[all-commits] [llvm/llvm-project] ac534d: [X86] combineArithReduction - use PACKUSWB directl...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ac534d2a16bb3c77eed050f1d43d4f87349ca097
      https://github.com/llvm/llvm-project/commit/ac534d2a16bb3c77eed050f1d43d4f87349ca097
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2023-10-25 (Wed, 25 Oct 2023)

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

  Log Message:
  -----------
  [X86] combineArithReduction - use PACKUSWB directly for PSADBW(TRUNCATE(v8i16 X)) reduction patterns

Avoids a crash in the D152928 patch due to a reduction pattern appearing after legalization

We can probably extend this further to avoid truncating to sub-128-bit vXi8 (and then calling WidenToV16I8) entirely, but we can't currently hit other cases.