[PATCH] D118977: [NVPTX] Add more FMA intriniscs/builtins

[Jakub Chlanda via Phabricator via llvm-commits]

jchlanda created this revision.
jchlanda added reviewers: tra, csigg, bkramer.
Herald added subscribers: asavonic, hiraditya, jholewinski.
jchlanda requested review of this revision.
Herald added subscribers: llvm-commits, cfe-commits, jdoerfert.
Herald added projects: clang, LLVM.

This patch adds builtins/intrinsics for the following variants of FMA:

- f16
  - rn
  - rn_ftz
  - rn_sat
  - rn_ftz_sat
  - rn_relu
  - rn_ftz_relu
- f16x2
  - rn
  - rn_ftz
  - rn_sat
  - rn_ftz_sat
  - rn_relu
  - rn_ftz_relu
- bf16
  - rn
  - rn_relu
- bf16x2
  - rn
  - rn_relu

They all require PTX 7.0, SM_80.

`ptxas` (Cuda compilation tools, release 11.0, V11.0.194) is happy with the generated assembly.

Depends on D<117887>


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D118977

Files:
  clang/include/clang/Basic/BuiltinsNVPTX.def
  clang/test/CodeGen/builtins-nvptx-native-half-type.c
  clang/test/CodeGen/builtins-nvptx.c
  llvm/include/llvm/IR/IntrinsicsNVVM.td
  llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
  llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.cpp
  llvm/test/CodeGen/NVPTX/math-intrins-sm80-ptx70-instcombine.ll
  llvm/test/CodeGen/NVPTX/math-intrins-sm80-ptx70.ll
  llvm/test/CodeGen/NVPTX/math-intrins-sm86-ptx72.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D118977.405891.patch
Type: text/x-patch
Size: 40262 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20220204/27834754/attachment.bin>