[all-commits] [llvm/llvm-project] 68ef68: [CostModel][X86] Improve accuracy of vXi8/vXi16 ve...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 68ef68f8ac5cad1fdcd9c7b0e2a8f134d9f595ae
      https://github.com/llvm/llvm-project/commit/68ef68f8ac5cad1fdcd9c7b0e2a8f134d9f595ae
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2021-05-25 (Tue, 25 May 2021)

  Changed paths:
    M llvm/lib/Target/X86/X86TargetTransformInfo.cpp
    M llvm/test/Analysis/CostModel/X86/div.ll
    M llvm/test/Analysis/CostModel/X86/fshl.ll
    M llvm/test/Analysis/CostModel/X86/fshr.ll
    M llvm/test/Analysis/CostModel/X86/rem.ll
    M llvm/test/Analysis/CostModel/X86/vshift-ashr-cost-inseltpoison.ll
    M llvm/test/Analysis/CostModel/X86/vshift-ashr-cost.ll
    M llvm/test/Analysis/CostModel/X86/vshift-lshr-cost-inseltpoison.ll
    M llvm/test/Analysis/CostModel/X86/vshift-lshr-cost.ll
    M llvm/test/Analysis/CostModel/X86/vshift-shl-cost-inseltpoison.ll
    M llvm/test/Analysis/CostModel/X86/vshift-shl-cost.ll

  Log Message:
  -----------
  [CostModel][X86] Improve accuracy of vXi8/vXi16 vector non-uniform shift costs on AVX2/AVX512 targets

Determined from llvm-mca analysis, AVX2+ capable targets have a higher throughput for VPBLENDVB and VPMOVZX ops, making it cheaper to perform shift+select patterns for vXi8 shifts or extend/shift/truncate for vXi16 shifts. Similarly AVX512BW can perform vXi8 as extend/shift/truncate patterns.


  Commit: ed14062be0c1769130b046880199bdba3c6a2ee2
      https://github.com/llvm/llvm-project/commit/ed14062be0c1769130b046880199bdba3c6a2ee2
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2021-05-25 (Tue, 25 May 2021)

  Changed paths:
    M llvm/lib/ExecutionEngine/JITLink/MachO_x86_64.cpp

  Log Message:
  -----------
  Fix MSVC "truncation of constant value" warning. NFCI.


Compare: https://github.com/llvm/llvm-project/compare/e3b8e6d48251...ed14062be0c1