VASP 4.6 segmentation fault with PGI compilers

[bproberts]

Hi,

I'm trying to compile VASP 4.6 on a Linux cluster using the PGI compilers. Unfortunately, it fails in either serial or parallel with a segmentation fault.

GDB reports the problem as follows:

((snip))
POSCAR, INCAR and KPOINTS ok, starting setup
WARNING: wrap around errors must be expected
FFT: planning ... 4
reading WAVECAR
prediction of wavefunctions initialized
entering main loop
N E dE d eps ncg rms rms(c)

Program received signal SIGSEGV, Segmentation fault.
0x0000000000663e90 in ATL_zdotc_xp0yp0aXbX ()
Missing separate debuginfos, use: debuginfo-install glibc-2.12-1.25.el6.x86_64
(gdb) bt
#0 0x0000000000663e90 in ATL_zdotc_xp0yp0aXbX ()
#1 0x000000000065f8fa in zdotc_ ()
#2 0x00000000005d1975 in zhetd2 (uplo='U', n=8, a=..., lda=40, d=..., e=..., tau=..., info=0) at ./lapack_atlas.f:3257
#3 0x00000000005eee8f in zhetrd (uplo='U', n=8, a=..., lda=40, d=..., e=..., tau=..., work=..., lwork=1528, info=0) at ./lapack_atlas.f:8670
#4 0x0000000000614247 in zheev (jobz='V', uplo='U', n=8, a=..., lda=40, w=..., work=..., lwork=1536, rwork=..., info=0) at ./lapack_atlas.f:15094
#5 0x0000000000649798 in zhegv (itype=1, jobz='V', uplo='U', n=8, a=..., lda=40, b=..., ldb=40, w=..., work=..., lwork=1536, rwork=..., info=0)
at ./lapack_atlas.f:22145
#6 0x00000000005666c5 in eddav () at ./davidson.f:577
#7 0x000000000056ae61 in elmin () at ./electron.f:359
#8 0x000000000041d1e1 in vamp () at ./main.f:2226
#9 0x00000000004086f4 in main ()
#10 0x00002aaaab395c9d in __libc_start_main () from /lib64/libc.so.6
#11 0x00000000004085e9 in _start ()
(gdb)

This is one of several Makefiles I've tried to use:

======== BEGIN MAKEFILE ========


.SUFFIXES: .inc .f .f90 .F
#-----------------------------------------------------------------------
# Makefile for Portland Group F90/HPF compiler release 3.0-1, 3.1
# and release 1.7
# (http://www.pgroup.com/ & ftp://ftp.pgroup.com/x86/, you need
# to order the HPF/F90 suite)
# we have found no noticable performance differences between
# any of the releases, even Athlon or PIII optimisation does
# not seem to improve performance
#
# The makefile was tested only under Linux on Intel platforms
# (Suse X,X)
#
# it might be required to change some of library pathes, since
# LINUX installation vary a lot
# Hence check ***ALL**** options in this makefile very carefully
#-----------------------------------------------------------------------
#
# Mind that some Linux distributions (Suse 6.1) have a bug in
# libm causing small errors in the error-function (total energy
# is therefore wrong by about 1meV/atom). The recommended
# solution is to update libc.
#
# Mind that some Linux distributions (Suse 6.1) have a bug in
# libm causing small errors in the error-function (total energy
# is therefore wrong by about 1meV/atom). The recommended
# solution is to update libc.
#
# BLAS must be installed on the machine
# there are several options:
# 1) very slow but works:
# retrieve the lapackage from ftp.netlib.org
# and compile the blas routines (BLAS/SRC directory)
# please use g77 or f77 for the compilation. When I tried to
# use pgf77 or pgf90 for BLAS, VASP hang up when calling
# ZHEEV (however this was with lapack 1.1 now I use lapack 2.0)
# 2) most desirable: get an optimized BLAS
# for a list of optimized BLAS try
# http://www.kachinatech.com/~hjjou/scilib/opt_blas.html
#
# the two most reliable packages around are presently:
# 3a) Intels own optimised BLAS (PIII, P4, Itanium)
# http://developer.intel.com/software/products/mkl/
# this is really excellent when you use Intel CPU's
#
# 3b) or obtain the atlas based BLAS routines
# http://math-atlas.sourceforge.net/
# you certainly need atlas on the Athlon, since the mkl
# routines are not optimal on the Athlon.
#
#-----------------------------------------------------------------------

# all CPP processed fortran files have the extension .f
SUFFIX=.f

#-----------------------------------------------------------------------
# fortran compiler and linker
#-----------------------------------------------------------------------
FC=pgf90
# fortran linker
FCL=$(FC)


#-----------------------------------------------------------------------
# whereis CPP ?? (I need CPP, can't use gcc with proper options)
# that's the location of gcc for SUSE 5.3
#
# CPP_ = /usr/lib/gcc-lib/i486-linux/2.7.2/cpp -P -C
#
# that's probably the right line for some Red Hat distribution:
#
# CPP_ = /usr/lib/gcc-lib/i386-redhat-linux/2.7.2.3/cpp -P -C
#
# SUSE 6.X, maybe some Red Hat distributions:

CPP_ = ./preprocess <$*.F | /usr/bin/cpp -P -C -traditional >$*$(SUFFIX)

#-----------------------------------------------------------------------
# possible options for CPP:
# possible options for CPP:
# NGXhalf charge density reduced in X direction
# wNGXhalf gamma point only reduced in X direction
# avoidalloc avoid ALLOCATE if possible
# IFC work around some IFC bugs
# CACHE_SIZE 1000 for PII,PIII, 5000 for Athlon, 8000 P4
# RPROMU_DGEMV use DGEMV instead of DGEMM in RPRO (usually faster)
# RACCMU_DGEMV use DGEMV instead of DGEMM in RACC (faster on P4)
# **** definitely use -DRACCMU_DGEMV if you use the mkl library
#-----------------------------------------------------------------------

CPP = $(CPP_) -DHOST=\"LinuxPgi\" \
-Dkind8 -DNGXhalf -DCACHE_SIZE=2000 -DPGF90 -Davoidalloc \
-DRPROMU_DGEMV

#-----------------------------------------------------------------------
# general fortran flags (there must a trailing blank on this line)
# the -Mx,119,0x200000 is required if you use older pgf90 versions
# on a more recent LINUX installation
# the option will not do any harm on other 3.X pgf90 distributions
#-----------------------------------------------------------------------

FFLAGS = -Mfree -Mx,119,0x200000

#-----------------------------------------------------------------------
# optimization,
# we have tested whether higher optimisation improves
# the performance, and found no improvements with -O3-5 or -fast
# (even on Athlon system, Athlon specific optimistation worsens performance)
#-----------------------------------------------------------------------

OFLAG = -O2 -tp=nehalem

OFLAG_HIGH = $(OFLAG)
OBJ_HIGH =
OBJ_NOOPT =
DEBUG = -g -O0
INLINE = $(OFLAG)


#-----------------------------------------------------------------------
# the following lines specify the position of BLAS and LAPACK
# what you chose is very system dependent
# P4: VASP works fastest with Intels mkl performance library
# Athlon: Atlas based BLAS are presently the fastest
# P3: no clue
#-----------------------------------------------------------------------

# Atlas based libraries
ATLASHOME= /share/apps/lib/atlas-3.9.78/lib
BLAS= -L$(ATLASHOME) -lf77blas -latlas

# use specific libraries (default library path points to other libraries)
#BLAS= $(ATLASHOME)/libf77blas.a $(ATLASHOME)/libatlas.a

# use the mkl Intel libraries for p4 (www.intel.com)
#INTEL_MKL_LIBDIR=/share/apps/intel/2011/mkl/lib/intel64
#BLAS=-L$(INTEL_MKL_LIBDIR) -lmkl_blas95_lp64 -lpthread

# LAPACK, simplest use vasp.4.lib/lapack_double
#LAPACK= ../vasp.4.lib/lapack_double.o

# use atlas optimized part of lapack
LAPACK= ../vasp.4.lib/lapack_atlas.o -llapack -lcblas

# use the mkl Intel lapack
#LAPACK= -L$(INTEL_MKL_LIBDIR) -lmkl_lapack95_lp64
#INTEL_MKL_LIBS=-Wl,--start-group $(INTEL_MKL_LIBDIR)/libmkl_intel_ilp64.a \
# $(INTEL_MKL_LIBDIR)/libmkl_sequential.a \
# $(INTEL_MKL_LIBDIR)/libmkl_core.a \
# -Wl,--end-group -lpthread -lm


#-----------------------------------------------------------------------

LIB = -L../vasp.4.lib -ldmy \
../vasp.4.lib/linpack_double.o $(LAPACK) \
$(BLAS)

# options for linking (none required)
LINK =

#-----------------------------------------------------------------------
# fft libraries:
# VASP.4.5 can use FFTW (http://www.fftw.org)
# since the FFTW is very slow for radices 2^n the fft3dlib is used
# in these cases
# if you use fftw3d you need to insert -lfftw in the LIB line as well
# please do not send us any querries reltated to FFTW (no support)
# if it fails, use fft3dlib
#-----------------------------------------------------------------------

FFT3D = fft3dfurth.o fft3dlib.o
#FFT3D = fftw3d+furth.o fft3dlib.o


#=======================================================================
# MPI section, uncomment the following lines
#
# one comment for users of mpich or lam:
# You must *not* compile mpi with g77/f77, because f77/g77
# appends *two* underscores to symbols that contain already an
# underscore (i.e. MPI_SEND becomes mpi_send__). The pgf90
# compiler however appends only one underscore.
# Precompiled mpi version will also not work !!!
#
# We found that mpich.1.2.1 and lam-6.5.X are stable
# mpich.1.2.1 was configured with
# ./configure -prefix=/usr/local/mpich_nodvdbg -fc="pgf77 -Mx,119,0x200000" \
# -f90="pgf90 -Mx,119,0x200000" \
# --without-romio --without-mpe -opt=-O \
#
# lam was configured with the line
# ./configure -prefix /usr/local/lam-6.5.X --with-cflags=-O -with-fc=pgf90 \
# --with-f77flags=-O --without-romio
#
# lam was generally faster and we found an average communication
# band with of roughly 160 MBit/s (full duplex)
#
# please note that you might be able to use a lam or mpich version
# compiled with f77/g77, but then you need to add the following
# options: -Msecond_underscore (compilation) and -g77libs (linking)
#
# !!! Please do not send me any queries on how to install MPI, I will
# certainly not answer them !!!!
#=======================================================================
#-----------------------------------------------------------------------
# fortran linker for mpi: if you use LAM and compiled it with the options
# suggested above, you can use the following lines
#-----------------------------------------------------------------------


#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=\"LinuxPgi\" \
# -Dkind8 -DNGZhalf -DCACHE_SIZE=2000 -DPGF90 -Davoidalloc -DRPROMU_DGEMV

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

BLACS=/usr/local/BLACS_lam
SCA_= /usr/local/SCALAPACK_lam

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

SCA=

#-----------------------------------------------------------------------
# libraries for mpi
#-----------------------------------------------------------------------

#LIB = -L../vasp.4.lib -ldmy \
# ../vasp.4.lib/linpack_double.o $(LAPACK) \
# $(SCA) $(BLAS)

# FFT: only option fftmpi.o with fft3dlib of Juergen Furthmueller

#FFT3D = fftmpi.o fftmpi_map.o fft3dlib.o

#-----------------------------------------------------------------------
# general rules and compile lines
#-----------------------------------------------------------------------
BASIC= symmetry.o symlib.o lattlib.o random.o

SOURCE= base.o mpi.o smart_allocate.o xml.o \
constant.o jacobi.o main_mpi.o scala.o \
asa.o lattice.o poscar.o ini.o setex.o radial.o \
pseudo.o mgrid.o mkpoints.o wave.o wave_mpi.o $(BASIC) \
nonl.o nonlr.o dfast.o choleski2.o \
mix.o charge.o xcgrad.o xcspin.o potex1.o potex2.o \
metagga.o constrmag.o pot.o cl_shift.o force.o dos.o elf.o \
tet.o hamil.o steep.o \
chain.o dyna.o relativistic.o LDApU.o sphpro.o paw.o us.o \
ebs.o wavpre.o wavpre_noio.o broyden.o \
dynbr.o rmm-diis.o reader.o writer.o tutor.o xml_writer.o \
brent.o stufak.o fileio.o opergrid.o stepver.o \
dipol.o xclib.o chgloc.o subrot.o optreal.o davidson.o \
edtest.o electron.o shm.o pardens.o paircorrection.o \
optics.o constr_cell_relax.o stm.o finite_diff.o \
elpol.o setlocalpp.o aedens.o

INC=

vasp: $(SOURCE) $(FFT3D) $(INC) main.o
rm -f vasp.pgi.serial
$(FCL) -o vasp.pgi.serial $(LINK) main.o $(SOURCE) $(FFT3D) $(LIB)
makeparam: $(SOURCE) $(FFT3D) makeparam.o main.F $(INC)
$(FCL) -o makeparam $(LINK) makeparam.o $(SOURCE) $(FFT3D) $(LIB)
zgemmtest: zgemmtest.o base.o random.o $(INC)
$(FCL) -o zgemmtest $(LINK) zgemmtest.o random.o base.o $(LIB)
dgemmtest: dgemmtest.o base.o random.o $(INC)
$(FCL) -o dgemmtest $(LINK) dgemmtest.o random.o base.o $(LIB)
ffttest: base.o smart_allocate.o mpi.o mgrid.o random.o ffttest.o $(FFT3D) $(INC)
$(FCL) -o ffttest $(LINK) ffttest.o mpi.o mgrid.o random.o smart_allocate.o base.o $(FFT3D) $(LIB)
kpoints: $(SOURCE) $(FFT3D) makekpoints.o main.F $(INC)
$(FCL) -o kpoints $(LINK) makekpoints.o $(SOURCE) $(FFT3D) $(LIB)

clean:
-rm -f *.g *.f *.o *.L *.mod ; touch *.F

main.o: main$(SUFFIX)
$(FC) $(FFLAGS) $(DEBUG) $(INCS) -c main$(SUFFIX)
xcgrad.o: xcgrad$(SUFFIX)
$(FC) $(FFLAGS) $(INLINE) $(INCS) -c xcgrad$(SUFFIX)
xcspin.o: xcspin$(SUFFIX)
$(FC) $(FFLAGS) $(INLINE) $(INCS) -c xcspin$(SUFFIX)

makeparam.o: makeparam$(SUFFIX)
$(FC) $(FFLAGS) $(DEBUG) $(INCS) -c makeparam$(SUFFIX)

makeparam$(SUFFIX): makeparam.F main.F
#
# MIND: I do not have a full dependency list for the include
# and MODULES: here are only the minimal basic dependencies
# if one strucuture is changed then touch_dep must be called
# with the corresponding name of the structure
#
base.o: base.inc base.F
mgrid.o: mgrid.inc mgrid.F
constant.o: constant.inc constant.F
lattice.o: lattice.inc lattice.F
setex.o: setexm.inc setex.F
pseudo.o: pseudo.inc pseudo.F
poscar.o: poscar.inc poscar.F
mkpoints.o: mkpoints.inc mkpoints.F
wave.o: wave.inc wave.F
nonl.o: nonl.inc nonl.F
nonlr.o: nonlr.inc nonlr.F

$(OBJ_HIGH):
$(CPP)
$(FC) $(FFLAGS) $(OFLAG_HIGH) $(INCS) -c $*$(SUFFIX)
$(OBJ_NOOPT):
$(CPP)
$(FC) $(FFLAGS) $(INCS) -c $*$(SUFFIX)

fft3dlib_f77.o: fft3dlib_f77.F
$(CPP)
$(F77) $(FFLAGS_F77) -c $*$(SUFFIX)

.F.o:
$(CPP)
$(FC) $(FFLAGS) $(OFLAG) $(INCS) -c $*$(SUFFIX)
.F$(SUFFIX):
$(CPP)
$(SUFFIX).o:
$(FC) $(FFLAGS) $(OFLAG) $(INCS) -c $*$(SUFFIX)

# these special rules are cummulative (that is once failed
# in one compiler version, stays in the list forever)
# -tpp5|6|7 P, PII-PIII, PIV
# -xW use SIMD (does not pay of on PII, since fft3d uses double prec)
# all other options do no affect the code performance since -O1 is used
#-----------------------------------------------------------------------

#xcgrad.o : xcgrad.F
# $(CPP)
# $(FC) -Mfree -Mnoupcase -O1 -c $*$(SUFFIX)

======== END MAKEFILE ========


Things I have tried to fix the problem include:

* Compiling "davidson.o" with reduced optimisation, i.e., -O1
* Compiling the whole thing (VASP executables and libraries) at -O0 with debug symbols
* Compiling with VASP lapack
* Compiling with the Intel MKL libraries (see commented out sections of the Makefile) - this produces a segfault at a different point

Does anyone have any other ideas I could try?

I'm using the PGI compilers version 12.5, though I've also used 12.4 which fails the same way. The cluster is running Red Hat Enterprise Linux 6.1. I have managed to compile VASP 4.6 successfully using the GNU compilers, after making some changes to a Makefile based on makefile.linux_gfortran.

[bproberts]

I should add that I started out trying ATLAS 3.9.69, and when that failed I tried ATLAS 3.9.78, which also produced this error.

[noel]

I have almost the same problem using openblas. I get a segmentation fault caused by zdotc, I tried mkl and get a segmentation fault in a different place in idamax called from dgetrf. Any help is appreciated.