<html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">Hello,<div class="">I’m solving the following dummy system <a href="http://jolivet.perso.enseeiht.fr/composite_ksp.tar.gz" class="">http://jolivet.perso.enseeiht.fr/composite_ksp.tar.gz</a></div><div class="">[A, B];[C, D], with a PCFIELDSPLIT. For the PC of D, I’m using a PCCOMPOSITE with two sub PCs. One of which is a PCKSP.</div><div class="">Could you please help me figure out what is wrong in the following piece of code, that may be launched with the following arguments:</div><div class="">$ mpirun -n 1 ./a.out -ksp_type preonly -pc_type fieldsplit -fieldsplit_1_pc_type composite -fieldsplit_1_sub_1_pc_type ksp -fieldsplit_1_sub_1_ksp_ksp_type gmres -fieldsplit_1_sub_1_ksp_pc_type gamg -fieldsplit_1_sub_1_ksp_ksp_converged_reason -fieldsplit_1_sub_1_ksp_pc_gamg_sym_graph 1 -fieldsplit_1_sub_1_ksp_pc_gamg_square_graph 10 -fieldsplit_1_sub_1_ksp_ksp_rtol 1e-8</div><div class=""><br class=""></div><div class="">It solves the dummy system twice, with a varying block D.</div><div class=""><br class=""></div><div class="">It should give you:</div><div class=""><div class=""> Linear fieldsplit_1_sub_1_ksp_ solve converged due to CONVERGED_RTOL iterations 8</div><div class="">solve #0: 16098.3</div><div class=""> Linear fieldsplit_1_sub_1_ksp_ solve did not converge due to DIVERGED_PC_FAILED iterations 0</div><div class=""> PC_FAILED due to SUBPC_ERROR</div><div class="">solve #1: inf</div></div><div class=""><br class=""></div><div class="">If I switch line 70 to #if 0, I get the expected output:</div><div class=""><div class=""> Linear fieldsplit_1_sub_1_ksp_ solve converged due to CONVERGED_RTOL iterations 8</div><div class="">solve #0: 16098.3</div><div class=""> Linear fieldsplit_1_sub_1_ksp_ solve converged due to CONVERGED_RTOL iterations 8</div><div class="">solve #1: 325.448</div></div><div class=""><br class=""></div><div class="">I’m realizing that this has probably nothing to do with the outer PCFIELDSPLIT, but this comes from a rather large FSI solver, so reproducing this behavior in “only” 97 SLOC is good enough for your I hope.</div><div class=""><br class=""></div><div class="">Thanks in advance,</div><div class="">Pierre</div></body></html>