Hello Daniel,
thanks for the observations. Interesting that the problem doesn't depend on
the hp framework.
I myself managed to debug the problem a little bit further with gdb on the
second running process (when starting with -np 2) and got the following
backtrace directly before crashing:
>>> bt
#0 Epetra_CrsGraph::MakeIndicesLocal (this=0x559811539590, domainMap=...,
rangeMap=...) at
/apps/spack/var/spack/stage/trilinos-master/packages/epetra/src/Epetra_CrsGraph.cpp:1823
#1 0x00007f3bdc4a48a2 in Epetra_CrsGraph::FillComplete
(this=0x559811539590, domainMap=..., rangeMap=...) at
/apps/spack/var/spack/stage/trilinos-master/packages/epetra/src/Epetra_CrsGraph.cpp:979
#2 0x00007f3bf4bcf089 in
dealii::TrilinosWrappers::SparsityPattern::compress (this=0x559811566ee0)
at
/apps/spack/var/spack/stage/dealii-9.1.0/source/lac/trilinos_sparsity_pattern.cc:820
#3 0x00007f3bf45a4983 in
dealii::BlockSparsityPatternBase<dealii::TrilinosWrappers::SparsityPattern>::compress
(this=0x7ffcb4c37918) at
/apps/spack/var/spack/stage/dealii-9.1.0/source/lac/block_sparsity_pattern.cc:172
#4 0x000055980fe1b3f8 in Problem_FeNothing::Demonstration<2>::setup_system
(this=0x7ffcb4c36690) at
~/Programming/problem_fe_nothing/problem_fe_nothing.cc:146
#5 0x000055980fe1953b in Problem_FeNothing::Demonstration<2>::run
(this=0x7ffcb4c36690) at
~/Programming/problem_fe_nothing/problem_fe_nothing.cc:163
#6 0x000055980fe12539 in main (argc=1, argv=0x7ffcb4c37c48) at
~/Programming/problem_fe_nothing/problem_fe_nothing.cc:183
The relevant point in the source code is at crashing time in
Epetra_CRSGraph.cpp around line 1823:
─── Source
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
1818 int GID = ColIndices[j];
1819 int LID = colmap.LID(GID);
1820 if(LID != -1)
1821 ColIndices[j] = LID;
1822 else
>>1823 throw ReportError("Internal error in FillComplete ",-1);
1824 }
1825 }
1826 }
1827 else if(RowMap().GlobalIndicesLongLong())
1828 {
the value of LID is indeed "-1" and therefore we land in line 1823 and the
error is thrown.
Since I'm really no expert in the underlying data structures in the deal.II
TrilinosWrapper and Trilinos itself, is there any way to get further help
in fixing this problem?
Thanks,
Mathias
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to the Google Groups
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/dealii/1d32e1c2-3a19-496a-bef3-a91ffbdec4ee%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
