| #
de3b81ad
|
| 12-Jan-2022 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'psanan-release-patch-77117' into 'release'
Docs: add link to MatPreallocatorPreallocate from MATPREALLOCATOR man page
See merge request petsc/petsc!4657
|
| #
c0aa6a63
|
| 09-Dec-2021 |
Jacob Faibussowitsch <jacob.fai@gmail.com> |
applied attribute format fixes to mat headers
|
| #
c74f52fc
|
| 17-Dec-2021 |
Patrick Sanan <patrick.sanan@gmail.com> |
Docs: add link to MatPreallocatorPreallocate from MATPREALLOCATOR man page
|
| #
d5ed94bb
|
| 14-Sep-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/jose/fix-manpages'
Fix manpages to avoid errors reported by doctext
See merge request petsc/petsc!4309
|
| #
d8d19677
|
| 12-Sep-2021 |
Jose E. Roman <jroman@dsic.upv.es> |
Fix manpages: Input/Output Parameter --> Parameters
|
| #
c1bf08c5
|
| 06-Sep-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'jose/spelling-errors' into 'main'
Fix many spelling errors in manpages and comments
See merge request petsc/petsc!4283
|
| #
a5b23f4a
|
| 03-Sep-2021 |
Jose E. Roman <jroman@dsic.upv.es> |
Fix spelling errors in manpages and comments
|
| #
777e0605
|
| 02-Sep-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'jed/mat-preallocator-sort' into 'main'
MatPreallocator: sort instead of inserting single entries
See merge request petsc/petsc!4273
|
| #
880e7892
|
| 31-Aug-2021 |
Jed Brown <jed@jedbrown.org> |
MatPreallocator: calculate row offsets to avoid sorting COO over rows
Commit-type: optimization
|
| #
146543f8
|
| 31-Aug-2021 |
Jed Brown <jed@jedbrown.org> |
MatPreallocator: sort instead of inserting single entries
The previous approach inserted one entry at a time from the hash. This
results in a lot of lookup and memmove overhead because the hashed ke
MatPreallocator: sort instead of inserting single entries
The previous approach inserted one entry at a time from the hash. This
results in a lot of lookup and memmove overhead because the hashed keys
are recovered in arbitrary order, so rows need to constantly be shifted.
Here, we sort all the row indices, then sort each row before inserting.
Further optimizations:
* The initial sort could be removed by scanning over dnz+onz (producing
row offsets) and dropping column indices directly into the appropriate
row.
* An alternative implementation (or related mode) would be to have a
mode that guarantees no duplicate entries and allow unsorted insertion,
with a final sort of each row in MatAssemblyBegin. This would only be
useful on initial assembly.
Commit-type: optimization
Reported-by: Gautam Bisht
show more ...
|
| #
16f8cdc7
|
| 09-Apr-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/barry/2021-03-28/fix-mpiu-allreduce-mpi-failure'
Fix capture of MPI error code in MPIU_Allreduce()
See merge request petsc/petsc!3777
|
| #
820f2d46
|
| 03-Apr-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
All MPIU_ functions except MPIU_File return MPI error codes for checking
Update checkbadSource to find use of CHKERRQ with MPIU_ functions
Commit-type: i.e. error-checking, optimization, bug-fix, p
All MPIU_ functions except MPIU_File return MPI error codes for checking
Update checkbadSource to find use of CHKERRQ with MPIU_ functions
Commit-type: i.e. error-checking, optimization, bug-fix, portability-fix, testing-fix, style-fix, feature, documentation, example
Funded-by:
Project:
Time: hours
Reported-by:
Thanks-to:
Development Tools: Vim, Emacs, Eclipse
show more ...
|
| #
f8f6e9ae
|
| 12-Oct-2020 |
Stefano Zampini <stefano.zampini@gmail.com> |
MatPreallocator: support NO_OFF_PROC_ENTRIES flag when assembling
|
| #
16924cb1
|
| 26-Aug-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'barry/2020-08-16/fix-source-formatting' into 'master'
Turn on checkbadSource test to generate an error when found; fix all source code that causes errors
See merge request petsc/petsc
Merge branch 'barry/2020-08-16/fix-source-formatting' into 'master'
Turn on checkbadSource test to generate an error when found; fix all source code that causes errors
See merge request petsc/petsc!3064
show more ...
|
| #
2da392cc
|
| 17-Aug-2020 |
Barry Smith <bsmith@mcs.anl.gov> |
Turn on checkbadSource test to generate an error when found; fix all source code that causes errors
Commit-type: portability-fix, testing-fix, style-fix, feature, maintainability
/spend 1.5h
|
| #
7b917211
|
| 24-Aug-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'dalcinl/fix-null' into 'master'
Fix 0 -> NULL for pointers
See merge request petsc/petsc!3086
|
| #
f4259b30
|
| 23-Aug-2020 |
Lisandro Dalcin <dalcinl@gmail.com> |
Mat: Replace 0 -> NULL for pointers
|
| #
833b33c1
|
| 29-May-2020 |
Satish Balay <balay@mcs.anl.gov> |
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
|
| #
050c3c49
|
| 10-Apr-2020 |
Stefano Zampini <stefano.zampini@gmail.com> |
MatPreallocatorPreallocate: call MatSetUp
|
| #
44117e68
|
| 16-Jul-2019 |
Karl Rupp <me@karlrupp.net> |
Merge branch 'jczhang/fix-matpreallocator-stash-bs' into maint [PR #1876]
* jczhang/fix-matpreallocator-stash-bs:
bs should be 1 for stash of MatPreallocator
|
| #
ae73880b
|
| 16-Jul-2019 |
Karl Rupp <me@karlrupp.net> |
Merge branch 'jczhang/fix-matpreallocator-stash-bs' [PR #1876]
* jczhang/fix-matpreallocator-stash-bs:
bs should be 1 for stash of MatPreallocator
|
| #
8011973b
|
| 15-Jul-2019 |
Junchao Zhang <jczhang@mcs.anl.gov> |
Fix: bs should be 1 for stash of MatPreallocator
|
| #
b707d744
|
| 07-Jul-2019 |
Barry Smith <bsmith@mcs.anl.gov> |
Merge branch 'maint'
|
| #
cc20cd54
|
| 07-Jul-2019 |
Barry Smith <bsmith@mcs.anl.gov> |
Merge branch 'dmay/fix-MatPreallocator/maint' into maint
|
| #
380bf85a
|
| 17-Jun-2019 |
Dave May <dave.mayhem23@gmail.com> |
MatPreallocatorPreallocate: Bug fix associated with the non-usage of the fill parameter
- The manpage indicates that if fill = PETSC_TRUE, the non-zero structure will be populated with zeros. This wa
MatPreallocatorPreallocate: Bug fix associated with the non-usage of the fill parameter
- The manpage indicates that if fill = PETSC_TRUE, the non-zero structure will be populated with zeros. This was not occurring before, nor was the fill flag being checked and throwing an error with code PETSC_ERR_SUP.
- This commit mixes this bug / oversight.
MatPreallocatorPreallocate_Preallocator: Added ierr, CHKERRQ() and a little style tweak
MatPreallocatorPreallocate_Preallocator: Updated to use MatSetValuesBlocked and PetscCalloc. Added missing ierr= and CHKERRQ() macros
MatPreallocatorPreallocate: Remove unnecessary variables
MatPreallocatorPreallocate: Man page updates
show more ...
|