rjmccall added a comment. The goal here seems to be to avoid the need to store pragma state in operator expressions. As I mentioned in another review, I'm not sure how directly interesting that goal is if we can avoid memory overhead in the common case. Storing pragma state in operators certainly has a substantial complexity cost, but forcing every AST client to track the currently-active pragma state also has a cost.
Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D77545/new/ https://reviews.llvm.org/D77545 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
