| #
c86c4108
|
| 30-Jul-2016 |
Matthew G. Knepley <knepley@gmail.com> |
Merge branch 'master' into mark/snes-ex56
* master: (93 commits)
Web: Added a tutorial
typo
rename X -> U
sowing: update to version that creates manpage URLs for 'enum' types/values
Increa
Merge branch 'master' into mark/snes-ex56
* master: (93 commits)
Web: Added a tutorial
typo
rename X -> U
sowing: update to version that creates manpage URLs for 'enum' types/values
Increase patchlevel to 3.7.3
move runex3_3 to NOTSINGLE [ partial backport 9ed0bd76d5c329eb91401ebabfbd851446214e20 ]
DMPlex: local coordinate vector should be a sequential vector
DM: Name coordinate vectors in subdomain and restrict hooks
tests: disable a couple of TS examples from single precision tests
reduce TS ex12_2 output
fix PetscFormatRealArray() to cast PetscReal in print statement to double so works for __float128 precision
Vec: enable Fortran interfaces for VecGetLocalVectorRead and VecRestoreLocalVectorRead
Bib: Added refs
PCShell: man page fixes
update handling of latex in example code to use https
fix typo in isltog comment
test: Avoid conversion of object <-> function pointer type [-Wpedantic]
fortran: enable interface for VecGetLocalVector(), VecRestoreLocalVector()
update citation for cmz2016
PCBDDC: silent compilation warnings from nightlybuilds
...
# Please enter a commit message to explain why this merge is necessary,
# especially if it merges an updated upstream into a topic branch.
#
# Lines starting with '#' will be ignored, and an empty message aborts
# the commit.
show more ...
|
| #
530f2ba0
|
| 22-Jul-2016 |
Karl Rupp <rupp@iue.tuwien.ac.at> |
Merge branch 'karlrupp/feature-viennacl-hybrid'
Conflicts:
src/mat/impls/aij/mpi/makefile
* karlrupp/feature-viennacl-hybrid:
Added hybrid capabilities (CUDA, OpenCL, OpenMP) for ViennaCL bindi
Merge branch 'karlrupp/feature-viennacl-hybrid'
Conflicts:
src/mat/impls/aij/mpi/makefile
* karlrupp/feature-viennacl-hybrid:
Added hybrid capabilities (CUDA, OpenCL, OpenMP) for ViennaCL bindings.
Depending on whether the user configures with CUDA, OpenCL, or OpenMP,
the respective backends will be available in ViennaCL.
Available options:
- viennacl_view: Lists available and selected backends
- viennacl_backed: One out of CUDA, OpenCL, OpenMP (if enabled in configure)
- viennacl_opencl_device_type: cpu, gpu, accelerator (OpenCL device type selection)
show more ...
|
| #
0cd40bc2
|
| 24-May-2016 |
Karl Rupp <rupp@iue.tuwien.ac.at> |
ViennaCL: Added hybrid capabilities (CUDA, OpenCL, OpenMP)
Depending on whether the user configures with CUDA, OpenCL, or OpenMP,
the respective backends will be available in ViennaCL.
CUDA has prec
ViennaCL: Added hybrid capabilities (CUDA, OpenCL, OpenMP)
Depending on whether the user configures with CUDA, OpenCL, or OpenMP,
the respective backends will be available in ViennaCL.
CUDA has precedence over OpenCL, which has precedence over OpenMP.
If none of the backends is selected, ViennaCL is built in sequential mode.
Remaining tasks to better leverage these hybrid capabilities:
- Select backend at runtime (API + options)
- Eliminate copy-overhead for OpenMP backend
- Documentation, examples, and tests to demonstrate usage
show more ...
|