| #
51bfd985
|
| 12-Jan-2022 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'barry/2021-12-21/fix-mpi-comm-dup-packages' into 'main'
Stash communicators previously used for external packages so that they may be reused.
See merge request petsc/petsc!4669
|
| #
57f21012
|
| 12-Jan-2022 |
Barry Smith <bsmith@mcs.anl.gov> |
Stash communicators previously used for external packages so that they may be reused.
Previously we used MPI_Comm_dup() and free() but for some MPI implementations including MPI Spectrum the MPI_Com
Stash communicators previously used for external packages so that they may be reused.
Previously we used MPI_Comm_dup() and free() but for some MPI implementations including MPI Spectrum the MPI_Comm_free() does not actually return the communicator so eventually they run out.
Commit-type: portability-fix
\spend 3h
Reported-by: Elias Karabelas <elias.karabelas@uni-graz.at>
show more ...
|
| #
59d34cac
|
| 10-Dec-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'adams/swarm-omp-pc' into 'main'
Adams/swarm omp pc
See merge request petsc/petsc!4620
|
| #
a52b435b
|
| 06-Dec-2021 |
Mark Adams <524115-markadams4@users.noreply.gitlab.com> |
PetscPragmaOMP
|
| #
c043d5d4
|
| 29-Nov-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/release'
|
| #
0569d543
|
| 29-Nov-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'balay/fix-PetscPragmaSIMD' into 'release'
PetscPragmaSIMD: fix for mingw/clang
See merge request petsc/petsc!4607
|
| #
e5518150
|
| 28-Nov-2021 |
Satish Balay <balay@mcs.anl.gov> |
PetscPragmaSIMD: fix for mingw/clang
|
| #
79982354
|
| 16-Nov-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'jacobf/2021-11-02/attribute-format-sys' into 'main'
Feature: PETSC_ATTRIBUTE_FORMAT - sys
See merge request petsc/petsc!4520
|
| #
3ca90d2d
|
| 15-Nov-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
apply PETSC_ATTRIBUTE_FORMAT and related fixes to sys sources
|
| #
33bb201b
|
| 15-Nov-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
Add PetscHasAttribute and PETSC_ATTRIBUTE_FORMAT macros
|
| #
c93fae50
|
| 15-Nov-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
move PETSc format specifiers to petscsystypes.h so they are defined as early as possible, and introduce PetscBLASInt_FMT, PETSC_MPI_COMM_FMT, and PETSC_MPI_WIN_FMT
|
| #
687f7014
|
| 04-Nov-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'barry/2021-11-02/rm-petsc-version-patch' into 'main'
Remove deprecated PETSC_VERSION_PATCH so as to not have confusion with patch...
See merge request petsc/petsc!4518
|
| #
cbc12506
|
| 02-Nov-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Remove deprecated PETSC_VERSION_PATCH so as to not have confusion with patch releases where the subminor version changes
Commit-type: documentation
\spend 5m
|
| #
08e3941c
|
| 25-Oct-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/release'
|
| #
6a490683
|
| 22-Oct-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'jczhang/fix-compilation-warnings' into 'release'
Fix some compilation warnings seen in users email
See merge request petsc/petsc!4465
|
| #
0a996781
|
| 21-Oct-2021 |
Junchao Zhang <jczhang@mcs.anl.gov> |
Config: also test CUDA C++ dialect for PETSC_NODISCARD
|
| #
ae77d22f
|
| 16-Oct-2021 |
Junchao Zhang <jczhang@mcs.anl.gov> |
CUDA: fix a type name typo in PetscCuBLASIntCast
|
| #
0baf1215
|
| 20-Oct-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'jacobf/2021-09-04/petscdevice-cupminitialization' into 'main'
Replace current CUPM initialization with unified PetscDevice version
See merge request petsc/petsc!4310
|
| #
a4af0cee
|
| 19-Oct-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
- Replace the current CUPM initialization infrastructure with PetscDevice.
- Remove -petsc_use_default_null_stream and -[cuda|hip]_synchronize options
- Remove PetscCUDASynchronize and PetscHIPSynchr
- Replace the current CUPM initialization infrastructure with PetscDevice.
- Remove -petsc_use_default_null_stream and -[cuda|hip]_synchronize options
- Remove PetscCUDASynchronize and PetscHIPSynchronize
- Remove PetscCUDAInitialize(), PetscCUDAInitializeCheck(), PetscHIPInitialize(), and
PetscHIPInitializeCheck()
- Remove PetscCUBLASInitializeHandle(), PetscCUSOLVERDnInitializeHandle(),
PetscHIPBLASInitializeHandle(), and PetscHIPSOLVERInitializeHandle().
- Remove petsc_gputimer_begin and petsc_gputimer_begin
- Add -device_enable, -device_select and -device_view options
- Replace -[cuda|hip]_device with split options -device_enable_[cuda|hip] and -device_select_[cuda|hip]
- Replace -[cuda|hip]_view with -device_view_[cuda|hip]
- Add PetscDeviceInitType
- Add PetscDeviceInitialize() and PetscDeviceInitialized()
- Add PetscDeviceView()
show more ...
|
| #
0609f53b
|
| 19-Oct-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
- Add PETSC_CONSTEXPR_[14|17] which are PETSC_CONSTEXPR if PETSc is configured with c++14
or c++17 support or later and empty otherwise
|
| #
193d546d
|
| 19-Oct-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
- Add PETSC_SINGLE_LIBRARY_INTERN which is PETSC_INTERN if building PETSc as a single
library and PETSC_EXTERN otherwise
|
| #
c7faf32e
|
| 19-Oct-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
- Add macro indirection layer to PetscDefined()
|
| #
3972cb20
|
| 19-Oct-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
- Add devicePackage attribute to all device-related libraries, which determines if
PETSC_HAVE_DEVICE should be defined
- Remove PETSC_HAVE_DEVICE definition from petscsys.h
|
| #
af9c4ebf
|
| 15-Oct-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'stefanozampini/fix-nec-release' into 'main'
NEC: few fixes
See merge request petsc/petsc!4417
|
| #
961fb248
|
| 05-Oct-2021 |
Stefano Zampini <stefano.zampini@gmail.com> |
NEC: few fixes
Configure: fix parsing of venumainfo
Remove misleading MPIEXECEXECUTABLE macro from petscvariables
parse NECMPI version from MPIEXEC
It appears they have fixed the MPI_Type_free issue
NEC: few fixes
Configure: fix parsing of venumainfo
Remove misleading MPIEXECEXECUTABLE macro from petscvariables
parse NECMPI version from MPIEXEC
It appears they have fixed the MPI_Type_free issue but many optimization faults persist even with 3.3.1
Disable O2 and use O1
show more ...
|