[llvm] [AMDGPU] Convert SVS form scratch frame index fallback into SV form (PR #219239)

[Lukas Sommer via llvm-commits]

================
@@ -3285,12 +3287,36 @@ bool SIRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator MI,
           auto [ImmOffset, RemainderOffset] =
               TII->splitFlatOffset(FullOffset, AMDGPUAS::PRIVATE_ADDRESS,
                                    AMDGPU::FlatAddrSpace::FlatScratch);
-          BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_MOV_B32_e32), TmpVGPR)
-              .addImm(RemainderOffset);
+
+          Register UsedVAddr;
+          if (MachineOperand *VAddr =
+                  TII->getNamedOperand(*MI, AMDGPU::OpName::vaddr)) {
+            MachineOperand *VData =
+                TII->getNamedOperand(*MI, AMDGPU::OpName::vdata);
+            bool CanReuseVAddr = VAddr->isKill() &&
+                                 !(VData && VAddr->getReg() == VData->getReg());
----------------
sommerlukas wrote:

Probably even more unlikely to happen, but should we use `regsOverlap` instead of `==` here to make sure `vdata` and `vaddr` don't overlap?

https://github.com/llvm/llvm-project/pull/219239