[kokkos] Disable Desul atomics temporarily to improve performance #309
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Outdated suggestions cannot be applied.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.
The Kokkos update to 3.5 turned out to decrease the throughput by 2-3x. Disabling "Desul atomics" (that Kokkos was changed to use by default in 3.5) appears to cure the issue, but this is only a temporary workaround, because the "old atomics implementation" is going to be phased out. The issue is being followed up with Kokkos developers in kokkos/kokkos#4780.
In addition, this PR adds support for using Kokkos' profiling tools via the
KOKKOS_PROFILE_LIBRARY
environment variable. (functionality that we were missing because of heavily customized initialization of Kokkos).,