MatCreateSubMatrix_MPIAIJ_SameRowColDist() is a new addition to petsc for a scalable implementation of MatCreateSubMatrix_MPIAIJ(), thus it is still buggy. I'm fixing a bug appeared with IS blocksize.
I cannot reproduce this valgrind error, neither from our nightly tests. Can I have your test that shows this error? Hong On Mon, Jun 19, 2017 at 5:38 AM, Lisandro Dalcin <[email protected]> wrote: > Dear Hong, any chance this is related to your work around commit > 8b3fa1f7362a032c26473e2c80cb80f09a5a21c7 > > This test is working just fine with PETSc 3.5, 3.6, and 3.7, but fails > with petsc/master > > $ git clone https://github.com/dalcinl/petiga > $ cd petiga > $ make > $ cd test > $ make IGAInputOutput > $ mpiexec -n 4 valgrind -q ./IGAInputOutput -iga_dim 2 > ==26589== Invalid write of size 8 > ==26589== at 0x5A7B8E7: ISGetSeqIS_SameColDist_Private (mpiaij.c:3026) > ==26589== by 0x5A7CFFF: MatCreateSubMatrix_MPIAIJ_SameRowColDist > (mpiaij.c:3130) > ==26589== by 0x5A7E1DC: MatCreateSubMatrix_MPIAIJ (mpiaij.c:3243) > ==26589== by 0x5B64553: MatCreateSubMatrix (matrix.c:7894) > ==26589== by 0x4E9A704: MatView_MPI_IGA (petigamat.c:75) > ==26589== by 0x5B19527: MatView (matrix.c:986) > ==26589== by 0x402F5D: main (IGAInputOutput.c:119) > > > > > -- > Lisandro Dalcin > ============ > Research Scientist > Computer, Electrical and Mathematical Sciences & Engineering (CEMSE) > Extreme Computing Research Center (ECRC) > King Abdullah University of Science and Technology (KAUST) > http://ecrc.kaust.edu.sa/ > > 4700 King Abdullah University of Science and Technology > al-Khawarizmi Bldg (Bldg 1), Office # 0109 > Thuwal 23955-6900, Kingdom of Saudi Arabia > http://www.kaust.edu.sa > > Office Phone: +966 12 808-0459 >
