changed prefixes of the two KSP in the PCFieldSplit Schur complement to fieldsplit_0_ and fieldsplit_1_if user provides only a single IS to PCFieldSplit then its complement is used for the other blo
changed prefixes of the two KSP in the PCFieldSplit Schur complement to fieldsplit_0_ and fieldsplit_1_if user provides only a single IS to PCFieldSplit then its complement is used for the other blockHg-commit: cb7e29354d6662e022e8b853277b051c3cb62f75
show more ...
commit after mergeHg-commit: dd840e3347a1e3546141c05fa3e28c14b3e8a313
added most of the appropriate PetscObjectIncrementTabLevel() calls on KSP objectsHg-commit: 55ad623694bf2fcd3544ef7f480b24298f32bf4c
basic Schur complement preconditioner is now working (no preconditioner on the Schur solve yet)Hg-commit: b92a24fa5a99cb9f0f1ebe423a22c2cab0402a12
change ISBlockGetSize() to return global number of blocks and add ISBlockGetLocalSize()Hg-commit: 6fa8486743cfd19d9386fa84119c3b6960268539
commit after mergeHg-commit: 09253a14fa13ed33fd4f0714e7098ca8ff9e44dd
I've noticed that recent commits are now accessing PetscObjectinternals like thisobj->hdr.commobj->hdr.type_nameI would really prefer to not do that, and instead use the more verbose((PetscOb
I've noticed that recent commits are now accessing PetscObjectinternals like thisobj->hdr.commobj->hdr.type_nameI would really prefer to not do that, and instead use the more verbose((PetscObject)obj)->comm((PetscObject)obj)->type_nameFixed all the ones I did.Hg-commit: 03c76b1cd4e2116af5998056afeb49537eef1fbc
Correction from [PETSC #17925]Hg-commit: 336f8437fc0fa2d5f7f0f4ac5f1727b61c605cfd
moved the const out of XXType definitions and into the Set and Get MethodsHg-commit: d4d9a237045a151847ef3121b00df73d952d0999
removed nonsense final argument I added to ISPartitioningCount()Hg-commit: f4e71c40b6fcb50d27a83b41493aa0911c342367
changed calling sequence of ISPartitioningCount() to (IS part,PetscInt len,PetscInt count[],PetscInt *npout) the final argument is optional.Hg-commit: f55b9a9d8c728d7e4ea3eff0d1c340467bfc89ed
fixed error in interface to 1wd ordering, did not malloc enough spaceHg-commit: 648582f97b5300c7ac3a65dc46ec0c51f60e0c7e
changed PetscEvent to PetscLogEventRemoved initialization of global variables PetscLogEvent and PetscCookie to 0 since no longer neededHg-commit: b49d69f3756a139801bd445902e6aa5221450631
removed no longer needed converts of Fortran to C MPI_Comm (since MPI 1.1 has a standard for this)Hg-commit: 263aca0fcfe7efcaed30a4abb46e648de3bb24ab
changed -np to -n for option used to pass number of processes to mpiexec in PETSc makefiles etcHg-commit: 32818c3c1dacc61b71d5dc1de8d364ff38c269b9
mv zpetsc.h (a meaningless name) to fortranimpl.h (a slightly more meaningful name)Hg-commit: d2687f12f92bff49ceff434100d018f6898829c3
bug fixes related to petsc-maint 16879 MPIAIJ matsetvalues did not properly reset ba = b->a array after disssembly SeqAIJ improperly increased "low" count by one if ignoredzero was ignored so
bug fixes related to petsc-maint 16879 MPIAIJ matsetvalues did not properly reset ba = b->a array after disssembly SeqAIJ improperly increased "low" count by one if ignoredzero was ignored so search did not locate next column MPIBAIJ and MPISBAIJ disassembly set column size to A->cmap.n when it should have been A->cmap.Nadded error checking for some (PetscMPIInt) castsHg-commit: a3c4121ea5c19e05c43e0d0492c8be7228c4775c
changed PetscTruth in Fortran from integer*4 to logical(kind=4) or logical*4 changed examples to reflect thisHg-commit: b6e99b1bf9c5c84144b543dacf084fc252312c84
manual pages for PetscRandom types and minor PetscRandom manual page updatesmoved IS initialization into ISInitializePackage()Hg-commit: d81416e49a52e842a1fc5d199928b8b819b9828d
Added much faster matrix converter to use hypre preconditioners for PETSc AIJ matricesHg-commit: 68b1593b74e668ad35839294cad5a7ed57cad6d9
mergeHg-commit: d507e9558c653b847e4e8b2a1687018f692e3610
petsc-maint 16799 ==6657== Conditional jump or move depends on uninitialised value(s) ==6657== at 0x8735AD1: petscoptionsgetstring_ (in /home/go
petsc-maint 16799 ==6657== Conditional jump or move depends on uninitialised value(s) ==6657== at 0x8735AD1: petscoptionsgetstring_ (in /home/gormo/fluidity/trunk/bin/petsc_readnsolve) ==6657== by 0x80F426D: petsc_readnsolve_options.878 (petsc_readnsolve.F90:264) ==6657== by 0x80F197C: petsc_readnsolve_ (petsc_readnsolve.F90:73) ==6657== by 0x80F18AD: main (petsc_readnsolve_main.cpp:28)moved zpetsc.h to private/zpetsc.hchanged Fortran interface function PetscOptionsGetXXX functions to allow passing PETSC_NULL_TRUTH as flag argumentHg-commit: a08dc4d2e15e97c0d3849b1cb02e6c1b0214b0c8
mergeHg-commit: b5a05984274c84515a795ed75eef9d3b27ba97f1
fixed many calls to PetscInfo to include the object the information is associated with. This should fix petsc-maint 16853 -info_exclude vec,matHg-commit: 76af22820baddbfcd9c2a4a4f02bebf64ea72ef9
changes for solving the strict aliasing issue in PetscObject and derivedHg-commit: 278faecc91564f5495fa9f7d48380d547995fa45
1...<<71727374757677787980>>...96