[PATCH] D60907: [OpenMP][WIP] Add math functions support in OpenMP offloading
Johannes Doerfert via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Fri Apr 19 13:48:38 PDT 2019
jdoerfert added a comment.
To follow up on my comment why this is NVPTX specific:
Is there a reason why this has to happen in the Cuda ToolChain part?
I would have assumed us to add the declarations similar to the ones provided in `__clang_openmp_math.h` whenever we may compile for a target.
So, if we have any OpenMP target related code in the TU, we add the header `__clang_openmp_target_math.h` which defines "common" math functions as you did in `__clang_openmp_math.h` (without the NVPTX guard). The runtime will then implement `__kmpc_XXXX` as it sees fit.
CHANGES SINCE LAST ACTION
More information about the cfe-commits