Christian Trott
Christian Trott
Ah thank you I will look into that.
std::thread::hardware_conccurency is bad I think. I thought we investigated this before. I.e. it did not seem to take into account stuff like process masks?
Ok I guess I remember wrong.
Ok there is usage out there. And it doesn't look like it has much of a crazy interface I guess. So maybe just keep it around, make sure our testing...
Yeah as far as I remember the attribute query does add some latency, so it's something we could do in debug mode. Might be worthwhile to check whether not using...
For clang-cuda support we probably can not require the same version of CUDA, since even with clang 22 they are only up to CUDA 12.1. For a long time they...
The icpx OpenMP build with sanitizer does have a real failure: ``` openmp.team_broadcast_long /usr/desul-install/include/desul/atomics/Lock_Based_Fetch_Op_Host.hpp:37:18: runtime error: load of value 4, which is not a valid value for type 'bool' SUMMARY:...
@janciesko can you rebase this and get this over the line?
I am semantically also ok with these checks. I.e. this all lead to aborts.
Here this might help to at least build again: -Xnvlink --suppress-stack-size-warning See: https://github.com/kokkos/kokkos/issues/2039