<div dir="ltr"><div dir="ltr"><br></div><div class="gmail_quote gmail_quote_container"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><div><br></div><div>If I use </div><div><br></div><div>-pc_type hypre</div><div>-pc_hypre_type boomeramg</div><div>-pc_hypre_boomeramg_coarsen_type hmis</div><div>-pc_hypre_boomeramg_interp_type ext+i</div><div>-pc_hypre_boomeramg_relax_type_all SOR/Jacobi</div><div>-pc_hypre_boomeramg_relax_type_coarse SOR/Jacobi</div><div>-pc_hypre_boomeramg_grid_sweeps_coarse 1</div><div>-pc_hypre_boomeramg_strong_threshold 0.25</div><div><br></div><div><br></div></div></blockquote><div><br></div><div>Some HYPRE options for the relaxation and coarsening type are not coded for GPU, that's probably the reason for the segfault.</div><div>I suggest you run without special options and start adding one by one until you find the ones are not supported.</div><div><br></div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div></div><div><br></div><div> </div><div>Instead if I use  only  </div><div><br></div><div>-pc_type hypre</div><div><br></div></blockquote><div><br></div><div>This probably leads to a successful setup of the preconditioner on the GPU, but then you most probably are passing a CPU vector for the RHS.</div><div>A convenient way to have always a consistent vector is to use MatCreateVecs</div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div></div><div>I get </div><div><br></div><div>                                                                                                                                                                                        </div><div>[0]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------</div><div>[0]PETSC ERROR: Invalid argument</div><div>[0]PETSC ERROR: HYPRE_MEMORY_DEVICE expects a device vector. You need to enable PETSc device support, for example, in some cases, -vec_type cuda</div><div><br></div><div><br></div><div>This helped my realise that hypre was trying to use the GPU despite in my code </div><div><br></div><div>Both runs are succesfull when compiling without -with-cuda</div><div><br></div><div>Best,</div><div>
<div dir="auto" style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none"><div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px"><div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px"><div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px"><div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px"><div>Herbert Owen<br>Senior Researcher, Dpt. Computer Applications in Science and Engineering<br>Barcelona Supercomputing Center (BSC-CNS)<br>Tel: +34 93 413 4038</div><div>Skype: herbert.owen<br><br><a href="https://urldefense.us/v3/__https://scholar.google.es/citations?user=qe5O2IYAAAAJ&hl=en__;!!G_uCfscf7eWS!f0Y_9okAiFSBeUBOlNplMe6jOCPtdRHfhX2s_rz3N9kCK4OJGiGWGjFyRd3JOwojw7MeNEA-VJtgdn8wBbCmRkKVuPTJ$" target="_blank">https://scholar.google.es/citations?user=qe5O2IYAAAAJ&hl=en</a></div><div><br></div></div><br></div><br></div><br></div><br></div><br><br>
</div>
<br></blockquote></div><div><br clear="all"></div><div><br></div><span class="gmail_signature_prefix">-- </span><br><div dir="ltr" class="gmail_signature">Stefano</div></div>