make error in 7.2.2, cannot find -ldmapp

[Ziyang Wei]

Dear NWchem team,

I'm trying to make NWchem on a linux cluster with intel compiler, but got an error message at a very late step, (I believe all the object files were compiled but ld reports an error): 

ifort -i8 -align -fpp -Qoption,fpp,-w0 -qno-openmp -fimf-arch-consistency=true -finline-limit=250 -O2 -g -fp-model source     -L/scratch/gpfs/zw7018/nwchem-7.2.2/lib/LINUX64 -L/scratch/gpfs/zw7018/nwchem-7.2.2/src/tools/install/lib   -o /scratch/gpfs/zw7018/nwchem-7.2.2/bin/LINUX64/nwchem nwchem.o stubs.o -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -lpspw -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lfcidump -lgwmol -lnwcutil -lga -larmci -lpeigs -lperfm -lcons -lbq -lnwcutil  -L/opt/intel/compilers_and_libraries_2019.5.281/linux/mkl/lib/intel64 -lmkl_scalapack_ilp64 -lmkl_intel_ilp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_ilp64 -lpthread -lm -L/opt/intel/compilers_and_libraries_2019.5.281/linux/mkl/lib/intel64 -lmkl_scalapack_ilp64 -lmkl_intel_ilp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_ilp64 -lpthread -lm -l64to32 -L/opt/intel/compilers_and_libraries_2019.5.281/linux/mkl/lib/intel64 -lmkl_scalapack_ilp64 -lmkl_intel_ilp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_ilp64 -lpthread -lm -L/opt/intel/compilers_and_libraries_2019.5.281/linux/mkl/lib/intel64 -lmkl_intel_ilp64 -lmkl_core -lmkl_sequential -lmkl_core -liomp5 -lpthread -ldmapp -lm  -L/opt/intel/compilers_and_libraries_2019.5.281/linux/mpi/intel64/lib/release -L/opt/intel/compilers_and_libraries_2019.5.281/linux/mpi/intel64/lib -lmpifort -lmpi -lrt -lpthread -ldl -lhwloc    -lcomex -lmpifort -lmpi -lrt -lpthread -ldl -lhwloc -lblas -lrt -lpthread -lpthread   

My detailed make process is (mostly adapted from https://nwchemgit.github.io/Compiling-NWChem.html#example-nersc-cori):

module purge

module load intel-mpi/intel/2019.5/64

module load intel-mkl/2019.5/5/64

module load intel/19.0/64/19.0.5.281

export NWCHEM_TOP=/scratch/gpfs/zw7018/nwchem-7.2.2

export NWCHEM_TARGET=LINUX64

export ARMCI_NETWORK=MPI-PR

export NWCHEM_MODULES="all"

export USE_MPI=y

make nwchem_config 

export USE_SCALAPACK=y  

export SCALAPACK="-L$MKLROOT/lib/intel64 -lmkl_scalapack_ilp64 -lmkl_intel_ilp64 -lmkl_core -lmkl_sequential -lmkl_blacs_intelmpi_ilp64 -lpthread -lm" 

export SCALAPACK_SIZE=8  

export SCALAPACK_LIB="$SCALAPACK" 

export BLAS_SIZE=8

export BLASOPT="-L$MKLROOT/lib/intel64 -lmkl_intel_ilp64 -lmkl_core -lmkl_sequential -lmkl_core -liomp5 -lpthread -ldmapp -lm"

export LAPACK_LIB="$SCALAPACK" 

export CC=icc

export CXX=icpc

export FC=ifort

make

It will be appreciated if suggestions can be provided. Thank you for your time. 

Best regards,

Ziyang Wei

[Edoardo Aprà]

Please get rid of "-ldmapp" since that was only used on Crays ages ago. It's not needed on any current system.

You might want to set ARMCI_NETWORK=MPI-PR for best parallel performances

https://nwchemgit.github.io/ARMCI.html

If you would like to use Intel compilers, I would recommend the following simple settings (that do not use MKL, in favor of automated builds of OpenBLAS and Scalapack)

export BLAS_SIZE=8

export SCALAPACK_SIZE=8

export BUILD_OPENBLAS=1

export BUILD_SCALAPACK=1

export USE_MPI=1

export ARMCI_NETWORK=MPI-PR

then compile with

make FC=ifort

[Ziyang Wei]

Dear Edoardo,

Many thanks for the advice. Removing -ldmapp indeed works. I have other problems when trying to build the openblas but using the mkl one works in the end.

Thank you for your time.

Best regards,

Ziyang Wei