Page 2 of 2

Re: VASP 6.2 compilation with OpenACC GPU port

Posted: Sun Aug 08, 2021 8:53 pm
by henrique_miranda
Could you indicate from where you took the "makefile.include" you are using from?
It seems to me that something is wrong with it.
You wrote that you want an ACC+OPENMP compilation but the -D_OPENMP preprocessor option is not set.

Have a look a the makefile we have on our wiki:
wiki/index.php/Makefile.include_nv_acc%2Bomp%2Bmkl

Re: VASP 6.2 compilation with OpenACC GPU port

Posted: Mon Aug 09, 2021 5:12 pm
by david_keller
This problem has been resolved.
Thanks,
Dave

Re: VASP 6.2 compilation with OpenACC GPU port

Posted: Tue Aug 31, 2021 1:56 pm
by ctetsass
I got it from the vasp wiki web site.


I still not able to compile vasp and here is the error.

I have enclosed the "makefile.include" let me know if something is wrong

Code: Select all

pif90 -acc -gpu=cc60,cc70,cc75,cc80,cuda11.4 -mp -Mfree -Mbackslash -Mlarge_arrays -fast -I/opt/nvidia/hpc_sdk/Linux_x86_64/21.7/compilers/extras/qd/include/qd   -I/usr/include   -c tutor.f90
NVFORTRAN-F-0004-Unable to open MODULE file base.mod (string.F: 6)
NVFORTRAN/x86-64 Linux 21.7-0: compilation aborted
make[2]: *** [makefile:181: string.o] Error 2
make[2]: *** Waiting for unfinished jobs....
NVFORTRAN-F-0004-Unable to open MODULE file string.mod (tutor.F: 202)
NVFORTRAN/x86-64 Linux 21.7-0: compilation aborted
make[2]: *** [makefile:181: tutor.o] Error 2
make[2]: Leaving directory '/home/ctetsass/Programs/vasp/6.2.1/build/gpu'
cp: cannot stat 'vasp': No such file or directory
make[1]: *** [makefile:146: all] Error 1
make[1]: Leaving directory '/home/ctetsass/Programs/vasp/6.2.1/build/gpu'
make: *** [makefile:17: gpu] Error 2

Code: Select all

# Precompiler options
CPP_OPTIONS= -DHOST=\"LinuxPGI\" \
             -DMPI -DMPI_BLOCK=8000 -DMPI_INPLACE -Duse_collective \
             -DscaLAPACK \
             -DCACHE_SIZE=4000 \
             -Davoidalloc \
             -Dvasp6 \
             -Duse_bse_te \
             -Dtbdyn \
             -Dqd_emulate \
             -Dfock_dblbuf \
             -D_OPENACC \
             -DUSENCCL -DUSENCCLP2P

CPP        = nvfortran -Mpreprocess -Mfree -Mextend -E $(CPP_OPTIONS) $*$(FUFFIX)  > $*$(SUFFIX)

FC         = mpif90 -acc -gpu=cc60,cc70,cc75,cc80,cuda11.4 -mp
FCL        = mpif90 -acc -gpu=cc60,cc70,cc75,cc80,cuda11.4 -c++libs -mp


FREE       = -Mfree

FFLAGS     = -Mbackslash -Mlarge_arrays

OFLAG      = -fast

DEBUG      = -Mfree -O0 -traceback

# Specify your NV HPC-SDK installation, try to set NVROOT automatically
#NVROOT     =$(shell which nvfortran | awk -F /compilers/bin/nvfortran '{ print $$1 }')
# ...or set NVROOT manually
#NVHPC      ?= /opt/nvidia/hpc_sdk
#NVVERSION  = 21.7
NVROOT     = /opt/nvidia/hpc_sdk/Linux_x86_64/21.7 #$(NVHPC)/Linux_x86_64/$(NVVERSION)

# Use NV HPC-SDK provided BLAS and LAPACK libraries
LIBDIR     = /opt/nvidia/hpc_sdk/Linux_x86_64/21.7/math_libs/lib64/
BLAS       = -L$(LIBDIR) -lblas
LAPACK     = -L$(LIBDIR) -llapack

BLACS      =

SCALAPACK  =  -L$(LIBDIR) -lscalapack #-Mscalapack

CUDA       = -cudalib=cublas,cusolver,cufft,nccl -cuda

LLIBS      = $(SCALAPACK) $(LAPACK) $(BLAS) $(CUDA)

# Software emulation of quadruple precsion
QD         = /opt/nvidia/hpc_sdk/Linux_x86_64/21.7/compilers/extras/qd
LLIBS      += -L/opt/nvidia/hpc_sdk/Linux_x86_64/21.7/compilers/extras/qd/lib -lqdmod -lqd #  -L$(QD)/lib -lqdmod -lqd
INCS       += -I/opt/nvidia/hpc_sdk/Linux_x86_64/21.7/compilers/extras/qd/include/qd  #-I$(QD)/include/qd

# Use the FFTs from fftw
#FFTW       ?= /opt/intel/oneapi/mkl/2021.3.0/include/fftw #/opt/gnu/fftw-3.3.6-pl2-GNU-5.4.0
#LLIBS      += -L #   -L$(FFTW)/lib -lfftw3
INCS       +=  -I/usr/include #-I$(FFTW)/include

OBJECTS    = fftmpiw.o fftmpi_map.o fftw3d.o fft3dlib.o

# Redefine the standard list of O1 and O2 objects
SOURCE_O1  := pade_fit.o
SOURCE_O2  := pead.o

# For what used to be vasp.5.lib
CPP_LIB    = $(CPP)
FC_LIB     = nvfortran
CC_LIB     = nvc
CFLAGS_LIB = -O
FFLAGS_LIB = -O1 -Mfixed
FREE_LIB   = $(FREE)

OBJECTS_LIB= linpack_double.o getshmem.o

# For the parser library
CXX_PARS   = nvc++ --no_warnings


# Normally no need to change this
SRCDIR     = ../../src
BINDIR     = ../../bin
 

Re: VASP 6.2 compilation with OpenACC GPU port

Posted: Mon Nov 22, 2021 7:18 am
by henrique_miranda
I am not able to find the 'makefile.include' in the post.
Are you sure you attached it?

What command did you use to run the make?

Re: VASP 6.2 compilation with OpenACC GPU port

Posted: Mon Nov 22, 2021 1:30 pm
by david_keller
This matter has been resolved and the issue should be closed.
Than you for the help.
Dave