[llvm] [AMDGPU] Generate waterfall for calls with SGPR(inreg) argument (PR #146997)

[via llvm-commits]

================
@@ -124,6 +124,12 @@ class SIFixSGPRCopies {
   SmallVector<MachineInstr*, 4> RegSequences;
   SmallVector<MachineInstr*, 4> PHINodes;
   SmallVector<MachineInstr*, 4> S2VCopies;
+  struct V2PysSCopyInfo {
----------------
Shoreshen wrote:

Hi @jmmartinez , this is for the convenient of calling loadMBUFScalarOperandsFromVGPR. If we put the pair, then we have to create 2 vectors for the call. 

Or maybe we should change the parameters of loadMBUFScalarOperandsFromVGPR?? Thanks

https://github.com/llvm/llvm-project/pull/146997