Merge branch 'knepley/feature-plex-assembly-cellrange'

* knepley/feature-plex-assembly-cellrange:
Plex: DMPlexComputeJacobian_Internal() defined - Give both t and X_tShift arguments - Give cell ra

Merge branch 'knepley/feature-plex-assembly-cellrange'

* knepley/feature-plex-assembly-cellrange:
Plex: DMPlexComputeJacobian_Internal() defined - Give both t and X_tShift arguments - Give cell range - Put in declaration
Plex: Now DMPlexComputeResidual_Internal() explicitly takes [cStart, cEnd)

show more ...

changed second argument of PetscObjectViewFromOptions() to PetscObject and all routines that use this routine

Merge branch 'knepley/feature-plex-bc'

* knepley/feature-plex-bc:
TS: Take out unnecessary declaration
XDMF: Allow missing time dimension
MatSchur: Better check for missing D matrix
TAO ex1:

Merge branch 'knepley/feature-plex-bc'

* knepley/feature-plex-bc:
TS: Take out unnecessary declaration
XDMF: Allow missing time dimension
MatSchur: Better check for missing D matrix
TAO ex1: Updated to new Plex interface
Plex: Fixed bug for FV BCs in DMCreateSection_Plex()
Plex+TS: FIxed DMPlexTSCheckFromOptions()
Plex: Changed the signature of pointwise input functions - Now the spatial dimension and number of fields is passed in
Plex: Fixed bugs with DMPlexCreateSection() - F90 declaration - Do not constrain more than the dog
DM: Can view the global section using -global_section_view
Plex: Now DMPlexCreateSection() also takes the constrained components
Plex: Now BC take constrained field components as well - Changed DMPlexAdd/GetBoundary() - Can use -bc_<name>_comp <nums> to override
Plex: Now -dm_plex_separate_marker works correctly in DMPlexCreateSquareBoundary()
DMLabel: Added DMLabelInsertIS()
SNES: Added DMSNESCheckFromOptions() - Should simplify examples

show more ...

Plex FVM: only add leaf face flux contributions

Plex FVM: don't interpolate face fields for parent faces

This would represent duplicated quadrature: only integrate over leaves

Plex FVM: refactor slope limiting for faces with children

There is now a recursive function for finding leaves to perform slope
limiting. For the most common case (tree depth 1) there's only one
le

Plex FVM: refactor slope limiting for faces with children

There is now a recursive function for finding leaves to perform slope
limiting. For the most common case (tree depth 1) there's only one
level of recursion, so we don't need to worry about the stack.

show more ...

Plex: Changed the signature of pointwise input functions
- Now the spatial dimension and number of fields is passed in

Plex: more avoiding parent faces in finite volume loops.

SNES: Added DMSNESCheckFromOptions()
- Should simplify examples

Plex: DMPlexComputeJacobian_Internal() defined
- Give both t and X_tShift arguments
- Give cell range
- Put in declaration

Plex: Now DMPlexComputeResidual_Internal() explicitly takes [cStart, cEnd)

Infrastructure that allows failures in PCSetUp(), PCApply(), MatMult() etc due to, for example a zero pivot or function evaluation outside its domain, to propagate up and become a KSP_DIVERGED inste

Infrastructure that allows failures in PCSetUp(), PCApply(), MatMult() etc due to, for example a zero pivot or function evaluation outside its domain, to propagate up and become a KSP_DIVERGED instead of generating an error that stops the program. In response to Issue 96. This includes failures in MatCreateSNESMF() applications due to domain errors. The mechanism to propagate some errors is by setting Info or Nan into the output vector and using the norm or inner product reductions in SNES or KSP to propagate the error condition to all processes and then handling them immediately after the norm or inner product.

This allows, for example, ODE integrators to try again with a smaller time-step if the PCSetUp failed instead of requiring a complete restart of the run with other options.

Currently some error conditions, such as function domain error in a line search may not get progated up using the correct SNESConvergedReason

See src/snes/examples/tests/ex69.c for the handling of several conditions

show more ...

PetscHeaderCreate: Shorten argument list and optimize memory allocation

* Remove type arguments from the macro
* Use a single memory allocation call instead of three

fixed missing CHKERRQ() detected by clang's --analyze

changed from CHKFORTRANNULLOBJECT() to CHKFORTRANNULLOBJECTDEREFERENCE() where appropriate in custom Fortran stub files

updated PETSc directory layout to match standard packaging strategies

include/petsc finclude,private,mpiuni
lib/petsc conf
bin/petsc*

Merge branch 'knepley/feature-viewer-partial-read'

* knepley/feature-viewer-partial-read:
Viewer: Fixed initialization
TS: Updated example to new interface
Plex: Fix compiler warning
Viewer:

Merge branch 'knepley/feature-viewer-partial-read'

* knepley/feature-viewer-partial-read:
Viewer: Fixed initialization
TS: Updated example to new interface
Plex: Fix compiler warning
Viewer: Now ViewerRead() takes an extra argument to return the number of items read - An error is thrown for insufficient data unless the extra argument is non-NULL
Viewer: Fixes for new Read interface - Fixed example - Fixed Fortran 90 binding - Fixed type
Builder: Allow a local dictionary for run parameters
Viewer: PetscViewerRead() should return the number of items actually read - This also applies to all its implementations

show more ...

when MPIU_REAL or MPIU_SCALAR is used then MPIU_SUM/MAX/MIN must be used

these bugs are why plex as crashing in quad precision

Viewer: Now ViewerRead() takes an extra argument to return the number of items read
- An error is thrown for insufficient data unless the extra argument is non-NULL

Viewer: PetscViewerRead() should return the number of items actually read
- This also applies to all its implementations

fixes links to "see also" methods
clarify arguments of func() by using variable names in prototype
corrects order of jac,jacpre in ex5.c to match doc

Contributed-by: Ed Bueler <elbueler@alaska.edu>

Plex+TS: Allow U_t to have positive sign in IFunction

Removed unused variables

Plex+SNES+TS: Better way to tell if we are implicit

Plex+SNES+TS: Use PetscDS interface for implicitness