Daniel Arndt

Does this still generate the correct result? Is it a problem if the `Kokkos` implementation is more flexible than `std::adjacent_difference`?

~~One way to resolve this issue would be to interpret `Kokkos::View` as a one-dimensional `View` with value type `Kokkos::Array` instead of as a two-dimensional `View` with value type `int`. Of...

Fixed in #6906.

We just don't treat this variable in `nvcc_wrapper`.

The alternative would be to just delete these lines, of course, since we don't have any control over the future (location) of this example anyway.

> Should these `[de]allocate` functions go through a proper deprecation process? They are technically public functions in a non-impl namespace, available to be called in places external to kokkos. Yeah,...

> Other than the need for a discussion for a potential deprecation, looks good to me. Let me add a commit with the deprecations.

> Since when do we have these MemSpace::[de]allocate overload with a label vs the ones w/o a label? The version with labels was introduced in https://github.com/kokkos/kokkos/pull/3084. The one without labels...

> is it a draft bec waiting for CI? Yes.

> Wanna update shift left for consistency? Sure, why not, see 89fcc46.