[llvm] [AArch64][SME] Allow memory operations lowering to custom SME functions. (PR #79263)

[Sander de Smalen via llvm-commits]

================
@@ -76,12 +76,74 @@ SDValue AArch64SelectionDAGInfo::EmitMOPS(AArch64ISD::NodeType SDOpcode,
   }
 }
 
+SDValue AArch64SelectionDAGInfo::EmitSpecializedLibcall(
+    SelectionDAG &DAG, const SDLoc &DL, SDValue Chain, SDValue Dst, SDValue Src,
+    SDValue Size, RTLIB::Libcall LC) const {
+  const AArch64Subtarget &STI =
+      DAG.getMachineFunction().getSubtarget<AArch64Subtarget>();
+  const AArch64TargetLowering *TLI = STI.getTargetLowering();
+  TargetLowering::ArgListTy Args;
+  TargetLowering::ArgListEntry Entry;
+  Entry.Ty = DAG.getDataLayout().getIntPtrType(*DAG.getContext());
+  Entry.Node = Dst;
+  Args.push_back(Entry);
+
+  enum { SME_MEMCPY = 0, SME_MEMMOVE, SME_MEMSET } SMELibcall;
----------------
sdesmalen-arm wrote:

I find this indirection using the enum and it indexing into an array very confusing, why wouldn't you just write:

```
case RTLIB::MEMCPY:
  Symbol = DAG.getExternalSymbol("__arm_sc_memcpy", PtrVT);
```

?

https://github.com/llvm/llvm-project/pull/79263