remove attaching batch size to Mat for batch lu solvers (cusporse and Kokkos)

Merge branch 'adams/aijkokkos-gpu-logging' into 'main'

Cleanup diagnostics and remove unneeded calls to WaitForKokkos

See merge request petsc/petsc!4748

MATAIJKOKKOS: support MatSetValuesCOO

Move CHKERRCUSPARSE() to petscdevice.h.

Moving to be consistent with where the other CUDA error checking macros are located.

Thanks-to: Jacob Faibussowitsch <jacob.fai@gmail.com>

VecKokkos+AIJKokkos: cleanup timers and remove WaitForKokkos

Fix code formating

Commit-type: style
/spend 5m

Merge remote-tracking branch 'origin/release'

Merge branch 'jczhang/fix-cuda11-kokkos' into 'main'

CI: add kokkos to the cuda-11 job

See merge request petsc/petsc!4734

Change "cuSPARSE error" to "cuSPARSE errorcode".

This is needed to avoid a false positive in the egrep command in the top
level makefile when building with some CUDA versions. Some warnings
about de

Change "cuSPARSE error" to "cuSPARSE errorcode".

This is needed to avoid a false positive in the egrep command in the top
level makefile when building with some CUDA versions. Some warnings
about deprecated cuSPARSE functions print this part of the PETSc source,
and the occurrence of "error " causes incorrect logging of a build
error. Changing to "errorcode" avoids this issue, and is consistent with
terminology used in the cuSPARSE documentation.

Thanks-to: Satish Balay <balay@mcs.anl.gov>

show more ...

Merge branch 'barry/2022-01-02/minor-sellmat' into 'main'

MatSELL maintenance and documentation

See merge request petsc/petsc!4715

AIJKOKKOS: disable cuda-11.x spgemm

Merge branch 'hzhang/feature-mumps-mem-estimate' into 'main'

Add a feature to mumps interface: skip operations of MatFactorSymbolic() if it is re-called without changes in original matrix

See merge

Merge branch 'hzhang/feature-mumps-mem-estimate' into 'main'

Add a feature to mumps interface: skip operations of MatFactorSymbolic() if it is re-called without changes in original matrix

See merge request petsc/petsc!4727

show more ...

add a feature to mumps interface: get memory estimate from MatFactorSymbolic(), then skip its operation when it is being called again.
Requested by: Varun Hiremath <varunhiremath@gmail.com>

Slightly improve the manual pages for MATSELL

Commit-type: documentation
\spend 45m

MATAIJKOKKOS: we use MatHeaderReplace

Stash communicators previously used for external packages so that they may be reused.

Previously we used MPI_Comm_dup() and free() but for some MPI implementations including MPI Spectrum the MPI_Com

Stash communicators previously used for external packages so that they may be reused.

Previously we used MPI_Comm_dup() and free() but for some MPI implementations including MPI Spectrum the MPI_Comm_free() does not actually return the communicator so eventually they run out.

Commit-type: portability-fix
\spend 3h
Reported-by: Elias Karabelas <elias.karabelas@uni-graz.at>

show more ...

Additional PetscInt_FMT empty string and SETERRQ trailing newline fixes

remove trailing newline from SETERRQ

applied attribute format fixes to mat headers

Merge branch 'jed/thrust-nosync' into 'main'

thrust: upgrade to 1.16 prerelease for thrust::cuda::par_nosync in vector operations

See merge request petsc/petsc!4676

thrust: upgrade to 1.16 prerelease for async on streams

Headers in this version are less promiscuous so we need more includes.

Merge branch 'stefanozampini/petsc4py' into 'main'

Some work on petsc4py

See merge request petsc/petsc!4644

Merge remote-tracking branch 'origin/release'

Revise a comment

Merge remote-tracking branch 'origin/release'