Michael Schellenberger Costa

Results 80 issues of Michael Schellenberger Costa

This adds both flavors of the new C++20 shift algorithms. Fixes #1484 NOTE: my stone aged desktop does not have a cuda capable device and it might explode if I...

type: enhancement
P2: nice to have

the `throw()` specification has been removed with C++20 and will error out on us. So rather than that, simply use noexcept, as C++03 is thankfully a thing of the past...

Our old implementation of `tuple` has a lot of limitations. So rather than using the old homegrown version reuse `cuda::std::tuple`

clang-format is generally confused by our `__host__` and `__device__` macros, severely messing up formatting. We can replace that by a single macro and tell clang-format to not mess with its...

Currently structured bindings for `cuda::std::tuple` and `cuda::std::array` were broken. The reason for that is that the standard requires, that the specializations of `tuple_size` and `tuple_element` reside in namespace std. whereas...

This implements the NVIDIA internal proposal for a `any_resource_ref` abstraction that facilitates the handling of memory allocations on host and device side. To facilitate adoption the proposal is backported to...

In addition to porting the span implementation from libc++ the PR does the following: * Remove outdated support for `tuple` interface * Remove outdated support for `const_iterator` * Change `index_type`...

This also ports all the standard concepts from libc++ preserving the new modular organization. This will be a base for further development such as `memory_resource` or in the far future...

This is meant as a PoC for integrating the new `memory_resource` design that we are currently implementing for libcu++ Everything here is meant as a basis for discussion and potential...

CMake
cpp