[求助] 用MPIF90编译后的Siesta并行出错，串行没事……

ifort (IFORT) 12.1.2 20111128
icc (ICC) 12.1.2 20111128
Copyright (C) 1985-2011 Intel Corporation.  All rights reserved.

.SUFFIXES:
.SUFFIXES: .f .F .o .a .f90 .F90

SIESTA_ARCH=x86_64-unknown-linux-gnu--Intel

FPP=
FPP_OUTPUT=
FC=mpif90
RANLIB=ranlib

SYS=nag

SP_KIND=4
DP_KIND=8
KINDS=$(SP_KIND) $(DP_KIND)

FFLAGS=-O2  
FPPFLAGS= -DMPI -DFC_HAVE_FLUSH -DFC_HAVE_ABORT
LDFLAGS=-static

ARFLAGS_EXTRA=

FCFLAGS_fixed_f=
FCFLAGS_free_f90=
FPPFLAGS_fixed_F=
FPPFLAGS_free_F90=

BLAS_LIBS=-L/opt/intel/mkl/lib/intel64/ -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lmkl_blas95_lp64
LAPACK_LIBS=-lmkl_lapack95_lp64
BLACS_LIBS=-lmkl_blacs_lp64
SCALAPACK_LIBS=-lmkl_scalapack_lp64

COMP_LIBS=

NETCDF_LIBS=
NETCDF_INTERFACE=

LIBS=$(SCALAPACK_LIBS) $(BLACS_LIBS) $(LAPACK_LIBS) $(BLAS_LIBS) $(NETCDF_LIBS)

#SIESTA needs an F90 interface to MPI
#This will give you SIESTA's own implementation
#If your compiler vendor offers an alternative, you may change
#to it here.
MPI_INTERFACE=libmpi_f90.a
MPI_INCLUDE=/home/dorm/ProgramFiles/MPICH/include

#Dependency rules are created by autoconf according to whether
#discrete preprocessing is necessary or not.
.F.o:
        $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_fixed_F)  $<
.F90.o:
        $(FC) -c $(FFLAGS) $(INCFLAGS) $(FPPFLAGS) $(FPPFLAGS_free_F90) $<
.f.o:
        $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_fixed_f)  $<
.f90.o:
        $(FC) -c $(FFLAGS) $(INCFLAGS) $(FCFLAGS_free_f90)  $<

* Maximum dynamic memory allocated =    13 MB
Fatal error in PMPI_Comm_size: Invalid communicator, error stack:
PMPI_Comm_size(111): MPI_Comm_size(comm=0x5b, size=0xc20c2c) failed
PMPI_Comm_size(69).: Invalid communicator
Fatal error in PMPI_Comm_size: Invalid communicator, error stack:
PMPI_Comm_size(111): MPI_Comm_size(comm=0x5b, size=0xc20c2c) failed
PMPI_Comm_size(69).: Invalid communicator