| #
c5eaf5f0
|
| 14-Feb-2017 |
Matthew G. Knepley <knepley@gmail.com> |
Merge branch 'knepley/feature-plex-tests-new' into knepley/feature-fe-dual-apply
* knepley/feature-plex-tests-new: (73 commits)
Plex: Fix overeager protection of function
Travis-CI: Attempt to s
Merge branch 'knepley/feature-plex-tests-new' into knepley/feature-fe-dual-apply
* knepley/feature-plex-tests-new: (73 commits)
Plex: Fix overeager protection of function
Travis-CI: Attempt to solve issues with gfortran
Travis-CI: Use newer macOS image
clarify damping and other issues related to SNESLINESEARCHBASIC
change example in new harness to use simple partitioner so gets consistent output on all systems
remove unused function from test file
cleanup formatting in example, plus myrank not rank
all requires need to go on the same line in test definition otherwise they are skipped
missing - makes the manual page formatting wrong
The developer manual clearly states that MPI_Comm_size() returns size it doesn't return numProcs or other nonsense. Pretty soon you'll start using mype instead of rank?
fix typo and cut and paste error in docs
bugfix for !HYPRE
fix warning for ';;'
may switch to scalable only when B->cmap->N > 100000
MatLRC: store in the context the parallel matrices rather than the sequential ones
Bib: Add ref
set default algorithm based on B->cmap->N for MatMatMult_MPIAIJ_MPIAIJ()
Error in VecScatterView_MPI() had the meaning of from and to backwards
fix build warning as reported by the nightly tests
Update alquimia.py to use Glenn's branch that updates alquimia for pflotran xsdk
...
show more ...
|
| #
cffa9059
|
| 14-Feb-2017 |
Matthew G. Knepley <knepley@gmail.com> |
Merge branch 'master' into knepley/feature-plex-tests-new
* master: (94 commits)
Plex: Fix overeager protection of function
Travis-CI: Attempt to solve issues with gfortran
Travis-CI: Use newe
Merge branch 'master' into knepley/feature-plex-tests-new
* master: (94 commits)
Plex: Fix overeager protection of function
Travis-CI: Attempt to solve issues with gfortran
Travis-CI: Use newer macOS image
clarify damping and other issues related to SNESLINESEARCHBASIC
change example in new harness to use simple partitioner so gets consistent output on all systems
remove unused function from test file
cleanup formatting in example, plus myrank not rank
missing - makes the manual page formatting wrong
The developer manual clearly states that MPI_Comm_size() returns size it doesn't return numProcs or other nonsense. Pretty soon you'll start using mype instead of rank?
fix typo and cut and paste error in docs
bugfix for !HYPRE
fix warning for ';;'
may switch to scalable only when B->cmap->N > 100000
MatLRC: store in the context the parallel matrices rather than the sequential ones
Bib: Add ref
set default algorithm based on B->cmap->N for MatMatMult_MPIAIJ_MPIAIJ()
Error in VecScatterView_MPI() had the meaning of from and to backwards
fix build warning as reported by the nightly tests
Update alquimia.py to use Glenn's branch that updates alquimia for pflotran xsdk
fix divide by zero exposed by 8534686b5096cd4c91a8f3895408b386c0628431 - and produced by petsc4py testsuite
...
Conflicts:
src/dm/impls/plex/examples/tests/ex13.c
src/dm/impls/plex/plexgmsh.c
src/snes/examples/tests/ex2.c
src/snes/examples/tests/output/ex2_3.out
src/snes/examples/tests/output/ex2_4.out
src/snes/examples/tests/output/ex2_5.out
show more ...
|
| #
0d0f1ffd
|
| 08-Feb-2017 |
Scott Kruger <scott.e.kruger@gmail.com> |
Merge branch 'master' into scott/test-harness-forloopfix
|
| #
31ae5339
|
| 07-Feb-2017 |
Satish Balay <balay@mcs.anl.gov> |
fix divide by zero exposed by 8534686b5096cd4c91a8f3895408b386c0628431 - and produced by petsc4py testsuite
|
| #
809b304d
|
| 26-Jan-2017 |
Barry Smith <bsmith@mcs.anl.gov> |
Merge branch 'master' of bitbucket.org:petsc/petsc
|
| #
96fc66b7
|
| 26-Jan-2017 |
Matthew G. Knepley <knepley@gmail.com> |
Merge remote-tracking branch 'origin/test-harness-replacefix' into knepley/feature-plex-tests-new
* origin/test-harness-replacefix:
Enable replace when file does not exist
Regenerate test harnes
Merge remote-tracking branch 'origin/test-harness-replacefix' into knepley/feature-plex-tests-new
* origin/test-harness-replacefix:
Enable replace when file does not exist
Regenerate test harness if sources change
bugfix for MatGetSubmatrices_MPIAIJ() Reported-by: Matthew Knepley
show more ...
|
| #
03e8432d
|
| 25-Jan-2017 |
Matthew G. Knepley <knepley@gmail.com> |
Merge branch 'master' into knepley/feature-plex-tests-new
* master: (37 commits)
fortran: include petscversion.h
test: c++ pipeline build fix for d2a797cfef64a79a330b420f01b2d605a8acc690
Extra
Merge branch 'master' into knepley/feature-plex-tests-new
* master: (37 commits)
fortran: include petscversion.h
test: c++ pipeline build fix for d2a797cfef64a79a330b420f01b2d605a8acc690
Extract the submatrices
Convert C Null pointer to Fortran in Fortran stub before passing back to user
PC Galekin and KSP should properly propagate calls to SetFromOptions to subsolvers
Converted Fortran PetscSF to Fortran type, added PETSC_NULL_SF
clarification in documentation
Update of petscdiff and associated harness hooks
Fixup more F90 interface definitions for users
More space-in-dirname enabling quotes
Enable paths with spaces
add cast; bugfix for 64bit
Typo -- initialized TESTFLAGS to -v by mistake
Adding new options to test framework
gtest: add valgrind support [when PETSC_ARCH has 'valgrind' in the name]
bugfix for unsorted ISCols
optimize matgetsubmatrices_mpiaij to use only one MPI_AllReduce
small cleanup MatGetSubMatrices_MPIAIJ_Local()
rm allcolumns from the input of MatGetSubmatrices_mpiaij_local()
add a test to ex4.c for reuse
...
show more ...
|
| #
e8e596da
|
| 25-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
Merge branch 'hzhang/fix-matgetsubmatrices-mpiaij'
|
| #
77b65b27
|
| 23-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
bugfix for MatGetSubmatrices_MPIAIJ()
Reported-by: Matthew Knepley
|
| #
dc84b4d1
|
| 20-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
Merge branch 'hzhang/opt-matgetsubmatrices-mpiaij'
|
| #
ec945f66
|
| 19-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
add cast; bugfix for 64bit
|
| #
f45aa8ac
|
| 18-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
bugfix for unsorted ISCols
|
| #
85f27616
|
| 17-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
optimize matgetsubmatrices_mpiaij to use only one MPI_AllReduce
|
| #
afd2212b
|
| 17-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
small cleanup MatGetSubMatrices_MPIAIJ_Local()
|
| #
a1e6e1df
|
| 17-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
rm allcolumns from the input of MatGetSubmatrices_mpiaij_local()
|
| #
89e74052
|
| 16-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
add a test for MatGetSubMatrix_XXX_All(); bugfix for noctable
|
| #
8534686b
|
| 12-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
reduce 2 MPI_Allreduce() in MatGetSubMatrices_MPIAIJ to one for each call
|
| #
93f4724e
|
| 12-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
rm 'sort the rows' because each row is obtained from C without changing column ordering
|
| #
553b068b
|
| 12-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
cleanup
|
| #
a46399cf
|
| 11-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
recover 'sort the rows' -- not sure if it is needed here, will check it!
|
| #
30882ffb
|
| 11-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
cleanup; fix indent
|
| #
b360545f
|
| 10-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
set error flag for (MAT_REUSE_MATRIX && !ismax)
|
| #
70de0b6f
|
| 10-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
replace previous error-prone #commit with this one: rm idex and idex2
|
| #
2d655906
|
| 09-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
reuse works now
|
| #
40ad5659
|
| 08-Jan-2017 |
Hong Zhang <hzhang@mcs.anl.gov> |
add id to Mat_Sub; works for the case of initial_matrix, not for reuse yet
|