[all-commits] [llvm/llvm-project] 306c8a: [SVE][CodeGen] Improve codegen of scalable masked ...

[kmclaughlin-arm via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 306c8ab20841aceef449dd27e1d0f46c6b85be44
      https://github.com/llvm/llvm-project/commit/306c8ab20841aceef449dd27e1d0f46c6b85be44
  Author: Kerry McLaughlin <kerry.mclaughlin at arm.com>
  Date:   2020-11-13 (Fri, 13 Nov 2020)

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

  Log Message:
  -----------
  [SVE][CodeGen] Improve codegen of scalable masked scatters

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.

Reviewed By: sdesmalen

Differential Revision: https://reviews.llvm.org/D90942