<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Jan 11, 2017 at 3:55 PM, Barry Smith <span dir="ltr"><<a href="mailto:bsmith@mcs.anl.gov" target="_blank">bsmith@mcs.anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="gmail-"><br>
> On Jan 11, 2017, at 2:41 PM, Kong, Fande <<a href="mailto:fande.kong@inl.gov">fande.kong@inl.gov</a>> wrote:<br>
><br>
> We do not run the code with Satish's fixes, instead, we are using petsc-3.7.3 and there are not such issues. I guess the issues are introduced recently.<br>
<br>
</span> Ok, then I am confused. Maybe master fixes the problem?<br>
<div class="gmail-HOEnZb"><div class="gmail-h5"><br></div></div></blockquote><div><br></div><div>It looks like master still has 'aN=A->cmap->N' and </div><div><br></div><div>ierr = PetscTableCreate(aN,aN,&ta);CHKERRQ(ierr);<br></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 class="gmail-HOEnZb"><div class="gmail-h5">
<br>
><br>
> Fande,<br>
><br>
> On Wed, Jan 11, 2017 at 1:31 PM, Mark Adams <<a href="mailto:mfadams@lbl.gov">mfadams@lbl.gov</a>> wrote:<br>
> Argh, yes that has to be the local size or a least not the global size.<br>
><br>
> On Wed, Jan 11, 2017 at 3:15 PM, Satish Balay <<a href="mailto:balay@mcs.anl.gov">balay@mcs.anl.gov</a>> wrote:<br>
> >>><br>
> aN=A->cmap->N<br>
> ierr = PetscTableCreate(aN,aN,&ta);<wbr>CHKERRQ(ierr);<br>
> <<<<br>
><br>
> Looks like thats a 'global' size. That doesn't look right.<br>
><br>
> <a href="https://bitbucket.org/petsc/petsc/commits/c373ccc6#chg-src/mat/impls/aij/mpi/mpimatmatmult.c" rel="noreferrer" target="_blank">https://bitbucket.org/petsc/<wbr>petsc/commits/c373ccc6#chg-<wbr>src/mat/impls/aij/mpi/<wbr>mpimatmatmult.c</a><br>
><br>
> I see Hong had previously updated this code. Will have to recheck..<br>
><br>
> Satish<br>
><br>
> On Wed, 11 Jan 2017, Mark Adams wrote:<br>
><br>
> > ><br>
> > ><br>
> > > Must be the same place Fande encountered this error.<br>
> > ><br>
> > ><br>
> > Satish, here is our stack trace. Do you have any thoughts vis-a-vis Fande's<br>
> > experience what the problem is here? My best guess is that the local<br>
> > matrix size is huge, but this code can have some very large stencils.<br>
> ><br>
> > Treb: about how many processors have this error? Is is it just one, a few,<br>
> > or tons?<br>
> ><br>
> > [5]PETSC ERROR: Argument out of range<br>
> > [5]PETSC ERROR: A really huge hash is being requested.. cannot process:<br>
> > 67108864<br>
> > [5]PETSC ERROR: See <a href="http://www.mcs.anl.gov/petsc/documentation/faq.html" rel="noreferrer" target="_blank">http://www.mcs.anl.gov/petsc/<wbr>documentation/faq.html</a> for<br>
><br>
> > trouble shooting.<br>
> > [5]PETSC ERROR: Petsc Development GIT revision: v3.7.4-2584-g7175242 GIT<br>
> > Date: 2016-12-21 13:19:16 -0600<br>
> > [5]PETSC ERROR: /tmp/targetExec on a arch-xc30-dbg64-intel named nid00013<br>
> > by u6338 Sat Jan 7 18:35:34 2017<br>
> > [5]PETSC ERROR: Configure options --COPTFLAGS="-O0 -g -no-ipo"<br>
> > --CXXOPTFLAGS="-O0 -g -no-ipo" --FOPTFLAGS="-O0 -g -ipo"<br>
> > --download-parmetis --download-metis --download-hypre --with-ssl=0<br>
> > --with-cc=cc --with-clib-autodetect=0 --with-cxx=CC<br>
> > --with-cxxlib-autodetect=0 --with-debugging=1 --with-fc=ftn<br>
> > --with-fortranlib-autodetect=0 --with-shared-libraries=0 --with-x=0<br>
> > --with-mpiexec=srun LIBS=-lstdc++ --with-64-bit-indices<br>
> > --prefix=/global/homes/m/<wbr>madams/petsc_install/petscv3.<wbr>7-xc3[1983]PETSC<br>
> > ERROR: --------------------- Error Message ------------------------------<br>
> > ------------------------------<wbr>--<br>
> ><br>
> > [5]PETSC ERROR: [522]PETSC ERROR: --------------------- Error Message<br>
> > ----------<br>
> ><br>
> > [5]PETSC ERROR: #2 PetscTableCreate() line 52 in<br>
> > /global/u2/m/madams/petsc_<wbr>install/petsc/src/sys/utils/<wbr>ctable.c<br>
> > [5]PETSC ERROR: #3 MatTransposeMatMultSymbolic_<wbr>MPIAIJ_MPIAIJ() line 1691 in<br>
> > /global/u2/m/madams/petsc_<wbr>install/petsc/src/mat/impls/<wbr>aij/<br>
> > mpi/mpimatmatmult.c<br>
> > [5]PETSC ERROR: #4 MatTransposeMatMult_MPIAIJ_<wbr>MPIAIJ() line 917 in<br>
> > /global/u2/m/madams/petsc_<wbr>install/petsc/src/mat/impls/<wbr>aij/<br>
> > mpi/mpimatmatmult.c<br>
> > [5]PETSC ERROR: [608]PETSC ERROR: --------------------- Error Message<br>
> > ----------<br>
> ><br>
> > [5]PETSC ERROR: #6 PCGAMGCoarsen_AGG() line 955 in<br>
> > /global/u2/m/madams/petsc_<wbr>install/petsc/src/ksp/pc/<wbr>impls/gamg/agg.c<br>
> > [5]PETSC ERROR: #7 PCSetUp_GAMG() line 524 in /global/u2/m/madams/petsc_inst<br>
> > all/petsc/src/ksp/pc/impls/<wbr>gamg/gamg.c<br>
> > [5]PETSC ERROR: #8 PCSetUp() line 968 in /global/u2/m/madams/petsc_inst<br>
> > all/petsc/src/ksp/pc/<wbr>interface/precon.c<br>
> > [5]PETSC ERROR: #9 KSPSetUp() line 393 in /global/u2/m/madams/petsc_inst<br>
> > all/petsc/src/ksp/ksp/<wbr>interface/itfunc.c<br>
> > [5]PETSC ERROR: [1640]PETSC ERROR: --------------------- Error Message -----<br>
> ><br>
> ><br>
> > > MatSetUpMultiply_MPIAIJ() does:<br>
> > ><br>
> > > ierr = PetscTableCreate(aij->B->rmap-<wbr>>n,mat->cmap->N+1,&gid1_lid1);<br>
> > > CHKERRQ(ierr);<br>
> > ><br>
> > > So the local matrix size (n) must be: 4294967296<br>
> > ><br>
> > > [assuming atleast one nonzero per row - the local matrix itself must use<br>
> > > more memory than the hash-table..]<br>
> > ><br>
> > > Satish<br>
> > ><br>
> ><br>
><br>
><br>
><br>
<br>
</div></div></blockquote></div><br></div></div>