| #
1e303875
|
| 24-Jul-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'scott/config-hip' into 'master'
HIP and SYCL configuration
See merge request petsc/petsc!2973
|
| #
28f796ea
|
| 14-Jul-2020 |
Scott Kruger <kruger@txcorp.com> |
Enable HIP and SYCL configuration
This adds support for HIP and SYCL configuration.
HIP is both a compiler (hipcc) and a package to be searched for (hipblas, hipsparse) similar to CUDA. Language:
Enable HIP and SYCL configuration
This adds support for HIP and SYCL configuration.
HIP is both a compiler (hipcc) and a package to be searched for (hipblas, hipsparse) similar to CUDA. Language: HIP; Compiler: HIPCC (even though it is C++)
SYCL is currently implemented for searching only the dpcpp compiler. Because Intel's oneAPI is very similar to MKL, Blas/Lapack is handled by those packages. Language: SYCL; Compiler: SYCLCXX
Although it is a C++ compiler, both hipcc and dpcpp are very slow. Currently using the .hip.cpp and .sycl.cxx extensions although more work will be needed for testings. See the kokkos work.
This includes preliminary changes for pipeline testing The HIP_PLATFORM is an environment variable that chooses the backend. So one configuration file, but the platform is specified in .gitlab-ci.yml.
This includes beta versions of configuration files for what could be used on Frontier. This is both the HIP version and CUDA versions for enabling comparisons.
show more ...
|
| #
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
|
| #
82707f21
|
| 13-Apr-2020 |
Stefano Zampini <stefano.zampini@gmail.com> |
Makefile: added CUFLAGS variable
this mirrors what is already there for the other compilers to have the user prescribing flags when running make
|
| #
528955be
|
| 04-Apr-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/maint'
|
| #
44774969
|
| 03-Apr-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'balay/ci-win32-workaround/maint' into 'maint'
ci: workaround for win32fe issue with simultaneous compile using the same obj file
See merge request petsc/petsc!2682
|
| #
06fe1a63
|
| 03-Apr-2020 |
Satish Balay <balay@mcs.anl.gov> |
ci: workaround for win32fe issue with simultaneous compile using the same obj file
https://gitlab.com/petsc/petsc/-/issues/591
|
| #
a742fcfe
|
| 02-Apr-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/maint'
|
| #
c49c71dd
|
| 02-Apr-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'scott/globsearch-long-arg' into 'maint'
Fixing the too many args problem
See merge request petsc/petsc!2663
|
| #
5b6dee57
|
| 02-Apr-2020 |
Scott Kruger <scott.e.kruger@gmail.com> |
Fixing the too many args problem
PETSc now has so many tests that the globsearch which used a python one-liner with the testlist coming in from stdin was failing. This upgrades the query_tests.py s
Fixing the too many args problem
PETSc now has so many tests that the globsearch which used a python one-liner with the testlist coming in from stdin was failing. This upgrades the query_tests.py script to enable doing the glob-style searching on just the names.
make: use $(info) instead of echo to work with large line lengths
show more ...
|
| #
39126ccf
|
| 29-Mar-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'balay/prefix-build-fixes' into 'master'
build: fixes for prefix build (and make check) from Jed
See merge request petsc/petsc!2651
|
| #
fca490e6
|
| 28-Mar-2020 |
Satish Balay <balay@mcs.anl.gov> |
build: fixes for prefix build (and make check) from Jed
|
| #
c11a6a3f
|
| 27-Mar-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/more-updates-for-missing-petsc_dir'
More support for not needing to set PETSC_DIR yet build with PETSc, also...
See merge request petsc/petsc!2626
|
| #
0e03b746
|
| 24-Mar-2020 |
Barry Smith <bsmith@mcs.anl.gov> |
build: add more support for not needing to set PETSC_DIR yet build with PETSc, also support for GNU make to directly use "make test" instead of "make -f gmakefile.test"
Still needs -include ../petsc
build: add more support for not needing to set PETSC_DIR yet build with PETSc, also support for GNU make to directly use "make test" instead of "make -f gmakefile.test"
Still needs -include ../petscdir.mk in many of the makefiles Still needs the link for ../petscdir.mk in many directories
GNUMakefile: call legacy makefile recursively instead of including it
GNUMakefile: promote default "all" target; "libs" builds libpetsc*
GNUMakefile: If gnumake < 4.2 prefer makefile wrapper interface over direct gmakefile interface to avoid the following warnings
gmakefile.test:92: arch-ci-linux-cxx-cmplx-pkgs-64idx/tests/testfiles: No such file or directory gmakefile:67: arch-ci-linux-cxx-cmplx-pkgs-64idx/lib/petsc/conf/files: No such file or directory
Commit-type: testing-fix, feature, usability Thanks-to: Jed Brown jed@jedbrown.org
show more ...
|
| #
c20d7725
|
| 22-Mar-2020 |
Jed Brown <jed@jedbrown.org> |
Merge branch 'jed/promote-examples-tests-tutorials' [petsc/petsc!2610]
* jed/promote-examples-tests-tutorials: Promote examples/{tests,tutorials}/ to {tests,tutorials}/
|
| #
c4762a1b
|
| 18-Mar-2020 |
Jed Brown <jed@jedbrown.org> |
Promote examples/{tests,tutorials}/ to {tests,tutorials}/
This shortens paths and improves consistency between test target names and paths to the source and output files. Most of the work was compl
Promote examples/{tests,tutorials}/ to {tests,tutorials}/
This shortens paths and improves consistency between test target names and paths to the source and output files. Most of the work was completed by this script, followed by mild cleanup of nonconforming cases.
for makefile in `git ls-files 'src/*makefile'`; do if rg -q 'DIRS.*\bexamples\b' $makefile; then base=$(dirname $makefile) dirs=$(cd $base/examples && ls -d tests tutorials 2>/dev/null | xargs echo) perl -pi -e "s#^(DIRS.*)\bexamples\b#\1${dirs}#" $makefile git rm $base/examples/makefile for t in $dirs; do git mv $base/examples/$t $base/ perl -pi -e "s#^(LOCDIR[[:space:]]*=).*#\1 $base/$t/#" $base/$t done fi done
git grep -l -E -z 'examples/(tutorials|tests)' | xargs -0 perl -pi -e 's#examples/(tutorials|tests)#\1#g' git checkout @ \ src/docs/website/documentation/changes/ \ src/benchmarks/results/
show more ...
|
| #
fac64830
|
| 17-Mar-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/maint'
|
| #
367d09cb
|
| 17-Mar-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/balay/fix-printcmd/maint' into maint
test-suite: fix unset variable error
See merge request petsc/petsc!2607
|
| #
e6bdf5b1
|
| 17-Mar-2020 |
Satish Balay <balay@mcs.anl.gov> |
test-suite: fix unset variable error
/bin/sh: 1: test: -ne: unexpected operator
|
| #
7d2962b0
|
| 17-Mar-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/maint'
|
| #
1781032e
|
| 17-Mar-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'scott/test-feature-printcmd' into 'maint'
PRINTONLY feature: print command to run
See merge request petsc/petsc!2603
|
| #
a6f3f80d
|
| 16-Mar-2020 |
Scott Kruger <scott.e.kruger@gmail.com> |
PRINTONLY feature: print command to run
Per suggestion of @jedbrown `make -f gmakefile.test test PRINTONLY=1` prints the command that can be run from the PETSC_DIR level for rapid debugging.
|
| #
099d8087
|
| 14-Feb-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/maint'
|
| #
398eebba
|
| 14-Feb-2020 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'balay/fix-make-do-not-ignore-error-from-gmakegentest/maint' into 'maint'
ci: fix gmakefiles not to ignore errors from gmakegentest.py [and gmakegen.py]...
See merge request petsc/pets
Merge branch 'balay/fix-make-do-not-ignore-error-from-gmakegentest/maint' into 'maint'
ci: fix gmakefiles not to ignore errors from gmakegentest.py [and gmakegen.py]...
See merge request petsc/petsc!2525
show more ...
|
| #
336d78f1
|
| 13-Feb-2020 |
Satish Balay <balay@mcs.anl.gov> |
ci: fix gmakefiles not to ignore errors from gmakegentest.py [and gmakegen.py] - otherwise CI thinks the job is a success.
And avoid errors with old make - when the files are not yet generated.
gma
ci: fix gmakefiles not to ignore errors from gmakegentest.py [and gmakegen.py] - otherwise CI thinks the job is a success.
And avoid errors with old make - when the files are not yet generated.
gmakefile.test:87: arch-ci-linux-ILP64/tests/testfiles: No such file or directory gmakefile:65: arch-ci-linux-ILP64/lib/petsc/conf/files: No such file or directory /usr/bin/python ./config/gmakegen.py --petsc-arch=arch-ci-linux-ILP64 /usr/bin/python /scratch/svcpetsc/glci-builds/GXbQf6e_/0/petsc/petsc/config/gmakegentest.py --petsc-dir=/scratch/svcpetsc/glci-builds/GXbQf6e_/0/petsc/petsc --petsc-arch=arch-ci-linux-ILP64 --testdir=./arch-ci-linux-ILP64/tests
show more ...
|