[PATCH] D111846: [LV] Drop integer poison-generating flags from instructions that need predication

[Diego Caballero via Phabricator via llvm-commits]

dcaballe updated this revision to Diff 384745.
dcaballe marked 2 inline comments as done.
dcaballe added a comment.

Addressed feedback.
Changed the approach to improve the complexity of detecting the
poison-generating recipes. We now gather these recipes before
executing the VPlan, starting from the recipes generating a widen
load/store and traversing the backward slice from their address
operand. In this way, recipes are only visited once.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D111846/new/

https://reviews.llvm.org/D111846

Files:
  llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
  llvm/test/Transforms/LoopVectorize/AArch64/sve-masked-loadstore.ll
  llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse-mask4.ll
  llvm/test/Transforms/LoopVectorize/AArch64/vector-reverse-mask4.ll
  llvm/test/Transforms/LoopVectorize/X86/drop-poison-generating-flags.ll
  llvm/test/Transforms/LoopVectorize/X86/gather_scatter.ll
  llvm/test/Transforms/LoopVectorize/X86/invariant-store-vectorization.ll
  llvm/test/Transforms/LoopVectorize/X86/load-deref-pred.ll
  llvm/test/Transforms/LoopVectorize/X86/masked_load_store.ll
  llvm/test/Transforms/LoopVectorize/X86/x86-interleaved-accesses-masked-group.ll
  llvm/test/Transforms/LoopVectorize/X86/x86-interleaved-store-accesses-with-gaps.ll
  llvm/test/Transforms/LoopVectorize/X86/x86-pr39099.ll
  llvm/test/Transforms/LoopVectorize/single-value-blend-phis.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D111846.384745.patch
Type: text/x-patch
Size: 198086 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20211104/b7045010/attachment-0001.bin>