Download and Install

Download

Compile

  • Before compile BerkeleyGW package, one should compile hdf5-1.12.0 code Download version 1.12.0 of HDF5.

2.1 Compile HDF5

./configure CC=mpiicc FC=mpiifort CXX=mpiicpc \
    --enable-fortran --enable-parallel --enable-shared \
    --prefix=/your_path
make

make install

2.2 Compile BerkeleyGW

  • Create arch.mk file, one can modifiy the ./config/frontera.tacc.utexas.edu_impi.mk file as follows.

# arch.mk for BerkeleyGW codes
#
# suitable for Frontera at TACC, Cascade Lake (CLX) architecture
# Uses intel compilers + impi
#
# BERKELEYGW DOES NOT WORK WITH INTEL/18 DUE TO A COMPILER BUG!
# ------------------------------------------------------------
#
# You'll need to run:
# module load arpack impi intel/19.0.5 phdf5
#
# Use 'make -j 8' for parallel build on Frontera
#
# Zhenglu Li
# July 2020, Berkeley

COMPFLAG  = -DINTEL
PARAFLAG  = -DMPI -DOMP
MATHFLAG  = -DUSESCALAPACK -DUNPACKED -DUSEFFTW3 -DHDF5
# Only uncomment DEBUGFLAG if you need to develop/debug BerkeleyGW.
# The output will be much more verbose, and the code will slow down by ~20%.
#DEBUGFLAG = -DDEBUG

FCPP    = cpp -C -nostdinc
F90free = mpiifort -free -qopenmp -ip -no-ipo
LINK    = mpiifort -qopenmp -ip -no-ipo
# We need the -fp-model procise to pass the testsuite.
FOPTS   = -O3 -fp-model source
FNOOPTS = -O2 -fp-model source -no-ip
#FOPTS   = -g -O0 -check all -Warn all -traceback
#FNOOPTS = $(FOPTS)
MOD_OPT = -module
INCFLAG = -I

C_PARAFLAG = -DPARA -DMPICH_IGNORE_CXX_SEEK
CC_COMP = mpiicpc
C_COMP  = mpiicc
C_LINK  = mpiicpc
C_OPTS  = -O3 -ip -no-ipo -qopenmp
C_DEBUGFLAG =

REMOVE  = /bin/rm -f

# Math Libraries
#
MKLPATH      = $(MKLROOT)/lib/intel64

FFTWLIB      =      -Wl,--start-group \
            $(MKLPATH)/libmkl_intel_lp64.a \
            $(MKLPATH)/libmkl_intel_thread.a \
            $(MKLPATH)/libmkl_core.a \
            -Wl,--end-group -liomp5 -lpthread -lm -ldl
FFTWINCLUDE  = $(MKLROOT)/include/fftw


LAPACKLIB    = -Wl,--start-group \
            $(MKLPATH)/libmkl_intel_lp64.a \
            $(MKLPATH)/libmkl_intel_thread.a \
            $(MKLPATH)/libmkl_core.a \
            $(MKLPATH)/libmkl_blacs_intelmpi_lp64.a \
            -Wl,--end-group -liomp5 -lpthread -lm -ldl
SCALAPACKLIB = $(MKLPATH)/libmkl_scalapack_lp64.a

HDF5PATH     =  /opt/software/BerkeleyGW-3.0.1/hdf5/lib
HDF5LIB      =      $(HDF5PATH)/libhdf5hl_fortran.a \
            $(HDF5PATH)/libhdf5_hl.a \
            $(HDF5PATH)/libhdf5_fortran.a \
            $(HDF5PATH)/libhdf5.a \
            -lz
HDF5INCLUDE  = $(HDF5PATH)/../include

TESTSCRIPT =

  • Compile

make -j N all-flavors (N is the number of cores)