[all-commits] [llvm/llvm-project] 0e6a48: [Hexagon] Optimize post-increment load and stores ...

[Sumanth Gundapaneni via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 0e6a48c3e8cc53f9eb5945ec04f8e03f6d2bae37
      https://github.com/llvm/llvm-project/commit/0e6a48c3e8cc53f9eb5945ec04f8e03f6d2bae37
  Author: Sumanth Gundapaneni <sgundapa at quicinc.com>
  Date:   2024-02-16 (Fri, 16 Feb 2024)

  Changed paths:
    M llvm/lib/Target/Hexagon/CMakeLists.txt
    M llvm/lib/Target/Hexagon/HexagonInstrInfo.cpp
    M llvm/lib/Target/Hexagon/HexagonInstrInfo.h
    A llvm/lib/Target/Hexagon/HexagonPostIncOpt.cpp
    M llvm/lib/Target/Hexagon/HexagonTargetMachine.cpp
    M llvm/lib/Target/Hexagon/MCTargetDesc/HexagonBaseInfo.h
    A llvm/test/CodeGen/Hexagon/post-inc-vec.mir
    A llvm/test/CodeGen/Hexagon/post_inc_store.mir
    A llvm/test/CodeGen/Hexagon/postincopt-crash.mir
    A llvm/test/CodeGen/Hexagon/postincopt-dcfetch.mir
    A llvm/test/CodeGen/Hexagon/valid-offset-loadbsw4.mir

  Log Message:
  -----------
  [Hexagon] Optimize post-increment load and stores in loops. (#82011)

This patch optimizes the post-increment instructions so that we can
packetize them together.
v1 = phi(v0, v3')
v2,v3  = post_load v1, 4
v2',v3'= post_load v3, 4

This can be optimized in two ways

v1 = phi(v0, v3')
v2,v3' = post_load v1, 8
v2' = load v1, 4



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications