[clang] [CIR] Upstream CIR codegen for align x86 builtins (PR #173788)

[Thibault Monnier via cfe-commits]

================
@@ -1331,13 +1332,47 @@ CIRGenFunction::emitX86BuiltinExpr(unsigned builtinID, const CallExpr *expr) {
   case X86::BI__builtin_ia32_permdf256:
   case X86::BI__builtin_ia32_permdi512:
   case X86::BI__builtin_ia32_permdf512:
-  case X86::BI__builtin_ia32_palignr128:
-  case X86::BI__builtin_ia32_palignr256:
-  case X86::BI__builtin_ia32_palignr512:
     cgm.errorNYI(expr->getSourceRange(),
                  std::string("unimplemented X86 builtin call: ") +
                      getContext().BuiltinInfo.getName(builtinID));
-    return {};
+    return mlir::Value{};
----------------
Thibault-Monnier wrote:

#171094 changed the return type to `std::optional<mlir::Value>`. This changed was made everywhere else, but seems to have been missed here.

https://github.com/llvm/llvm-project/pull/173788