OpenMx installation in Linux CentOS 6.8

Posted on
No user picture. Anbupalam Joined: 09/01/2012
I am not able to compile openmx under 3.3.0. I was advised to change the compiler by including
CXX=g++ in the ~/.R/Makevars file.

But the openMx is still being compiled using icpc instead of g++ and I was asked by the help desk to contact the developers.
Any help?

Thanks
Anbu.

Replied on Tue, 03/14/2017 - 08:43
Picture of user. jpritikin Joined: 05/23/2012

You're on the right track. Try CXX1X
Replied on Tue, 03/14/2017 - 17:51
No user picture. Anbupalam Joined: 09/01/2012

In reply to by jpritikin

Tried CXX1X but still the compilation didn't go through:

NOTE: ./configure is not an autoconf generated script.
Change default C/C++ compiler and default compile flags by editing ~/.R/Makevars
** libs
g++ -std=c++11 -I/apps/R/3.3.0/lib64/R/include -DNDEBUG -I/usr/local/include -I"/home/561/aat561/R/x86_64-pc-linux-gnu-library/3.3/Rcpp/include" -I"/home/561/aat561/R/x86_64-pc-linux-gnu-library/3.3/RcppEigen/include" -I"/home/561/aat561/R/x86_64-pc-linux-gnu-library/3.3/StanHeaders/include" -I"/home/561/aat561/R/x86_64-pc-linux-gnu-library/3.3/BH/include" -I"/home/561/aat561/R/x86_64-pc-linux-gnu-library/3.3/rpf/include" -fopenmp -Wall -Wextra -Wno-unknown-pragmas -Wno-unused-parameter -Wvla -Wno-unused-local-typedefs -Wno-deprecated-declarations -fdiagnostics-color=auto -DSHADOW_DIAG -fpic -g -O2 -c Compute.cpp -o Compute.o
cc1plus: error: unrecognized command line option "-std=c++11"
cc1plus: error: unrecognized command line option "-fdiagnostics-color=auto"
cc1plus: warning: unrecognized command line option "-Wno-unused-local-typedefs"
make: *** [Compute.o] Error 1
ERROR: compilation failed for package 'OpenMx'
* removing '/home/561/aat561/R/x86_64-pc-linux-gnu-library/3.3/OpenMx'

The downloaded source packages are in
'/short/ey6/aat561/tmp/RtmpKeaBDy/downloaded_packages'
Warning message:
In install.packages("OpenMx") :
installation of package 'OpenMx' had non-zero exit status

Replied on Wed, 03/15/2017 - 19:19
No user picture. Anbupalam Joined: 09/01/2012

In reply to by jpritikin

I changed the compiler to version 6.2.0, the compilation went on for sometime and finally I got the error message:

Is it due to the /usr/bin/ld:../src/libnpsol.a ? If so, please let me know how to fix this.

g++ -std=c++11 -shared -L/apps/R/3.3.0/lib64/R/lib -L/usr/local/lib64 -o OpenMx.so Compute.o ComputeGD.o ComputeNR.o ComputeSD.o FellnerFitFunction.o ba81quad.o dmvnorm.o fitMultigroup.o general.o glue.o matrix.o merge.o nanotimer.o nloptcpp.o omxAlgebra.o omxAlgebraFitFunction.o omxCsolnp.o omxData.o omxExpectation.o omxExpectationBA81.o omxExportBackendState.o omxFIMLFitFunction.o omxFIMLSingleIteration.o omxFitFunction.o omxFitFunctionBA81.o omxGREMLExpectation.o omxGREMLfitfunction.o omxHessianCalculation.o omxImportFrontendState.o omxLISRELExpectation.o omxMLFitFunction.o omxMatrix.o omxNPSOLSpecific.o omxNormalExpectation.o omxRAMExpectation.o omxRFitFunction.o omxRowFitFunction.o omxState.o omxStateSpaceExpectation.o omxSymbolTable.o omxWLSFitFunction.o optimize.o options.o sadmvn.o slsqp.o ssMLFit.o stop.o subnp.o -fopenmp -lifport -lifcore -limf -lsvml -lm -lipgo -lirc -lpthread -lirc_s -ldl -L/apps/intel-ct/16.0.3.210/mkl/lib/intel64 -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -liomp5 -lpthread -L../src -lnpsol -lifport -lifcore -limf -lsvml -lm -lipgo -lirc -lpthread -lirc_s -ldl -L/apps/R/3.3.0/lib64/R/lib -lR
/usr/bin/ld:../src/libnpsol.a: file format not recognized; treating as linker script
/usr/bin/ld:../src/libnpsol.a:1: syntax error
collect2: error: ld returned 1 exit status
make: *** [OpenMx.so] Error 1
ERROR: compilation failed for package 'OpenMx'
* removing '/home/561/aat561/R/x86_64-pc-linux-gnu-library/3.3/OpenMx'

Replied on Thu, 03/16/2017 - 09:22
Picture of user. jpritikin Joined: 05/23/2012

In reply to by Anbupalam

Do you need npsol? If not, use 'make cran-install' instead of 'make install'. Otherwise remove src/npsol.a and get a fresh copy from http://openmx.psyc.virginia.edu/OpenMx2/npsol/linux/x86_64/gcc4.8/