https://github.com/ojhunt commented:
I can't look at this too hard right now, but I _think_ this should be ok with pointer auth atm. There is an option that's not yet upstreamed that will impact this, but I think the implementation already neuters the "this type can be zero-inited" logic that influences this logic. *thinking* Does this have any impact on copy/move semantics? (Again, no time to read spec atm) https://github.com/llvm/llvm-project/pull/185886 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
