mgehre added a comment. I see. Thinking about this, the same issue should also apply to a throw inside the compound stmt. And it's not only about suppressing the dtors but any CFG entries after return/throw. I'll prepare an updated patch.
Manuel Klimek <kli...@google.com> schrieb am So., 1. Nov. 2015 23:04: > klimek added a reviewer: jordan_rose. > klimek added a comment. > > +jordan for an opinion on whether we want to fix the more general case. > > My main problem is that while we're at it we need to fully understand the > change anyway, and if somebody runs into this later, they'll need a long > time debugging this surprising effect (like what happened to you here ;) > > http://reviews.llvm.org/D13973 http://reviews.llvm.org/D13973 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
