Skip to content
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

Merge trilinos/develop into e3sm-project/develop #101

Merged
merged 26 commits into from
Feb 21, 2024

Conversation

E3SM-Bot
Copy link
Collaborator

Created by Github action

thomasguillet and others added 26 commits February 2, 2024 14:56
Rather, perform the scaling as part of perturbX.update()
Instead of copying X to Y and then reading from Y.
Since scaleFactor == 1.0 in Centered mode, this does not impact the
results, but makes all the finite difference formulas in the code
all consistent with each other.
…_20240216_175856

Automatically Merged using Trilinos Master Merge AutoTester
PR Title: b'Trilinos Master Merge PR Generator: Auto PR created to promote from master_merge_20240216_175856 branch to master'
PR Author: trilinos-autotester
Automatically Merged using Trilinos Pull Request AutoTester
PR Title: b'Tpetra: Add more warnings about complex'
PR Author: cgcgcg
For a template class `X<T>`, the name of the constructor may be spelled as `X()` or `X<T>()`, but some compilers balk at the latter and common practice is to use the former.

In this case, I run into the following compiler error:
```
include/Teuchos_StandardParameterEntryValidators.hpp:2016:52: error: expected unqualified-id before ‘)’ token
 2016 |   AbstractArrayValidator<ValidatorType, EntryType>();
      |                                                    ^
```
This is easily fixed by dropping the (unnecessary) template arguments.
…arameters

The simple_fields transition for all stk::mesh::Fields involves removing
all template parameters beyond the datatype (e.g. things like
stk::mesh::Cartesian3d).  As a deprecation strategy, we spun up a
differently-namespace version of all of these tag types (e.g.
stk::mesh::legacy::Cartesian3d) that were not deprecated, and we
deprecated the original versions.  We could then use the legacy::
versions internally in things like IO where we had to preserve old
behavior but couldn't tolerate deprecation build warnings in Sierra.
Customers through Trilinos would see deprecation warnings with their
use of the original tag types if their code had not been converted yet.

It turns out that this can't work for Trilinos users.  If a Field
uses the legacy:: tag type, it is a completely different Field type
from the non-legacy:: version and our storage and retrieval machinery
can't see that they are supposed to be the same.  Things like the
coordinates field that is auto-registered in IO is a good example
of the problem.  Any attempt at internal conversion of the datatypes
would necessarily have to refer to the deprecated types, so we
can't build this in Sierra.

So, we have to drop deprecation markings on these tag types and get
rid of the legacy:: versions.  A run-time warning is now printed at
construction time, which will hopefully still get noticed by
Trilinos users.
…field_deprecation_fix

Automatically Merged using Trilinos Pull Request AutoTester
PR Title: b'stk: Changed the deprecation strategy for the legacy Field template parameters'
PR Author: djglaze
Automatically Merged using Trilinos Pull Request AutoTester
PR Title: b'Fix naming of constructor of template class.'
PR Author: bangerth
…ix-NOX-MatrixFree-backward

Automatically Merged using Trilinos Pull Request AutoTester
PR Title: b'NOX: add tests, fix NOX::Epetra::MatrixFree for backward difference (12715)'
PR Author: thomasguillet
@github-actions github-actions bot merged commit f528a4c into develop Feb 21, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

8 participants