bugfix for -pc_factor_shift_type nonzero.(see [petsc-maint #45421] Problem with ICC on SEQBAIJ)Hg-commit: c6c6f641c1bbfbd11575aaba0155c7b59a657064
general cleanup of petscsys.hremoved PetscStrfree() since PetscFree() does the same thingHg-commit: 383ece197d31b92016b12ff3a62a2e744f8d981d
changed cookie to classid cause that is what it is.Hg-commit: ffa25d1edea7afc409fcc13e7f2b63587b1494c1
[petsc-maint #44397] seg fault in PCView_Factor in PETSc-3.1 PCView_Factor used factored matrix which doesn't exist before factorization hence crash. Fix by adding field to PC_Factor containing the m
[petsc-maint #44397] seg fault in PCView_Factor in PETSc-3.1 PCView_Factor used factored matrix which doesn't exist before factorization hence crash. Fix by adding field to PC_Factor containing the matfactortypefixed minor typo in manual page for PetscLogEventBegin()Hg-commit: 4921590835475b04d3f3e57015a8d9ec589593a4
show more ...
updated MatFactorInfo to use a PetscReal for shifttype so that it can be used from Fortran and PythonHg-commit: 6c65af8d77207f21d4c5751ca2d8dacb3c75581f
Factor same statements of PCView_LU/ILU/ICC/Cholesky() into PCView_Factor() for reducing code duplicationHg-commit: 839497342f67ab6feffce0c4a6c0661ea44d6070
continue replacing PCFactorSetShiftxxx() with PCFactorSetShiftType() and PCFactorSetShiftAmount()Hg-commit: b64c17c9fbc009eefb8072446fc4ef9e80146566
replace shiftpd, shiftnz, shiftinblocs with shifttype and shiftamountonly partially done for seqaij format.Hg-commit: e8ba899db4d012e7c2e1e0b2afe23bf80c595415
Add PCFactorSetShiftType() and PCFactorSetShiftAmount()Hg-commit: 11b2d9da5dd711b89de2445adb08f29eaa980990
Bugfix for shift schemes used in numerical factorization.Turn-off default shiftnz/shiftpd in icc and ilu until a better design is determinedOnly one of shiftnz, shiftpd and shiftinblocks is allowed
Bugfix for shift schemes used in numerical factorization.Turn-off default shiftnz/shiftpd in icc and ilu until a better design is determinedOnly one of shiftnz, shiftpd and shiftinblocks is allowed to be turned on.Hg-commit: 9fab294fdd73e23ef93a1c65bfd18772aaab2b91
removed PETSC_ARCH/conf/base andadded support for not setting PETSC_ARCH and PETSC_DIR with --prefix optionHg-commit: e478f4df238e5763b64567336d7cf94ca5d89d58
factor common options for PCSetFromOptions_ILU(), LU() etc into PCSetFromOptions_Factor()rename PCFactorSetPivoting -> PCFactorSetColumnPivotHg-commit: 9fc226d4711fc60f38c040374fd14f9fe3620f06
fixed two errors introduced when changing PetscOptionsHasName to PetscOptionsGetTruth()Hg-commit: a8176fa5bda99bab0dbeaf44fbb1e40534c57e0b
changed PetscOptionsName() and PetscOptionsHasName() to return true if option is set, regardless of the value it is set to, include 0 and false changed most uses of these routines to the Truth vari
changed PetscOptionsName() and PetscOptionsHasName() to return true if option is set, regardless of the value it is set to, include 0 and false changed most uses of these routines to the Truth varientHg-commit: fd6366abd7eb1f3ec925faa0ecd078d5cc025eb2
merged hdf4 viewer code into hd5 viewer code. Basic vector viewing and loading name work with hdf5 viewer (using name of vector as name in hdf5 file). Vector viewing also works for DA vectors, but st
merged hdf4 viewer code into hd5 viewer code. Basic vector viewing and loading name work with hdf5 viewer (using name of vector as name in hdf5 file). Vector viewing also works for DA vectors, but still uses GlobalToNatural and so does not provide multi-dimensional DA array info into HDF5 file yet. Will be done next.Hg-commit: 3a20bd7aed8a5ccc4310dfbef8f234ed655cbbbf
fixes, cleanups, and heavy review of the meany XXXRegisterAll() callsHg-commit: a22e06998b33293c00422f45f4010e18e1dfb00e
removed -I$PETSC_DIR from make system and changed the corresponding#include "src/dsdsdsds" to "../src/sdsdsds" in the source codeHg-commit: 6e3c4cc4b4bfcb366bae9b71afcb5009c1f0429d
added PCFactorGetMatSolverPackage()Hg-commit: a123d5b4378bd8a00fad3da3e3525ceb2ae3ebe5
moved much of the PC_Factor methods inside the individual ilu/lu/icc/cholesky files into a single copy in the factor directoryfixed output from some example tests that do -ksp_view to contain the
moved much of the PC_Factor methods inside the individual ilu/lu/icc/cholesky files into a single copy in the factor directoryfixed output from some example tests that do -ksp_view to contain the new solver package information that is now printedHg-commit: f303c70be042abf5c2b17ed52c7a90c376106dc8
merged common fields of Factor preconditioner into PC_Factor base classHg-commit: e4d76e70e3f11265d8bb736f6d921cb6c5c43fbd
* Using MUMPS, MatGetInfo(...,&info), info.nz_allocated is given for theinitial matrix (not the factorized one). Currently I use :- Mat_MUMPS *lu=(Mat_MUMPS*)mat->spptr;- lu->id.INFOG(20)to get
* Using MUMPS, MatGetInfo(...,&info), info.nz_allocated is given for theinitial matrix (not the factorized one). Currently I use :- Mat_MUMPS *lu=(Mat_MUMPS*)mat->spptr;- lu->id.INFOG(20)to get the same information. It's a dirty solution...Added MatFactorGetSolverPackage(), also properly printed with -mat_view_info and -ksp_viewHg-commit: d45534d976bbd57e4a4d4c4173150bb1403ccc76
pass MatFactorInfo as const into factorization routines, they shouldn't be changing it.Hg-commit: 06e09652bd7d2f02616c7256594ec565736aabbf
1) The first argument to MatXXXFactorSymbolic/Numeric() is now the factored matrix2) The assigment of the MatXXXFactorNumeric ops pointer is now done in the symbolic factorization the assigment o
1) The first argument to MatXXXFactorSymbolic/Numeric() is now the factored matrix2) The assigment of the MatXXXFactorNumeric ops pointer is now done in the symbolic factorization the assigment of the MatSolve pointer is now done in the numeric factorization (This is because only then do you know the right pointer that should be used)Hg-commit: efc00afbc7c5103f2f56fd96963db2c4cf4e1ed4
a.peyser@umiami.edu Subject: MPIDense row distribution on transpose. Date: September 4, 2008 10:06:36 PM CDT To: petsc-dev@mcs.anl.gov Cc: petsc-users@mcs.anl.gov Reply-To: petsc-dev@mcs.
a.peyser@umiami.edu Subject: MPIDense row distribution on transpose. Date: September 4, 2008 10:06:36 PM CDT To: petsc-dev@mcs.anl.gov Cc: petsc-users@mcs.anl.gov Reply-To: petsc-dev@mcs.anl.gov used PETSC_DECIDE for local row/column layout of MatTranspose for MPIDense instead of using the transpose of the given layout, like all the other formats do.Hg-commit: 799c15ac905a55d3ddab0c3b2506cd4f2f95ec94
fix memory leak in calling PetscStrallocpy()Hg-commit: 34c1975360b8fcfa3da6181c3b10c443734e1771
12345678