[RegCNET] A Problem During Make RegCM-4.3.5

干豆腐 164993534 at qq.com
Wed Jan 2 10:33:06 CET 2013


Dear All
    I have some problems when compiling the newest  version of RegCM-4.3.5 . I download the .tar.gz file on  the website and compile the model. After ./configure  --enable-clm CC=icc FC=ifort MPIFC=mpiifort ,and make . 
  
 1. When it making /Main/mpplib/mod_mppparam.F90 ,an  error occured . 
 Here are the details:
  
 mod_mppparam.F90(464): error #6404: This name does not have a type, and must have an explicit type.   [MPI_STATUS_IGNORE]
                   cartesian_communicator,mpi_status_ignore,mpierr)
 -----------------------------------------^
 mod_mppparam.F90(588): error #6404: This name does not have a type, and must have an explicit type.   [MPI_PROC_NULL]
     ma%top         = mpi_proc_null
 ---------------------^
 compilation aborted for mod_mppparam.F90 (code 1)
  
 Then I comment out the two lines in the mod_mppparam.F90 
  
 !#ifdef MPI_SERIAL
   integer(ik4) mpi_status_ignore(mpi_status_size)
   integer(ik4) , parameter :: mpi_proc_null = -2
 !#endif
  
 and compiler can continue.
  
 2.When it making /Main/mod_regcm_interface.F90 , another error occured  .
  Here are the details:
  
  mod_regcm_interface.F90(90): error #6404: This name does not have a type, and must have an explicit type.   [MPI_ERRORS_RETURN]
     call mpi_comm_set_errhandler(mycomm, mpi_errors_return, ierr)
  

 I searched all the directories, the .F90 files and the .f90 files  ,and I cannot find where the subroutine is. That is very odd.
 My compiler is Intel Composer XE  2011 sp1.8.273 , MPIFC is  mpiifort .  I also used openmpi and mpich to compile, which yielded the same errors.  
 Why I cannot compile the model. Any help or suggestions will be  greatly appreciated. Thank you very much !
  
 Best regards !
 2013-01-02
  
 
 XingLi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ictp.it/pipermail/regcnet/attachments/20130102/acb34908/attachment-0001.htm>


More information about the RegCNET mailing list