Live patching is absolutely possible technically. There are some architectural barriers in the design of the current system that make it harder to achieve than it should be. There's disagreement on this point, but I'm of the opinion that it should have been a feature from day one. An architecture that allows for live patching has a bunch of other benefits as well that make the system easier to work with. It forces the patcher to be more decoupled from the firmware. It forces us to have a clearly defined DSP API. I also think it could be done in a way that preserves overall performance; that's a common criticism of the idea. It really comes down to committing developer time to do the changes completely and not trying to graft them onto the existing architecture.
I'll leave it there for now, but if you're interested in more detail about what I think the correct architecture is, let me know.