Tobias Fuchs

Resolved in PR #237

@devreal `MPI_Accumulate` does not reduce a sequence to a single value, if I get the docs right: It performs an element-wise atomic update operation, similar to a put, but reduces...

Will be fixed along with introduction of execution- and launch policies, see PR #300

@fuerlinger Solving this using view expressions now. The changes proposed in #347 have been rejected as commented here: https://github.com/dash-project/dash/pull/347#issuecomment-289970209 In brief: It doesn't work for non-contiguous elements (e.g. cyclic distribution)...

Addressed in #410 New implementation uses view expressions. If find this diff quite convincing:

Among the topics discussed in the DASH team summit this week, I found this problem the most intriguing. It is orthogonal to any common conceptual category of parallel programming. I...

The changes in this PR are not related to memory spaces or Mephisto but on iterator concepts. Global memory concepts, maybe (both developed and implemented by myself, but why should...

This defect only has effect in builds with `DASH_ENABLE_THREADSUPPORT`, right? So previously existing functionality is not affected? We could live with that.

@devreal I'd be fine with that. Known issues are to be documented in `CHANGELOG.md`.

Global-to-global `dash::copy` is a feature that depends on the stale `nviews` feature-branch. It already works as-is for one-dimensional ranges. I will fork a feature branch named something like `rviews` or...