[llvm] 62d1fb8 - AMDGPU/GlobalISel: Use unmerge instead of extract in addrspace queries

[Matt Arsenault via llvm-commits]

Author: Matt Arsenault
Date: 2020-08-24T11:07:51-04:00
New Revision: 62d1fb828f918fef0c139028fed6e81c5ba1fa41

URL: https://github.com/llvm/llvm-project/commit/62d1fb828f918fef0c139028fed6e81c5ba1fa41
DIFF: https://github.com/llvm/llvm-project/commit/62d1fb828f918fef0c139028fed6e81c5ba1fa41.diff

LOG: AMDGPU/GlobalISel: Use unmerge instead of extract in addrspace queries

This is a bit more consistent with regular operation legalization.

Added: 
    

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

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
index 17ac0c026912..7734d11fb2a5 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
@@ -3442,7 +3442,9 @@ bool AMDGPULegalizerInfo::legalizeIsAddrSpace(MachineInstr &MI,
                                               MachineIRBuilder &B,
                                               unsigned AddrSpace) const {
   Register ApertureReg = getSegmentAperture(AddrSpace, MRI, B);
-  auto Hi32 = B.buildExtract(LLT::scalar(32), MI.getOperand(2).getReg(), 32);
+  auto Unmerge = B.buildUnmerge(LLT::scalar(32), MI.getOperand(2).getReg());
+  Register Hi32 = Unmerge.getReg(1);
+
   B.buildICmp(ICmpInst::ICMP_EQ, MI.getOperand(0), Hi32, ApertureReg);
   MI.eraseFromParent();
   return true;