Commits

Show all
Author Commit Message Labels Comments Date
Matt Knepley
Fix for complex+boost
Jed Brown
Remove erroneous call to PetscObjectProcessOptionsHandlers from PetscBagSetFromOptions, bug was introduced in fdc7be12de37.
Lisandro Dalcin
setup.py: change version format
Lisandro Dalcin
setup.py: change version format
Jed Brown
Fix typo in man page, #95584
Satish Balay
#95342 fix compiler warning
Matt Knepley
Fixed F90 complex support
Matt Knepley
Documentation fix
Matt Knepley
Documentation fix
Satish Balay
Increase patchlevel to 3.2-p5
Jed Brown
Fix MatIncreaseOverlap (-pc_type asm) with MPIAIJ and bs>1 by reverting PetscLayoutFindOwner to original API.
Jed Brown
Fix info printing, args did not match format string
Jed Brown
Fix MatInvertBlockDiagonal_SeqAIJ, the old code produced the transpose so it was worthless for nonsymmetric problems
Jed Brown
Fix more C@*/ mistakes
Jed Brown
Add PetscObjectCopyFortranFunctionPointers() and fix DMRefine() and DMCoarsen() to forward fortran function pointers. fixes #90287
Jed Brown
[petsc-maint #91419] Bug in MUMPS build script (John Fettig)
Satish Balay
Increase patchlevel to 3.2-p4
Satish Balay
fix formatted comment
BarryFSmith
added DMSetFunction, DMSetInitialGuess, DMSetJacobian custom Fortran stubs and test example
ecoon
Fixes to python PetscBinaryIO to allow multiple precision, complex scalars, and 64 bit indices. Also adds crufty petsc_conf.py which parses petscconf.h and petscvariables to find current PETSC_DIR's settings
Jed Brown
cmakeboot.py: refuse to use CMake version < 2.6.2, fall back to legacy
Jed Brown
Zero the array pointer in VecRestoreArray and VecRestoreArrayRead
Jed Brown
Backport CMake version check for pre-2.8 with clanguage=C++
Jose E. Roman
Protect missing Lapack function (this was already fixed in petsc-dev)
Lisandro Dalcin
Ignore options MAT_KEEP_NONZERO_PATTERN for seq/mpi dense matrix types
Lisandro Dalcin
Fix option name in MatOptions list
Lisandro Dalcin
Fallback to set local to global mappings in DMGetMatrix_DA
Lisandro Dalcin
Missing ISDestroy() in MatGetSubMatrix_MPIDense()
ecoon
bug fix in when indices get altered to look like 2D for 3D das with 1 node in the z-direction
BarryFSmith
updated faq for more details on computing singular values when solving linear systems
  1. Prev
  2. Next