[PATCH] D78492: [ARM][ParallelDSP] Handle squaring multiplies

[Sam Parker via Phabricator via llvm-commits]

samparker created this revision.
samparker added reviewers: olista01, SjoerdMeijer, efriedma.
Herald added subscribers: danielkiss, hiraditya, kristof.beyls.
Herald added a project: LLVM.

The logic in ARMParallelDSP is setup to merge two 16-bits loads into a 32-bit load and feed them into the smlads. This requires that four loads are combined for the four inputs, but there wasn't actually a check for this.

Bugzilla: https://bugs.llvm.org/show_bug.cgi?id=45616


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D78492

Files:
  llvm/lib/Target/ARM/ARMParallelDSP.cpp
  llvm/test/CodeGen/ARM/ParallelDSP/squaring.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D78492.258730.patch
Type: text/x-patch
Size: 14060 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200420/66b21447/attachment.bin>