maddyscientist

Implement `std::array`

3

Static array storage is commonly used in CUDA programs, and many folks simply resort to reinventing the wheel of `std::array` on the device. It would be highly desirable to have...

libcu++ atomic fails to compile with address sanitizer (ASAN) enabled

2

In testing libcu++ for deployment in QUDA, I found that libcu++ atomic fails to compile if address sanitizer is enabled. ```bash $ nvcc bug.cu -o bug -arch=sm_70 -Xcompiler -fsanitize=address -Ilibcudacxx/include...

Jitify is using both `std::map` and `std::unordered_map` with no real thought as to which is more suitable. This is a request to clean this up to use the most appropropriate...

XLC isn't the most robust C++11 compiler, so perhaps not surprising. ``` jitify.hpp:2115:8: error: expression contains unexpanded parameter pack 'args' return (this->launch(std::vector< void *> ({((void *)(&args))}), {reflection::reflect< ArgTypes> ()...})); ```...

This shouldn't be too hard to do. At the moment these functions are only used in the multigrid coarse grid construction, and we should really test these operations separately in...

The QUDA Chroma wiki [page](https://github.com/lattice/quda/wiki/Chroma-with-QUDA) is out of date for the latest CMake driven build approach and should be updated.

Add Citation file

1

And maybe look into correct Zenodo attribution

We don't need to tune starting from grid size 1.

Requested by @jcosborn and @jxy. I'll do this post GK merge.

Use a macro to replace `__device__ __host__` for marking parallel functions

1

This is something to do post GK merge. Some possible options include ``` quda_parallel quda_offload QUDA_PARALLEL QUDA_OFFLOAD ``` For aesthetic reasons I prefer lower case, but understand the reasoning for...