[PATCH] D71067: [AArch64] Fix MUL/SUB fusing

[Sanne Wouda via Phabricator via llvm-commits]

sanwou01 created this revision.
sanwou01 added reviewers: dmgreen, SjoerdMeijer, fhahn, Gerolf, mstorsjo, asbirlea.
Herald added subscribers: llvm-commits, hiraditya, kristof.beyls.
Herald added a project: LLVM.

When MUL is the first operand to SUB, we can't use MLS because the accumulator
should be negated.  Emit a NEG of the accumulator and an MLA instead, similar to
what we do for FMUL / FSUB fusing.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D71067

Files:
  llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
  llvm/test/CodeGen/AArch64/neon-mla-mls.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D71067.232344.patch
Type: text/x-patch
Size: 11395 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191205/e1cc4035/attachment.bin>