[CP2K:3381] Re: segmentation fault

[Peter Mamonov]

Hi Nikita!

Correct me if I'm wrong, but as i can see you are trying to build CP2K
for SKIF `Chebyshev`. If so, you can try this (working) arch file for
Chebyshev (see below). Also be sure to switch to intel's compiler
using command:

mpi-selector --set intel_mpi_intel64-4.0.0.025

and run your task with `-as intel` option:

mpirun -as intel -blah -blah blah

Best regards,
Peter
--

# Chebyshev ARCH file

MKL_ROOT  = /opt/intel/mkl/10.2.4.032
MKL_LIB = $(MKL_ROOT)/lib/em64t
MKL_INCLUDE = $(MKL_ROOT)/include

CC       = cc
CPP      =
FC       = mpif90
LD       = mpif90
AR       = ar -r

DFLAGS   = -D__INTEL -D__FFTSG -D__FFTMKL -D__parallel -D__BLACS
-D__SCALAPACK -D__HAS_NO_ISO_C_BINDING

CPPFLAGS = -traditional -C $(DFLAGS) -I$(MKL_INCLUDE)/include

FCFLAGS  = -fc=ifort $(DFLAGS) -I$(MKL_INCLUDE) -O2 -xHost
-heap-arrays 64 -fpp -free

LDFLAGS  = $(FCFLAGS) -L$(MKL_LIB)

LIBS     = -Wl,--start-group \
$(MKL_LIB)/libmkl_scalapack_lp64.a \
$(MKL_LIB)/libmkl_blacs_intelmpi_lp64.a \
$(MKL_LIB)/libmkl_intel_lp64.a \
$(MKL_LIB)/libmkl_sequential.a \
$(MKL_LIB)/libmkl_core.a \
-static-mpi \
-Wl,--end-group

OBJECTS_ARCHITECTURE = machine_intel.o

# End of Chebyshev ARCH file


On Fri, Jul 22, 2011 at 2:56 PM, Jörg Saßmannshausen
<j.sassma... at ucl.ac.uk> wrote:
> Dear Nikita,
>
> did you use that before you started a run as well?
>
> Unfortunately, it is some time that I have compiled cp2k so I don't know on
> top of my head. The limit of the stack-size is usually the culprit here.
>
> All the best from a sunny London
>
> Jörg
>
> On Friday 22 July 2011 11:06:51 Nikita wrote:
>> Hi Jörg,
>>
>> thanks for your suggestion! But I had already tried this before
>> compilation, but the error occured all the same.
>> Maybe there are some other tricks to overcome this obstacle?
>>
>> Thank you in advance,
>> Nikita
>>
>> On Jul 22, 1:36 pm, Jörg Saßmannshausen <j.sassma... at ucl.ac.uk>
>>
>> wrote:
>> > Hi Nikita,
>> >
>> > the intel compiler is notorious for doing segfaults.
>> >
>> > Have you tried:
>> >
>> > $ ulimit -s unlimited
>> >
>> > That might cure the problem.
>> >
>> > Regards
>> >
>> > Jörg
>> >
>> > On Friday 22 July 2011 10:13:53 Nikita wrote:
>> > > Dear CP2K users and developers,
>> > >
>> > > I have recently compiled CP2K on our university cluster. But the
>> > > calculation failed with the following message:
>> > > ********************************************
>> > > --------------------------
>> > >  OPTIMIZATION STEP:      2
>> > >  --------------------------
>> > > forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> > > Image              PC                Routine            Line
>> > > Source
>> > > cp2k.popt          0000000001195583  Unknown               Unknown
>> > > Unknown
>> > > cp2k.popt          000000000104A51A  Unknown               Unknown
>> > > Unknown
>> > > cp2k.popt          00000000004CEDA1  Unknown               Unknown
>> > > Unknown
>> > > cp2k.popt          000000000040E631  Unknown               Unknown
>> > > Unknown
>> > > cp2k.popt          00000000004096BE  Unknown               Unknown
>> > > Unknown
>> > > cp2k.popt          000000000040868C  Unknown               Unknown
>> > > Unknown
>> > > libc.so.6          00007FED46734CF4  Unknown               Unknown
>> > > Unknown
>> > > cp2k.popt          0000000000408599  Unknown               Unknown
>> > > Unknown
>> > > ************************************************
>> > >
>> > > Could you suggest any solutions to my problem? By the way, here is my
>> > > arch file:
>> > >
>> > > **********************************
>> > > INTEL_MKL = /opt/intel/mkl/10.2.4.032
>> > > INTEL_INC = $(INTEL_MKL)/include/fftw
>> > > INTEL_LIB = $(INTEL_MKL)/lib/em64t
>> > >
>> > > CC       = mpicc
>> > > CPP      =
>> > > FC =  mpif90
>> > > LD = mpif90
>> > > AR       = /usr/bin/ar -r
>> > > DFLAGS   = -D__INTEL -D__FFTSG -D__parallel -D__SCALAPACK -D__BLACS -
>> > > D__FFTW3 -D__LIBINT  -D__HAS_NO_ISO_C_BINDING
>> > > CPPFLAGS = -C -traditional $(DFLAGS) -I$(INTEL_INC)
>> > > FCFLAGS  = $(DFLAGS) -I$(INTEL_INC) \
>> > >            -O2 -xW -funroll-loops -fpp -free -heap-arrays 64
>> > > LDFLAGS  = $(FCFLAGS) -I$(INTEL_INC) -i-static
>> > > LIBS     = -L$(INTEL_LIB) -lmkl_scalapack_lp64 -
>> > > lmkl_blacs_intelmpi_lp64 \
>> > >            -lmkl_intel_lp64 -lmkl_sequential -lmkl_core \
>> > >            /home/vakula/CP2K_NEW/fftw3_compiled/lib/libfftw3.a \
>> > >            /home/vakula/CP2K_NEW/cp2k/tools/hfx_tools/libint_tools/
>> > > libint_cpp_wrapper.o \
>> > >            /home/vakula/CP2K_NEW/libint_compiled/lib/libderiv.a \
>> > >            /home/vakula/CP2K_NEW/libint_compiled/lib/libint.a \
>> > >            -lstdc++
>> > >
>> > > OBJECTS_ARCHITECTURE = machine_intel.o
>> > > ******************************
>> > >
>> > > Thank you in advance,
>> > > Nikita Vakula
>> > > PhD student, Moscow State University
>> >
>> > --
>> > *************************************************************
>> > Jörg Saßmannshausen
>> > University College London
>> > Department of Chemistry
>> > Gordon Street
>> > London
>> > WC1H 0AJ
>> >
>> > email: j.sassma... at ucl.ac.uk
>> > web:http://sassy.formativ.net
>> >
>> > Please avoid sending me Word or PowerPoint attachments.
>> > Seehttp://www.gnu.org/philosophy/no-word-attachments.html
>
> --
> *************************************************************
> Jörg Saßmannshausen
> University College London
> Department of Chemistry
> Gordon Street
> London
> WC1H 0AJ
>
> email: j.sassma... at ucl.ac.uk
> web: http://sassy.formativ.net
>
> Please avoid sending me Word or PowerPoint attachments.
> See http://www.gnu.org/philosophy/no-word-attachments.html
>
> --
> You received this message because you are subscribed to the Google Groups "cp2k" group.
> 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.
>
>