| #
6f9ccc09
|
| 02-Jan-2018 |
Barry Smith <bsmith@mcs.anl.gov> |
Merge branch 'barry/more-to-new-test-harness' into barry/all-new-test-harness-branches
|
| #
9b58da5a
|
| 02-Jan-2018 |
Barry Smith <bsmith@mcs.anl.gov> |
Merge branch 'barry/vec-to-test-harness' into barry/all-new-test-harness-branches
|
| #
35e5de05
|
| 02-Jan-2018 |
Barry Smith <bsmith@mcs.anl.gov> |
Merge branch 'jed/tests-cxx-linker' into barry/more-to-new-test-harness
|
| #
967fd949
|
| 31-Dec-2017 |
Richard Tran Mills <rmills@rmills.org> |
Merge branch 'rmills/fix-aijmkl-style'
|
| #
9c46acdf
|
| 26-Dec-2017 |
Richard Tran Mills <rmills@rmills.org> |
Fix several minor code issues in AIJMKL and BAIJMKL code.
Remove improper returning of error codes from PetscFunctionReturn(); remove unused variable; fix some code formatting.
Commit-type: style-f
Fix several minor code issues in AIJMKL and BAIJMKL code.
Remove improper returning of error codes from PetscFunctionReturn(); remove unused variable; fix some code formatting.
Commit-type: style-fix Reported-by: Barry Smith <bsmith@mcs.anl.gov>
show more ...
|
| #
551aa5c8
|
| 06-Dec-2017 |
Richard Tran Mills <rmills@rmills.org> |
AIJMKL now uses PetscObjectState to determine if MKL matrix handle must be rebuilt.
|
| #
1fa399ad
|
| 04-Dec-2017 |
Matthew G. Knepley <knepley@gmail.com> |
Merge branch 'master' into knepley/fix-plex-bc-multiple
* master: (347 commits) tests: add 'PETSC_HAVE_C99' flag - so that 'requires: c99' can be used in the test suite PGI: fix build error with
Merge branch 'master' into knepley/fix-plex-bc-multiple
* master: (347 commits) tests: add 'PETSC_HAVE_C99' flag - so that 'requires: c99' can be used in the test suite PGI: fix build error with updated 17.10 compilers Fix for test dirs that are not tutorials/tests fix use of double in PetscAbsReal() which produces compiler warnings with single precision build elemental: remove info on conflict of -with-cxx-dialect=C++11 on osx - it works now. Cleanup .bib files Fix allfortranstubs rule to rm generated files from correct directory, not outdated location ierr set but not checked Remove initial refine so test runs faster and does not time out on some machines Gail's copyedits for bibtex items Docs: manual: add further description of 3 different current testing options DM: check for ops->localtolocalbegin/end before calling PetscSF: fix typo in tutorial ex1 Vec+HDF5: Fix output for complex - Need to check for empty group name SNES: SNESDestroy() can be called with snes->dm == NULL SNES: SNESDestroy() can be called with snes->dm == NULL Vec+HDF5: Fix output for complex - Need to check for empty group name Docs: Manual: Update tests section gmakegentest: also exclude .#filenames created by emacs DMPlex: Fix output of test ex1 after changes to Gmsh reader ...
show more ...
|
| #
c1d5218a
|
| 13-Nov-2017 |
Richard Tran Mills <rmills@rmills.org> |
mkl: fix compiler warning with --with-mkl_sparse_optimize=0
Reported-by: Eric Chamberland <Eric.Chamberland@giref.ulaval.ca>
|
| #
4a940b00
|
| 13-Nov-2017 |
Satish Balay <balay@mcs.anl.gov> |
mkl: fix breakage with --with-mkl_sparse_optimize=0
Also change one the tests using MKL to use the above option
Reported-by: Jose E. Roman <jroman@dsic.upv.es> Reported-by: Eric Chamberland <Eric.C
mkl: fix breakage with --with-mkl_sparse_optimize=0
Also change one the tests using MKL to use the above option
Reported-by: Jose E. Roman <jroman@dsic.upv.es> Reported-by: Eric Chamberland <Eric.Chamberland@giref.ulaval.ca>
show more ...
|
| #
0df32d94
|
| 11-Nov-2017 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/rmills/feature-aijmkl-matmatmult'
|
| #
66b7eeb6
|
| 02-Nov-2017 |
Richard Tran Mills <rmills@rmills.org> |
Document -mat_aijmkl_eager_inspection option and update formatting for SeqAIJMKL manual page.
|
| #
3af10221
|
| 02-Nov-2017 |
Richard Tran Mills <rmills@rmills.org> |
Updated SeqAIJMKL documentation: MatMatMult() and MatTransposeMatMult() are now supported.
|
| #
aab60f1b
|
| 27-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
MatSeqAIJMKL_create_from_mkl_handle() now simply destroys the old matrix if MAT_REUSE_MATRIX is specified.
Previously, we attempted to save some expense by directly copying exported CSR arrays from
MatSeqAIJMKL_create_from_mkl_handle() now simply destroys the old matrix if MAT_REUSE_MATRIX is specified.
Previously, we attempted to save some expense by directly copying exported CSR arrays from MKL, but this causes problems because the MKL and PETSc representations differ in small ways (e.g., more explicit nonzeros in the PETSc representation due to preallocation.)
show more ...
|
| #
a557fde5
|
| 25-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Fixed wrong operation being set for SeqAIJMKL MatTransposeMatMult.
|
| #
7225e97a
|
| 24-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Fixed missing ')' in comment.
|
| #
886913bf
|
| 24-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Destroy existing optimized MKL matrix handle in SeqAIJMK assembly end if using lazy inspection.
|
| #
f3fd1758
|
| 24-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Fixed missing setup of aijmkl->descr in MatSeqAIJMKL_create_from_mkl_handle().
|
| #
372ec6bb
|
| 24-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Added MatTransposeMatMult_SeqAIJMKL_SeqAIJMKL_SpMV2().
|
| #
6c87cf42
|
| 24-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Added code to handle MAT_REUSE_MATRIX when constructing SEQAIJMKL matrix from MKL matrix handle.
|
| #
45fbe478
|
| 24-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Added initial MatMatMult_SeqAIJMKL_SeqAIJMKL_SpMV2() implementation.
|
| #
19afcda9
|
| 20-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Added routine to create sequential AIJMKL matrix from an MKL sparse matrix handle.
|
| #
62916e1b
|
| 17-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Merge branch 'rmills/aijmkl-lazy-inspection'
|
| #
5b49642a
|
| 16-Oct-2017 |
Richard Tran Mills <rmills@rmills.org> |
Can now choose between lazy or eager inspection for AIJMKL matrices; the default is lazy (i.e., does not happen until an operation like MatMult() is requested).
Commit-type: feature Funded-by: ECP T
Can now choose between lazy or eager inspection for AIJMKL matrices; the default is lazy (i.e., does not happen until an operation like MatMult() is requested).
Commit-type: feature Funded-by: ECP Time: 1.5 hours
show more ...
|
| #
13befd7a
|
| 14-Sep-2017 |
Karl Rupp <me@karlrupp.net> |
Merge branch 'pr746/irinasok/feature-baijmkl/master'
* pr746/irinasok/feature-baijmkl/master: Implementation of BAIJMKL matrix type similar to AIJMKL. Supports MKL calls for MatMult (add and transpo
Merge branch 'pr746/irinasok/feature-baijmkl/master'
* pr746/irinasok/feature-baijmkl/master: Implementation of BAIJMKL matrix type similar to AIJMKL. Supports MKL calls for MatMult (add and transposed). In case if used MKL version does not support mkl_sparse_optimize() (MKL 11.2 and earlier) PETSc default kernels are used. Test added on configuration step to check if used version of MKL supports column major layout with zero based indexing. All released versions of MKL do not support such cross case and MatSeqBAIJMKL_create_mkl_handle() converts matrix arrays to one based indexing before passing them to mkl routines.
show more ...
|
| #
80095d54
|
| 14-Sep-2017 |
Irina Sokolova <irina.sokolova@intel.com> |
Bug fixes Fix for test 107_2. To prevent MKL routine returning error in case when MatSetOption(mat,MAT_STRUCTURE_ONLY,PETSC_TRUE) is called. In this case matrix will be optimized before call to mkl_s
Bug fixes Fix for test 107_2. To prevent MKL routine returning error in case when MatSetOption(mat,MAT_STRUCTURE_ONLY,PETSC_TRUE) is called. In this case matrix will be optimized before call to mkl_sparse_x_mv. The same problem will appear in AIJMKl case. Also added MatCreateBAIJMKL and MatCreateSeqBAIJMKL to external include to call them from examples
show more ...
|