| #
22f8e9db
|
| 18-Oct-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'adams/gamg-reduce-opt-cuda' into 'master'
Adams/gamg reduce opt cuda
See merge request petsc/petsc!3339
|
| #
c9567895
|
| 18-Oct-2020 |
Mark <cal2princeton@yahoo.com> |
Reducing first coarse grid to one process/device in Cuda and manual override of this and other heuristics for process reduction in GAMG coarse grids.
|
| #
efba6e65
|
| 10-Oct-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/release' into master
|
| #
d1debf72
|
| 09-Oct-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'stefanozampini/fix-gamg-transposegencuda' into 'release'
Fix CUDA performance regression and KOKKOS bugfixes
See merge request petsc/petsc!3310
|
| #
d98d7c49
|
| 07-Oct-2020 |
Stefano Zampini <stefano.zampini@gmail.com> |
MATMPIAIJCUSPARSE: fix conversions
|
| #
c58ef05e
|
| 06-Oct-2020 |
Stefano Zampini <stefano.zampini@gmail.com> |
MatInode: set it to false when using device matrices
|
| #
12c380df
|
| 28-Sep-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'adams/feature-mat-cuda' into 'master'
Adams/feature mat cuda
See merge request petsc/petsc!3137
|
| #
3fa6b06a
|
| 02-Sep-2020 |
Mark Adams <mfadams@lbl.gov> |
add cuda matrix for meta data method, assembly
|
| #
95d17da4
|
| 28-Sep-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'barry/2020-09-25/do-not-log-gpu-transfer-time' into 'master'
Do not log transfer time as GPU time. GPU time is used to get performance ON...
See merge request petsc/petsc!3247
|
| #
ae74c241
|
| 26-Sep-2020 |
Barry Smith <bsmith@mcs.anl.gov> |
Do not log transfer time as GPU time. GPU time is used to get performance ON the GPU (for example flop rates)
logging transfer times distorts these numbers since they would include the time to copy d
Do not log transfer time as GPU time. GPU time is used to get performance ON the GPU (for example flop rates)
logging transfer times distorts these numbers since they would include the time to copy down the matrix.
The transfer times are already logged in the copy to and from events and the performance flop rates including the
copy times are already logged in the overall flop rate.
Commit-type: bug-fix, style-fix
/spend 5m
show more ...
|
| #
6f7728a9
|
| 22-Sep-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'jczhang/support-cuda-11' into 'master'
Add support of CUDA-11
See merge request petsc/petsc!3207
|
| #
afb2bd1c
|
| 17-Sep-2020 |
Junchao Zhang <jczhang@mcs.anl.gov> |
Add support of CUDA-11
|
| #
fd7c363c
|
| 17-Sep-2020 |
Satish Balay <balay@mcs.anl.gov> |
Revert "CUDA: Get PETSc to compile with CUDA 11."
This reverts commit 21a6055d1ede0202436f221d06f48a7b9836052d.
|
| #
cf5f8798
|
| 09-Sep-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'karlrupp/fix-cuda-11-compilation' into 'master'
CUDA: Get PETSc to compile with CUDA 11.
See merge request petsc/petsc!3154
|
| #
21a6055d
|
| 08-Sep-2020 |
Karl Rupp <me@karlrupp.net> |
CUDA: Get PETSc to compile with CUDA 11.
CUSPARSE has changed significantly in CUDA 11.
This has lead to a lot of broken code.
This commit disables the AIJCUSPARSE class so that PETSc at least compi
CUDA: Get PETSc to compile with CUDA 11.
CUSPARSE has changed significantly in CUDA 11.
This has lead to a lot of broken code.
This commit disables the AIJCUSPARSE class so that PETSc at least compiles with other GPU features enabled.
show more ...
|
| #
fd2eec0d
|
| 23-Aug-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'jczhang/add-kokkos-init' into 'master'
Add Kokkos initialization
See merge request petsc/petsc!3048
|
| #
05035670
|
| 19-Aug-2020 |
Junchao Zhang <jczhang@mcs.anl.gov> |
Add support of CUDA/HIP initialization at the same time
|
| #
721b214e
|
| 02-Aug-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'jczhang/feature-delay-cuda-init' into 'master'
Do not init cuda if a code does not create any petsc cuda objects
See merge request petsc/petsc!2987
|
| #
cae85d06
|
| 21-Jul-2020 |
Junchao Zhang <jczhang@mcs.anl.gov> |
Do lazy petsc cuda initialization and validation
So that we can avoid overburdening the GPU in CI.
See more at https://gitlab.com/petsc/petsc/-/issues/360#note_373723821
|
| #
dadd21d5
|
| 04-Jul-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'barry/2020-06-23/pcfactor-no-ordering-external' into 'master'
Most external factorization packages do not use the ordering provided by...
See merge request petsc/petsc!2913
|
| #
2c7c0729
|
| 24-Jun-2020 |
Barry Smith <bsmith@mcs.anl.gov> |
Most external factorization packages do not use the ordering provided by PETSc, therefor only provide it when needed.
Adds MatOrderingType external to indicate not to generate an ordering and use wh
Most external factorization packages do not use the ordering provided by PETSc, therefor only provide it when needed.
Adds MatOrderingType external to indicate not to generate an ordering and use what the package needs
Now -pc_view will not print the wrong PETSc ordering when the ordering is done externally
Saves some memory and compute time.
Commit-type: optimization
Time: 1.2 hours
Reported-by: Junchao Zhang <junchao.zhang@gmail.com>
Thanks-to: Stefano Zampini <stefano.zampini@gmail.com>
show more ...
|
| #
7f3f1200
|
| 01-Jul-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'stefanozampini/feature-cusparse-notransgen-herm' into 'master'
feature cusparse notransgen herm
See merge request petsc/petsc!2915
|
| #
54da937a
|
| 28-Jun-2020 |
Stefano Zampini <stefano.zampini@gmail.com> |
MATSEQAIJCUSPARSE: test multiple transpose generation flags
|
| #
85ba7357
|
| 28-Jun-2020 |
Stefano Zampini <stefano.zampini@gmail.com> |
MATSEQAIJCUSPARSE: log transpose generation
|
| #
51c6d536
|
| 27-Jun-2020 |
Stefano Zampini <stefano.zampini@gmail.com> |
MATSEQAIJCUSPARSE: remove wrong CPUtoGPU memory logging
|