Mikael Simberg

@sithhell - For --hpx:attach-debugger, I suppose you'd want something more than the help text that is printed at startup/exception by hpx? - For the usage of sanitizers, would it be...

Btw, I started this over [here](https://stellar-group.github.io/hpx/docs/sphinx/branches/master/html/manual/debugging_hpx_applications.html). I'll try to add a section on `--hpx:attach-debugger` soonish.

I'm not against this idea at all, but I think for example the `future_overhead` test is somewhat artificial because there we're really just interested in testing task spawning. In practice...

I think your key point is this: > synchronize cheaply on the completion no? Is the code you posted above with the latch your ideal way of writing that or...

With the risk of getting sidetracked, why would you call `async` inside the parallel for loop (if all you want to do is call `task(stuff)` for each index)?

cublas/hipblas is optional because hip installations do not necessarily include hipblas (https://github.com/STEllAR-GROUP/hpx/blob/581d3fb174e55eb1a8ba68d54dc9108e3cbac82b/cmake/HPX_SetupCUDA.cmake#L47-L53). @diehlpk can you try a newer version of CMake? CMake has been notoriously bad at finding various CUDA...

I suppose this could do with a HEP (like #4593) to work out the details. Before that I'd like to go one step further than what I proposed above, and...

If we agree to do the above, the follow-up is to make an explicit list of what the public API contains. I would keep that list in the documentation with...

> Why does the shared priority scheduler cause problems with the suspension tests? I was not aware of this issue, perhaps I can help solve that... You tell me ;)...

I meant to add: the fact that the shared priority scheduler sometimes ignores to return a thread object is related to suspension, but that doesn't only affect the background thread....