convergence problem in vasp.4.6.35

[deltz]

Dear All,

I compiled vasp.4.6.35 by change the makefile.linux_ifc_P4 to the end of this thread as i pasted, this makefile works fine in other vasp versions, 4.6.31 for example, there's no error during the compilation, but when i try to test the efficiency and validity by running the same job which have done by 4.6.31, i found the electronic steps in the new version is hard to converge, for example, the new version does not converge after 400 steps which is the NELM, but for vasp.4.6.31, it converge soon after about 40 steps. And i have tried AMIX, BMIX, and LMAXMIX, the situation remains. There maybe something wrong with my compilation, since i have not see any big changes between 4.6.31 and 4.6.35. Any suggestions? Thanks.




#########
.SUFFIXES: .inc .f .f90 .F
# all CPP processed fortran files have the extension .f90
SUFFIX=.f90

#-----------------------------------------------------------------------
# fortran compiler and linker
#-----------------------------------------------------------------------
#FC=fort
# fortran linker
#FCL=$(FC)
CPP_ = ./preprocess <$*.F | /usr/bin/cpp -P -C -traditional >$*$(SUFFIX)
FFLAGS = -FR -lowercase -assume byterecl
OFLAG=-O3 -xW -tpp7

OFLAG_HIGH = $(OFLAG)
OBJ_HIGH =

OBJ_NOOPT =
DEBUG = -FR -O0
INLINE = $(OFLAG)
BLAS=-L/opt/intel/mkl/10.0.1.014/lib/em64t -lguide -lpthread
LAPACK= -lmkl_lapack -lmkl
LIB = -I/opt/intel/mkl/10.0.1.014/include/ -L../vasp.4.lib -ldmy \
../vasp.4.lib/linpack_double.o $(BLAS) \
$(LAPACK)
LINK =
FC=mpif90
FCL=$(FC)

#-----------------------------------------------------------------------
# additional options for CPP in parallel version (see also above):
# NGZhalf charge density reduced in Z direction
# wNGZhalf gamma point only reduced in Z direction
# scaLAPACK use scaLAPACK (usually slower on 100 Mbit Net)
#-----------------------------------------------------------------------

CPP = $(CPP_) -DMPI -DHOST=\"LinuxIFC\" -DIFC \
-Dkind8 -DNGZhalf -DCACHE_SIZE=4000 -DPGF90 -Davoidalloc \
-DMPI_BLOCK=500 \
## -DRPROMU_DGEMV -DRACCMU_DGEMV

#-----------------------------------------------------------------------
# location of SCALAPACK
# if you do not use SCALAPACK simply uncomment the line SCA
#-----------------------------------------------------------------------

BLACS=$(HOME)/archives/SCALAPACK/BLACS/
SCA_=$(HOME)/archives/SCALAPACK/SCALAPACK

SCA= $(SCA_)/libscalapack.a \
$(BLACS)/LIB/blacsF77init_MPI-LINUX-0.a $(BLACS)/LIB/blacs_MPI-LINUX-0.a $(BLACS)/LIB/blacsF77init_MPI-LINUX-0.a

SCA=

FFT3D = fftmpi.o fftmpi_map.o fft3dlib.o


<span class='smallblacktext'>[ Edited ]</span>

[deltz]

Dear all,
i solved this problem by downscale O3 to O1 in OFLAG, and O2 failed too, maybe it is caused by ifort 10, as in ifort 9 (at least in my case), O2 and O3 works fine.