History log of /petsc/src/mat/impls/sell/mpi/mpisell.c (Results 51 – 75 of 229)
Revision Date Author Comments
# 2398a04f 09-Feb-2024 Satish Balay <balay@mcs.anl.gov>

Merge branch 'adams/feature-jacobi-l1' into 'main'

Mat: GetAbsSum (l1) row norms; PCJacobi: add scaled L1 diagonal approximation

See merge request petsc/petsc!7260


# eede4a3f 09-Feb-2024 Mark Adams <524115-markadams4@users.noreply.gitlab.com>

Mat: GetAbsSum (l1) row norms; PCJacobi: add scaled L1 diagonal approximation


# 942d34d0 03-Jan-2024 Satish Balay <balay@mcs.anl.gov>

Merge branch 'jolivet/fix-ubsan' into 'main'

Pacify UBSan

See merge request petsc/petsc!7158


# 8e3a54c0 24-Dec-2023 Pierre Jolivet <pierre@joliv.et>

Sys: add PetscSafePointerPlusOffset() to pacify UBSan


# 7b506345 14-Nov-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'barry/2023-10-19/fix-pcview-redundant/release' into 'main'

Fix bugs in handling PetscViewerGetSubViewer() and tabing in ASCII viewers

See merge request petsc/petsc!6948


# fe8fb074 19-Oct-2023 Barry Smith <bsmith@mcs.anl.gov>

Fix bugs in handling PetscViewerGetSubViewer() and tabing in ASCII viewers

Reported-by: Pierre Jolivet


# a7898f52 28-Oct-2023 Satish Balay <balay@mcs.anl.gov>

Merge remote-tracking branch 'origin/release'


# 48b8f0a5 28-Oct-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'barry/2023-10-25/fix-offdiagonal/release' into 'release'

Fix use of offdiagonal to off-diagonal

See merge request petsc/petsc!6962


# 4cf0e950 25-Oct-2023 Barry Smith <bsmith@mcs.anl.gov>

Fix use of offdiagonal to off-diagonal


# 17e6ea21 21-Aug-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'jacobf/2023-08-18/undelete-some-mpisell-code' into 'main'

Undelete some mpisell code

See merge request petsc/petsc!6827


# 43b34f9d 18-Aug-2023 Jacob Faibussowitsch <jacob.fai@gmail.com>

Make some functions in mpisell.c static


# fa078d78 18-Aug-2023 Jacob Faibussowitsch <jacob.fai@gmail.com>

Undelete some mpisell code


# 9c5460f9 17-Aug-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'jacobf/2023-08-09/linter-detect-static-function-candidates' into 'main'

Linter detect possible static functions

See merge request petsc/petsc!6796


# ba38deed 15-Aug-2023 Jacob Faibussowitsch <jacob.fai@gmail.com>

Delete unused code


# 6c37f76f 27-Jul-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'tisaac/feature-log-handler' into 'main'

Deglobalize logging into PetscLogHandler and PetscLogState

See merge request petsc/petsc!6709


# 2611ad71 14-Jul-2023 Toby Isaac <toby.isaac@gmail.com>

Profiling: Improve !defined(PETSC_USE_LOG) #defines

- If we cast macro arguments to void, we can avoid unused errors that have
forced lots of conditionally included code like this:

```c
#if def

Profiling: Improve !defined(PETSC_USE_LOG) #defines

- If we cast macro arguments to void, we can avoid unused errors that have
forced lots of conditionally included code like this:

```c
#if defined(PETSC_USE_LOG)
PetscLogEvent event;
#endif
```

- Having global counters like petsc_TotalFlops externed event
if !PetscDefined(USE_LOG) helps to remove a lot of needless
`#ifdef`s around code that can instead be placed in
`if (PetscDefined(USE_LOG)) {}` blocks.

show more ...


# 02477ebb 15-Jul-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'jacobf/2023-06-30/enable-doc-lint' into 'main'

Enable docs linting

See merge request petsc/petsc!6667


# fe59aa6d 12-Jul-2023 Jacob Faibussowitsch <jacob.fai@gmail.com>

Lint apply: mat


# b3f1e23b 31-May-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'jacobf/2023-05-25/no-memcpy-ops' into 'main'

Don't use PetscMemcpy() or PetscArraycpy() to copy the ops struct

See merge request petsc/petsc!6507


# aea10558 25-May-2023 Jacob Faibussowitsch <jacob.fai@gmail.com>

Don't use PetscMemcpy() or PetscArraycpy() to copy the ops struct


# 5bb1e8a9 30-May-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'hongzh/fix-matconvert-sellcuda' into 'main'

Fix MatConvert for SELLCUDA

See merge request petsc/petsc!6519


# b5917f1b 29-May-2023 Hong Zhang <hongzhang@anl.gov>

Fix MatConvert for SELLCUDA

MatConvert functions for SELLCUDA did not work properly. MatConvert_Basic was picked.


# dd874c20 10-Apr-2023 Satish Balay <balay@mcs.anl.gov>

Merge branch 'hongzh/sell-cuda' into 'main'

SELL-based SpMV

See merge request petsc/petsc!3428


# 4f8ff0b3 08-Apr-2023 Hong Zhang <hongzhang@anl.gov>

Change pointer to arrarray


# 07e43b41 10-Sep-2020 Hong Zhang <hongzhang@anl.gov>

Further optimization of MatMult_SeqSELLCUDA

- Add more kernels
- Use multiple threads per row for matrices with narrow slices
- Use multiple blocks per slice for matrices with wide slices
- Add thre

Further optimization of MatMult_SeqSELLCUDA

- Add more kernels
- Use multiple threads per row for matrices with narrow slices
- Use multiple blocks per slice for matrices with wide slices
- Add three new APIs to return the irregularity ratio, the maximum slice width and the average slice width

Experiments show that column blocking gives much worse performance for wide matrices and permulation based on slice width has almost no impact on the performance.

show more ...


12345678910