CP2K compilation errors

Eric Shamay eric.... at gmail.com
Wed May 30 00:37:08 UTC 2007


I've managed to get the entire program to compile. Here are the
pertinent stats of the system:

firstly, my arch file
---------------------------------------------
PERL     = perl
CC       = mpicc
CPP      = cpp
FC       = mpif90 -FR
FC_fixed = mpif90 -FI
LD       = mpif90
AR       = ar -r
MKLPATH  = $(MKL)/lib/64
DFLAGS   = -D__INTEL -D__FFTSG -D__FFTW3 -D__parallel -D__BLACS -
D__SCALAPACK -D__FFTMKL -I$(FFTW)/include -I$(MKL)/include
CPPFLAGS = -C $(DFLAGS) -P -traditional
FCFLAGS  = -O2 -Vaxlib -tpp2
LDFLAGS  = $(FCFLAGS) -L$(HOME)/lib -L$(FFTW)/lib -L$(MPI)/lib -L$
(MKL)/lib/64
#LIBS     = -lmkl_scalapack -lmkl_blacsF77init -lmkl_blacs\
#           -lmkl_lapack -lmkl -lguide -lmpi -lfftw3
LIBS     = $(MKLPATH)/libmkl_blacs.a\
                   $(MKLPATH)/libmkl_scalapack.a\
                   $(MKLPATH)/libmkl_blacsF77init.a\
                   $(MKLPATH)/libmkl_lapack.a\
                   $(MKLPATH)/libmkl_blacs_intelmpi.a\
                   $(MKLPATH)/libguide.a\
                   -lpthread -lmpi -lmkl -lfftw3

OBJECTS_ARCHITECTURE = machine_intel.o
---------------------------------------------

intel/9.1 (for the compilers)

> mpif90 -V
Intel(R) Fortran Itanium(R) Compiler for Itanium(R)-based applications
Version 9.1    Build 20070320 Package ID: <installpackageid>

>mpicc -V
Intel(R) C Itanium(R) Compiler for Itanium(R)-based applications
Version 9.1    Build 20070320 Package ID: <installpackageid>

mkl/7.2cluster_64

fftw/3.1.1

mpi/openmpi-1.1.5_intel-9.1

This setup finishes through the entire compilation and exits without
error. I then attempt to run a test as per the INSTALL file but by
running the command:

> mpirun -np 8 $PATH_TO_CP2KEXE H2O.inp > H2O.out

and the program produces the following output:
--------------------------------------------------------------------------------------------------------------------------------------------
 username at optix has created process number
0     14411
 username at optix has created process number
1     14412
 username at optix has created process number
2     14413
 username at optix has created process number
3     14414
 username at optix has created process number
4     14415
 username at optix has created process number
5     14416
 username at optix has created process number
6     14417
 username at optix has created process number
7     14418

  **** **** ******  **  PROGRAM STARTED AT              Tue May 29
17:26:39 2007
 ***** ** ***  *** **   PROGRAM STARTED
ON                                 optix
 **    ****   ******    PROGRAM STARTED
BY                              username
 ***** **    ** ** **   PROGRAM PROCESS
ID                                 14411
  **** **  *******  **  PROGRAM STARTED IN /mnt/netapp/home/users/
eshamay/tmp/cp
                                           2k/tests/QS

 CP2K| Program compiled at                          Tue May 29
17:12:39 PDT 2007
 CP2K| Program compiled
on                                                 optix
 CP2K| Program compiled for                                     Linux-
ia64-intel
 CP2K| Last CVS entry         ur_center_integrals.F/1.19/Fri May 18
22:07:04 200
 CP2K| Input file
name                                                   H2O.inp

 GLOBAL| Force Environment
number                                              1
 GLOBAL| Basis set file name
BASIS_SET
 GLOBAL| Potential file name
POTENTIAL
 GLOBAL| MM Potential file name
MM_POTENTIAL
 GLOBAL| Restart file
name                                               RESTART
 GLOBAL| Coordinate file
name                                              UNSET
 GLOBAL| Method
name                                                        CP2K
 GLOBAL| Project
name                                                        H2O
 GLOBAL| Default FFT
library                                               FFTW3
 GLOBAL| Run
type                                                             MD
 GLOBAL| All-to-all communication in single
precision                          F
 GLOBAL| Global print
level                                                    1
 GLOBAL| Total number of message passing
processes                             8
 GLOBAL| Number of threads for this
process                                    1
 GLOBAL| This output is from
process                                           0

6 additional processes aborted (not shown)
--------------------------------------------------------------------------------------------------------------------------------------------

This is followed by a series of mpi segfault/crash error that look
like :


17:35:12  eshamay at optix (~/tmp/cp2k/tests/QS) Signal:11 info.si_errno:
0(Success) si_code:2(SEGV_ACCERR)
Failing at addr:0xf3
[0] func:/usr/local/packages/openmpi/1.1_intel-9.1/lib/libopal.so.0
[0x200000000044c800]
[1] func:[0xa0000000000107e0]
[2] func:/usr/local/packages/openmpi/1.1_intel-9.1/lib/libmpi.so.0
[0x2000000000182dd0]
[3] func:/usr/local/packages/openmpi/1.1_intel-9.1/lib/libmpi.so.
0(MPI_Comm_size-0x9d3c20) [0x2000000000182c20]
[4] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(Cblacs_pinfo+0x2000000001798240) [0x40000000022eea90]
[5] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(blacs_gridmap_+0x2000000001140bb0) [0x4000000001c97410]
[6] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(blacs_gridinit_+0x20000000011409d0) [0x4000000001c97240]

<snip>

*** End of error message ***
6] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(main+0x1fffffffff523f30) [0x4000000000076d40]
[17] func:/lib/tls/libc.so.6.1(__libc_start_main-0x2bccc0)
[0x2000000000899c50]
[18] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(_start+0x1fffffffff520160) [0x4000000000076a80]
*** End of error message ***
6] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(main+0x1fffffffff523f30) [0x4000000000076d40]
[17] func:/lib/tls/libc.so.6.1(__libc_start_main-0x2bccc0)
[0x2000000000899c50]
[18] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(_start+0x1fffffffff520160) [0x4000000000076a80]
*** End of error message ***
6] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(main+0x1fffffffff523f30) [0x4000000000076d40]
[17] func:/lib/tls/libc.so.6.1(__libc_start_main-0x2bccc0)
[0x2000000000899c50]
[18] func:/home/users/eshamay/tmp/cp2k/exe/Linux-ia64-intel/
cp2k.popt(_start+0x1fffffffff520160) [0x4000000000076a80]
*** End of error message ***
mpirun noticed that job rank 1 with PID 14565 on node "localhost"
exited on signal 11.


So far all the help was right on track and I've finally been able to
get it to compile, but now I'm stuck at this near-immediate crash
after trying to run the program in parallel. Any further help would be
greatly appreciated.

~Eric





On May 28, 12:12 pm, akohlmey <akoh... at gmail.com> wrote:
> On May 28, 2:57 pm, Eric Shamay <eric.... at gmail.com> wrote:
>
> > Thank you for the help with the compiler. I installed a newer version
> > and overcame the previous error. Now I'm stuck with an error involving
> > libmkl_scalapack:
>
> > IPO link: can not find -lmkl_scalapack
> > ifort: error: problem during multi-file optimization compilation (code
> > 1)
>
> > Installing the latest version of Intel's mkl does not include
> > libmkl_scalapack, and so I'm left wondering where I can find that
> > particular library.
>
> eric,
> you have to either use intel's cluster mkl or compile
> BLACS/ScaLAPACK yourself.
>
> you can see my notes on how i compiled both packages for our group at:https://www.liniac.upenn.edu//wiki/tiki-index.php?page=scalapack+for+CMM
>
> you will have to adapt the configuration files for your MPI
> installation.
> when compiling BLACS/ScaLAPACK it is essential to verify your
> compilation
> with the corresponding testers. BLACS should pass completely, in
> ScaLAPACK
> you may get away with a few tests failing (minor). the success of the
> latter
> depends also a lot on the optimizations in the BLAS/LAPACK that is
> used
> with ScaLAPACK. i had the best successes with mkl.
> it is important not to overoptimize.
>
> hope this helps,
>    axel.
>
> > Thanks,
> > ~Eric
>
> > On May 24, 11:54 am, "Nichols A. Romero" <naro... at gmail.com> wrote:
>
> > > 9.1.041 also works for me.
>
> > > On 5/24/07, Teodoro Laino <teodor... at gmail.com> wrote:




More information about the CP2K-user mailing list