forked from trilinos/Trilinos
-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Zoltan2: IdentifierAdapters test #173
Open
cwschilly
wants to merge
550
commits into
NGA-FY23-develop
Choose a base branch
from
id_adapters_test
base: NGA-FY23-develop
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ements / deep copies
Add a no-package-enables config without deprecated code for future PR build. User Support Ticket(s) or Story Referenced: TRILFRAME-578
KokkosKernels: patch in PR 1927
…omp-with-no-zoltan Automatically Merged using Trilinos Pull Request AutoTester PR Title: SEACAS: decomp, nem_slice better handling of no-zoltan case PR Author: gsjaardema
Call out scalar fills
In response to review of PR trilinos#12036.
In response to review comment for PR trilinos#12036.
…_20230804_175808 Automatically Merged using Trilinos Master Merge AutoTester PR Title: Trilinos Master Merge PR Generator: Auto PR created to promote from master_merge_20230804_175808 branch to master PR Author: trilinos-autotester
Tpetra: Update deep_copy diagnostics to distinguish scalar fill
This essentially removes all usage of the template parameters beyond just the datatype in stk::mesh::Fields. This makes setting the size of a Field a run-time configuration option through calls to stk::mesh::put_field_on_mesh() and keeps the datatype (double, int, etc.) a compile-time configuration option. Functions like stk::mesh::MetaData::declare_field() and stk::mesh::MetaData::get_field() now just pass the datatype as a template parameter instead of the whole templated Field type itself. The stk::mesh::MetaData::use_simple_fields() method is called early in MetaData construction to force the new behavior, to prevent any accidental regressions and to ensure that any auto-registered Fields (like coordinates) will be sized consistently by things like stk::mesh::StkMeshIoBroker. Any calls to the legacy functions will trigger a run-time error. There should be no user-visible behavior changes, and external applications that use Panzer should not need to make any changes as long as they consistently use Panzer's typedefs to refer to higher- dimensional Fields like panzer_stk::VectorFieldType. If external apps grab a stk::mesh::MetaData and register or manipulate their own Fields, there is a chance that they will encounter a new error. The error message should guide them to the correct solution.
Automatically Merged using Trilinos Pull Request AutoTester PR Title: Framework: Add deprecated-off nopackages config PR Author: sebrowne
Changed function names for lowCommMakeColMap (TEMPORARY)
…g-factors Automatically Merged using Trilinos Pull Request AutoTester PR Title: b'Panzer: add scaling factors to the stk io closure model factory' PR Author: rppawlo
…y23-pr-candidate-12 Automatically Merged using Trilinos Pull Request AutoTester PR Title: b'NOX: Remove use of hardcoded MPI_COMM_WORLD from NOX package' PR Author: thearusable
Automatically Merged using Trilinos Pull Request AutoTester PR Title: b'Ifpack2: Adds missing cmake logic to address test dependency on Xpetra' PR Author: hkthorn
This reverts commit 0e4cfcb.
…ch-3 Automatically Merged using Trilinos Pull Request AutoTester PR Title: b'Framework: Prohibit MPI_COMM_WORLD usages' PR Author: sebrowne
Automatically Merged using Trilinos Pull Request AutoTester PR Title: b'MueLu: Adding finer grained profiling regions for H2D/D2H diagnosis' PR Author: csiefer2
KokkosKernels 4.2 will add some new overloads for spmv that accept execution space instances. The Stokhos specializations for spmv, where Scalar is PCE or MPVector, won't have them yet. So make sure that the existing Stokhos specializations still work by calling a Stokhos implementation (and not a KokkosKernels implementation).
Stokhos: update for new KokkosSparse::spmv overloads
…epetra-disabled-cuda-builds Automatically Merged using Trilinos Pull Request AutoTester PR Title: b'Disable Epetra stack in CUDA PR builds' PR Author: srbdev
Automatically Merged using Trilinos Pull Request AutoTester PR Title: b'NOX: Fix font and formatting in screen output' PR Author: mayrmt
…pleTesting Framework: Add SimpleTesting to Trilinos
PR trilinos#12190 actually failed to fix KokkosSparse::spmv for Sacado scalar types, when building with Kokkos/KokkosKernels develop branch. This actually fixes that issue (tested with develop and master KokkosKernels) and is quite a bit cleaner (though it uses version macros that can be taken out for the 4.2 release).
Automatically Merged using Trilinos Pull Request AutoTester PR Title: b'Stokhos: fix KokkosKernels trilinos#1959' PR Author: brian-kelley
* #128: specify cuda dir * #128: temporarily run GPU on every push; stop MPI builds * #128: specify correct dockerfile * #128: provide correct build/source dirs * #128: rework MPI in build script * #128: rework MPI in other gpu build script * #128: CI build script try another configuration and fix invalid path * #128: fix missing letter in path for a gpu build * #128: add newlines to end of files * #128: add spack find -p to find cuda root * #128: only run one pipeline; add cuda paths * #128: add kokkos variables * #128: add Tpetra_INST_SERIAL:BOOL=ON * #128: add CUDA root flag * #128: use correct kokkos architecture * #128: enable cusolver and cusparse * #128: emulate local build * #128: try with different docker image * #128: update cuda path * Try adding debug flag for Buildx * Tpetra: Disable cudaMemcpyAsync for Intercept.cpp * #128: lower -j * #128: use defaul kokkos architecture * #128: use fewer processes for GPU testing * 128: re-enable Kokkos_ARCH_AMPERE86 * #128: add cuda sample build to CI to validate CUDA * #128: run cuda test on NGA host * #128: update jobs dependency on CI for cuda * #128: add CUD sample run * #128: remove command not existing in CI * #128: change cuda path * #128: try to display information about driver * #128: change bad command in CI * #128: fix command * #128: try install nvidia util in Docker container * #128: remove commands * #128: fix cuda path * #128: fix dockerfile * #128: add different cuda test images * Run container in separate step * Remove not needed code * Apply changes to Epetra=OFF * Try to build and run docker within same step * #128: remove unused old CI files * #128: check both gpu pipelines * #128: Tpetra_INST_SERIAL=ON * #128: fix workflow name * #128: rework with cuda 11.4 dockerfile * #168: try to simplify CI sheel script * #128: try simplify shell script * #128: remove librairies path for blas and lapack to check if resolved * #128: try remove Lapack and blas lib paths from cmake call * #128: try again changing path dynamically * #128: fix another path * #128: fix blas path * #128: apply working conffiguration to other build scripts * #128: restore triggering workflows on PR * #128: disable GPU build job for PR having `EpetraMPI T1` label * #128: enable GPU build only with EpetraMPI T2 and EpetraMPI T3 labels * #128: upload test log * #128: fix typo * #128: fix artifacts * #128: add junit report for tests * #128: add junit reporting in CI and set * #128: fix artifact name * #128: fix artifacts missing * #128 fix extra slach char in path * #128: fix artifacts path * #128: fix path in gitbub action * #128: try mounting artifacts folder into the host runner * #128: use same logic for gpu or non-gpu pipelines * 128: Finalize pipelines (GPU on push, MPI cancellations) * 128: remove label requirements * Revert "Tpetra: Disable cudaMemcpyAsync for Intercept.cpp" This reverts commit 5db2d5d. * #128: test intercept reversion * Revert "Revert "Tpetra: Disable cudaMemcpyAsync for Intercept.cpp"" This reverts commit de87a22. * #128: fix underscore * #128: run GPU pipeline on merge to fy23 develop --------- Co-authored-by: Thomas Dutheillet-Lamonthézie <[email protected]> Co-authored-by: Jacob Domagala <[email protected]>
@stmcgovern This PR can likely be closed |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Test CI failures on PR trilinos#12082