[libc-commits] [PATCH] D152603: [libc] Add math functions to AMD/NVPTX libm

Siva Chandra via Phabricator via libc-commits libc-commits at lists.llvm.org
Wed Jun 21 15:49:13 PDT 2023


sivachandra added a comment.

In D152603#4439280 <https://reviews.llvm.org/D152603#4439280>, @jhuber6 wrote:

> A lot of the intrinsics being implemented by @arsenm seem to be porting `ocml` functions to the backend. I wonder if a better approach for `libc` specifically would be to port those implementations to C++ so that we can share it between NVPTX and AMDGPU.

If you can actually use the existing C++ implementations or extend/tweak the existing ones to suit GPU architecture, we would not have had to use the builtins and vendor implementations in the first place?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D152603/new/

https://reviews.llvm.org/D152603



More information about the libc-commits mailing list