add support for single char argument fortran bindings
Sys: better mismatched MPIU_Allreduce error diagnosisIf allreduce mismatches at different code lines, better to print out the line numbers on all processes (vs. rank 0) for diagnosis.In addition,
Sys: better mismatched MPIU_Allreduce error diagnosisIf allreduce mismatches at different code lines, better to print out the line numbers on all processes (vs. rank 0) for diagnosis.In addition, line numbers are good enough, we don't need to check petsc functions (if functions are different, lines are defintely different)Use a function instead of inlined macros to save stack space in PetscLogHandlerView_Default_Info(), where MPIU_Allreduce() is called many times.If with inlined macros, we encoutered __chkstk segfault in PetscLogHandlerView_Default_Info() with mswin-mpiuni.Inspired by !8583
show more ...
Merge remote-tracking branch 'origin/release'
Logging - update test ex68 to have nested stages and events
Remove explicitly listed empty files and switch to output/empty.out
fix 'make check' run targets [that don't use test harness] to not rely on empty output files
Fortran: Fix HDF5 attribute writing interface- Ensure real attribute writing routine is available unconditionally in real-only builds- Add new wrappers in zhdf5f.c for writing: - Integer attribu
Fortran: Fix HDF5 attribute writing interface- Ensure real attribute writing routine is available unconditionally in real-only builds- Add new wrappers in zhdf5f.c for writing: - Integer attributes - Scalar (PetscScalar) attributes - Real attributes- Update petscviewer.h90 to expose the real version consistently- Adds a minimal test program (ex80f.F90) to verify functionality- Update test case to reflect proper namesFixes: #194Thanks-to: Martin Diehladd dependency for HDF5 in testfix the c stubs to remove garbage valuesupdate the test case to reflect proper namesfix clang format
Sys: more fixes for PetscLogView()Reported-by: Volker Jacht @jac_grs
Sys: check in PetscLogView() only when in PetscFinalize()Fixes #1761Reported-by: Volker Jacht @jac_grs
PetscOptionsDeprecated: do not overwrite new option if it is provided
Fortran 90: fully embrace After 34 years!- deprecate use of 'F90' in Fortran function names- use Fortran pointers when appropriate- the new Fortran API is not backward compatible with previous ve
Fortran 90: fully embrace After 34 years!- deprecate use of 'F90' in Fortran function names- use Fortran pointers when appropriate- the new Fortran API is not backward compatible with previous versions!- also clean up inconsistent PETSc code detected by new Fortran generation tools- drop use of bfort- automatically generate all the Fortran PETSc objects, enums etc from the include files- generate most of the Fortran interface definitions and functions from the source code- simplify the number and organization of Fortran modulesCo-authored-by: Jose E. Roman <jroman@dsic.upv.es>
CI: enable compilation of examples in some stage-1 jobs
Fix MPIU_* routines to always return MPI error codes and thus be usable with PetscCallMPI()Introduce MPIU_Count to be used when MPI_Count is not available
Add PetscCallExternalAbort()
Add length argument to PetscMPIErrorString() to prevent potential array overflowAdd test to loop over MPI error codes. Reveals known bug in MPICH implementation
Add shared memory support for PCMPI using both shmget() and MPI_Win_allocate_shared()
The use case of PetscOptionsGetViewer() has evolved to no longer be the get/restore modelHence the code needs to be refactoredIntroduce PetscCallNull() and PetscCallMPINull() to simplify PETSC_VI
The use case of PetscOptionsGetViewer() has evolved to no longer be the get/restore modelHence the code needs to be refactoredIntroduce PetscCallNull() and PetscCallMPINull() to simplify PETSC_VIEWER_XXXX_() routinesReported-by: James WrightThanks-to: Jeremy L Thompson
Document that PetscViewerGetSubViewer() is collective on the outer viewer and fix bugs.Reported-by: Pierre Jolivet
Use strerror(errno) on fprintf() when possible to provide more detail on why the system call failed
docs: misc fixes
Profiling: Create PETSCLOGHANDLERNVTX
Merge branch 'tisaac/feature-no-exit-traceback' into 'main'Sys: Do not exit in PetscTraceBackErrorHandler(); use TLS to silence non-root tracebacksSee merge request petsc/petsc!6846
Profiling: Add test of dropped PetscLogEventEnd()
Sys: Do not exit in PetscTraceBackErrorHandler(); use TLS to silence non-root tracebacks
Profiling: add sys/tests/ex71.c
123