There appear to be several examples with "the wrong partitioning" in the output files? I think you need to list in the args: the partitioning you want otherwise it depends on what you currently have installed; which is worthless for regression tests. See attached: all the different iteration counts are due to partitioning. Barry
On Feb 6, 2017, at 7:20 PM, Matthew Knepley <[email protected]> wrote:
On Mon, Feb 6, 2017 at 6:30 PM, Barry Smith <[email protected]> wrote:
Matt,
How come? I check and it is using the simple partitioner and no random number generator is called. How come when I run it it gives a different partitioning than in the output file? The vectors also have different partitionings. Are the output files out of date?
In my branch, ex15 has not had its partitioning replaced. It is still using ParMetis.
I can do it.
Matt
~/Src/petsc/src/dm/impls/plex/examples/tests (knepley/feature-plex-tests-new=) arch-double $ petscmpiexec -n 2 ./ex15 -verbose -globaltonatural_sf_view > joe ~/Src/petsc/src/dm/impls/plex/examples/tests (knepley/feature-plex-tests-new=) arch-double $ paste output/ex15_0.out joe | more PetscSF Object: 2 MPI processes PetscSF Object: 2 MPI processes type not yet set type not yet set [0] Number of roots=6, leaves=6, remote ranks=1 [0] Number of roots=4, leaves=4, remote ranks=2 [0] 0 <- (1,0) [0] 0 <- (0,0) [0] 1 <- (1,1) [0] 1 <- (0,1) [0] 2 <- (1,2) [0] 2 <- (1,0) [0] 3 <- (1,3) [0] 3 <- (1,1) [0] 4 <- (1,4) [1] Number of roots=14, leaves=14, remote ranks=2 [0] 5 <- (1,5) [1] 0 <- (1,2) [1] Number of roots=12, leaves=12, remote ranks=2 [1] 1 <- (1,3) [1] 0 <- (1,6) [1] 2 <- (0,2) [1] 1 <- (1,7) [1] 3 <- (0,3) [1] 2 <- (1,8) [1] 4 <- (1,4) [1] 3 <- (1,9) [1] 5 <- (1,5) [1] 4 <- (0,0) [1] 6 <- (1,6) [1] 5 <- (0,1) [1] 7 <- (1,7) [1] 6 <- (1,10) [1] 8 <- (1,8) [1] 7 <- (1,11) [1] 9 <- (1,9) [1] 8 <- (0,2) [1] 10 <- (1,10) [1] 9 <- (0,3) [1] 11 <- (1,11) [1] 10 <- (0,4) [1] 12 <- (1,12) [1] 11 <- (0,5) [1] 13 <- (1,13) ==== original V in global ordering. size==18 block size=2 ==== original V in global ordering. size==18 block size=2 Vec Object: V 2 MPI processes Vec Object: V 2 MPI processes type: mpi type: mpi Process [0] Process [0]
-- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener