[all-commits] [llvm/llvm-project] 140680: [X86] Add peephole for (add (concat_vectors vpmadd...

Simon Pilgrim via All-commits all-commits at lists.llvm.org
Fri Dec 6 08:25:45 PST 2024


  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 140680c5c8560f03b7a4b3f7db5a7f3c158b938a
      https://github.com/llvm/llvm-project/commit/140680c5c8560f03b7a4b3f7db5a7f3c158b938a
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-12-06 (Fri, 06 Dec 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/vpdpwssd.ll

  Log Message:
  -----------
  [X86] Add peephole for (add (concat_vectors vpmaddwd, vpmaddwd)) -> vpdpwssd on VNNI targets

Cleanup for #118433


  Commit: 6bc3c9ee6bfbace7275dedfddef1cb8f177aa491
      https://github.com/llvm/llvm-project/commit/6bc3c9ee6bfbace7275dedfddef1cb8f177aa491
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-12-06 (Fri, 06 Dec 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-5.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-6.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-5.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-6.ll

  Log Message:
  -----------
  [X86] combineX86ShuffleChain - always create VPERMV3 nodes if started from a VPERMV3 node

If the root shuffle node was a VPERMV3 node, then we can always replace it with a new VPERMV3 node - it doesn't matter if other variable shuffles in the chain had multiple uses.


  Commit: 9ad22cf0ee9eb686ea9fd76fd9cc41f4c006ccbc
      https://github.com/llvm/llvm-project/commit/9ad22cf0ee9eb686ea9fd76fd9cc41f4c006ccbc
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-12-06 (Fri, 06 Dec 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/vector-shuffle-512-v64.ll

  Log Message:
  -----------
  [X86] lowerV32I16Shuffle - attempt to fold unary shuffle to lane permute + repeated mask

Fixes #79799


Compare: https://github.com/llvm/llvm-project/compare/d6ec7c82f383...9ad22cf0ee9e

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


More information about the All-commits mailing list