| 3f81ab83 | 26-Jun-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Check that SuperLU_DIST integer size matches PETSc indices size
Commit-type: error-checking, configure \spend 30m Reported-by: Drew Parsons (@RizzerAtGitLab) |
| d90cfc44 | 23-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/jczhang/fix-kokkos-test'
Fix some config issues wrt Kokkos
See merge request petsc/petsc!4056 |
| 84c7a2fd | 02-Jun-2021 |
Pierre Jolivet <pierre@joliv.et> |
configure: HYPRE 2.22.0 |
| cc82f273 | 21-Jun-2021 |
Junchao Zhang <jczhang@mcs.anl.gov> |
Config: also check -Wl to include rpath in MPICXX_LIBS |
| 4715b3ca | 21-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/balay/add-valgrind.h'
valgrind.h: add to petsc repo
See merge request petsc/petsc!4094 |
| bf113f49 | 21-Jun-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
fix pragma poison and introduce new extraCompilerFlags ctx manager |
| 6edef35e | 14-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
valgrind.h: add to petsc at include/private/valgrind |
| 98e73e17 | 11-Jun-2021 |
Pierre Jolivet <pierre@joliv.et> |
MatHtool: new option -mat_htool_clustering |
| a000ad5a | 12-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
mpi4py: add with-mpi=1, --with-mpi4py-dir options
mpi4py: set PETSC_MPI4PY_PYTHONPATH etc in configureLibrary() - not in Install() |
| 03d0e64f | 11-Jun-2021 |
Barry Smith <bsmith@petsc.dev> |
configure: print in summary when mpich or openmpi is used |
| e0f85f4d | 08-Jun-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Add configure support for GSL and example that utilizes GSL
\spend 4h |
| 087fd826 | 12-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/release' |
| 26535a54 | 10-Jun-2021 |
Junchao Zhang <jczhang@anl.gov> |
Config: error out if mumps is used with 64-bit BLAS indices |
| d15db81b | 10-Jun-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Add the ability to not remove the directory when printing with logPrintBox()
This is used by mpi4py to print the needed PYTHONPATH argument
Commit-type: configure \spend 20m Thanks-to: Satish Balay
Add the ability to not remove the directory when printing with logPrintBox()
This is used by mpi4py to print the needed PYTHONPATH argument
Commit-type: configure \spend 20m Thanks-to: Satish Balay <sbalay@mcs.anl.gov>
show more ...
|
| 504e95fa | 27-May-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Change --download-mpi4py to install immediately instead of waiting until PETSc is built
This is to allow other external packages that depend on mpi4py to be built immediately instead of also waiting
Change --download-mpi4py to install immediately instead of waiting until PETSc is built
This is to allow other external packages that depend on mpi4py to be built immediately instead of also waiting until PETSc is built
Fix the messages regarding PYTHONPATH and the mpi4py make check to work with prefix installs
Commit-type: configure, enhancement Funded-by: Brain project Project: Brain \spend 1.5h
show more ...
|
| 042217e8 | 10-Jun-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
MatSetValuesDevice: Cleanup and simplify code, including example
User reported crash of example code. Kernel was passed an ierr that lived in CPU memory
MatSetValuesDevice: do not include private h
MatSetValuesDevice: Cleanup and simplify code, including example
User reported crash of example code. Kernel was passed an ierr that lived in CPU memory
MatSetValuesDevice: do not include private headers from public headers
Feature: MatSetValuesDevice determines automatically from the context (where it is included from) if it is being used from C, CUDA, or Kokkos, PETSC_DEVICE_FUNC_DEC no longer needs to be set before including petscaijdevice.h
Feature: MatSetValuesDevice() now ignores all values outside the global column range.
PetscSplitCSRDataStructure is now a pointer, not a struct, like most PETSc objects, please leave it that way.
Fix all uses of CTABLE that were related to the original MatSetValuesDevice()
Have atomicAdd use Kokkos atomic-add with CPU build when building with Kokkos.
Cuda should now work with --download-openmpi, this is done by updating updateCompilers() to rerun portions of packages/cuda.py after the compilers are reset to use MPI wrappers. This is needed because the resetting of the compilers removes all the compiler flags and packages/cuda.py sets certain values into these flags that was previously lost.
Add MPICXX_INCLUDES, MPICXX_LIBS to fix compile targets for Kokkos examples
'make check' now runs properly for Kokkos test of src/snes/ex3k, fixed bug in the makefile wrt MPI_IS_MPIUNI check
Testing makefile rules: add ex*cu binaries to clean rule
Reported-by: Sam Fagbemi <samkorede24@gmail.com> Thanks-to: Stefano Zampini <stefano.zampini@gmail.com> Thanks-to: Mark Adams <mfadams@lbl.gov>
/spend 16h
show more ...
|
| 71c41f4f | 14-May-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Add sowing doctext -Wargdesc argument to error on missing argument descriptions
Save sowing doctext output to file so it can be checked after all manualpages are built
Update sowing URl to retreive
Add sowing doctext -Wargdesc argument to error on missing argument descriptions
Save sowing doctext output to file so it can be checked after all manualpages are built
Update sowing URl to retreive Sowing with additional error checking
Still to do is compare the name of the manual page with the name of the function
Commit-type: error-checking, docs /spend 35m
show more ...
|
| 647d62ba | 08-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/release' |
| 029e6311 | 07-Jun-2021 |
Pierre Jolivet <pierre@joliv.et> |
configure: missing filter rule
ld: skipping incompatible /soft/gcc-7.5.0/lib/libcilkrts.so when searching for -lcilkrts |
| 1bde7f05 | 04-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'barry/2021-05-26/debug-o0' into 'main'
Pass -O0 in when -with-debugging is on for configure to make the code truly debugable.
See merge request petsc/petsc!4030 |
| ad67e154 | 04-Jun-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Magma seems to use both NVCCFLAGS and DEVCFLAGS which causes repeat of -g -O0 and complaints on a CI machine
\spend 20m |
| 2cc4dfa6 | 27-May-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Update -with-debugging=1 defaults for COPTFLAGS, FOPTFLAGS, CXXOPTFLAGS to use -O0 (or -Od for windows compilers) along with -g to make the code truly debuggable.
Commit-type: debuggability \spend 1
Update -with-debugging=1 defaults for COPTFLAGS, FOPTFLAGS, CXXOPTFLAGS to use -O0 (or -Od for windows compilers) along with -g to make the code truly debuggable.
Commit-type: debuggability \spend 10m
show more ...
|
| b9bb70c4 | 03-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'barry/2021-05-26/handle-dylib-like-so' into 'main'
Handle .dylib files like .so when generating argument lists
See merge request petsc/petsc!4025 |
| 5c6407d7 | 03-Jun-2021 |
Vaclav Hapla <vaclav.hapla@erdw.ethz.ch> |
packages/hdf5.py: disable --download-hdf5-fortran-bindings by default |
| 51664b96 | 02-Jun-2021 |
Satish Balay <balay@mcs.anl.gov> |
gitcommit: follow-up to 3c43290cbaa09cbcaf486ab9c4f7393eb19a428c for changes in main |