MatMultHermitianTranspose: use MatTranspose directly for real builds

MATSEQAIJCUSPARSE: implement hermitian transpose operation

Simplify the class, using a single routine for all the MatMult variants
Make transpose generation optional (not default)

Merge branch 'stefanozampini/feature-matconvert-seqaijcusparse' into 'master'

support matconvert with cusparse

See merge request petsc/petsc!2880

MATSEQAIJCUSPARSE: allow MatConvert with reuse != MAT_INPLACE_MATRIX

Merge branch 'stefanozampini/fix-matproduct-logging' into 'master'

fix matproduct logging

See merge request petsc/petsc!2857

MATSEQAIJCUSPARSE: add logging support for MatProduct with dense matrix

Merge branch 'jose/compile-warning' into 'master'

Fix compile warning

See merge request petsc/petsc!2823

Fix compile warning

Merge branch 'stefanozampini/feature-add-hpackages-rebased' into 'master'

Fixes to MatProduct + many bug fixes + new hierarchical matrix support

See merge request petsc/petsc!2800

MATSEQAIJCUSPARSE: support for MatMat operations

fix issue with ILU reusage (leaks and not setting the offdiagonal flag properly)
add missing nullify of function composition

MatConvert_SeqAIJ_SeqAIJCUSPARSE: fix bug with reuse != MAT_INPLACE_MATRIX

Merge branch 'jczhang/tune-aijcusparse-compressedrow' into 'master'

Do not force compressed row on GPU, instead use whatever on CPU

See merge request petsc/petsc!2707

MAT: do not force compressed row on GPU, instead use whatever on CPU

Merge branch 'rmills/fix-aijcusparse-bindtocpu' into 'master'

Fix errors in MatBindToCPU_SeqAIJCUSPARSE().

See merge request petsc/petsc!2627

Fix errors in MatBindToCPU_SeqAIJCUSPARSE().

Reported-by: Junchao Zhang <jczhang@mcs.anl.gov>

Merge branch 'rmills/fix-pinnedmemory-vecgetlocalvector' into 'master'

Rmills/fix pinnedmemory vecgetlocalvector

See merge request petsc/petsc!2621

Fix a few warnings.

Reported-by: Jose E. Roman <jroman@dsic.upv.es>

Merge remote-tracking branch 'origin/maint'

Merge branch 'jczhang/fix-aijcusparse-memory-leak' into 'maint'

Fix memory leaks and bugs in MATAIJCUSPARSE

See merge request petsc/petsc!2575

Merge branch 'rmills/add-matbindtocpu-aijcusparse' into 'master'

Add MatBindToCPU_SeqAIJCUSPARSE().

See merge request petsc/petsc!2597

Fix bugs and simplify MatMultTransposeAdd_SeqAIJCUSPARSE

MatT always uses non-compressed row format on GPU. We should not use cusparsestruct->workVector, which is created for Mat instead of MatT, an

Fix bugs and simplify MatMultTransposeAdd_SeqAIJCUSPARSE

MatT always uses non-compressed row format on GPU. We should not use cusparsestruct->workVector, which is created for Mat instead of MatT, and might be less than A->cmap->n (when Mat has compressed rows)

show more ...

Save rowoffsets_gpu and use A->was_assembled to easily update Mat and MatT when matrix was assembled

Old code used if (A->assembled ..), which is wrong and could not take advantage of the fact that

Save rowoffsets_gpu and use A->was_assembled to easily update Mat and MatT when matrix was assembled

Old code used if (A->assembled ..), which is wrong and could not take advantage of the fact that A's non-zero pattern is not changed

show more ...

Fix memory leak: free matTranspose and workVector when rebuilding cusparsestruct

Reported-by: jordic <jordic@cttc.upc.edu>

Add MatBindToCPU_SeqAIJCUSPARSE().

Merge branch 'jczhang/feature-cuda-error-string' into 'master'

Map a cuda error code to its name and description

See merge request petsc/petsc!2228