andykaylor wrote: > #166036 removed that semicolon, but it also added a new test. Maybe we should > copy that over, too.
The new test there looks a bit dubious. It's only checking that a void return is generated, which doesn't really have anything to do with the expression handling. The AST tells us the return type and sets up the implicit cast to void in this case. It could be interesting to verify that we are emitting code for any sub-expression with side effects, but I think that is covered elsewhere. https://github.com/llvm/llvm-project/pull/167000 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
