libcudacxx
libcudacxx copied to clipboard
NVIDIA
[ARCHIVED] The C++ Standard Library for your entire system. See https://github.com/NVIDIA/cccl
@griwes has a plan for this. This will let us solve a number of problems, such as enabling CUDA packaging to glob our headers instead of adding them manually.
Especially after the refactoring away from submodules, I find the name of this directory very confusing; the upstream tests (e.g. raw libc++ tests) live in `libcxx/test`, not `.upstream-tests`.
https://nvidia.github.io/libcudacxx/releases/versioning.html#libcu-abi-versioning says: > A snapshot of the codebase may support multiple ABI versions at the same time. They always use the latest available ABI version by default. The macro `_LIBCUDACXX_CUDA_ABI_VERSION_LATEST`...
From internal GitLab issue 1: > __memcpy_async for large extents, especially with group, should opportunistically widen the LDGSTS. For example, sizeof(T) == 4 but extents are actually aligned on 16bytes...
Closes #137.
Design: #129 Depends on #105 Building of off [`cuda::memory_resource`](https://github.com/NVIDIA/libcudacxx/pull/105) to introduce `cuda::polymorphic_allocator`-like types
Hello, suggested by @allisonvacanti I am opening an issue for information gathering. I am wondering if libcu++ can support half complex natively or not. We are evaluating the possibility of...
