Jan Stephan

Please rebase!

It seems that Catch2 v3.3.2 fixed this somehow, I can no longer reproduce this.

Ah, sorry for the noise. I looked at the wrong test cases...

I finally had the time to investigate a bit more. Since the issue persists in CUDA 12.1 I opened an issue on NVIDIA's bug tracker (NVIDIA ID `4160692`). Reproducer: ```c++...

NVIDIA were able to reproduce the issue and are now looking into it.

AMD calls them `Matrix Cores`, at least one GPU (MI100) has them already. I haven't found the accompanying API in HIP yet, though.

Since we switched to C++17 last year: Which useful parts of (for example) Mp11 are we missing that can't be easily implemented through fold expressions etc.?

> I think alpaka's `ViewSubView` should be entirely replaced by `std::mdspan` No. This would break user code that cannot depend on `std::mdspan`. Edit: At least not until `std::mdspan` becomes available...

See also #403. I'm more in favour of the boolean solution.

Yes. Or just `queue.isBlocking()`.