Fixed a variety of issues related to iOS builds

Some developers did not properly use relative paths for private include files
file that is not normally compiled had errors due to outdated API in it

Fixed a variety of issues related to iOS builds

Some developers did not properly use relative paths for private include files
file that is not normally compiled had errors due to outdated API in it (why is it compiled with iOS?)
some iOS build scripts had outdated paths for various items
iSO version of makeall implicitly assumed PETSC_DIR had been set (OSX version previously fixed by Satish Balay)

Reported-by: Francisco Schuetze <fjschuetze@gmail.com>

show more ...

All kinds of goofy errors with captilization and functon names

Reported-by: Adrian Croucher <a.croucher@auckland.ac.nz>

moved #define of function names to before formated bfort comments otherwise bfort crashes

Reported-by: Adrian Croucher <a.croucher@auckland.ac.nz>

PCTELESCOPE: add option to change subcommunicator type (interlaced or contiguous)

Add a new data field to PCTELESCOPE which allows, before setup only, the user to specify
which type of subcommunicat

PCTELESCOPE: add option to change subcommunicator type (interlaced or contiguous)

Add a new data field to PCTELESCOPE which allows, before setup only, the user to specify
which type of subcommunicator to create. Under the assumption of sequential ordering on
equally-sized compute nodes, this allows the user to specify whether their subcommunicator
lives on a subset of nodes (with the same number of ranks/node), or whether it is striped across nodes (the default).

This commit also adds some missing __FUNCT__ definitions

An example is added which calls KSPView. This isn't ideal, so a better idea of how to test this functionality would be welcome.

show more ...

Let's not test our pull requests at all

PCTelescope:
* Corrected bug fix and should hopefully also make the static analyser happy

PCTelescope:
* Bug fixes for nullspace

some uses of PetscTryMethod() actually require that the

object is correct and the function returns an appropriate value.
Change those calls to PetscUseMethod().

Merge branch 'pr426/dmay/dmay/pctelescope-mods/master'

PCTelescope - fix for static analysis warnings:
* I think that this commit will remove the last warning about indexing a null pointer

PCTelescope - Fixes reported by static analysis:
* Initialize variables which are otherwise set inside if (isActiveRank(sred->psubcomm)){} statements

few more clang static analyzer fixes

PCTelescopeRichardson:
* Removed set but unused variable

PCTelescope:
* Added PCApplyRichardson

Telescope:
* Removed support for non-zero initial guess

PCTelescope: Removed unused variable

PCTelescopeDMDA permutation matrix:
* Corrected pre-allocation of the permutation matrix (twice as many non-zeros were requested than needed)

PCTelescopeMatCreate_dmda:
* Removed comment code, leave comment regarding lazy assembly

PCTelescope: Removed unused block size variable

PCTelescope:
* Removed spaces from blank lines

PCTelescope manpage:
* Updated doc for PCTelescopeSetIgnoreKSPComputeOperators

PCTelescope - New API
* Added API for Get/Set flags indicating whether KSPSetComputeOperators should be used or ignored.

PCTelescope: * Use flag to ignore kspcompute operators when defining sub-KSP

PCTelescope: * Copy method and context to remove nullspace from parent

PCTelescope: * Changed prefix on sub-DM, now it inherits the name from the sub-KSP * Added flag to optionally ignore KSPComputeOperators