[CP2K:4034] Re: Segmentation fault problem

Alin M Elena alinm... at gmail.com
Sat Sep 15 22:34:02 UTC 2012


Hi Zhiping,

Here is a Makefile content that I run without problems

CC       = icc
CPP      =
FC       = mpif90 
LD       = mpif90
AR       = ar -r
DFLAGS   = -D__INTEL -D__FFTSG -D__parallel -D__BLACS -D__SCALAPACK -D__FFTMKL 
CPPFLAGS =
FCFLAGS  = $(DFLAGS)  -O3 -xSSE4.2 -heap-arrays  -fpp -free
FCFLAGS2 = $(DFLAGS)  -O1  -heap-arrays  -fpp -free
LDFLAGS  = $(FCFLAGS) -L${MKLROOT}/lib/intel64
LIBS     = -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_sequential -lmkl_core -
lmkl_blacs_intelmpi_lp64 

OBJECTS_ARCHITECTURE = machine_intel.o


graphcon.o: graphcon.F
        $(FC) -c $(FCFLAGS2) $<


you may need to add the libint part.... is a good idea to use the fftw that 
comes with mkl... Also check that your nodes really support SSE4.2

Alin
On Saturday 15 Sep 2012 15:00:22 Zhiping wrote:

I solved this by change '-O3' into '-O2' in FCFLAGS.
it works now for QS benchmarks with 256 water molecules.
however, it still fails with the H2O-hybrid-b3lyp and some other hybrid tests, 
with error information 'Segmentation fault'
any suggestions? thank you.
Zhiping
On Sunday, September 16, 2012 4:02:35 AM UTC+8, Zhiping wrote:
Dear all:
I compile cp2k using intel compiler and MKL 11.1.069. Parallel compilation is 
done with intel mpi 4.0.2.003.
While no error is reported during the compilation and I obtain the binaries. I 
found that serial executable works but the parallel one fails with 
'Segmentation fault (174)' on the test runs.
I follow the threads in several forums and tries several tricks such as 
'ulimit -s unlimited', but still cannot get it working. Could anybody here 
give some suggestions?
Thank you.
Best,
Zhiping
--
App. 1: OUTPUT of uname 0a
Linux ln1 2.6.18-194.17.1.0.1.el5_lustre.1.8.5 #1 SMP Wed Aug 3 18:09:09 CST 
2011 x86_64 x86_64 x86_64 GNU/Linux



App. 2: FILE: Linux-x86_64-intel.popt
#
INTEL_INC=/apps/intel/Compiler/11.1/069/c/mkl/include/fftw 
INTEL_LIB=/apps/intel/Compiler/11.1/069/c/mkl/lib/em64t 
FFTW3_INC=/home/user/bin/lib/fftw3/include
FFTW3_LIB=/home/user/bin/lib/fftw3/lib
INT_INC=/home/user/bin/lib/libint/include
INT_LIB=/home/user/bin/lib/libint/lib


CC       = cc
CPP      = 
FC       = mpiifort
LD       = mpiifort
AR       = ar -r
DFLAGS   = -D__INTEL -D__FFTSG -D__parallel -D__BLACS -D__SCALAPACK -D__FFTW3 
-D__LIBINT -D__HAS_NO_ISO_C_BINDING
CPPFLAGS = -C -traditional $(DFLAGS) -I$(INTEL_INC) -I$(INT_INC) -
I$(FFTW3_INC)


FCFLAGS  = $(DFLAGS) -I$(INTEL_INC) -I$(INT_INC) -I$(FFTW3_INC) -O3 -xW -heap-
arrays 64 -funroll-loops -fpp -free
FCFLAGS2 = $(DFLAGS) -I$(INTEL_INC) -I$(INT_INC) -I$(FFTW3_INC) -O1 -xW -heap-
arrays 64 -fpp -free
LDFLAGS  = $(FCFLAGS) -I$(INTEL_LIB) -L$(INT_LIB) -L$(FFTW3_LIB)
LIBS     = $(LDFLAGS) -lmkl_scalapack_lp64 -lmkl_intel_lp64 -
lmkl_blacs_intelmpi_lp64 -lmkl_lapack95_lp64 -lmkl_sequential -lmkl_core -
lguide -lfftw3 /home/user/bin/lib/libint/lib/libint_cpp_wrapper.o -lderiv -lint 
-lr12 -lstdc++


OBJECTS_ARCHITECTURE = machine_intel.o


graphcon.o: graphcon.F
        $(FC) -c $(FCFLAGS2) $< 
-- 
You received this message because you are subscribed to the Google Groups 
"cp2k" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/cp2k/-/J6EH6lKxC4QJ.
To post to this group, send email to cp... at googlegroups.com.
To unsubscribe from this group, send email to 
cp2k+uns... at googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/cp2k?hl=en.




-- 
Without Questions there are no Answers!
______________________________________________________________________
Alin Marin ELENA
Advanced Molecular Simulation Research Laboratory
School of Physics, University College Dublin
----
Ardionsamblú Móilíneach Saotharlann Taighde
Scoil na Fisice, An Coláiste Ollscoile, Baile Átha Cliath

-----------------------------------------------------------------------------------
http://alin.elenaworld.net
______________________________________________________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20120915/713b35cb/attachment.htm>


More information about the CP2K-user mailing list