<html><head><meta http-equiv="Content-Type" content="text/html; charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><div class=""><br class=""></div> Mark,<div class=""><br class=""></div><div class=""> PCFIELDSPLIT PCFieldSplitSetDefaults uses the block size of the matrix to determine the number of splits. </div><div class=""><br class=""></div><div class=""> It could be that the matrices returned by PCASM have a default block size of 1 (instead of inheriting the block size of the matrix the sub matrix was obtained with). </div><div class=""><br class=""></div><div class=""> MatCreateSubMatrices_MPIAIJ_Local() has err = MatSetBlockSizes(submats[i],rbs,cbs);CHKERRQ(ierr); where rbs and cbs get their block sizes from the IS. You would need to dig into the code, with the debugger? to see if the block size is propagated to the sub matrices and why not. It should be, but likely that is never specifically tested.</div><div class=""><br class=""></div><div class=""><br class=""></div><div class=""> Barry</div><div class=""><br class=""><div><br class=""><blockquote type="cite" class=""><div class="">On Dec 31, 2020, at 7:40 AM, Mark Adams <<a href="mailto:mfadams@lbl.gov" class="">mfadams@lbl.gov</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div dir="ltr" class="">Humm, I have two fields (Nf) here:<div class=""><br class=""></div><div class="">/usr/local/Cellar/mpich/3.3.2_1/bin/mpiexec -n 1 ./ex2 -ex2_test_type spitzer -ex2_connor_e_field_units -dm_landau_Ez 0 -petscspace_degree 3 -dm_landau_type p4est -dm_landau_ion_masses 1 -dm_landau_ion_charges 1 -dm_landau_thermal_temps 4,4 -dm_landau_n 1,1 -ts_monitor -ts_adapt_monitor -snes_rtol 1.e-14 -snes_stol 1.e-14 -snes_monitor -snes_converged_reason -snes_max_it 15 -ts_type arkimex -ts_exact_final_time stepover -ts_arkimex_type 1bee -ts_max_snes_failures -1 -ts_rtol 1e-3 -ts_dt .1e-1 -ts_adapt_clip .25,1.1 -ts_max_time 100 -ts_max_steps 1 -ts_adapt_scale_solve_failed 0.75 -ts_adapt_time_step_increase_delay 5 -pc_type lu -ksp_type preonly -dm_landau_amr_levels_max 8 -dm_landau_amr_post_refine 0 -dm_landau_domain_radius 5 -ex2_plot_dt 1 -info :dm -ts_adapt_dt_min 1e-4 -ts_adapt_dt_max 1. -dm_preallocate_only -ex2_impurity_source_type pulse -ex2_pulse_start_time 32 -ex2_pulse_width_time 15 -ex2_pulse_rate 2e+0 -ex2_t_cold 1 -dm_landau_device_type cpu -pc_type asm -sub_pc_type fieldsplit -sub_pc_fieldsplit_type additive -sub_fieldsplit_0_pc_type lu -ksp_monitor<br class="">[0] ProcessOptions(): num_species set to number of thermal temps provided (2)<br class="">masses: e= 9.109e-31; ions in proton mass units: 1.000e+00 0.000e+00 ...<br class="">charges: e=-1.602e-19; charges in elementary units: 1.000e+00 0.000e+00<br class="">thermal T (K): e= 4.642e+07 i= 4.642e+07 imp= 0.000e+00. v_0= 2.652e+07 n_0= 1.000e+20 t_0= 4.630e-05 domain= 5.000e+00<br class="">[0] ProcessREOptions(): Num electrons from ions=1., T_cold= 1.160e+07, ion potential= 1.500e+01, E_z= 0.000e+00 v_0= 2.652e+07<br class="">CalculateE j0=0. Ec = 0.0509908<br class="">[0] DMGetDMSNES(): Creating new DMSNES<br class="">[0] DMGetDMTS(): Creating new DMTS<br class="">[0] DMGetDMSNESWrite(): Copying DMSNES due to write<br class="">0 TS dt 0.01 time 0.<br class=""> <b class="">0) species-0: charge density= -1.6022862392985e+01 z-momentum= -1.4067414979736e-19 energy= 9.6063873494138e+04<br class=""> 0) species-1: charge density= 1.6029890912678e+01 z-momentum= -8.1631830255319e-18 energy= 9.6333618295472e+04</b><br class=""> 0) Total: charge density= 7.0285196929767e-03, momentum= -8.3038571753292e-18, energy= 1.9239749178961e+05 (m_i[0]/m_e = 1835.47, 50 cells)<br class="">testSpitzer 0) time= 0.000e+00 n_e= 1.000e+00 E= 0.000e+00 J= -1.495e-07 J_re= -0.000e+00 0 % Te_kev= 3.995e+00 Z_eff=1. E/J to eta ratio=-0. (diff=1e+10) constant E<br class="">[0] DMGetDMKSP(): Creating new DMKSP<br class="">[0]FormLandau: 800 IPs, 50 cells, totDim=32, Nb=16, Nq=16, elemMatSize=1024, dim=2, Tab: Nb=16 <b class=""><font color="#ff0000" class="">Nf=2 </font></b>Np=16 cdim=2 N=896 N+hang=1176<br class=""> 0 SNES Function norm 4.974988592014e-03<br class="">[0]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------<br class="">[0]PETSC ERROR: Petsc has generated inconsistent data<br class=""><b class="">[0]PETSC ERROR: Unhandled case, must have at least two fields, not 1<br class=""></b>[0]PETSC ERROR: See <a href="https://www.mcs.anl.gov/petsc/documentation/faq.html" class="">https://www.mcs.anl.gov/petsc/documentation/faq.html</a> for trouble shooting.<br class="">[0]PETSC ERROR: Petsc Development GIT revision: v3.14.2-353-gc57b1f101b GIT Date: 2020-12-27 11:51:24 -0600<br class="">[0]PETSC ERROR: ./ex2 on a arch-macosx-gnu-g named MarksMac-302.local by markadams Thu Dec 31 08:35:49 2020<br class="">[0]PETSC ERROR: Configure options --with-mpi-dir=/usr/local/Cellar/mpich/3.3.2_1 COPTFLAGS="-g -O0" CXXOPTFLAGS="-g -O0" --download-metis=1 --download-parmetis=1 --download-kokkos=1 --download-kokkos-kernels=1 --download-p4est=1 --with-zlib=1 --with-make-np=4 --download-hdf5=1 -with-cuda=0 --with-x=0 --with-debugging=1 PETSC_ARCH=arch-macosx-gnu-g --with-64-bit-indices=0 --with-openmp=0 --with-ctable=0<br class="">[0]PETSC ERROR: #1 PCFieldSplitSetDefaults() line 571 in /Users/markadams/Codes/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c<br class="">[0]PETSC ERROR: #2 PCSetUp_FieldSplit() line 614 in /Users/markadams/Codes/petsc/src/ksp/pc/impls/fieldsplit/fieldsplit.c<br class="">[0]PETSC ERROR: #3 PCSetUp() line 1015 in /Users/markadams/Codes/petsc/src/ksp/pc/interface/precon.c<br class="">[0]PETSC ERROR: #4 KSPSetUp() line 406 in /Users/markadams/Codes/petsc/src/ksp/ksp/interface/itfunc.c<br class="">[0]PETSC ERROR: #5 PCSetUpOnBlocks_ASM() line 437 in /Users/markadams/Codes/petsc/src/ksp/pc/impls/asm/asm.c<br class="">[0]PETSC ERROR: #6 PCSetUpOnBlocks() line 1046 in /Users/markadams/Codes/petsc/src/ksp/pc/interface/precon.c<br class="">[0]PETSC ERROR: #7 KSPSetUpOnBlocks() line 214 in /Users/markadams/Codes/petsc/src/ksp/ksp/interface/itfunc.c<br class="">[0]PETSC ERROR: #8 KSPSolve_Private() line 727 in /Users/markadams/Codes/petsc/src/ksp/ksp/interface/itfunc.c<br class="">[0]PETSC ERROR: #9 KSPSolve() line 963 in /Users/markadams/Codes/petsc/src/ksp/ksp/interface/itfunc.c<br class="">[0]PETSC ERROR: #10 SNESSolve_NEWTONLS() line 225 in /Users/markadams/Codes/petsc/src/snes/impls/ls/ls.c<br class="">[0]PETSC ERROR: #11 SNESSolve() line 4563 in /Users/markadams/Codes/petsc/src/snes/interface/snes.c<br class="">[0]PETSC ERROR: #12 TSStep_ARKIMEX() line 845 in /Users/markadams/Codes/petsc/src/ts/impls/arkimex/arkimex.c<br class="">[0]PETSC ERROR: #13 TSStep() line 3771 in /Users/markadams/Codes/petsc/src/ts/interface/ts.c<br class="">[0]PETSC ERROR: #14 TSSolve() line 4168 in /Users/markadams/Codes/petsc/src/ts/interface/ts.c<br class="">[0]PETSC ERROR: #15 main() line 774 in ex2.c<br class="">[0]PETSC ERROR: PETSc Option Table entries:<br class="">[0]PETSC ERROR: -dm_landau_amr_levels_max 8<br class="">[0]PETSC ERROR: -dm_landau_amr_post_refine 0<br class="">[0]PETSC ERROR: -dm_landau_device_type cpu<br class="">[0]PETSC ERROR: -dm_landau_domain_radius 5<br class="">[0]PETSC ERROR: -dm_landau_Ez 0<br class="">[0]PETSC ERROR: -dm_landau_ion_charges 1<br class="">[0]PETSC ERROR: -dm_landau_ion_masses 1<br class="">[0]PETSC ERROR: -dm_landau_n 1,1<br class="">[0]PETSC ERROR: -dm_landau_thermal_temps 4,4<br class="">[0]PETSC ERROR: -dm_landau_type p4est<br class="">[0]PETSC ERROR: -dm_preallocate_only<br class="">[0]PETSC ERROR: -ex2_connor_e_field_units<br class="">[0]PETSC ERROR: -ex2_impurity_source_type pulse<br class="">[0]PETSC ERROR: -ex2_plot_dt 1<br class="">[0]PETSC ERROR: -ex2_pulse_rate 2e+0<br class="">[0]PETSC ERROR: -ex2_pulse_start_time 32<br class="">[0]PETSC ERROR: -ex2_pulse_width_time 15<br class="">[0]PETSC ERROR: -ex2_t_cold 1<br class="">[0]PETSC ERROR: -ex2_test_type spitzer<br class="">[0]PETSC ERROR: -info :dm<br class="">[0]PETSC ERROR: -ksp_monitor<br class="">[0]PETSC ERROR: -ksp_type preonly<br class="">[0]PETSC ERROR: -pc_type asm<br class="">[0]PETSC ERROR: -petscspace_degree 3<br class="">[0]PETSC ERROR: -snes_converged_reason<br class="">[0]PETSC ERROR: -snes_max_it 15<br class="">[0]PETSC ERROR: -snes_monitor<br class="">[0]PETSC ERROR: -snes_rtol 1.e-14<br class="">[0]PETSC ERROR: -snes_stol 1.e-14<br class="">[0]PETSC ERROR: -sub_fieldsplit_0_pc_type lu<br class="">[0]PETSC ERROR: -sub_pc_fieldsplit_type additive<br class="">[0]PETSC ERROR: -sub_pc_type fieldsplit<br class="">[0]PETSC ERROR: -ts_adapt_clip .25,1.1<br class="">[0]PETSC ERROR: -ts_adapt_dt_max 1.<br class="">[0]PETSC ERROR: -ts_adapt_dt_min 1e-4<br class="">[0]PETSC ERROR: -ts_adapt_monitor<br class="">[0]PETSC ERROR: -ts_adapt_scale_solve_failed 0.75<br class="">[0]PETSC ERROR: -ts_adapt_time_step_increase_delay 5<br class="">[0]PETSC ERROR: -ts_arkimex_type 1bee<br class="">[0]PETSC ERROR: -ts_dt .1e-1<br class="">[0]PETSC ERROR: -ts_exact_final_time stepover<br class="">[0]PETSC ERROR: -ts_max_snes_failures -1<br class="">[0]PETSC ERROR: -ts_max_steps 1<br class="">[0]PETSC ERROR: -ts_max_time 100<br class="">[0]PETSC ERROR: -ts_monitor<br class="">[0]PETSC ERROR: -ts_rtol 1e-3<br class="">[0]PETSC ERROR: -ts_type arkimex<br class="">[0]PETSC<br class=""></div></div><br class=""><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Dec 31, 2020 at 12:30 AM Jed Brown <<a href="mailto:jed@jedbrown.org" class="">jed@jedbrown.org</a>> wrote:<br class=""></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Mark Adams <<a href="mailto:mfadams@lbl.gov" target="_blank" class="">mfadams@lbl.gov</a>> writes:<br class="">
<br class="">
> I think the first thing that I need to do is get ASM working on the CPU in<br class="">
> my code as a direct solver.<br class="">
><br class="">
> FieldSplit was mentioned the way to do this. How would I do that?<br class="">
<br class="">
-pc_type asm -sub_pc_type fieldsplit -sub_pc_fieldsplit_type additive -sub_fieldsplit_0_pc_type lu<br class="">
</blockquote></div>
</div></blockquote></div><br class=""></div></body></html>