[llvm] [XCore] Fix crash lowering unsized/zero-length-array thread_local globals (PR #220543)

[Bill Wendling via llvm-commits]

================
@@ -121,21 +122,24 @@ static bool rewriteNonInstructionUses(GlobalVariable *GV, Pass *P) {
   return true;
 }
 
-static bool isZeroLengthArray(Type *Ty) {
-  ArrayType *AT = dyn_cast<ArrayType>(Ty);
-  return AT && (AT->getNumElements() == 0);
-}
-
 bool XCoreLowerThreadLocal::lowerGlobal(GlobalVariable *GV) {
   Module *M = GV->getParent();
   if (!GV->isThreadLocal())
     return false;
 
-  // Skip globals that we can't lower and leave it for the backend to error.
-  if (!rewriteNonInstructionUses(GV, this) ||
-      !GV->getType()->isSized() || isZeroLengthArray(GV->getType()))
+  if (!rewriteNonInstructionUses(GV, this))
     return false;
 
+  // The lowered representation needs an ArrayType of the value type, which
+  // requires a known per-element stride: reject anything that can't provide
+  // one now, with a clear diagnostic, rather than emitting a malformed GEP
+  // that only fails much later (and much less clearly) in instruction
+  // selection.
+  if (!GV->getValueType()->isSized() ||
+      GV->getGlobalSize(M->getDataLayout()) == 0)
+    reportFatalUsageError("Size of thread local object '" + GV->getName() +
----------------
isanbard wrote:

Okay, I'm just having trouble with how it's represented in C or C++...(Well, C++ is weird enough that it's probably very simple to do...)

https://github.com/llvm/llvm-project/pull/220543