[llvm] [AMDGPU][CodeGen][True16] Track waitcnt as vgpr32 instead of vgpr16 for D16 Instructions in GFX11 (PR #157795)

[Brox Chen via llvm-commits]

================
@@ -845,6 +845,15 @@ RegInterval WaitcntBrackets::getRegInterval(const MachineInstr *MI,
     assert(Result.first >= 0 && Result.first < SQ_MAX_PGM_VGPRS);
     assert(Size % 16 == 0);
     Result.second = Result.first + (Size / 16);
+
+    if (Size == 16 && Context->ST->has16bitD16HWBug()) {
----------------
broxigarchen wrote:

I think we currently only have D16 load and store (flat/global/scratch/ds) that creates 16bit vgpr dependency here. Hope I didn't miss other cases

https://github.com/llvm/llvm-project/pull/157795