René Widera
René Widera
MAybe we should think about adding debug lvl 3 where all entries and exits of alpaka functions will be visible and disable this for debug lvl 2.
The bug will be fixed with #2213 and I keep this PR to add icpx 204.0 to the CI. We still need to think about how we can enable OpenMP...
IMO caching makes sense, we should store the value during the device creation then there will be no need for a mutex.
Partly solved by #2246. Never the less we should cache all over runtime constant device properties within the device, than there is no need to query the API multiple times.
> By the way, I do like the idea, but IMHO "visibility" is not enough for this: @fwyzard I agree with you that this is not enough for optimized code...
we need to fix the OSX CI before we can merge this.
@mehmetyusufoglu could you please rebase this against the develop branch? I merged the OSX CI fix
> Seems @bernhardmgruber's idea works fine too. Does anyone have tips on what kind of additional tests to write for this? Maybe extend the [current tests](https://github.com/alpaka-group/alpaka/blob/develop/test/unit/vec/src/VecTest.cpp) so that they'd consturct...
@sliwowitz ping
suggestion for the interface change: ```C++ template< typename TAcc, typename TDev, typename TGridElemExtent = Vec, typename TThreadElemExtent = Vec> ALPAKA_FN_HOST auto getValidWorkDiv( [[maybe_unused]] TDev const& dev, [[maybe_unused]] TGridElemExtent const& gridElemExtent...