Damien L-G comments

Results 491 comments of


                                            Damien L-G

Refactor Desul atomics to support compiling CUDA with nvc++

> Did you do a plain copy from `desul`? I cleaned up history * The first commit is a copy from desul [f5b9eeb](https://github.com/desul/desul/commit/f5b9eeb740ed855ff47ba006f27921734a382737), * the second one a bug fix...

Warnings in cuda builds: Kokkos_ViewCtor.hpp(304): warning: missing return statement at end of non-void function

Same issue as #5432

Warnings in cuda builds: Kokkos_ViewCtor.hpp(304): warning: missing return statement at end of non-void function

This is develop only right? In that case please close it.

Improve `ExecutionSpace::print_configuration` in all backends

> For SYCL+CUDA, we currently get output such as \ which would be plenty for a bug report but most of the information only appears relevant if people are to...

Improve `ExecutionSpace::print_configuration` in all backends

@Rombur do you think including managed memory support info would be useful?

Improve `ExecutionSpace::print_configuration` in all backends

@e10harvey for instance `KOKKOS_ENABLE_IMPL_CUDA_MALLOC_ASYNC` is missing

`kokkosp_end_deep_copy` needs some more arguments

> Do all profiling tools fence internally? No

Pass FunctorAnalysis to ParallelReduce

Need to rebase

OpenMP Target reduction with result in host rank 1 view fails to compile

It would help if you can provide a minimal reproducer. Consider opening a draft PR so we can better understand the changes you are suggesting.

Adding OpenACC support for Makefiles (Makefile.kokkos and Makefile.targets)

Must keep an eye on #5451 which introduces a new `KOKKOS_ENABLE_OPENACC_COLLAPSE_HIERARCHICAL_CONSTRUCTS` macro