[PATCH] D96599: [AArch64] Allow folding FMUL/FADD into FMA for FP16 types

[Bradley Smith via Phabricator via llvm-commits]

bsmith created this revision.
bsmith added reviewers: joechrisellis, peterwaller-arm, dmgreen.
Herald added subscribers: danielkiss, hiraditya, kristof.beyls.
bsmith requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

isFMAFasterThanFMulAndFAdd should return true for FP16 types when
HasFullFP16 is present, since we have the instructions to handle it for
both SVE and NEON.

Depends on D96424 <https://reviews.llvm.org/D96424>


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D96599

Files:
  llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
  llvm/test/CodeGen/AArch64/f16-instructions.ll
  llvm/test/CodeGen/AArch64/sve-fp-combine.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D96599.323301.patch
Type: text/x-patch
Size: 7057 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210212/546982e5/attachment.bin>