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

[arsenm]

Summary

This is a bit more consistent the regular operation legalization.

[foad]

Looks obvious, but aren't there any tests that show the legalized GMIR for this case?

[arsenm]

In D86246#2227115, @foad wrote:

Looks obvious, but aren't there any tests that show the legalized GMIR for this case?

The only test for it goes to the ISA, and doesn't produce an observable change there

[arsenm]

62d1fb828f918fef0c139028fed6e81c5ba1fa41

llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp

bool AMDGPULegalizerInfo::legalizeImplicitArgPtr(MachineInstr &MI,
   return true;
 }
 
 bool AMDGPULegalizerInfo::legalizeIsAddrSpace(MachineInstr &MI,
                                               MachineRegisterInfo &MRI,
                                               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;
 }
 
 // The raw.(t)buffer and struct.(t)buffer intrinsics have two offset args:
 // offset (the offset that is included in bounds checking and swizzling, to be
 // split between the instruction's voffset and immoffset fields) and soffset