[all-commits] [llvm/llvm-project] 1df379: [X86] early-ifcvt-remarks.ll - use i64 arithmetic ...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 1df37980c296ab33e96038c1daa06d580ae8b925
      https://github.com/llvm/llvm-project/commit/1df37980c296ab33e96038c1daa06d580ae8b925
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-06-11 (Tue, 11 Jun 2024)

  Changed paths:
    M llvm/test/CodeGen/X86/early-ifcvt-remarks.ll

  Log Message:
  -----------
  [X86] early-ifcvt-remarks.ll - use i64 arithmetic to ensure ifcvt doesn't drop below threshold

Upcoming SimplifyDemandedBits support for CMOV will simplify the code and reduce the critical path below the threshold if we stick with i32 multiplies


  Commit: 464eb648fa33c46f51f7c3d1f4ab3eef7ec12750
      https://github.com/llvm/llvm-project/commit/464eb648fa33c46f51f7c3d1f4ab3eef7ec12750
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-06-11 (Tue, 11 Jun 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/widen-load-of-small-alloca.ll

  Log Message:
  -----------
  [X86] SimplifyDemandedBitsForTargetNode - add basic X86ISD::CMOV handling

Add basic pass through handling - we could extend this to truncate CMOVQ to CMOVL in a future patch


Compare: https://github.com/llvm/llvm-project/compare/c0e1ad779f8b...464eb648fa33

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