• View
• Edit
• History
• Print

Examples

• SERIAL VERSION:

64 bit linux machine using the Intel compiler, the MKL:
MKLPATH=/path/to/the/mkl FC=ifort CC=icc ./configure --with-blas="$MKLPATH/libmkl_solver_ilp64.a -Wl,--start-group $MKLPATH/libmkl_intel_ilp64.a $MKLPATH/libmkl_intel_thread.a $MKLPATH/libmkl_core.a -Wl,--end-group -openmp -lpthread" --prefix=/data/share/cfour (all in one line !!!)

64 bit linux machine using the Portland compiler, the ACML:
FC=pgf90 CC=pgcc MPIFC=mpif90 ./configure --prefix=/data/share/cfour (all in one line !!!)

on an Intel Xeon featuring EM64T:
FC=ifort CC=icc ./configure --with-blas='-L/opt/intel/mkl721/lib/em64t -lmkl_solver -lmkl_lapack -lmkl_em64t -lmkl -lguide -lpthread' (all in one line !!!)

on a DEC alpha:
./configure

under HPUX:
./configure

under IBM AIX:
./configure --with-blas=blashc
cp blas_handcoded/dgemm.pwr5 blas_handcoded/dgemm.f

under Mac OS X 10.6.7, with Xcode4 and Intel® Fortran Composer XE for Mac OS X 2011 (64 bit version):
FC=ifort ./configure --with-blas="-lmkl_blas95_ilp64 -lmkl_lapack95_ilp64 /opt/intel/mkl/lib/libmkl_intel_ilp64.a /opt/intel/mkl/lib/libmkl_sequential.a /opt/intel/mkl/lib/libmkl_core.a -lpthread" LDFLAGS="-L/opt/intel/mkl/lib -L/Xcode4/SDKs/MacOSX10.6.sdk/usr/lib" FCFLAGS="-warn noalign -I/opt/intel/mkl/include/intel64/ilp64 -DINT64 -i8"

• PARALLEL VERSION:
  

32 bit linux machine using the Intel compiler, the MKL and LAM/MPI:
FC=ifort CC=icc MPIFC=mpif77 ./configure --with-blas-dir=/opt/intel/mkl/10.0.011/lib/32 --with-blas='/opt/intel/mkl/10.0.011/lib/32/libmkl_ia32.a /opt/intel/mkl/10.0.011/lib/32/libguide.a -lpthread' --with-lapack='/opt/intel/mkl/10.0.011/lib/32/libmkl_ia32.a /opt/intel/mkl/10.0.011/lib/32/libguide.a -lpthread' --enable-mpi=lam (all in one line !!!)

64 bit linux machine using the Intel compiler, the MKL and LAM/MPI:
FC=ifort CC=icc MPIFC=mpif77 ./configure --with-blas-dir=/opt/intel/mkl/9.0/lib/em64t --with-blas='/opt/intel/mkl/9.0/lib/em64t/libmkl_em64t.a /opt/intel/mkl/9.0/lib/em64t/libguide.a -lpthread' --enable-mpi=lam (all in one line !!!)

64 bit linux machine using the Intel compiler, the MKL and OpenMPI:
MKLPATH=/path/to/the/mkl FC=ifort CC=icc MPIFC=mpif90./configure --with-blas="$MKLPATH/libmkl_solver_ilp64.a -Wl,--start-group $MKLPATH/libmkl_intel_ilp64.a $MKLPATH/libmkl_intel_thread.a $MKLPATH/libmkl_core.a -Wl,--end-group -openmp -lpthread" --prefix=/data/share/cfour_mpi --enable-mpi=openmpi --with-mpirun="mpirun -np \$CFOUR_NUM_CORES" --with-exenodes="mpirun -np \$CFOUR_NUM_CORES" (all in one line !!!)

64 bit linux machine using the Portland compiler, the ACML and OpenMPI:
FC=pgf90 CC=pgcc MPIFC=mpif90 ./configure --prefix=/data/share/cfour_mpi --enable-mpi=openmpi --with-mpirun="mpirun -np \$CFOUR_NUM_CORES" --with-exenodes="mpirun -np \$CFOUR_NUM_CORES" (all in one line !!!)

under Mac OS X 10.6.7, with Xcode4 and Intel® Fortran Composer XE for Mac OS X 2011 (64 bit version with OpenMPI):
FC=ifort ./configure --with-blas="-lmkl_blas95_ilp64 -lmkl_lapack95_ilp64 /opt/intel/mkl/lib/libmkl_intel_ilp64.a /opt/intel/mkl/lib/libmkl_sequential.a /opt/intel/mkl/lib/libmkl_core.a -lpthread" --enable-mpi=openmpi --with-mpirun="/opt/openmpi/bin/mpirun -np \$CFOUR_NUM_CORES" --with-exenodes="/opt/openmpi/bin/mpirun -np \$CFOUR_NUM_CORES" LDFLAGS="-L/opt/intel/mkl/lib -L/opt/openmpi/lib -L/Xcode4/SDKs/MacOSX10.6.sdk/usr/lib" FCFLAGS="-warn noalign -I/opt/intel/mkl/include/intel64/ilp64 -DINT64 -i8" (all in one line !!!)