[PATCH] D31951: TableGen support for parametrized register class information

[Alex Bradbury via Phabricator via llvm-commits]

asb added a comment.

Phew, this is a monster patch to review but clearly there's no real way it could be reduced further. I have no real experience with the TableGen implementation, which limits my ability to contribute to this code review. I have at least carefully read through the code to try to spot any obvious logic errors or issues.

One question I'd welcome a second opinion on - is the introduction of the `vi::` namespace justified?



================
Comment at: utils/TableGen/CodeGenDAGPatterns.cpp:495
+      if (B.empty()) {
+        TP.error("empty2a");
+        return Changed;
----------------
Is it worth having a more meaningful error message for these? Maybe "Set unexpectedly empty (ref 2a)"?


Repository:
  rL LLVM

https://reviews.llvm.org/D31951