<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="">The fix suggested by Mark resolves the error:</div><div class=""><div class="">#define GAMG_HASH(key) (<span style="background-color: rgb(255, 255, 255);" class="">PetscInt</span>)((((PetscInt64)7)*<span style="background-color: rgb(255, 255, 255);" class="">(PetscInt64)</span>key)%<span style="background-color: rgb(255, 255, 255);" class="">(PetscInt64)</span>a_tab->size)<br class=""></div><div class=""><br class=""></div><div class="">I’m guessing this is similar to the fix suggested by Barry:</div><div class="">#define GAMG_HASH(key) ((((PetscInt64)7)*key)%a_tab->size)</div><div><br class=""></div><div>FWIW, my problem has around 1.2B d.o.f., so if the key is something like a column or row index, that could explain the error when doing 7*key</div><div><br class=""></div><div>Thanks,</div><div>Pierre</div><div><br class=""><blockquote type="cite" class=""><div class="">On 24 Jun 2018, at 11:28 PM, Smith, Barry F. <<a href="mailto:bsmith@mcs.anl.gov" class="">bsmith@mcs.anl.gov</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div class=""><br class=""> Pierre,<br class=""><br class=""> Can you confirm if my suggested fix resolved the error message or not? If so, we'll add it to the maint branch.<br class=""><br class=""> Thanks<br class=""><br class=""> Barry<br class=""><br class=""><br class=""><a href="https://stackoverflow.com/questions/11720656/modulo-operation-with-negative-numbers" class="">https://stackoverflow.com/questions/11720656/modulo-operation-with-negative-numbers</a><br class=""><br class=""><br class=""><br class=""><blockquote type="cite" class="">On Jun 24, 2018, at 4:19 PM, Mark Adams <mfadams@lbl.gov> wrote:<br class=""><br class=""><br class=""><br class="">On Sun, Jun 24, 2018 at 2:09 PM Smith, Barry F. <bsmith@mcs.anl.gov> wrote:<br class=""><br class=""><br class=""><blockquote type="cite" class="">On Jun 24, 2018, at 11:15 AM, Mark Adams <mfadams@lbl.gov> wrote:<br class=""><br class="">First, -pc_gamg_threshold 0.8 is crazy. 0.08 is very high. probably best to start with 0.<br class=""><br class="">This threshold will delete all of your edges in the coarsening graph, probably, and GAMG will switch to a one level solve (silently although -info will tell you).<br class=""><br class="">And yes, this looks like a 64 bit indices problem, we have:<br class=""><br class="">#define GAMG_HASH(key) ((((PetscInt)7)*key)%a_tab->size)<br class=""><br class="">We should probably cast everything to (long long) or (unsigned PetscInt) I guess ... I'm not sure how this should be done exactly. Any ideas from the peanut gallery?<br class=""></blockquote><br class="">Mark,<br class=""><br class=""> It should only be changed if we know the exact mechanism that causes the failure. It is not obvious to me that just changing ((((PetscInt)7)*key)%a_tab->size) to ((((PetscInt64)7)*key)%a_tab->size) should matter.<br class=""><br class="">If 7*key > 2B then you can get a negative number and %a_tab->size will still be a negative number, which triggers:<br class=""><br class="">ksp/pc/impls/gamg/agg.c: if (flid < 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_PLIB,"Cannot find gid1 in table”);<br class=""><br class="">which is the second error that Pierre gets. I don't fully understand the first error but a messed up hash table is certainly a possibility.<br class=""><br class="">And maybe we should cast the hell out of it to be safe:<br class=""><br class="">#define GAMG_HASH(key) (PetscInt)((((PetscInt64)7)*(PetscInt64)key)%(PetscInt64)a_tab->size)<br class=""><br class="">I am a bit puzzled that this works. I would think this would be more of a problem but maybe everyone just uses 64 bit ints. I have dealt with this many times and I am puzzled that I left it in a state that looks so wrong to me now. Perhaps Pierre's complicated thing is causing this somehow.<br class=""><br class=""><br class=""><br class=""> Barry<br class=""><br class=""><blockquote type="cite" class=""><br class=""><br class=""><br class="">On Sun, Jun 24, 2018 at 2:34 AM Pierre Jolivet <pierre.jolivet@enseeiht.fr> wrote:<br class="">Hello,<br class="">Any ideas on how to debug this?<br class="">[406]PETSC ERROR: key 328214936 not found in table<br class="">[406]PETSC ERROR: See http://www.mcs.anl.gov/petsc/documentation/faq.html for trouble shooting.<br class="">[406]PETSC ERROR: Petsc Development GIT revision: v3.9.2-603-gceafe64 GIT Date: 2018-06-10 12:46:16 -0500<br class="">[406]PETSC ERROR: /ccc/work/cont003/rndm/rndm/FreeFem-sources/src/mpi/FreeFem++-mpi on a arch-linux2-c-debug-bullxmpi named curie2698 by jolivetp Sun Jun 24 07:06:07 2018<br class="">[406]PETSC ERROR: Configure options --with-scalar-type=real --with-hypre-dir=arch-linux2-c-opt-bullxmpi --with-metis-dir=arch-linux2-c-opt-bullxmpi --with-mumps-dir=arch-linux2-c-opt-bullxmpi --with-mkl_cpardiso=1 --with-mkl_pardiso=1 --with-parmetis-dir=arch-linux2-c-opt-bullxmpi --with-ptscotch-dir=arch-linux2-c-opt-bullxmpi --with-sowing-dir=arch-linux2-c-opt-bullxmpi --with-blacs-include=/ccc/products/mkl-18.0.1.163/default/18.0.1.163/mkl/include --with-blacs-lib=/ccc/products/mkl-18.0.1.163/default/18.0.1.163/mkl/lib/intel64/libmkl_blacs_openmpi_lp64.so --with-blaslapack-dir=/ccc/products/mkl-18.0.1.163/default/18.0.1.163/mkl/lib/intel64 --with-x=0 --with-debugging=1 --with-errorchecking=1 --with-mkl_pardiso-dir=/ccc/products/mkl-18.0.1.163/default/18.0.1.163/mkl --with-mkl_cpardiso-dir=/ccc/products/mkl-18.0.1.163/default/18.0.1.163/mkl --with-scalapack-include=/ccc/products/mkl-18.0.1.163/default/18.0.1.163/mkl/include --with-scalapack-lib="[/ccc/products/mkl-18.0.1.163/default/18.0.1.163/mkl/lib/intel64/libmkl_scalapack_lp64.so,/ccc/products/mkl-18.0.1.163/default/18.0.1.163/mkl/lib/intel64/libmkl_blacs_openmpi_lp64.so]" --with-fortran-bindings=0 PETSC_ARCH=arch-linux2-c-debug-bullxmpi<br class="">[406]PETSC ERROR: #1 PCGAMGHashTableFind() line 106 in /ccc/work/cont003/rndm/rndm/petsc/include/petsc/private/pcgamgimpl.h<br class="">[406]PETSC ERROR: #2 smoothAggs() line 526 in /ccc/work/cont003/rndm/rndm/petsc/src/ksp/pc/impls/gamg/agg.c<br class="">[406]PETSC ERROR: #3 PCGAMGCoarsen_AGG() line 970 in /ccc/work/cont003/rndm/rndm/petsc/src/ksp/pc/impls/gamg/agg.c<br class="">[406]PETSC ERROR: #4 PCSetUp_GAMG() line 518 in /ccc/work/cont003/rndm/rndm/petsc/src/ksp/pc/impls/gamg/gamg.c<br class="">[406]PETSC ERROR: #5 PCSetUp() line 932 in /ccc/work/cont003/rndm/rndm/petsc/src/ksp/pc/interface/precon.c<br class="">[406]PETSC ERROR: #6 KSPSetUp() line 381 in /ccc/work/cont003/rndm/rndm/petsc/src/ksp/ksp/interface/itfunc.c<br class=""><br class="">For smaller problems, this works fine. Should I compile using 64 bit indices?<br class="">I’m using -pc_type gamg -pc_gamg_threshold 0.8 -pc_gamg_agg_nsmooths 1 -pc_gamg_sym_graph true -pc_gamg_square_graph 10<br class=""><br class="">Thanks,<br class="">Pierre<br class=""></blockquote></blockquote><br class=""></div></div></blockquote></div><br class=""></div></body></html>