[lldb] [llvm] [AArch64][llvm] Tighten SYSP parsing; don't disassemble invalid encodings (PR #182410)

[Jonathan Thackray via llvm-commits]

================
@@ -8701,52 +8700,56 @@ ParseStatus AArch64AsmParser::tryParseGPRSeqPair(OperandVector &Operands) {
   MCRegister FirstReg;
   ParseStatus Res = tryParseScalarRegister(FirstReg);
   if (!Res.isSuccess())
-    return Error(S, "expected first even register of a consecutive same-size "
-                    "even/odd register pair");
+    return Error(S, FirstRegExpected);
 
   const MCRegisterClass &WRegClass =
       AArch64MCRegisterClasses[AArch64::GPR32RegClassID];
   const MCRegisterClass &XRegClass =
       AArch64MCRegisterClasses[AArch64::GPR64RegClassID];
 
-  bool isXReg = XRegClass.contains(FirstReg),
-       isWReg = WRegClass.contains(FirstReg);
-  if (!isXReg && !isWReg)
-    return Error(S, "expected first even register of a consecutive same-size "
-                    "even/odd register pair");
+  bool IsXZRPair = AllowXZRPair && FirstReg == AArch64::XZR;
+  bool IsXReg = XRegClass.contains(FirstReg);
----------------
jthackray wrote:

In the function header a few lines above, the argument is `std::optional<unsigned> WPairClass`, whereas we always have `unsigned XPairClass` passed in. So we don't need it.

https://github.com/llvm/llvm-project/pull/182410