[PATCH] D154738: [SLP]Introduce isValiVectorOp to check if the vector instruction is going to be scalarized.

[Alexey Bataev via Phabricator via llvm-commits]

ABataev created this revision.
ABataev added reviewers: RKSimon, reames.
Herald added subscribers: luke, vporpo, frasercrmck, kerbowa, luismarques, apazos, sameer.abuasal, s.egerton, Jim, jocewei, PkmX, the_o, brucehoult, MartinMosbeck, rogfer01, edward-jones, zzheng, jrtc27, niosHD, sabuasal, simoncook, johnrusso, rbar, asb, hiraditya, jvesely.
Herald added a project: All.
ABataev requested review of this revision.
Herald added subscribers: wangpc, MaskRay.
Herald added a project: LLVM.

If the going-to-be-generated vector instruction is going to be
scalarized, not need to try to vectorize it. Instead, better to generate
buildvector/gather node and try-to-vectorize operands independently. It
shall improve vectorization quality/performance.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D154738

Files:
  llvm/include/llvm/Analysis/TargetTransformInfo.h
  llvm/include/llvm/Analysis/TargetTransformInfoImpl.h
  llvm/include/llvm/CodeGen/BasicTTIImpl.h
  llvm/lib/Analysis/TargetTransformInfo.cpp
  llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
  llvm/test/Transforms/SLPVectorizer/AArch64/accelerate-vector-functions-inseltpoison.ll
  llvm/test/Transforms/SLPVectorizer/AArch64/accelerate-vector-functions.ll
  llvm/test/Transforms/SLPVectorizer/AMDGPU/packed-math.ll
  llvm/test/Transforms/SLPVectorizer/RISCV/math-function.ll
  llvm/test/Transforms/SLPVectorizer/X86/arith-div-undef.ll
  llvm/test/Transforms/SLPVectorizer/X86/control-dependence.ll
  llvm/test/Transforms/SLPVectorizer/X86/multi-nodes-to-shuffle.ll
  llvm/test/Transforms/SLPVectorizer/X86/sin-sqrt.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D154738.538224.patch
Type: text/x-patch
Size: 64566 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230707/64dc1541/attachment.bin>