AaronBallman wrote: > After it's done, the builtin provided here can be either removed or replaced > with a version wrapping the more efficient one
Apologies if I've missed something in the various conversations, but this catches my eye as a concern. Once we release the builtin to the wild, people will start using it, which means removing or replacing it becomes more effort for us and more pain for our users. So I'm worried we'll end up in the very bad situation where we have two builtins that do very similar things and users have to figure out which to use and why. Is there something holding up #106730? It'd be better for us to land the correct solution instead of a stopgap that may end up having unfortunate extra costs associated with it? https://github.com/llvm/llvm-project/pull/139730 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
