[petsc-users] Compilation error of test program: "MatDenseGetArray" was not declared in this scope

Matthew Knepley knepley at gmail.com
Tue Jan 19 07:55:28 CST 2021


On Tue, Jan 19, 2021 at 7:28 AM Roland Richter <roland.richter at ntnu.no>
wrote:

> Hei,
>
> I just encountered an issue which I can not solve on my own. I wrote a
> test program on my development machine, and compilation works fine, but
> when trying to compile it on the production machine compilation fails with
>
> I believe your production machine has PETSc version < 3.14, since
MatDenseGetArrayWrite() was not introduced until 3,14.

  Thanks,

     Matt

> *[ 25%] Building CXX object
> CMakeFiles/test_program.dir/source/core/tester_helper.cpp.o*
> */share/apps/software/OpenMPI/3.1.3-gcccuda-2019a/bin/mpicxx
> -DBOOST_NO_AUTO_PTR -DUSE_CUDA -DUSE_PHYSICS_VARIABLES
> -Dtest_program_EXPORTS
> -I~/work_directory/full_physics/Test-Controller/library/include
> -I~/work_directory/full_physics/physics_variables/include
> -I~/local_opt_GCC7/interp2d/include -I/usr/include/eigen3
> -I/share/apps/software/Compiler/GCC/8.2.0-2.31.1/GSL/2.5/include
> -I/share/apps/software/MPI/intel/2018.3.222-GCC-7.3.0-2.30/impi/2018.3.222/imkl/2018.3.222/compilers_and_libraries_2018.3.222/linux/mkl/include
> -isystem ~/local_opt_GCC7/dealii/include -isystem
> ~/local_opt_GCC7/dealii/include/deal.II/bundled -isystem
> /share/apps/software/Core/icc/2019.1.144-GCC-8.2.0-2.31.1/tbb/include
> -isystem ~/local_opt_GCC7/boost/include -isystem
> ~/local_opt_GCC7/petsc/include -isystem ~/local_opt_GCC7/trilinos/include
> -isystem ~/local_opt_GCC7/arrayfire/include -isystem
> /share/apps/software/imkl/2019.5.281-iimpi-2019b/mkl/include -isystem
> ~/local_opt_GCC7/suitesparse/include -isystem
> /share/apps/software/MPI/GCC-CUDA/8.2.0-2.31.1-10.1.105/OpenMPI/3.1.3/HDF5/1.10.5/include
> -isystem /share/apps/software/Compiler/GCCcore/8.2.0/zlib/1.2.11/include
> -isystem /share/apps/software/Compiler/GCCcore/8.2.0/Szip/2.1.1/include
> -isystem
> /share/apps/software/Compiler/GCC/8.2.0-2.31.1/CUDA/10.1.105/include
> -isystem
> /share/apps/software/Compiler/GCC/8.2.0-2.31.1/OpenMPI/3.1.3/include
> -isystem ~/local_opt_GCC7/adol-c/include -isystem
> ~/local_opt_GCC7/p4est/include -isystem ~/local_opt_GCC7/slepc/include
> -isystem ~/local_opt_GCC7/armadillo/include -isystem
> /share/apps/software/MPI/GCC-CUDA/8.2.0-2.31.1-10.1.105/OpenMPI/3.1.3/FFTW/3.3.8/include
> -march=native -fopenmp-simd  -fPIC -flto -march=native -fopenmp -O2
> -funroll-loops -funroll-all-loops -fstrict-aliasing -DMKL_LP64 -m64 -O3
> -DNDEBUG -fPIC -pedantic -fPIC -Wall -Wextra -Wmissing-braces
> -Woverloaded-virtual -Wpointer-arith -Wsign-compare -Wswitch -Wsynth
> -Wwrite-strings -Wno-placement-new -Wno-literal-suffix -Wno-psabi
> -Wno-class-memaccess -fopenmp-simd -march=native -mavx
> -Wno-unused-local-typedefs -O2 -funroll-loops -funroll-all-loops
> -fstrict-aliasing -Werror -flto -std=gnu++17 -MD -MT
> CMakeFiles/test_program.dir/source/core/tester_helper.cpp.o -MF
> CMakeFiles/test_program.dir/source/core/tester_helper.cpp.o.d -o
> CMakeFiles/test_program.dir/source/core/tester_helper.cpp.o -c
> ~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:
> In member function 'void tester_helper::nonlinear_step(double&, double)':*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:765:5:
> error: 'MatDenseGetArrayWrite' was not declared in this scope*
> *     MatDenseGetArrayWrite (petsc_field.pc_spectral, &write_ptr);*
> *     ^~~~~~~~~~~~~~~~~~~~~*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:765:5:
> note: suggested alternative: 'MatDenseGetArrayRead'*
> *     MatDenseGetArrayWrite (petsc_field.pc_spectral, &write_ptr);*
> *     ^~~~~~~~~~~~~~~~~~~~~*
> *     MatDenseGetArrayRead*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:772:5:
> error: 'MatDenseRestoreArrayWrite' was not declared in this scope*
> *     MatDenseRestoreArrayWrite (petsc_field.pc_spectral, &write_ptr);*
> *     ^~~~~~~~~~~~~~~~~~~~~~~~~*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:772:5:
> note: suggested alternative: 'MatDenseRestoreArrayRead'*
> *     MatDenseRestoreArrayWrite (petsc_field.pc_spectral, &write_ptr);*
> *     ^~~~~~~~~~~~~~~~~~~~~~~~~*
> *     MatDenseRestoreArrayRead*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:
> In member function 'void
> tester_helper::nonlinear_step_material_transition()':*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:870:4:
> error: 'MatDenseGetArrayWrite' was not declared in this scope*
> *    MatDenseGetArrayWrite(split_mat, &write_mat_ptr);*
> *    ^~~~~~~~~~~~~~~~~~~~~*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:870:4:
> note: suggested alternative: 'MatDenseGetArrayRead'*
> *    MatDenseGetArrayWrite(split_mat, &write_mat_ptr);*
> *    ^~~~~~~~~~~~~~~~~~~~~*
> *    MatDenseGetArrayRead*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:874:4:
> error: 'MatDenseRestoreArrayWrite' was not declared in this scope*
> *    MatDenseRestoreArrayWrite(split_mat, &write_mat_ptr);*
> *    ^~~~~~~~~~~~~~~~~~~~~~~~~*
> *~/work_directory/full_physics/Test-Controller/library/source/core/tester_helper.cpp:874:4:
> note: suggested alternative: 'MatDenseRestoreArrayRead'*
> *    MatDenseRestoreArrayWrite(split_mat, &write_mat_ptr);*
> *    ^~~~~~~~~~~~~~~~~~~~~~~~~*
> *    MatDenseRestoreArrayRead*
>
> Where does that error come from, and how can I fix it/circumvent it? I
> compile the program using gcc 8.2.0 (including mpi-wrappers) on the
> production machine, and with gcc 9.3.1 on the dev machine.
>
> Thanks,
>
> regards,
>
> Roland
>


-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener

https://www.cse.buffalo.edu/~knepley/ <http://www.cse.buffalo.edu/~knepley/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20210119/709105cd/attachment-0001.html>


More information about the petsc-users mailing list