------------------------------^
fortcom: Error: analytical.f90, line 659: The shape matching rules of actual arguments and dummy arguments have been violated.   [BOTTOM]
     &                        OCEAN(ng) % bottom)
------------------------------------------^
compilation aborted for analytical.f90 (code 1)
make: *** [/home/dnavidad/ROMS/PROJECTS/Inlet/Build/analytical.o] Error 1

how could i solve it?

thaks

[jcwarner]

i just did an svn update, it compiled, and ran fine.
so i can not reproduce your error. was there more info at the start of the build?

[dnavidad]

Hi, thanks for your reply.

after compiling and run the upwelling case, this is the next step i have done with ROMS, so i have no many experience.

1. how can i get this svn update, could you give me the link?

before this error there is no other message exccept the typical stuff when compiling a lot of information but nothing wrong.

Maybe i am doing something wrong in the build bash. In the tutorial it is explained that, in such test "To run this application the user needs to activate INLET_TEST. It only can be run in distributed-memory (MPI) ....".
2. Where i have to activate this INLET_TEST? in the source code?.

Please let me post here part of my build.bash

Code: Select all

#export      MY_CPP_FLAGS="-DAVERAGES"

# Other user defined environmental variables. See the ROMS makefile for
# details on other options the user might want to set here. Be sure to 
# leave the switched meant to be off set to an empty string or commented
# out. Any string value (including off) will evaluate to TRUE in
# conditional if-stamentents.

 export           USE_MPI=
 export        USE_MPIF90=
 export              FORT=ifort

#export        USE_OpenMP=on

 export         USE_DEBUG=
#export         USE_LARGE=on
 export       USE_NETCDF4=  

thanks, and i hope you provide me some help.

[jcwarner]

you should have your own copy of the code. contact Rutgers to get your own version, via the rutgers website.
The inlet test only works in mpi. It also needs the model coupling toolkit, to exchange data between roms and swan.
so you need
export USE_MPI=on
and maybe
export USE_MPIF90=on

[dnavidad]

Thanks,

i have my copy of the source code.

where should i activate the INLET_TEST? i am working in a cluster i think that all libraries for shared memory are installed.

i changed the option to "on" in the
.....
.....
.....
.....
.....
/Biology -IROMS/Nonlinear/Sediment -IROMS/Utility -IROMS/Drivers -IROMS/Functionals -I/home/dnavidad/ROMS/PROJECTS/Inlet -IWaves/SWAN/Src -IMaster -ICompilers -D'HEADER_DIR="/home/dnavidad/ROMS/PROJECTS/Inlet"' ROMS/Modules/mod_kinds.F > /home/dnavidad/ROMS/PROJECTS/Inlet/Build/mod_kinds.f90
ROMS/Bin/cpp_clean /home/dnavidad/ROMS/PROJECTS/Inlet/Build/mod_kinds.f90
cd /home/dnavidad/ROMS/PROJECTS/Inlet/Build; /opt/mpich/gcc/current/bin/mpif90 -c -heap-arrays -fp-model precise -g -check uninit -ftrapuv -traceback -I/usr/local/mct/include mod_kinds.f90
No Fortran 90 compiler specified when mpif90 was created,
or configuration file does not specify a compiler.
make: *** [/home/dnavidad/ROMS/PROJECTS/Inlet/Build/mod_kinds.o] Error 1


that's mean that i should compile the source with mpif90 or something like that?

[jcwarner]

"No Fortran 90 compiler specified when mpif90 was created,
or configuration file does not specify a compiler."

This is a problem with the installation of mpi on your system. You need to work this out with your system administrator. Did you run the upwelling test with mpi? if not, keep working until you can get upwelling to run in mpi. then you can try more complicated tests like inlet_test.

[dnavidad]

thanks, i will.

ok i think your comments are very usefull, reasonable.

i ran then upwelling test case withouth mpi and also serial.

i will try that you are recommending me and also i will check out mpi- installation with the administrator.

i would very please you keep for your help in further question.

thanks

[dnavidad]

Hi again,

it was not a problem with the installation but a problem in specifying where mpif90 is in the cluster.

i put the correct path for the mpif90 in the linux-ifort.mk of my /src/Compiler folder


ifdef USE_MPIF90
FC := /opt/mpich/intel/current/bin/mpif90
LD := $(FC)


I have other question. This is part of my build.bash

if [ -n "${USE_MPIF90:+1}" ]; then
case "$FORT" in
ifort )
# export PATH=/opt/intelsoft/mpich/bin:$PATH
# export PATH=/opt/intelsoft/mpich2/bin:$PATH
export PATH=/opt/intelsoft/openmpi/bin:$PATH


the uncoment line it is about openmpi. i wonder if i should comment this one and uncomment the first one changing also the path for the one in my cluster? "/opt/mpich/intel/current/bin"



In any case i get the following error message when compiling

cd /home/dnavidad/ROMS/PROJECTS/upwelling_parallel/Build; /opt/mpich/intel/current/bin/mpif90 -c -heap-arrays -fp-model precise -ip -O3 -xW -free analytical.f90
fortcom: Error: analytical.f90, line 325: Syntax error, found ':' when expecting one of: => = . %
ana_grid.h: no values provided for Xsize, Esize, depth, f0, beta.
----------------^
fortcom: Error: analytical.f90, line 1107: Syntax error, found ':' when expecting one of: => = . %
ana_vmix.h: no values provided for Akv.
----------------^
fortcom: Error: analytical.f90, line 1120: Syntax error, found ':' when expecting one of: => = . %
ana_vmix.h: no values provided for Akt.
----------------^
compilation aborted for analytical.f90 (code 1)
make: *** [/home/dnavidad/ROMS/PROJECTS/upwelling_parallel/Build/analytical.o] Error 1

Is it a real error in the code or i am still doing something wrong? i will check out the analitical.f90 and look for ':'

thanks in advance

[dnavidad]

Just to say that i tried first again with upwelling case.

[jcwarner]

something is not set correctly in your build.bash. make sure that you are pointing to the correct ROMS application, code path, functionals path, and include path.
here are some of the parts that are important:

export ROMS_APPLICATION= _____

export MY_ROOT_DIR=_____________
export MY_PROJECT_DIR=${MY_ROOT_DIR}

export MY_HEADER_DIR=${MY_PROJECT_DIR}/ROMS/Include
export MY_ANALYTICAL_DIR=${MY_PROJECT_DIR}/ROMS/Functionals

your setttings may need to be different, it depends on your system and how you manage your file locations.

[dnavidad]

Thanks again. You were right.

Please, let me post part of the build.bash

[dnavidad@darkstar upwelling_parallel]$ joe .build.bash

export ROMS_APPLICATION=UPWELLING_parallel
export MY_ROOT_DIR=/home/dnavidad/ROMS
export MY_PROJECT_DIR=${MY_ROOT_DIR}/PROJECTS/upwelling_parallel
export MY_ROMS_SRC=${MY_ROOT_DIR}/src

here was the error
export MY_HEADER_DIR=${MY_PROJECT_DIR}
export MY_ANALYTICAL_DIR=${MY_PROJECT_DIR}

i changed it to

export MY_HEADER_DIR=${MY_ROMS_SRC}/ROMS/Include
export MY_ANALYTICAL_DIR=${MY_ROMS_SRC}/ROMS/Functionals

[dnavidad]

export MY_HEADER_DIR=${MY_ROMS_SRC}/ROMS/Include

this reported me error message


MY_HEADER_DIR=${MY_PROJECT_DIR} is right, my header file upwelling_parallel.h it is founded here.

[dnavidad]

Hi

i get the same error.

cd /home/dnavidad/ROMS/PROJECTS/upwelling_parallel/Build; /opt/mpich/intel/current/bin/mpif90 -c -heap-arrays -fp-model precise -ip -O3 -xW -free analytical.f90
fortcom: Error: analytical.f90, line 325: Syntax error, found ':' when expecting one of: => = . %
ana_grid.h: no values provided for Xsize, Esize, depth, f0, beta.
----------------^
fortcom: Error: analytical.f90, line 1107: Syntax error, found ':' when expecting one of: => = . %
ana_vmix.h: no values provided for Akv.
----------------^
fortcom: Error: analytical.f90, line 1120: Syntax error, found ':' when expecting one of: => = . %
ana_vmix.h: no values provided for Akt.
----------------^
compilation aborted for analytical.f90 (code 1)
make: *** [/home/dnavidad/ROMS/PROJECTS/upwelling_parallel/Build/analytical.o] Error 1

[kate]

You have to tell your build.bash where to find your ana_grid.h. I don't find it an entirely trivial matter, so I usually prefer to deal directly with the makefile. You can always check the resulting analytical.f90 to see what it has picked up and keep trying until it's right.

[dnavidad]

thanks, please let me post

>[dnavidad@darkstar Functionals]$ pwd
/home/dnavidad/ROMS/src/ROMS/Functionals (where /home/dnavidad/ROMS/src=MY_ROMS_SRC)

[dnavidad@darkstar Functionals]$ ls
ana_biology.h ana_humid.h ana_nudgcoef.h ana_smflux.h ana_tair.h
ana_btflux.h ana_initial.h ana_pair.h ana_specir.h ana_tclima.h
ana_cloud.h analytical.F ana_passive.h ana_spinning.h ana_tobc.h
ana_diag.h ana_m2clima.h ana_perturb.h ana_srflux.h ana_vmix.h
ana_drag.h ana_m2obc.h ana_psource.h ana_ssh.h ana_winds.h
ana_fsobc.h ana_m3clima.h ana_rain.h ana_sss.h ana_wwave.h
ana_grid.h ana_m3obc.h ana_scope.h ana_sst.h Module.mk
ana_hmixcoef.h ana_mask.h ana_sediment.h ana_stflux.h

and my build.bash
....
....
export MY_ROMS_SRC=${MY_ROOT_DIR}/src

export MY_ANALYTICAL_DIR=${MY_ROMS_SRC}/ROMS/Functionals

so, i think that the path for ana_grid.h is correct.

Any idea?

[dnavidad]

Hi,

the problem was a simple one, there was something wrong between my buuild.bash and the ana_grid.h.

in the build.bash was written

export ROMS_APPLICATION=UPWELLING_PARALLEL

my header was upwelling_parallel.h also, but the problem is that upwelling_parallel was not defined in ana_grid.h. Just copying the same parameter in ana_grid.h or just leaving the same name for the parallel case the test compile.

thanks.

[dnavidad]

Hello,

i am trying to compile the inlet test case. that is the error i get

thanks in advance

cd /home/dnavidad/ROMS/PROJECTS/Inlet/Build; /opt/mpich/intel/current/bin/mpif90 -c -heap-arrays -fp-model precise -g -check uninit -ftrapuv -traceback -I/usr/local/mct/include ocean_coupler.f90
fortcom: Error: ocean_coupler.f90, line 18: Error in opening the compiled module file. Check INCLUDE paths. [M_MCTWORLD]
USE m_MCTWorld, ONLY : MCTWorld_init => init
----------^
fortcom: Error: ocean_coupler.f90, line 23: Error in opening the compiled module file. Check INCLUDE paths. [M_GLOBALSEGMAP]
USE m_GlobalSegMap, ONLY : GlobalSegMap
----------^
fortcom: Error: ocean_coupler.f90, line 24: Error in opening the compiled module file. Check INCLUDE paths. [M_GLOBALSEGMAP]
USE m_GlobalSegMap, ONLY : GlobalSegMap_init => init
----------^
fortcom: Error: ocean_coupler.f90, line 25: Error in opening the compiled module file. Check INCLUDE paths. [M_GLOBALSEGMAP]
USE m_GlobalSegMap, ONLY : GlobalSegMap_lsize => lsize
----------^
fortcom: Error: ocean_coupler.f90, line 26: Error in opening the compiled module file. Check INCLUDE paths. [M_GLOBALSEGMAP]
USE m_GlobalSegMap, ONLY : GlobalSegMap_clean => clean
----------^
fortcom: Error: ocean_coupler.f90, line 27: Error in opening the compiled module file. Check INCLUDE paths. [M_GLOBALSEGMAP]
USE m_GlobalSegMap, ONLY : GlobalSegMap_Ordpnts => OrderedPoints
----------^
fortcom: Error: ocean_coupler.f90, line 31: Error in opening the compiled module file. Check INCLUDE paths. [M_ATTRVECT]
USE m_AttrVect, ONLY : AttrVect
----------^
fortcom: Error: ocean_coupler.f90, line 32: Error in opening the compiled module file. Check INCLUDE paths. [M_ATTRVECT]
USE m_AttrVect, ONLY : AttrVect_init => init
----------^
fortcom: Error: ocean_coupler.f90, line 33: Error in opening the compiled module file. Check INCLUDE paths. [M_ATTRVECT]
USE m_AttrVect, ONLY : AttrVect_zero => zero
----------^
fortcom: Error: ocean_coupler.f90, line 34: Error in opening the compiled module file. Check INCLUDE paths. [M_ATTRVECT]
USE m_AttrVect, ONLY : AttrVect_lsize => lsize
----------^
fortcom: Error: ocean_coupler.f90, line 35: Error in opening the compiled module file. Check INCLUDE paths. [M_ATTRVECT]
USE m_AttrVect, ONLY : AttrVect_clean => clean
----------^
fortcom: Error: ocean_coupler.f90, line 36: Error in opening the compiled module file. Check INCLUDE paths. [M_ATTRVECT]
USE m_AttrVect, ONLY : AttrVect_importRAttr => importRAttr
----------^
fortcom: Error: ocean_coupler.f90, line 37: Error in opening the compiled module file. Check INCLUDE paths. [M_ATTRVECT]
USE m_AttrVect, ONLY : AttrVect_exportRAttr => exportRAttr
----------^
fortcom: Error: ocean_coupler.f90, line 41: Error in opening the compiled module file. Check INCLUDE paths. [M_ROUTER]
USE m_Router, ONLY : Router
----------^
fortcom: Error: ocean_coupler.f90, line 42: Error in opening the compiled module file. Check INCLUDE paths. [M_ROUTER]
USE m_Router, ONLY : Router_init => init
----------^
fortcom: Error: ocean_coupler.f90, line 43: Error in opening the compiled module file. Check INCLUDE paths. [M_ROUTER]
USE m_Router, ONLY : Router_clean => clean
----------^
fortcom: Error: ocean_coupler.f90, line 47: Error in opening the compiled module file. Check INCLUDE paths. [M_TRANSFER]
USE m_Transfer, ONLY: MCT_Send => send
----------^
fortcom: Error: ocean_coupler.f90, line 48: Error in opening the compiled module file. Check INCLUDE paths. [M_TRANSFER]
USE m_Transfer, ONLY: MCT_Recv => recv
----------^
fortcom: Error: ocean_coupler.f90, line 52: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIX]
USE m_SparseMatrix, ONLY : SparseMatrix
----------^
fortcom: Error: ocean_coupler.f90, line 53: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIX]
USE m_SparseMatrix, ONLY : SparseMatrix_init => init
----------^
fortcom: Error: ocean_coupler.f90, line 54: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIX]
USE m_SparseMatrix, ONLY : SparseMatrix_importGRowInd => &
----------^
fortcom: Error: ocean_coupler.f90, line 56: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIX]
USE m_SparseMatrix, ONLY : SparseMatrix_importGColInd => &
----------^
fortcom: Error: ocean_coupler.f90, line 58: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIX]
USE m_SparseMatrix, ONLY : SparseMatrix_importMatrixElts => &
----------^
fortcom: Error: ocean_coupler.f90, line 60: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIXPLUS]
USE m_SparseMatrixPlus, ONLY : SparseMatrixPlus
----------^
fortcom: Error: ocean_coupler.f90, line 61: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIXPLUS]
USE m_SparseMatrixPlus, ONLY : SparseMatrixPlus_init => init
----------^
fortcom: Error: ocean_coupler.f90, line 62: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIXPLUS]
USE m_SparseMatrixPlus, ONLY : SparseMatrixPlus_clean => clean
----------^
fortcom: Error: ocean_coupler.f90, line 66: Error in opening the compiled module file. Check INCLUDE paths. [M_SPARSEMATRIXPLUS]
USE m_SparseMatrixPlus, ONLY : Xonly
----------^
fortcom: Error: ocean_coupler.f90, line 70: Error in opening the compiled module file. Check INCLUDE paths. [M_MATATTRVECTMUL]
USE m_MatAttrVectMul, ONLY : MCT_MatVecMul => sMatAvMult
----------^
fortcom: Error: ocean_coupler.f90, line 81: Conflicting attributes or multiple declaration of name. [ATTRVECT]
TYPE(AttrVect) :: wav2ocn_AV ! AttrVect variables
-----------^
fortcom: Error: ocean_coupler.f90, line 82: Conflicting attributes or multiple declaration of name. [ATTRVECT]
TYPE(AttrVect) :: ocn2wav_AV
-----------^
fortcom: Severe: Too many errors, exiting
compilation aborted for ocean_coupler.f90 (code 1)
make: *** [/home/dnavidad/ROMS/PROJECTS/Inlet/Build/ocean_coupler.o] Error 1

[jcwarner]

what is in
/usr/local/mct/include

[dnavidad]

there is nothing.

but there is in /opt/MCT/intel_mpich

so i changed the following path in the build.bash

export MCT_INCDIR=/opt/MCT/intel_mpich/include/
export MCT_LIBDIR=/opt/MCT/intel_mpich/lib/


after that i get a error regarding radiation_stress.F and with a little of help

i changed

#define NEARSHORE_MELLOR for #define NEARSHORE_MELLOR05

in the inlet_test.h.

now it is working,

Many thanks

[jcwarner]

yes. There are times when "the powers that be" just go ahead and change things just because they think it should be a different way. Glad you were able to identify this and get it to work.

[dnavidad]

thanks. i will continue asking.

unfortunately the range of kind of errors you can get is very wide but fortunately you usually find someone with the same problem or who can give you the right indications to solve it.

[Manel_Grifoll]

Recently I have run the inlet case SWAN-ROMS and I had a similar error.
After to explore parallization libraries and compilation options I solved using the headers file from my repository (i.e. myroms) instead the file from wikiroms which are both different. Perhaps could be your case.

Manel

[dnavidad]

Hola Manel, saludos from USA para Cataluña.

In somehow the problem is the same, but finally it is solved.ROMS requires its time to get the understanding and the path and libraries stuff usually gives a lot of problem.

gracias for your post.