[PATCH] D99281: [AArch64][GlobalISel] Combine funnel shifts to AArch64 rotate opcodes.

[Amara Emerson via Phabricator via llvm-commits]

aemerson created this revision.
aemerson added reviewers: paquette, arsenm, foad.
aemerson added a project: LLVM.
Herald added subscribers: danielkiss, hiraditya, kristof.beyls, rovka.
aemerson requested review of this revision.
Herald added a subscriber: wdng.

Adds AArch64::G_ROR opcode to allow us to import patterns for selection.

This fixes the 0.5% size regression on ClamAV introduced when G_FSHL/G_FSHR started being lowered instead of falling back.

The 0.7% regression on consumer-typeset is still remaining.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D99281

Files:
  llvm/include/llvm/CodeGen/GlobalISel/CombinerHelper.h
  llvm/lib/Target/AArch64/AArch64Combine.td
  llvm/lib/Target/AArch64/AArch64InstrGISel.td
  llvm/lib/Target/AArch64/GISel/AArch64PreLegalizerCombiner.cpp
  llvm/test/CodeGen/AArch64/GlobalISel/prelegalizercombiner-funnel-shifts-to-rotates.mir
  llvm/test/CodeGen/AArch64/GlobalISel/select-rotates.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D99281.333035.patch
Type: text/x-patch
Size: 11593 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210324/e9a8b049/attachment.bin>