[all-commits] [llvm/llvm-project] 0ba781: [X86] LowerSelect - generalize "select icmp(x, 0), ...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 0ba78182b975d8ccd8ca42b33fbf038a85a44747
      https://github.com/llvm/llvm-project/commit/0ba78182b975d8ccd8ca42b33fbf038a85a44747
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-09-05 (Thu, 05 Sep 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp

  Log Message:
  -----------
  [X86] LowerSelect - generalize "select icmp(x,0), lhs, rhs" folding patterns [NFC] (#107374)

This patch proposes we add a LowerSELECTWithCmpZero helper, which allows us to fold the compare-with-zero from different condition nodes with minimal duplication.

So far I've only handled the simple no-cmov case for or/xor nodes, but the intention is to convert more folds in future PRs.

NFC preliminary patch for #107272



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