Compiling issue.


Jump to page 1Prev 16
Clicked A Few Times
This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.

It was created by Global Arrays (GA) configure 5.3, which was
generated by GNU Autoconf 2.69. Invocation command line was

 $ ../ga-5-3/configure --prefix=/apps/.oldfslapps/chem/nwchem-6.5/src/tools/install --with-tcgmsg --with-mpi=-I/apps/openmpi/1.6.5_intel-13.0.1/include -L/apps/openmpi/1.6.5_intel-13.0.1/lib -lmpi -lmpi_f90 -lmpi_f77 -lpthread --enable-peigs --enable-underscoring --disable-mpi-tests --without-scalapack --without-lapack --with-blas8=-L/apps/intel_compilers/2013.1.117/mkl/lib/intel64 -L/apps/intel_compilers/2013.1.117/mkl/lib/intel64/libmkl_blas95_ilp64.a -lmkl_intel_lp64 -lmkl_sequential -lmkl_core -lpthread -lm --with-sockets CC=icc F77=ifort ARMCI_DEFAULT_SHMMAX_UBOUND=131072

    1. --------- ##
    2. Platform. ##
    3. --------- ##

hostname = m7int02
uname -m = x86_64
uname -r = 2.6.32-504.el6.x86_64
uname -s = Linux
uname -v = #1 SMP Tue Sep 16 01:56:35 EDT 2014

Clicked A Few Times
more detail
I can confirm one more thing. If I run exactly the same input file with the same executable with 1 processor, it runs fine, but if I use 2 it fails with the singularity of the Pulay matrix.

Forum Vet
Could you try to set BLASOPT to the following value and relink
 -Wl,--start-group ${MKLROOT}/lib/intel64/libmkl_intel_ilp64.a ${MKLROOT}/lib/intel64/libmkl_core.a ${MKLROOT}/lib/intel64/libmkl_sequential.a -Wl,--end-group -lpthread -lm

Clicked A Few Times
That seems to have fixed it.
It is running with on a 96 proc DFT job, so that seems to have fixed it.

Forum Vet
Quote:Mattasplund Jul 23rd 7:25 am
It is running with on a 96 proc DFT job, so that seems to have fixed it.

Good
I got this recipe from the Intel Math Kernel Library Link Line Advisor website

https://software.intel.com/en-us/articles/intel-mkl-link-line-advisor


Forum >> NWChem's corner >> Compiling NWChem
Jump to page 1Prev 16