Tpetra: Pass initialization tests with CUDA_LAUNCH_BLOCKING off #7967
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Outdated suggestions cannot be applied.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.
Since packages are getting updated to work with CUDA_LAUNCH_BLOCKING=0
and Tpetra is completed, this allows the unit test suite to all pass for Tpetra.
The tests were checking for CUDA_LAUNCH_BLOCKING off and failing based on std::cerr output.
This keeps the warning but changes it to std::cout so the Tpetra test won't detect it.
I expect we'll want to update this warning again when Trilinos is all working with launch blocking off. Perhaps warn that user code needs fencing if they have it off and warn that they might want to turn it off for efficient Trilinos code, if they have it on.
Note we'll also need #7964 to get the tests passing on white Pascal nodes with blocking off.
@trilinos/tpetra
Motivation
Get Tpetra all passing for CUDA_LAUNCH_BLOCKING=0
Testing
Cuda white Pascal node - Tpetra Core tests