[PATCH] D67968: [TableGen] Introduce a generic automaton (DFA) backend

[Mikael Holmén via Phabricator via llvm-commits]

uabelho added a comment.

clang-cuda-build buildbot failed too:
http://lab.llvm.org:8011/builders/clang-cuda-build/builds/37865/steps/ninja%20check%201/logs/stdio

I think it can be fixed with

  @@ -371,20 +371,20 @@ uint64_t Transition::transitionFrom(uint64_t State) {
   void CustomDfaEmitter::printActionType(raw_ostream &OS) { OS << TypeName; }
   
   void CustomDfaEmitter::printActionValue(action_type A, raw_ostream &OS) {
     const ActionTuple &AT = Actions[A];
     if (AT.size() > 1)
  -    OS << "{";
  +    OS << "std::make_tuple(";
     bool First = true;
     for (const auto &SingleAction : AT) {
       if (!First)
         OS << ", ";
       First = false;
       SingleAction.print(OS);
     }
     if (AT.size() > 1)
  -    OS << "}";
  +    OS << ")";
   }
   
   namespace llvm {
   
   void EmitAutomata(RecordKeeper &RK, raw_ostream &OS) {

similar to the fix in r372384 (169cb6347 <https://reviews.llvm.org/rG169cb63478aa047451786d8ccf6af4b721e3b271>).


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D67968/new/

https://reviews.llvm.org/D67968