[all-commits] [llvm/llvm-project] b0fb93: [X86] LowerStore - peek through oneuse bitcasts to...

Simon Pilgrim via All-commits all-commits at lists.llvm.org
Tue Sep 29 08:23:42 PDT 2026


  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b0fb9358d471a0436b72b0f117bf9d1a1ca3216e
      https://github.com/llvm/llvm-project/commit/b0fb9358d471a0436b72b0f117bf9d1a1ca3216e
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2026-09-29 (Tue, 29 Sep 2026)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/frem.ll
    M llvm/test/CodeGen/X86/half.ll
    M llvm/test/CodeGen/X86/i64-mem-copy.ll
    M llvm/test/CodeGen/X86/oddshuffles.ll
    M llvm/test/CodeGen/X86/subvectorwise-store-of-vector-splat.ll
    M llvm/test/CodeGen/X86/vector-half-conversions.ll
    M llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-3.ll
    M llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-4.ll
    M llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-2.ll
    M llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-4.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-4.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-6.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-8.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-5.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-6.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-8.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-4.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-6.ll
    M llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-8.ll
    M llvm/test/CodeGen/X86/vector-shuffle-512-v64.ll
    M llvm/test/CodeGen/X86/vector-shuffle-concatenation.ll
    M llvm/test/CodeGen/X86/x86-interleaved-access.ll

  Log Message:
  -----------
  [X86] LowerStore - peek through oneuse bitcasts to see if vector is freely splittable. (#227282)

Peek through bitcasts when seeing if we can avoid an unnecessary
concat_vectors and store the subvectors directly - if the concat had
been worth it, combineConcatVectorOps would have moved the concat
further up.

Helps avoid more cases of unnecessary vzeroupper, 256-bit usage etc. in
particular on AVX1 targets (Sandybridge, Jaguar and Bulldozer all
benefit from this), but AVX2/512 targets as well.



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