root DMView() and DMLoad() handle classid and typename
PCView binary does NOT view matrices

Hg-commit: 8fd378babe626ea07f0585f9209c640de6c87424

DM: Fix const problems

Hg-commit: e053c2b14b55691e25a0f8f70190b106ec10320c

DMDA: More changes to this crappy function to support Hypre

Hg-commit: 16d74ff52f4f5e947d30268e7aed543081d94efb

DM: Added MatGet/SetDM() and updated DMs to handle Vec and Mat for this

Hg-commit: d20ec0ce94bab2e32af8971314fa1b0124689ebf

reemove OBJSC and OBJSF from invididual makefiles and auto-define in conf/variables

Hg-commit: 2f01c17f3c19dc4b89aaf306cfd60a12312c270d

fixed bug introduced in VecView_MPI_Draw_DA2d() that did not use separate window for each field

Hg-commit: 532903956d0485dec11bdc703d1006d0fd2fab02

drawzoom clears window at start therefor VecView_MPI_da2_draw should not clear window
or you get blank images in movies

Hg-commit: a3ef8f2e3e2012243eceba8b141aeedf28a5940f

commit after merge

Hg-commit: ed1a46afbb3ca05b94b623600c6c387ba23868ed

VecView_MPI_DA1d did not handle periodic value correctly on multiprocesses for draw viewer

Hg-commit: 3aa13963d117a0b525d3490ed7fca9badff3221d

Const-correctness for MatGetRowIJ()/MatGetColumnIJ()

The arrays have always been read-only according to the
documentation. This patch should cause no behavioral change.

This const-correctness casca

Const-correctness for MatGetRowIJ()/MatGetColumnIJ()

The arrays have always been read-only according to the
documentation. This patch should cause no behavioral change.

This const-correctness cascaded into the private ordering and coloring
routines.

Hg-commit: 7efa5f739222584a785675705479b4e17c346e6e

show more ...

fixed formating of variable declarations messed up by putting const int XXXType
fixed DMDAGet/SetCoordinates() that matt did not remove

Hg-commit: de8db80506612f87fb4b9c77dd85f76c605a3942

changed #define char* XXXType to typedef const char* XXXType

Hg-commit: 2f1984a65b4f21a886ea1c854235ca6d51deefad

Merge

Hg-commit: e67798811b9e9715a27e1162d786ebd65ddc7d21

DM: Moved coordinates to DM, fixed default SF creation
- Now we have DMGetCoordinateDM() which describes coordinate layout and does G2L
DMGet/SetCoordinates() and DMGet/SetCoordinatesLocal()
- One

DM: Moved coordinates to DM, fixed default SF creation
- Now we have DMGetCoordinateDM() which describes coordinate layout and does G2L
DMGet/SetCoordinates() and DMGet/SetCoordinatesLocal()
- One can be made from the other (global or local) automatically
- Added dm->ops->createcoordinatedm()
- DMComplex no longer creates the coordinate Vec up front
- DM default SF can now handle global vets without constraints

Hg-commit: 6226b76f6e469184f69d66f3475966ce9ac97872

show more ...

added DMDAGetLocalBlockInfo() which includes overlap

Hg-commit: f3bef0e03e3ad693aa13e9fedc4d36fbba37b2d9

cleanup of phasefield examples, btw they don't work

Hg-commit: c8049baf05f70f0b83f50268daaee005a2b44135

commit after merge

Hg-commit: bee0a1be7a80c0df3b9f1d2c288f971e7235f1d1

removed trailing blanks from source code
added rule rmtrailingblanks accessible via make ACTION=remtrailingblanks tree

Hg-commit: 6e3652165d48aa5f3e94dd66bf607adea6fac27a

#132641 win32/c compile fix

Hg-commit: 748e275d3dd4c9dbb9d0ed1dd0a463933e01f14b

commit after merge

Hg-commit: 164e59610053be83dd5b3675a7005b6300e65292

code to match PETSc style guide for if (xxx) for (xxx) sizeof(xxx)

Hg-commit: 01e0d19bcd83a980fb5e8e32b294a95da72b405f

DM: Many simple compile bugs fixed
- Fixed Fortran stub in PetscSection

Hg-commit: b649896ec9c0ec4feaae1d975a40d272797ee7ff

updated DMSlicedCreate() to take ghost point and coupling information
cleaned up docs for DMSlicedCreate

Hg-commit: 6301f3ffea760b7fe7ad5493469b1d249d9048bf

DMDA+PetscSection: Fixed ghosts to be one-sided

Hg-commit: 2d93bc609e6722f4daeb9ce93b7f83a604e26f7e

DMDA+PetscSection: Put in support for vertex communication

Hg-commit: 2d10f703fe530523d5d2812371ee01cc3d50a7fc