{"id":1224,"date":"2014-08-19T14:37:36","date_gmt":"2014-08-19T06:37:36","guid":{"rendered":"http:\/\/www.hsyyf.me\/?p=1224"},"modified":"2014-08-19T14:37:36","modified_gmt":"2014-08-19T06:37:36","slug":"ifortvasp%e7%bc%96%e8%af%91%e5%ae%89%e8%a3%85","status":"publish","type":"post","link":"https:\/\/www.hsyyf.me\/?p=1224","title":{"rendered":"ifort+vasp\u7f16\u8bd1\u5b89\u88c5"},"content":{"rendered":"

vasp\u7684\u7f16\u8bd1\u5b89\u88c5\u53ef\u4ee5\u8bf4\u662f\u5343\u5947\u767e\u602a\uff0c\u6211\u91c7\u7528\u7684\u662fivf\u7684\u7f16\u8bd1\u5668\u548cmpich2\u8fdb\u884c\u6700\u7b80\u5355\u7684\u7f16\u8bd1\uff0c\u7cfb\u7edf\u662fubuntu14.04\uff0c\u4e0d\u5f97\u4e0d\u8bf4\uff0c\u5c31\u662f\u56e0\u4e3a\u8fd9\u4e2a\u7cfb\u7edf\uff0c\u5f53\u65f6\u6d6a\u8d39\u4e86\u6211\u4e00\u4e2a\u661f\u671f\u7684\u65f6\u95f4\u8fdb\u884c\u7f16\u8bd1\u3002<\/p>\n

\u9996\u5148\uff0c\u5b89\u88c5ifort\uff0c\u9009\u62e9ifort\u7684\u4e00\u5927\u539f\u56e0\u662f\u56e0\u4e3aifort\u6709\u514d\u8d39\u9002\u7528\u7248\uff0c\u53ef\u4ee5\u7533\u8bf7\u5230\u4e00\u5e74\u7684\u9002\u7528\u6743\u3002\u5148\u7528\u90ae\u7bb1\u7533\u8bf7\u4e00\u4e2alic\uff0cIntel\u4f1a\u628aID\u548c\u4e0b\u8f7d\u5730\u5740\u901a\u8fc7\u90ae\u7bb1\u544a\u8bc9\u4f60\uff0c\u4e0b\u8f7d\u4e0b\u6765\u6b63\u5e38\u5b89\u88c5\u5373\u53ef\uff08\u6559\u80b2\u7f51\u901f\u5ea6\u5c1a\u53ef\uff0c\u80fd\u670910M\/s\u7684\u901f\u5ea6\uff0c\u4e1a\u754c\u826f\u5fc3\u554a\u3002\uff09<\/p>\n

\u5176\u6b21\uff0c\u5b89\u88c5mpich2\uff0cubuntu\u6e90\u91cc\u81ea\u5e26
\nsudo apt-get install mpich2<\/code><\/p>\n

 <\/p>\n

\u5bf9\u4e8evasp\u7684\u6e90\u4ee3\u7801\uff0c\u5206\u4e24\u6b65\u7f16\u8bd1\uff0c\u4e00\u6b65\u662f\u7f16\u8bd1vasp.5.lib\uff0c\u8fd9\u4e2a\u53ea\u9700\u8981\u628amakefile_ivf\u590d\u5236\u6210makefile\u6587\u4ef6\uff0c\u628agcc\u548civf\u6362\u6210icc\u548cifort\u5373\u53ef\u3002\u8fd9\u90e8\u5206\u6ca1\u4ec0\u4e48\u96be\u5f97\u3002<\/p>\n

\u7f16\u8bd1\u5e76\u884cvasp\u7a0b\u5e8f\u7684makefile\u6587\u4ef6\u9996\u5148\u7528makefile_linux_ivf\u590d\u5236\u6210makefile\u6587\u4ef6\u3002\u505a\u5982\u4e0b\u4fee\u6539\uff1a<\/p>\n

53~55\u884c\u6ce8\u89c6\u6389\uff1b70\u884c\uff08CPP_\u5f00\u5934\uff09\u628a\u201c-C\u201d\u53c2\u6570\u5220\u6389\uff0c\u5426\u5219\u65e0\u6cd5\u6b63\u5e38\u7f16\u8bd1\uff0c\u8fd9\u4e2a\u95ee\u9898\u7ea0\u7ed3\u6211\u4e00\u4e2a\u591a\u661f\u671f\u7684\u5173\u952e\u6240\u5728\uff0c\u800c\u4e14\u7f51\u4e0a\u4e5f\u6ca1\u6709\u8bf4\u660e\uff1b\u6ce8\u91ca\u70b9BALS\u884c\uff0c\u6dfb\u52a0\u81ea\u5df1intel\u7f16\u8bd1\u5668mkl\u5e93\u7684\u4f4d\u7f6e\uff1bLAPACK\u884c\u7559\u7a7a\uff1b\u4ecempi\u5f00\u59cb\uff0c\u6253\u5f00\u76f8\u5e94\u7684\u7f16\u8bd1\u9879\u3002<\/p>\n

\u6700\u7ec8\u7684vasp\u5e76\u884c\u7f16\u8bd1\u7684makefile\u6a21\u7248\u5982\u4e0b\uff1a
\n.SUFFIXES: .inc .f .f90 .F
\n#-----------------------------------------------------------------------
\n# Makefile for Intel Fortran compiler for Pentium\/Athlon\/Opteron
\n# bases systems
\n# we recommend this makefile for both Intel as well as AMD systems
\n# for AMD based systems appropriate BLAS and fftw libraries are
\n# however mandatory (whereas they are optional for Intel platforms)
\n#
\n# The makefile was tested only under Linux on Intel and AMD platforms
\n# the following compiler versions have been tested:
\n# - ifc.7.1 works stable somewhat slow but reliably
\n# - ifc.8.1 fails to compile the code properly
\n# - ifc.9.1 recommended (both for 32 and 64 bit)
\n# - ifc.10.1 partially recommended (both for 32 and 64 bit)
\n# tested build 20080312 Package ID: l_fc_p_10.1.015
\n# the gamma only mpi version can not be compiles
\n# using ifc.10.1
\n#
\n# it might be required to change some of library pathes, since
\n# LINUX installation vary a lot
\n# Hence check ***ALL*** options in this makefile very carefully
\n#-----------------------------------------------------------------------
\n#
\n# BLAS must be installed on the machine
\n# there are several options:
\n# 1) very slow but works:
\n# retrieve the lapackage from ftp.netlib.org
\n# and compile the blas routines (BLAS\/SRC directory)
\n# please use g77 or f77 for the compilation. When I tried to
\n# use pgf77 or pgf90 for BLAS, VASP hang up when calling
\n# ZHEEV (however this was with lapack 1.1 now I use lapack 2.0)
\n# 2) more desirable: get an optimized BLAS
\n#
\n# the two most reliable packages around are presently:
\n# 2a) Intels own optimised BLAS (PIII, P4, PD, PC2, Itanium)
\n# http:\/\/developer.intel.com\/software\/products\/mkl\/
\n# this is really excellent, if you use Intel CPU's
\n#
\n# 2b) probably fastest SSE2 (4 GFlops on P4, 2.53 GHz, 16 GFlops PD,
\n# around 30 GFlops on Quad core)
\n# Kazushige Goto's BLAS
\n# http:\/\/www.cs.utexas.edu\/users\/kgoto\/signup_first.html
\n# http:\/\/www.tacc.utexas.edu\/resources\/software\/
\n#
\n#-----------------------------------------------------------------------<\/p>\n

# all CPP processed fortran files have the extension .f90
\nSUFFIX=.f90<\/p>\n

#-----------------------------------------------------------------------
\n# fortran compiler and linker
\n#-----------------------------------------------------------------------
\n#FC=ifort
\n# fortran linker
\n#FCL=$(FC)<\/p>\n

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

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

#-----------------------------------------------------------------------
\n# possible options for CPP:
\n# NGXhalf charge density reduced in X direction
\n# wNGXhalf gamma point only reduced in X direction
\n# avoidalloc avoid ALLOCATE if possible
\n# PGF89 work around some for some PGF90 \/ IFC bugs
\n# CACHE_SIZE 1000 for PII,PIII, 5000 for Athlon, 8000-12000 P4, PD
\n# RPROMU_DGEMV use DGEMV instead of DGEMM in RPRO (depends on used BLAS)
\n# RACCMU_DGEMV use DGEMV instead of DGEMM in RACC (depends on used BLAS)
\n# tbdyn MD package of Tomas Bucko
\n#-----------------------------------------------------------------------<\/p>\n

#CPP = $(CPP_) -DHOST=\\\"LinuxIFC\\\" \\
\n# -DCACHE_SIZE=12000 -DPGF90 -Davoidalloc -DNGXhalf \\
\n# -DRPROMU_DGEMV -DRACCMU_DGEMV<\/p>\n

#-----------------------------------------------------------------------
\n# general fortran flags (there must a trailing blank on this line)
\n# byterecl is strictly required for ifc, since otherwise
\n# the WAVECAR file becomes huge
\n#-----------------------------------------------------------------------<\/p>\n

FFLAGS = -FR -lowercase -assume byterecl <\/p>\n

#-----------------------------------------------------------------------
\n# optimization
\n# we have tested whether higher optimisation improves performance
\n# -axK SSE1 optimization, but also generate code executable on all mach.
\n# xK improves performance somewhat on XP, and a is required in order
\n# to run the code on older Athlons as well
\n# -xW SSE2 optimization
\n# -axW SSE2 optimization, but also generate code executable on all mach.
\n# -tpp6 P3 optimization
\n# -tpp7 P4 optimization
\n#-----------------------------------------------------------------------<\/p>\n

# ifc.9.1, ifc.10.1 recommended
\nOFLAG=-O2 -ip -ftz <\/p>\n

OFLAG_HIGH = $(OFLAG)
\nOBJ_HIGH =
\nOBJ_NOOPT =
\nDEBUG = -FR -O0
\nINLINE = $(OFLAG)<\/p>\n

#-----------------------------------------------------------------------
\n# the following lines specify the position of BLAS and LAPACK
\n# VASP works fastest with the libgoto library
\n# so that's what we recommend
\n#-----------------------------------------------------------------------<\/p>\n

# mkl.10.0
\n# set -DRPROMU_DGEMV -DRACCMU_DGEMV in the CPP lines
\n#BLAS=-L\/opt\/intel\/mkl100\/lib\/em64t -lmkl -lpthread<\/p>\n

# even faster for VASP Kazushige Goto's BLAS
\n# http:\/\/www.cs.utexas.edu\/users\/kgoto\/signup_first.html
\n# parallel goto version requires sometimes -libverbs
\n#BLAS= \/opt\/libs\/libgoto\/libgoto.so
\nBLAS=-L\/opt\/intel\/composer_xe_2013_sp1.2.144\/mkl\/lib\/intel64 -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lpthread
\n# LAPACK, simplest use vasp.5.lib\/lapack_double
\nLAPACK= <\/p>\n

# use the mkl Intel lapack
\n#LAPACK= -lmkl_lapack<\/p>\n

#-----------------------------------------------------------------------<\/p>\n

#LIB = -L..\/vasp.5.lib -ldmy \\
\n# ..\/vasp.5.lib\/linpack_double.o $(LAPACK) \\
\n# $(BLAS)
\n#
\n# options for linking, nothing is required (usually)
\nLINK = <\/p>\n

#-----------------------------------------------------------------------
\n# fft libraries:
\n# VASP.5.2 can use fftw.3.1.X (http:\/\/www.fftw.org)
\n# since this version is faster on P4 machines, we recommend to use it
\n#-----------------------------------------------------------------------<\/p>\n

#FFT3D = fft3dfurth.o fft3dlib.o<\/p>\n

# alternatively: fftw.3.1.X is slighly faster and should be used if available
\n#FFT3D = fftw3d.o fft3dlib.o \/opt\/libs\/fftw-3.1.2\/lib\/libfftw3.a<\/p>\n

#=======================================================================
\n# MPI section, uncomment the following lines until
\n# general rules and compile lines
\n# presently we recommend OPENMPI, since it seems to offer better
\n# performance than lam or mpich
\n#
\n# !!! Please do not send me any queries on how to install MPI, I will
\n# certainly not answer them !!!!
\n#=======================================================================
\n#-----------------------------------------------------------------------
\n# fortran linker for mpi
\n#-----------------------------------------------------------------------<\/p>\n

FC=mpif90 -f90=ifort
\nFCL=$(FC)<\/p>\n

#-----------------------------------------------------------------------
\n# additional options for CPP in parallel version (see also above):
\n# NGZhalf charge density reduced in Z direction
\n# wNGZhalf gamma point only reduced in Z direction
\n# scaLAPACK use scaLAPACK (usually slower on 100 Mbit Net)
\n# avoidalloc avoid ALLOCATE if possible
\n# PGF90 work around some for some PGF90 \/ IFC bugs
\n# CACHE_SIZE 1000 for PII,PIII, 5000 for Athlon, 8000-12000 P4, PD
\n# RPROMU_DGEMV use DGEMV instead of DGEMM in RPRO (depends on used BLAS)
\n# RACCMU_DGEMV use DGEMV instead of DGEMM in RACC (depends on used BLAS)
\n# tbdyn MD package of Tomas Bucko
\n#-----------------------------------------------------------------------<\/p>\n

#-----------------------------------------------------------------------<\/p>\n

CPP = $(CPP_) -DMPI -DHOST=\\\"LinuxIFC\\\" -DIFC \\
\n -DCACHE_SIZE=4000 -DPGF90 -Davoidalloc\\
\n -DMPI_BLOCK=8000
\n## -DRPROMU_DGEMV -DRACCMU_DGEMV -DNGZhalf<\/p>\n

#-----------------------------------------------------------------------
\n# location of SCALAPACK
\n# if you do not use SCALAPACK simply leave that section commented out
\n#-----------------------------------------------------------------------<\/p>\n

BLACS=$(HOME)\/archives\/SCALAPACK\/BLACS\/
\nSCA_=$(HOME)\/archives\/SCALAPACK\/SCALAPACK<\/p>\n

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

SCA=<\/p>\n

#-----------------------------------------------------------------------
\n# libraries for mpi
\n#-----------------------------------------------------------------------<\/p>\n

LIB = -L..\/vasp.5.lib -ldmy \\
\n ..\/vasp.5.lib\/linpack_double.o $(LAPACK) \\
\n $(SCA) $(BLAS)<\/p>\n

# FFT: fftmpi.o with fft3dlib of Juergen Furthmueller
\nFFT3D = fftmpi.o fftmpi_map.o fft3dfurth.o fft3dlib.o <\/p>\n

# alternatively: fftw.3.1.X is slighly faster and should be used if available
\n#FFT3D = fftmpiw.o fftmpi_map.o fftw3d.o fft3dlib.o \/opt\/libs\/fftw-3.1.2\/lib\/libfftw3.a<\/p>\n

#-----------------------------------------------------------------------
\n# general rules and compile lines
\n#-----------------------------------------------------------------------
\nBASIC= symmetry.o symlib.o lattlib.o random.o <\/p>\n

SOURCE= base.o mpi.o smart_allocate.o xml.o \\
\n constant.o jacobi.o main_mpi.o scala.o \\
\n asa.o lattice.o poscar.o ini.o mgrid.o xclib.o vdw_nl.o xclib_grad.o \\
\n radial.o pseudo.o gridq.o ebs.o \\
\n mkpoints.o wave.o wave_mpi.o wave_high.o \\
\n $(BASIC) nonl.o nonlr.o nonl_high.o dfast.o choleski2.o \\
\n mix.o hamil.o xcgrad.o xcspin.o potex1.o potex2.o \\
\n constrmag.o cl_shift.o relativistic.o LDApU.o \\
\n paw_base.o metagga.o egrad.o pawsym.o pawfock.o pawlhf.o rhfatm.o paw.o \\
\n mkpoints_full.o charge.o Lebedev-Laikov.o stockholder.o dipol.o pot.o \\
\n dos.o elf.o tet.o tetweight.o hamil_rot.o \\
\n steep.o chain.o dyna.o sphpro.o us.o core_rel.o \\
\n aedens.o wavpre.o wavpre_noio.o broyden.o \\
\n dynbr.o rmm-diis.o reader.o writer.o tutor.o xml_writer.o \\
\n brent.o stufak.o fileio.o opergrid.o stepver.o \\
\n chgloc.o fast_aug.o fock.o mkpoints_change.o sym_grad.o \\
\n mymath.o internals.o dynconstr.o dimer_heyden.o dvvtrajectory.o vdwforcefield.o \\
\n hamil_high.o nmr.o pead.o mlwf.o subrot.o subrot_scf.o \\
\n force.o pwlhf.o gw_model.o optreal.o davidson.o david_inner.o \\
\n electron.o rot.o electron_all.o shm.o pardens.o paircorrection.o \\
\n optics.o constr_cell_relax.o stm.o finite_diff.o elpol.o \\
\n hamil_lr.o rmm-diis_lr.o subrot_cluster.o subrot_lr.o \\
\n lr_helper.o hamil_lrf.o elinear_response.o ilinear_response.o \\
\n linear_optics.o linear_response.o \\
\n setlocalpp.o wannier.o electron_OEP.o electron_lhf.o twoelectron4o.o \\
\n ratpol.o screened_2e.o wave_cacher.o chi_base.o wpot.o local_field.o \\
\n ump2.o bse_te.o bse.o acfdt.o chi.o sydmat.o dmft.o \\
\n rmm-diis_mlr.o linear_response_NMR.o<\/p>\n

vasp: $(SOURCE) $(FFT3D) $(INC) main.o
\nrm -f vasp
\n$(FCL) -o vasp main.o $(SOURCE) $(FFT3D) $(LIB) $(LINK)
\nmakeparam: $(SOURCE) $(FFT3D) makeparam.o main.F $(INC)
\n$(FCL) -o makeparam $(LINK) makeparam.o $(SOURCE) $(FFT3D) $(LIB)
\nzgemmtest: zgemmtest.o base.o random.o $(INC)
\n$(FCL) -o zgemmtest $(LINK) zgemmtest.o random.o base.o $(LIB)
\ndgemmtest: dgemmtest.o base.o random.o $(INC)
\n$(FCL) -o dgemmtest $(LINK) dgemmtest.o random.o base.o $(LIB)
\nffttest: base.o smart_allocate.o mpi.o mgrid.o random.o ffttest.o $(FFT3D) $(INC)
\n$(FCL) -o ffttest $(LINK) ffttest.o mpi.o mgrid.o random.o smart_allocate.o base.o $(FFT3D) $(LIB)
\nkpoints: $(SOURCE) $(FFT3D) makekpoints.o main.F $(INC)
\n$(FCL) -o kpoints $(LINK) makekpoints.o $(SOURCE) $(FFT3D) $(LIB)<\/p>\n

clean:
\n-rm -f *.g *.f *.o *.L *.mod ; touch *.F<\/p>\n

main.o: main$(SUFFIX)
\n$(FC) $(FFLAGS)$(DEBUG) $(INCS) -c main$(SUFFIX)
\nxcgrad.o: xcgrad$(SUFFIX)
\n$(FC) $(FFLAGS) $(INLINE) $(INCS) -c xcgrad$(SUFFIX)
\nxcspin.o: xcspin$(SUFFIX)
\n$(FC) $(FFLAGS) $(INLINE) $(INCS) -c xcspin$(SUFFIX)<\/p>\n

makeparam.o: makeparam$(SUFFIX)
\n$(FC) $(FFLAGS)$(DEBUG) $(INCS) -c makeparam$(SUFFIX)<\/p>\n

makeparam$(SUFFIX): makeparam.F main.F
\n#
\n# MIND: I do not have a full dependency list for the include
\n# and MODULES: here are only the minimal basic dependencies
\n# if one strucuture is changed then touch_dep must be called
\n# with the corresponding name of the structure
\n#
\nbase.o: base.inc base.F
\nmgrid.o: mgrid.inc mgrid.F
\nconstant.o: constant.inc constant.F
\nlattice.o: lattice.inc lattice.F
\nsetex.o: setexm.inc setex.F
\npseudo.o: pseudo.inc pseudo.F
\nposcar.o: poscar.inc poscar.F
\nmkpoints.o: mkpoints.inc mkpoints.F
\nwave.o: wave.F
\nnonl.o: nonl.inc nonl.F
\nnonlr.o: nonlr.inc nonlr.F<\/p>\n

$(OBJ_HIGH):
\n$(CPP)
\n$(FC) $(FFLAGS) $(OFLAG_HIGH) $(INCS) -c $*$(SUFFIX)
\n$(OBJ_NOOPT):
\n$(CPP)
\n$(FC) $(FFLAGS) $(INCS) -c $*$(SUFFIX)<\/p>\n

fft3dlib_f77.o: fft3dlib_f77.F
\n$(CPP)
\n$(F77) $(FFLAGS_F77) -c $*$(SUFFIX)<\/p>\n

.F.o:
\n$(CPP)
\n$(FC) $(FFLAGS) $(OFLAG) $(INCS) -c $*$(SUFFIX)
\n.F$(SUFFIX):
\n$(CPP)
\n$(SUFFIX).o:
\n$(FC) $(FFLAGS) $(OFLAG) $(INCS) -c $*$(SUFFIX)<\/p>\n

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

fft3dlib.o : fft3dlib.F
\n$(CPP)
\n$(FC) -FR -lowercase -O2 -c $*$(SUFFIX)<\/p>\n

fft3dfurth.o : fft3dfurth.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

fftw3d.o : fftw3d.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

wave_high.o : wave_high.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

radial.o : radial.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

symlib.o : symlib.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

symmetry.o : symmetry.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

wave_mpi.o : wave_mpi.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

wave.o : wave.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

dynbr.o : dynbr.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

asa.o : asa.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

broyden.o : broyden.F
\n$(CPP)
\n$(FC) -FR -lowercase -O2 -c $*$(SUFFIX)<\/p>\n

us.o : us.F
\n$(CPP)
\n$(FC) -FR -lowercase -O1 -c $*$(SUFFIX)<\/p>\n

LDApU.o : LDApU.F
\n$(CPP)
\n$(FC) -FR -lowercase -O2 -c $*$(SUFFIX)<\/code>
\n <\/p>\n","protected":false},"excerpt":{"rendered":"

vasp\u7684\u7f16\u8bd1\u5b89\u88c5\u53ef\u4ee5\u8bf4\u662f\u5343\u5947\u767e\u602a\uff0c\u6211\u91c7\u7528\u7684\u662fivf\u7684\u7f16\u8bd1\u5668\u548cmpich2\u8fdb\u884c\u6700\u7b80\u5355\u7684\u7f16\u8bd1\uff0c\u7cfb\u7edf\u662fubuntu […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0,"footnotes":""},"categories":[3],"tags":[108,170,169],"class_list":["post-1224","post","type-post","status-publish","format-standard","hentry","category-3","tag-fortran","tag-intel","tag-vasp"],"_links":{"self":[{"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=\/wp\/v2\/posts\/1224","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1224"}],"version-history":[{"count":0,"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=\/wp\/v2\/posts\/1224\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1224"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1224"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.hsyyf.me\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1224"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}