[PATCH] D90942: [SVE][CodeGen] Improve codegen of scalable masked scatters

[Kerry McLaughlin via Phabricator via llvm-commits]

kmclaughlin created this revision.
kmclaughlin added reviewers: sdesmalen, efriedma, fhahn, craig.topper.
Herald added subscribers: llvm-commits, ecnelises, psnobl, hiraditya, tschuett.
Herald added a project: LLVM.
kmclaughlin requested review of this revision.

If the scatter store is able to perform the sign/zero extend of
its index, this is folded into the instruction with refineIndexType().
Additionally, refineUniformBase() will return the base pointer and index
from an add + splat_vector.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D90942

Files:
  llvm/include/llvm/CodeGen/TargetLowering.h
  llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
  llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
  llvm/lib/Target/AArch64/AArch64ISelLowering.h
  llvm/test/CodeGen/AArch64/sve-masked-scatter-32b-scaled.ll
  llvm/test/CodeGen/AArch64/sve-masked-scatter-32b-unscaled.ll
  llvm/test/CodeGen/AArch64/sve-masked-scatter-64b-unscaled.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D90942.303439.patch
Type: text/x-patch
Size: 41824 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201106/3fe63db9/attachment.bin>