[all-commits] [llvm/llvm-project] f1bbab: [ARM] Lower arm_neon_vbsl to ARMISD::VBSP and fold...

[David Green via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: f1bbabd6289a351430657a2eb3b57fffa8c6d248
      https://github.com/llvm/llvm-project/commit/f1bbabd6289a351430657a2eb3b57fffa8c6d248
  Author: David Green <david.green at arm.com>
  Date:   2024-09-25 (Wed, 25 Sep 2024)

  Changed paths:
    M llvm/lib/Target/ARM/ARMISelLowering.cpp
    M llvm/lib/Target/ARM/ARMInstrNEON.td
    M llvm/test/CodeGen/ARM/vbsl.ll

  Log Message:
  -----------
  [ARM] Lower arm_neon_vbsl to ARMISD::VBSP and fold (vbsl x, y, y) to y (#109761)

This helps clean up the patterns a little and will help share combines
on both the intrinsic and VBSP. A combine is then added to fold away the
VBSP if both the selected operands are the same.



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