[PATCH] D116529: [GlobalISel] Fold or of shifts with constant amount to funnel shift.

[Abinav Puthan Purayil via Phabricator via llvm-commits]

abinavpp created this revision.
abinavpp added reviewers: aemerson, arsenm, foad, paquette.
Herald added subscribers: kerbowa, hiraditya, rovka, nhaehnle, jvesely.
abinavpp requested review of this revision.
Herald added subscribers: llvm-commits, wdng.
Herald added a project: LLVM.

This change folds (or (shl x, C0), (lshr y, C1 <https://reviews.llvm.org/C1>)) to funnel shift iff C0
and C1 <https://reviews.llvm.org/C1> are constants where C0 + C1 <https://reviews.llvm.org/C1> is the bit-width of the shift
instructions.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D116529

Files:
  llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
  llvm/test/CodeGen/AArch64/arm64-rev.ll
  llvm/test/CodeGen/AMDGPU/GlobalISel/combine-fsh.mir
  llvm/test/CodeGen/AMDGPU/GlobalISel/combine-rot.mir
  llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll
  llvm/test/CodeGen/AMDGPU/GlobalISel/uaddsat.ll
  llvm/test/CodeGen/AMDGPU/GlobalISel/usubsat.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D116529.397024.patch
Type: text/x-patch
Size: 32563 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20220103/a87c9f59/attachment.bin>