[llvm] [AMDGPU] Add regbankselect rules for G_ADD/SUB and variants (PR #159860)

[Anshil Gandhi via llvm-commits]

================
@@ -500,6 +500,24 @@ void RegBankLegalizeHelper::lowerUnpackMinMax(MachineInstr &MI) {
   MI.eraseFromParent();
 }
 
+void RegBankLegalizeHelper::lowerScalarizeV2S16(MachineInstr &MI) {
----------------
gandhi56 wrote:

Right, I will use a G_BITCAST to cast from V2S16 to S32 for the G_ADD then.

https://github.com/llvm/llvm-project/pull/159860