voyager-jhk wrote: @vgvassilev @vbvictor Yes, getting this into Clang proper is definitely the ideal goal. As vbvictor noted though, the core lifetime analysis doesn't model these escapes yet.
Doing this natively would likely require extending annotations like [[clang::lifetimebound]] or noescape. That’d be a much heavier lift. https://github.com/llvm/llvm-project/pull/203757 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
