This thread is locked.Only browsing is available.
Top Page > Browsing
NEGF calculation of OpenMX Ver 3.9
Date: 2019/12/29 03:40
Name: Xingqiang SHI   <>

Dear All

I am using the Ver 3.9 to do the test calculations. All the simple calculations goes well, and the "Lead" tests well for the directory "openmx3.9/work/negf_example". However, all the "NEGF" tests in the same directory fails, with the message below:

[node36:64221] *** Process received signal ***
[node36:64221] Signal: Segmentation fault (11)
[node36:64221] Signal code: Address not mapped (1)
[node36:64221] Failing at address: 0x65400
[node36:64221] [ 0] /lib64/[0x3f2280f710]
[node36:64221] [ 1] /public/home/shixq/SXQ/openmx3.9/work/openmx[0x5fea2f]
[node36:64221] [ 2] /public/home/shixq/SXQ/openmx3.9/work/openmx[0x809c3e]
[node36:64221] [ 3] /public/home/shixq/SXQ/openmx3.9/work/openmx[0x498776]
[node36:64221] [ 4] /public/home/shixq/SXQ/openmx3.9/work/openmx[0x7c11e7]
[node36:64221] [ 5] /public/home/shixq/SXQ/openmx3.9/work/openmx[0x7c0150]
[node36:64221] [ 6] /public/home/shixq/SXQ/openmx3.9/work/openmx[0x432f88]
[node36:64221] [ 7] /lib64/[0x3f2241ed1d]
[node36:64221] [ 8] /public/home/shixq/SXQ/openmx3.9/work/openmx[0x431829]
[node36:64221] *** End of error message ***

The makefile is as following:

MKLROOT = /public/software/compiler/intel/composer_xe_2013_sp1.0.080/mkl
CC = mpicc -O3 -openmp -I/public/home/shixq/VASP/fftw-3.3.4/fftw/include -I/public/Soft_Install/Intel/icc/composer_xe_2013_sp1/include -I/usr/local/include
FC = mpif90 -O3 -openmp -I/usr/local/include -I/public/home/shixq/VASP/fftw-3.3.4/fftw/include -I/public/Soft_Install/Intel/icc/composer_xe_2013_sp1/include -heap-arrays 64
LIB= ${MKLROOT}/lib/intel64/libmkl_scalapack_lp64.a -Wl,--start-group ${MKLROOT}/lib/intel64/libmkl_intel_lp64.a ${MKLROOT}/lib/intel64/libmkl_intel_thread.a ${MKLROOT}/lib/intel64/libmkl_core.a ${MKLROOT}/lib/intel64/libmkl_blacs_openmpi_lp64.a -Wl,--end-group -L/public/software/mpi/openmpi/1.8.4/intel/lib -lmpi_usempif08 -lmpi_usempi_ignore_tkr -lmpi_mpifh -liomp5 -lpthread -lm -ldl

What's the reason for the fail of the NEGF calculations?

With best regards
Xingqiang SHI
Page: [1]

Re: NEGF calculation of OpenMX Ver 3.9 ( No.1 )
Date: 2019/12/30 13:26
Name: T. Ozaki


In fact, we have found there is a bug related to the NEGF calculation.

Please download a corrected routine at

and recompile OpenMX again.

If your problem is solved by the correction, I will release a patch to Ver. 3.9.

Thank you very much for your cooperation in advance.


Re: NEGF calculation of OpenMX Ver 3.9 ( No.2 )
Date: 2019/12/30 16:39
Name: Xingqiang SHI  <>

Dear T. Ozaki

The problem is solved by the correction. Thank you very much.

Best wishes,

Page: [1]