Hello!
I'm new to ROMS and have started trying the upwelling example, but I'm receiving this strange error message:
:>make upwelling
makefile:226: INCLUDING FILE Build/make_macros.mk WHICH CONTAINS APPLICATION-DEPENDENT MAKE DEFINITIONS
make: *** No rule to make target `upwelling'. Stop.
Does anyone have a clue ? I'm really kind a lost around here... :
To compile the model you don't have to use the command "make" directly.
Instead edit and use the script build.bash (or build.sh, depending on your preferences) in your ~/ROMS-3/ROMS/Bin directory.
If you do use make, the syntax is simply "make". You would have to edit the makefile so that your app is set to UPWELLING there (or in the build script). Either way, there are some variables that you have to set: MPI or not, compiler name, application name, etc.
thank you for your help!
I think I've fixed it, but now I'm facing this problem under the command " make"(upw case)
makefile:226: INCLUDING FILE Build/make_macros.mk WHICH CONTAINS APPLICATION-DEPENDENT MAKE DEFINITIONS
cd Build; ifort -c -ip -O3 -xW white_noise.f90
Invalid_sym_id for MP_SCATTER, -352301747
fortcom: Severe: **Internal compiler error: segmentation violation signal raised** Please report this error along with the circumstances in which it occurred in a Software Problem Report. Note: File and line given may not be explicit cause of this error.
compilation aborted for white_noise.f90 (code 3)
Them I've tried it on another machine and received back:
fortcom: Error: mod_parallel.f90, line 18: Cannot open include file 'mpif.h'
> include 'mpif.h'
> ----------------^
> fortcom: Error: mod_parallel.f90, line 76: This symbol must be a defined parameter or an argument of an inquiry
> function that evaluates to a compile-time constant. [MPI_DOUBLE_PRECISION]
> integer, parameter :: MP_FLOAT = MPI_DOUBLE_PRECISION
> -----------------------------------------^
> fortcom: Error: mod_parallel.f90, line 76: This name does not have a type, and must have an explicit type.
> [MPI_DOUBLE_PRECISION]
> integer, parameter :: MP_FLOAT = MPI_DOUBLE_PRECISION
> -----------------------------------------^
> compilation aborted for mod_parallel.f90 (code 1)
> make: *** [Build/mod_parallel.o] Error 1
Does anyone have a good tip about what's happening at any of this cases?
macassis wrote:
fortcom: Severe: **Internal compiler error: segmentation violation signal raised** Please report this error along with the circumstances in which it occurred in a Software Problem Report. Note: File and line given may not be explicit cause of this error.
compilation aborted for white_noise.f90 (code 3)
Don't you love it? This is saying your compiler can't handle compiling white_noise.f90. Have you looked at white_noise.f90 in the Build directory? It's likely that you don't actually need this file so you could try renaming white_noise.F to white_noise.F.orig in the ROMS/Utility directory which will prevent make from trying to compile it.
fortcom: Error: mod_parallel.f90, line 18: Cannot open include file 'mpif.h'
> include 'mpif.h'
> ----------------^
> fortcom: Error: mod_parallel.f90, line 76: This symbol must be a defined parameter or an argument of an inquiry
>
Do you have MPI installed on this second computer? Why not try serial mode first?
uops,thanx, but nope! tried renaming the file at the first machine and it gave me back:
"
makefile:226: INCLUDING FILE Build/make_macros.mk WHICH CONTAINS APPLICATION-DEPENDENT MAKE DEFINITIONS
cd Build; ifort -c -ip -O3 -xW white_noise.f90
Invalid_sym_id for MP_SCATTER, -352301747
fortcom: Severe: **Internal compiler error: segmentation violation signal raised** Please report this error along with the circumstances in which it occurred in a Software Problem Report. Note: File and line given may not be explicit cause of this error.
compilation aborted for white_noise.f90 (code 3)
make: *** [Build/white_noise.o] Error 3
"
do you have another tip?
You are building the model with MPI enabled. As Kate has asked, do you really want to do that? Even if you do want to use MPI eventually, you should try serial mode first.
If you are not deliberately trying to build the model with MPI enabled, then something is happening behind the scenes and you need to find out what. MPI mode is controlled by the preprocessor macro USE_MPI. Are you defining it to a non-empty value?
? I'm really kind a lost around here... : 
