Remove carriage return in Petsc(Assert|Check)()git grep -l -E "Petsc(Assert|Check)\(.*\\\n\"[,)]" | xargs sed -r -i'' 's#Petsc(Assert|Check)\((.*)\\n"([,)])#Petsc\1\(\2\"\3#g'Suggested-by: Stefan
Remove carriage return in Petsc(Assert|Check)()git grep -l -E "Petsc(Assert|Check)\(.*\\\n\"[,)]" | xargs sed -r -i'' 's#Petsc(Assert|Check)\((.*)\\n"([,)])#Petsc\1\(\2\"\3#g'Suggested-by: Stefano Zampini
show more ...
Remove trailing whitespace
Fix double whitespace after colon
checkbadSource: rules for PetscFunctionBegin and derivatives
SNES ex15: Set particles from options
ex13: add general positions for parallel HEM
Fix "Extra spaces in test harness rules" in checkbadSource
Merge branch 'barry/2023-12-21/rm-unneeded-mansec' into 'main'Remove MANSEC from makefiles it is not needed in, tutorials and testsSee merge request petsc/petsc!7137
Remove MANSEC from makefiles it is not needed in, tutorials and tests
Tests: tweak tests to make them less sensitive to the GEMV optimization
DMPLEXSNES: support objective functions
SNESNEWTONTR: code improvementssupport QN modelsaccepts Cauchy point if it is on the boundaryEnlarge trust region only if direction is on the boundaryUse previously estimated largest eigenval
SNESNEWTONTR: code improvementssupport QN modelsaccepts Cauchy point if it is on the boundaryEnlarge trust region only if direction is on the boundaryUse previously estimated largest eigenvalue of hessian in minimum decreaseCustomize trust region normFix dogleg path computationUpdate output files and add tests
Fix PetscOptionsGetViewer() so it works in a thread-safe wayWhen PetscOptionsGetViewer() returns, for example, PETSC_VIEWER_STDOUT_SELF the same object is sharedby all threads that call PetscOptio
Fix PetscOptionsGetViewer() so it works in a thread-safe wayWhen PetscOptionsGetViewer() returns, for example, PETSC_VIEWER_STDOUT_SELF the same object is sharedby all threads that call PetscOptionsGetViewer(). In the previous code PetscObjectReference() anddereference (via PetscViewerDestroy() where called on the viewer by multiple threads after the viewerwas accessed thus corrupting the object.This branch introduces a PetscOptionsRestoreViewer() that allows not doing the referencing and dereferencingon persistant viewers, thus removing the specific race condition above.Other possible solutions includeUse a lock on each PetscObjectReference/Dereference() just in case they are shared. Could be time consuming so onecould mark each object as either threadshared or independent and only do the locks on threadshared objects. This lowersthe cost but would require more extensive changes to PETSc infrastructure. And I do not know if this shared/not shared modelis what we need in the long run, though it might be.
Plex: Fix DMInterpolateEvaluate() for FV
Fix bugs in handling PetscViewerGetSubViewer() and tabing in ASCII viewersReported-by: Pierre Jolivet
cleanup ex13 for square graph
Remove first and last empty lines
add old aggressive coarsening algorithm as option, switch default MIS ordering algorithm back to old one
Profiling: Improve !defined(PETSC_USE_LOG) #defines- If we cast macro arguments to void, we can avoid unused errors that have forced lots of conditionally included code like this:```c #if def
Profiling: Improve !defined(PETSC_USE_LOG) #defines- If we cast macro arguments to void, we can avoid unused errors that have forced lots of conditionally included code like this:```c #if defined(PETSC_USE_LOG) PetscLogEvent event; #endif```- Having global counters like petsc_TotalFlops externed event if !PetscDefined(USE_LOG) helps to remove a lot of needless `#ifdef`s around code that can instead be placed in `if (PetscDefined(USE_LOG)) {}` blocks.
matrix free in docs and comments should be matrix-free
Replace LINPACKcgtql1 with LAPACKstev for computing sym tri eigenvaluesSpecifically, removes translated Fortran and Algol routines which had non-terminating loops if assumptions were violated with
Replace LINPACKcgtql1 with LAPACKstev for computing sym tri eigenvaluesSpecifically, removes translated Fortran and Algol routines which had non-terminating loops if assumptions were violated with a standard LAPACK eigenvalue solver routine. Add new alternatives due to minor differences in eigenvalue estimates.
Add PetscCheck() and PetscCheckA() for FortranAlso fix incorrect use of " in Fortran examples that should be '
Support fp_trap in CI
SNES test ex4: disable fpe trap when computing infinity
SNES test ex69: disable fp_trap for some tests
12345678