[libc-commits] [PATCH] D152486: [libc] Begin implementing a 'libmgpu.a' for math on the GPU

Joseph Huber via Phabricator via libc-commits libc-commits at lists.llvm.org
Mon Jun 12 05:34:54 PDT 2023


jhuber6 updated this revision to Diff 530475.
jhuber6 added a comment.

Tweak a few things.

I think what would be valuable moving forward is a way to choose between the
implementation in `libc` and simply using the vendor libraries. That would allow
us to provide a `libm.a` that is immediately useful and then perform an
iterative implementation of other math functions.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D152486/new/

https://reviews.llvm.org/D152486

Files:
  libc/config/gpu/entrypoints.txt
  libc/config/gpu/headers.txt
  libc/src/math/gpu/CMakeLists.txt
  libc/src/math/gpu/amdgpu/amdgpu.h
  libc/src/math/gpu/amdgpu/declarations.h
  libc/src/math/gpu/amdgpu/platform.h
  libc/src/math/gpu/common.h
  libc/src/math/gpu/nvptx/declarations.h
  libc/src/math/gpu/nvptx/nvptx.h
  libc/src/math/gpu/sin.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D152486.530475.patch
Type: text/x-patch
Size: 11899 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/libc-commits/attachments/20230612/92fc6f51/attachment.bin>


More information about the libc-commits mailing list