[Wien] compile problem mpi parallel

[Chandrika]

Hello Laurence,
Thanks for the reply. Since my parrallelisation is with Oscar default, it is 
not compiled with mpif90. I am trying to get around this without disturbing 
the cluster. I will be glad for any suggestions on this.
Also I am not sure about- MPIRUN parameters. What do -np _NP_ -machinefile 
_HOSTS_ _EXEC_ stand for?
Is the machinefile the one in mpich directory or the Wien directory? Do I 
give the full path?
Thank you for your time.
Chandrika

----- Original Message ----- 
From: "Laurence Marks" <L-marks at northwestern.edu>
To: "A Mailing list for WIEN2k users" <wien at zeus.theochem.tuwien.ac.at>
Sent: Wednesday, September 19, 2007 5:58 PM
Subject: Re: [Wien] compile problem mpi parallel


> Did you define the appropriate value for parallel compilation in the
> setup, e.g. mpif90 or are you using ifort for the mpi compilation?
> (Check the value of MPF in the Makefile.) You have to use mpif90 which
> should have the directory where mpi.h embedded in it as well as the
> locations of needed libraries. If you are using a proper mpif90 then
> something might be broke in your mpi compilation -- test it with
> something simple, e.g. "Hello World".
>
> On 9/19/07, Chandrika <rcais at cal3.vsnl.net.in> wrote:
>>
>>
>> Dear Wien users,
>> I am having a problem in compiling for mpi parallel. I have a cluster of
>> eight P IV machines with fedora core 5 and Intel Cluster mkl. I have
>> parallelised using Oscar 5 (mpich).
>> K point parallel runs successfully. The options I used for mpi are
>>
>>
>> FOPT:-FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML
>>
>> LDFLAGS:-L/opt/intel_fc_80/lib -i-static -lguide_stats -lpthread
>>
>> DPARALLEL:'-DParallel'
>>
>> R_LIBS:-L/opt/intel/mkl72cluster/lib/32 -lmkl_lapack
>> -lmkl_ia32 -lguide_stats -lpthread
>>
>> RP_LIBS:-L /opt/intel/mkl72cluster/lib/32 -lmkl_sclapack -lmkl_blacs
>> -lmkl_lapack -lmkl_ia32 -lguide_stats -lpthread -lpblas -lmpich
>>
>> MPIRUN:mpirun -np _NP_ -machinefile _HOSTS_ _EXEC_
>>
>>
>>
>> The last two options are probably wrong. The compile error for lapw0 says 
>> :
>>
>>
>>
>> SRC_lapw0/compile.msg:fortcom: Error: modules.F, line 19: Cannot open
>> include file 'mpif.h'
>>
>> SRC_lapw0/compile.msg:fortcom: Error: modules.F, line 42: Cannot open
>> include file 'mpif.h'
>>
>> SRC_lapw0/compile.msg:fortcom: Error: modules.F, line 20: This name does 
>> not
>> have a type, and must have an explicit type.   [MPI_STATUS_SIZE]
>>
>> SRC_lapw0/compile.msg:fortcom: Error: modules.F, line 20: The highest 
>> data
>> type rank permitted is INTEGER(KIND=8).   [MPI_STATUS_SIZE]
>>
>> SRC_lapw0/compile.msg:fortcom: Error: modules.F, line 22: This name does 
>> not
>> have a type, and must have an explicit type.   [MPI_SUCCESS]
>>
>> SRC_lapw0/compile.msg:fortcom: Error: modules.F, line 26: This name does 
>> not
>> have a type, and must have an explicit type.   [MPI_COMM_WORLD]
>>
>> SRC_lapw0/compile.msg:fortcom: Error: modules.F, line 44: This name does 
>> not
>> have a type, and must have an explicit type.   [MPI_UNDEFINED]
>>
>>
>>
>> I have included mpich library path into basrc and ld.so.conf.
>>
>> Please suggest how to solve this.
>>
>> Thank you for your time.
>>
>> Chandrika
>>
>>
>>
>>
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>>
>
>
> -- 
> Laurence Marks
> Department of Materials Science and Engineering
> MSE Rm 2036 Cook Hall
> 2220 N Campus Drive
> Northwestern University
> Evanston, IL 60208, USA
> Tel: (847) 491-3996 Fax: (847) 491-7820
> email: L-marks at northwestern dot edu
> Web: www.numis.northwestern.edu
> EMM2007 http://ns.crys.ras.ru/EMMM07/
> Commission on Electron Diffraction of IUCR
> www.numis.northwestern.edu/IUCR_CED
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien