On Tue, May 28, 2024 at 10:20:21AM +0200, Christian Brauner wrote: > > This would solve all the problems in this proposal as well as all the > > obvious ones it doesn't solve. > > As I've said multiple times on the thread I agree that this is what we > should do next and I would be happy to take patches for this. But > exposing the 64bit mount id doesn't impede or complicate that work. It's > a simple and useful addition that can be done now and doesn't prevent us > from doing the proposed work. I really confuses the user story as we now have not only one but two broken modes for the open by handle ops. We just further go down the rabbit hole of mixing a non-persistent identifiers with a persistent one. > Hell, if you twist my arm I'll even write the patches for this. I'm also happy to help with that despite very limited time, but I'd rather avoid doing the misguided mount ID thing.
