[all-commits] [llvm/llvm-project] b02b5b: [AMDGPU] Simplify use of hasMovrel and hasVGPRInde...

[Jay Foad via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b02b5b7b598ff146f8d5ed529412236533429403
      https://github.com/llvm/llvm-project/commit/b02b5b7b598ff146f8d5ed529412236533429403
  Author: Jay Foad <jay.foad at amd.com>
  Date:   2024-08-23 (Fri, 23 Aug 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/GCNSubtarget.cpp
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/indirect-addressing-si.ll

  Log Message:
  -----------
  [AMDGPU] Simplify use of hasMovrel and hasVGPRIndexMode (#105680)

The generic subtarget has neither of these features. Rather than forcing
HasMovrel on, it is simpler to expand dynamic vector indexing to a
sequence of compare/select instructions.

NFC for real subtargets.



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