Problem running GW calculations with nsspol 2
Posted: Wed Jun 01, 2011 11:51 am
Dear ABINITers,
I have compiled ABINIT 6.6.3 using Intel compilers, MKL, and Intel MPI. When I now try to run an input file that used to work correctly with version 6.0.3 on a different machine, I get segmentation faults in the last step of the GW run. As the calculation is performed in a spin-polarized way, that is nsspol 2, I changed bdgw from
bdgw4 380 384
to
bdgw4 380 384
380 384
I don't get this error with nsspol 1. The problem occurs with both version 6.6.3 and 6.4.3. I have also tried to reduce the optimization to standard level. What else can be the source of this error? I attached the input and log file.
Thanks!
Best regards,
Martin Haeufel
The build information reads:
=== Build Information ===
Version : 6.6.3
Build target : x86_64_linux_intel12.0
Build date : 20110520
=== Compiler Suite ===
C compiler : intel12.0
CFLAGS : -g -O3 -vec-report0
C++ compiler : gnu12.0
CXXFLAGS : -m64 -g -O3 -mtune=native -march=native
Fortran compiler : intel12.0
FCFLAGS : -g -extend-source -vec-report0 -noaltparam -nofpscomp
FC_LDFLAGS : -static-intel -static-libgcc
=== Optimizations ===
Debug level : basic
Optimization level : aggressive
Architecture : amd_opteron
=== MPI ===
Parallel build : yes
Parallel I/O : yes
Time tracing : no
GPU support : no
=== Connectors / Fallbacks ===
Connectors on : yes
Fallbacks on : yes
DFT flavor :
FFT flavor : fftw3
LINALG flavor : mkl+scalapack
MATH flavor : none
TIMER flavor : abinit
TRIO flavor :
=== Experimental features ===
Bindings : no
Exports : no
GW double-precision : no
I have compiled ABINIT 6.6.3 using Intel compilers, MKL, and Intel MPI. When I now try to run an input file that used to work correctly with version 6.0.3 on a different machine, I get segmentation faults in the last step of the GW run. As the calculation is performed in a spin-polarized way, that is nsspol 2, I changed bdgw from
bdgw4 380 384
to
bdgw4 380 384
380 384
I don't get this error with nsspol 1. The problem occurs with both version 6.6.3 and 6.4.3. I have also tried to reduce the optimization to standard level. What else can be the source of this error? I attached the input and log file.
Thanks!
Best regards,
Martin Haeufel
The build information reads:
=== Build Information ===
Version : 6.6.3
Build target : x86_64_linux_intel12.0
Build date : 20110520
=== Compiler Suite ===
C compiler : intel12.0
CFLAGS : -g -O3 -vec-report0
C++ compiler : gnu12.0
CXXFLAGS : -m64 -g -O3 -mtune=native -march=native
Fortran compiler : intel12.0
FCFLAGS : -g -extend-source -vec-report0 -noaltparam -nofpscomp
FC_LDFLAGS : -static-intel -static-libgcc
=== Optimizations ===
Debug level : basic
Optimization level : aggressive
Architecture : amd_opteron
=== MPI ===
Parallel build : yes
Parallel I/O : yes
Time tracing : no
GPU support : no
=== Connectors / Fallbacks ===
Connectors on : yes
Fallbacks on : yes
DFT flavor :
FFT flavor : fftw3
LINALG flavor : mkl+scalapack
MATH flavor : none
TIMER flavor : abinit
TRIO flavor :
=== Experimental features ===
Bindings : no
Exports : no
GW double-precision : no