[petsc-users] Running KSP of preonly doesn't make sense with nonzero initial guess

[Barry Smith]

   You must be seting KSPSetInitialGuessNonzero() in your code. So either remove that line or use -ksp_type richardson instead of -ksp_type preonly.

  Barry


> On Jul 22, 2015, at 1:49 PM, Jin, Shuangshuang <Shuangshuang.Jin at pnnl.gov> wrote:
> 
> Hi, I’m trying to use PETSc ksp to solve a linear equation with the following option: “-ksp_type preonly -pc_type lu -pc_factor_mat_ordering_type amd”
>  
> I got an error message below. Does anyone have an idea how it comes from?
>  
> [0]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------
> [0]PETSC ERROR:
> [0]PETSC ERROR: Running KSP of preonly doesn't make sense with nonzero initial guess
>                you probably want a KSP type of Richardson
> [0]PETSC ERROR: See http://www.mcs.anl.gov/petsc/documentation/faq.html for trouble shooting.
> [0]PETSC ERROR: Petsc Release Version 3.6.0, Jun, 09, 2015
> [0]PETSC ERROR: Unknown Name on a olympus-openmpi_f-gnu-cxx-complex-opt named olympus.local by d3m956 Wed Jul 22 11:37:26 2015
> [0]PETSC ERROR: Configure options PETSC_ARCH=olympus-openmpi_f-gnu-cxx-complex-opt --with-prefix=./ --with-mpi=1 --with-cc=mpicc --with-fc=mpif90 --with-cxx=mpicxx --with-c++-support=1 --with-c-support=0 --with-c-language=C++ --with-fortran=0 --with-scalar-type=complex --with-fortran-kernels=generic --download-superlu_dist --download-superlu --download-parmetis --download-metis --download-f2cblaslapack=1 --download-suitesparse --with-clanguage=c++ --with-shared-libraries=0 --with-x=0 --with-mpirun=mpirun --with-mpiexec=mpiexec --with-debugging=0
> [0]PETSC ERROR: #1 KSPSolve_PREONLY() line 23 in /pic/projects/gridpack/software_f/petsc-3.6.0/src/ksp/ksp/impls/preonly/preonly.c
> [0]PETSC ERROR: #2 KSPSolve() line 604 in /pic/projects/gridpack/software_f/petsc-3.6.0/src/ksp/ksp/interface/itfunc.c
> [0]PETSC ERROR: #3 p_solve() line 120 in /pic/projects/gridpack/gridpack-Github/src/math/petsc/petsc_linear_solver_implementation.hpp
> terminate called after throwing an instance of 'gridpack::math::PETScException'
>   what():  PETSc error (0): Error detected in C PETSc
> [olympus:27408] *** Process received signal ***
> [olympus:27408] Signal: Aborted (6)
> [olympus:27408] Signal code:  (-6)
> [olympus:27408] [ 0] /lib64/libpthread.so.0 [0x3d2920eca0]
> [olympus:27408] [ 1] /lib64/libc.so.6(gsignal+0x35) [0x3d2862ffc5]
> [olympus:27408] [ 2] /lib64/libc.so.6(abort+0x110) [0x3d28631a70]
> [olympus:27408] [ 3] /share/apps/gcc/4.6.2/lib64/libstdc++.so.6(_ZN9__gnu_cxx27__verbose_terminate_handlerEv+0x11d) [0x7f4aa53705bd]
> [olympus:27408] [ 4] /share/apps/gcc/4.6.2/lib64/libstdc++.so.6 [0x7f4aa536e766]
> [olympus:27408] [ 5] /share/apps/gcc/4.6.2/lib64/libstdc++.so.6 [0x7f4aa536e793]
> [olympus:27408] [ 6] /share/apps/gcc/4.6.2/lib64/libstdc++.so.6 [0x7f4aa536e88e]
> [olympus:27408] [ 7] ./ds2.x(_ZNK8gridpack4math31PETScLinearSolverImplementationISt7complexIdEiE7p_solveERKNS0_7VectorTIS3_iEERS6_+0x4cf) [0x74634f]
> [olympus:27408] [ 8] ./ds2.x(_ZNK8gridpack4math25BaseLinearSolverInterfaceISt7complexIdEiE5solveERKNS0_7VectorTIS3_iEERS6_+0x37) [0x65e84f]
> [olympus:27408] [ 9] ./ds2.x(_ZNK8gridpack4math13LinearSolverTISt7complexIdEiE7p_solveERKNS0_7VectorTIS3_iEERS6_+0x3b) [0x68fb89]
> [olympus:27408] [10] ./ds2.x(_ZNK8gridpack4math25BaseLinearSolverInterfaceISt7complexIdEiE5solveERKNS0_7VectorTIS3_iEERS6_+0x37) [0x65e84f]
> [olympus:27408] [11] ./ds2.x(_ZN8gridpack18dynamic_simulation6DS2App5solveENS0_9DS2Branch5EventE+0x8f0) [0x656386]
> [olympus:27408] [12] ./ds2.x(main+0x41d) [0x64d9b1]
> [olympus:27408] [13] /lib64/libc.so.6(__libc_start_main+0xf4) [0x3d2861d9f4]
> [olympus:27408] [14] ./ds2.x [0x64d4dd]
> [olympus:27408] *** End of error message ***
> Abort
>  
> Thanks,
> Shuangshuang