<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/4.6.6">
</HEAD>
<BODY>
Hi,<BR>
<BR>
I encountered the following error while solving a symmetric positive defined system:<BR>
<BR>
Linear solve did not converge due to DIVERGED_PCSETUP_FAILED iterations 0<BR>
PCSETUP_FAILED due to SUBPC_ERROR <BR>
<BR>
This error appears only if I use the optimized version of both petsc and my code ( compiler: gfortran, flags: -O3 ).<BR>
It is weird since I am solving a time-dependent problem and everything, i.e. results and convergence rate, are as expected until the above error shows up. If I run both petsc and my code in debug mode, everything goes smooth till the end of the simulation.<BR>
However, if I reduce the ksp_rtol, even the debug run fails, after running as expected for a while, because of a KSP_DIVERGED_INDEFINITE_PC . <BR>
The options I am using are:<BR>
<BR>
-ksp_type cg<BR>
-ksp_norm_type unpreconditioned<BR>
-ksp_rtol 1e-8<BR>
-ksp_lag_norm<BR>
-ksp_initial_guess_nonzero yes<BR>
-pc_type mg<BR>
-pc_mg_galerkin<BR>
-pc_mg_levels 4<BR>
-mg_levels_ksp_type richardson<BR>
-mg_coarse_ksp_constant_null_space<BR>
-mg_coarse_pc_type lu<BR>
-mg_coarse_pc_factor_mat_solver_package superlu_dist<BR>
-options_left<BR>
<BR>
I attached a copy of ksp_view. I am currently using petsc-master (last updated yesterday).<BR>
I would appreciate any suggestion on this matter.<BR>
<BR>
Thanks,<BR>
Michele<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
</BODY>
</HTML>