| 4:57:01 AM PST - Thu, Jan 22nd 2015 |
|
Hello,
I got NWChem 6.5 installed on Helios system with Intel Xeon Phi. Here is input file. Same input works well on CRAY NWChem 6.3.
Here are errors that I get after launching nwchem:
1) 0:Floating Point Exception error, status=: 8
2) (rank:0 hostname:helios775 pid:31625):ARMCI DASSERT fail. ../../ga-5 3/armci/src/common/signaltrap.c:SigFpeHandler():249 cond:0
3) Last System Error Message from Task 0:: No such file or directory
Each error is repeated 16 times. Output
Compilation parameters:
Quote:
NWCHEM_TOP /csc/home2/$usr/NWChem-6.5
NWCHEM_TARGET LINUX64
ARMCI_NETWORK OPENIB
- MPI settings
USE_MPI y
USEMPIF y
USE_MPIF4 y
MPI_LOC /opt/mpi/bullxmpi/1.2.8.2
MPI_LIB $MPI_LOC/lib
MPI_INCLUDE $MPI_LOC/include
LIBMPI "-lmpi_f90 -lmpi_f77 -lmpi -ldl -lm -lnuma -Wl,--export-dynamic -lrt -lnsl -lutil -lm -ldl"
- Modules
NWCHEM_MODULES all
- Additional environmental variables
LARGE_FILES TRUE
USE_NOFSCHECK TRUE
MRCC_THEORY TRUE
- Intel Xeon Phi
USE_OPENMP 1
USE_OFFLOAD 1
- Compilation
cd $NWCHEM_TOP/src
make nwchem_config
make FC=ifort CC=icc >& make.log
Here is script, which starts NWChem:
Quote: #!/bin/bash
#SBATCH -J water_test # jobname
#SBATCH -A NAME # project name
#SBATCH -N 1 # number of nodes
#SBATCH -n 16 # number of tasks
#SBATCH -o %j.out # strout filename (%j is jobid)
#SBATCH -e %j.err # stderr filename (%j is jobid)
#SBATCH -t 12:00:00 # execute time
# mpirun -np
export OMP_NUM_THREADS=4
export MIC_USE_2MB_BUFFER=16K
export ARMCI_OPENIB_DEVICE=mlx4_0
module load intel bullxmpi srun
mpirun -np 16 /csc/scratch2/kotomin/nwchem nwchem.nw
• MPI - bullxmpi/1.2.8.2
Thanks in advance!
|