[llvm] 3cf4896 - [AMDGPU] Remove old isCheapToSpeculateCttz FIXME

[Simon Pilgrim via llvm-commits]

Author: Simon Pilgrim
Date: 2022-08-24T15:53:38+01:00
New Revision: 3cf48963ff5a6df0c41c0a3a7afd4a0f6a2ffeab

URL: https://github.com/llvm/llvm-project/commit/3cf48963ff5a6df0c41c0a3a7afd4a0f6a2ffeab
DIFF: https://github.com/llvm/llvm-project/commit/3cf48963ff5a6df0c41c0a3a7afd4a0f6a2ffeab.diff

LOG: [AMDGPU] Remove old isCheapToSpeculateCttz FIXME

As confirmed on D132520 - this should always return true

Added: 
    

Modified: 
    llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index 73bd9817f3903..3f16367add784 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -692,7 +692,6 @@ bool AMDGPUTargetLowering::isLoadBitCastBeneficial(EVT LoadTy, EVT CastTy,
 // SI+ has instructions for cttz / ctlz for 32-bit values. This is probably also
 // profitable with the expansion for 64-bit since it's generally good to
 // speculate things.
-// FIXME: These should really have the size as a parameter.
 bool AMDGPUTargetLowering::isCheapToSpeculateCttz() const {
   return true;
 }