[llvm] [AMDGPU] Correctly insert s_nops for implicit read of SDWA (PR #100276)

[Matt Arsenault via llvm-commits]

================
@@ -1040,6 +1096,37 @@ int GCNHazardRecognizer::checkInlineAsmHazards(MachineInstr *IA) {
     if (Op.isReg() && Op.isDef()) {
       WaitStatesNeeded =
           std::max(WaitStatesNeeded, checkVALUHazardsHelper(Op, MRI));
+
+      if (!TRI.isVectorRegister(MRI, Op.getReg()))
+        continue;
+
+      if (ST.hasDstSelForwardingHazard()) {
----------------
arsenm wrote:

Is this redundant with the early exit above? How does this interact with ST.has12DWordStoreHazard, which seems to be ignored 

https://github.com/llvm/llvm-project/pull/100276