Vec: make lock available in optimized mode

Having the lock available is important to take runtime decisions
on how to export the vector data in python

VecISXXX: require same communicator between Vecs and IS

this allows to skip operations if the global size of the IS is zero

DM: Add interface to query length of a sequence in HDF5

Plex: Fix refinement of cylinder mesh

DM+Plex: Now we can register different geometric models for mesh refinement
- Add a registration list for geometry model functions
- Add DM method snaptogeommodel

Merge remote-tracking branch 'origin/release'

CI: add lib/petsc/bin/maint/runjobs.py script that allows developers to select exact jobs to run within a pipeline

Merge branch 'barry/2024-04-12/improve-pcfieldsplit-matnest' into 'main'

Support PCFieldSplitSetFields() and it's command line form for MATNEST

See merge request petsc/petsc!7464

Plex: Major changes to the behavior of cohesive cells
- Ghost cohesive cells now have no special marking (was shift3)
- By default, impinging faces separating fault sides do not split, but instead fo

Plex: Major changes to the behavior of cohesive cells
- Ghost cohesive cells now have no special marking (was shift3)
- By default, impinging faces separating fault sides do not split, but instead force unsplit nodes on the fault

show more ...

Plex: Extensive reworking of orientation
- This allows an interior surface to be oriented
- Added DMPlexOrientLabel()

Allow -pc_fieldsplit_%d_fields i,j to work intuitively as it works for matrices with a block size

Label: Add DMLabelGetValueBounds()

DM: Add sparse localization interface

TSRosW: announce newly-added methods

revert sources to dev
- reset PETSC_VERSION_RELEASE
- reset doc/changes/dev.rst

release: set petsc v3.21.0 strings

doc: fix format

Merge branch 'jrwrigh/multiple_isoperiodic' into 'main'

Plex Isoperiodic: Allow multiple periodic directions

See merge request petsc/petsc!7305

Plex Isoperiodic: Allow multiple Face SFs

This is important for cases of multiple directions of periodicity, whose
transforms are necessarily distinct from each other.

HYPRE and MFEM: support for single precision

Add HYPRE in single precision to the testsuite

Add SELLHIP

- The HIP kernels are converted directly from their CUDA version
- AMD GPUs and NVIDIA GPUs use different warp sizes. We set the warp size to 64 by default for AMD GPUs to faciliate comp

Add SELLHIP

- The HIP kernels are converted directly from their CUDA version
- AMD GPUs and NVIDIA GPUs use different warp sizes. We set the warp size to 64 by default for AMD GPUs to faciliate compile-time code optimization

show more ...

PC: Add PCJacobiGetDiagonal

Merge branch 'balay/require-F2003-GET_COMMAND_ARGUMENT' into 'main'

Fortran: require GET_COMMAND_ARGUMENT() from F2003, and remove F77 iargc etc..

See merge request petsc/petsc!7394

Merge branch 'knepley/feature-partitioner-edge-weights' into 'main'

Partitioner: Support edge weights

See merge request petsc/petsc!7395

Fortran: require GET_COMMAND_ARGUMENT() from F2003, and remove F77 iargc etc..