Daniel Arndt

Introduce modernize-concat-nested-namespaces

1

Let's see if people like us commit to this style.

Deadlock ThreadVectorRange for SYCL+CUDA in incremental Tests 12a and 12b

5

After implementing `TeamPolicy` for `vector_size>1` (see https://github.com/kokkos/kokkos/pull/4183#discussion_r680212952), incremental tests 12a and 12b are deadlocking in the inner-most loop, i.e. ThreadVectorRange, in the subgroup barrier, when running with the `SYCL` `CUDA`...

Use Kokkos::Tools::Experimental::device_id for all Profiling events

3

As noticed in #5106, we don't use Kokkos::Tools::Experimental::device_id for all profiling events which means that, e.g., the number of the device isn't reported for `fence` events.

GitHub: | os | compiler | | ----- | ---- | |fedora:latest | gcc 12.1.1| |fedora:latest | clang 14.0.0| |fedora:rawhide | gcc 12.1.1| |fedora:rawhide | clang 14.0.5| |ubuntu:latest | gcc...

Fixes https://github.com/kokkos/kokkos-tools/issues/40. By default, `SpaceTimeStack` sets `USE_MPI=1` and assumes that the application uses MPI and initializes it. As also described in the link issue, this assumption is oftentimes not true....

Use Kokkos::view_alloc with Kokkos::resize/realloc and Kokkos::create_mirror* for Kokkos 3.7

3

Fixes #641. Only files in `src` are changed.

Implement sortObjects in terms of sortByKey

2

Mostly for CI for now.

Add DistributedTree Nearest query with callback

6

The strategy here is to do a regular query that gets us the MPI ranks, and primitive indices for each query, send them back to the MPI rank owning the...