Mark await expressions as caught or uncaught

src/ast/ast-numbering.cc

Index: src/ast/ast-numbering.cc
diff --git a/src/ast/ast-numbering.cc b/src/ast/ast-numbering.cc
index d0101828d6897f99c91e069024460c4aae10ebf6..6c37128941a9fcd6b2f76e00908469d7b6fd0be2 100644
--- a/src/ast/ast-numbering.cc
+++ b/src/ast/ast-numbering.cc
@@ -249,6 +249,12 @@ void AstNumberingVisitor::VisitCallRuntime(CallRuntime* node) {
   IncrementNodeCount();
   node->set_base_id(ReserveIdRange(CallRuntime::num_ids()));
   VisitArguments(node->arguments());
+  // Indicate the catch prediction to the running async program
+  if (node->is_jsruntime() &&
+      node->context_index() == Context::ASYNC_FUNCTION_AWAIT_CAUGHT_INDEX &&
+      catch_prediction_ == HandlerTable::ASYNC_AWAIT) {
+    node->set_context_index(Context::ASYNC_FUNCTION_AWAIT_UNCAUGHT_INDEX);

[jgruber, 2016/09/08 10:19:59]

Changing the called runtime function from the AstNumberingVisitor seems very
unexpected to me. At the very least there should be a huge comment about this
where the runtime call is created in the parser. 
But maybe someone else should weigh in here that knows more about this than me.

Is it not possible to do this from the parser?

+  }
 }