[Pw_forum] Compile Hitachi

WANG Yuan Xu wang.yuanxu at nims.go.jp
Fri Oct 15 03:08:32 CEST 2004

Dear Paolo Giannozzi,

	Compile on Hitachi is still not successful. I do not define -D_shmem in make.sys. But it still compile it, the shmem_include.f90.

 I use "./configure.old hitachi"   get make.sys and then change some content. 
 Another the file "Modules/.dependences" is from above command. I am not sure if it is correct. My Hitachi is SR11000 and it is different hitachi sr8000.
 Could you help me to find some reason?

# System-dependent definitions for Hitachi SR8000
# contributed by Noejung Park
# Use precompiled fftw library (version <= 2.1.5, NOT v.3!)
# In this case, specify also  how to load the fftw library (FFTW_LIB)
# and the path to the fftw.h include file (FFTW_INC_DIR). Example:
 FFTW_LIB=-L/home/araimasa/src/fftw-2.1.5/fftw/.libs -lfftw

# Use the local copy of fftw

# Fortran compiler:
F90 = mpif90_r 
F77 = f90 -hf77
CC = mpicc_r
FFLAGS = -free 
#FFLAGS = -O4 -32 -noparallel -precexp=basic -conti199 -nosaveallocate \
#         -I$(OSHOME)/include -I/usr/mpi/include
# This is needed to tell the compiler where modules are
# Libraries:
MYLIB = lapack_ibm 
LAPACKdir = /usr/local/lib
#MPIdir = /opt/hitachi/matmpp/lib/
MPIdir = /usr/include/
LIBS= -L$(LAPACKdir) -llapack -lblas -lmpi\
      -L$(MPIdir)  -lfmpi -lmpi $(FFTW_LIB) -lm

LDFLAGS = $(OSHOME)/flib/ptools.a $(OSHOME)/flib/flib.a $(OSHOME)/clib/clib.a $(LIBS)
# ar:
AR = ar
RANLIB         = echo

 .SUFFIXES : .o .c .f .f90

	$(F90) $(F90FLAGS) $(MODULEFLAG) -c $<

	$(F77) $(F77FLAGS) -c $<

	$(CC) $(CCFLAGS) -c $<

WANG Yuan Xu
National Institute for Materials Science
Computational Materials Science Center
Namiki 1-1, Tsukuba 305-0044, Japan
Phone +81-29-8513354-8092; Fax +81-29-8541207
wang.yuanxu at nims.go.jp

======= 2004-09-30 20:18:45 you wrote=======

>On Thursday 30 September 2004 07:55, WANG Yuan Xu wrote:
>> I use the Ba pseudo file from PWscf web site.  And O an Ti was
>> Ti.vdb.UPF and O.vdb.UPF
>> I get the error information "from readpp # 2  inconsistent DFT read"
>for PWscf, the exchange-correlation used in the calculation is read from
>the PP files, and it must be the same in all PP's (of course). One can
>edit the PP files and modify the exch-corr field, but it is not a good idea.
>Note that FPMD uses a different approach: you have to write the exch-corr 
>you want in the input data (variable "xc_type" in namelist &system)
>Does PWscf work now on the Hitachi machine, by the way?
>Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
>Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
>Piazza dei Cavalieri 7      I-56126 Pisa, Italy
>Pw_forum mailing list
>Pw_forum at pwscf.org

= = = = = = = = = = = = = = = = = = = =

More information about the users mailing list