You could do worse than Mach ports

[Peter da Silva]

Thinking about this message more, and the thread about using L4:

Every attempt to start by taking an existing "micro"kernel, and an existing monolithic kernel, and bring them together has produced the same sequence of stages: first, you get the monolithic kernel running under as a server under the microkernel, and performance sucks. Then through a fairly significant effort the performance is brought up to scratch. At this point you tend to get a breathing space. If there's something people wanted to do under the microkernel, they implement servers to do it.


And that seems to be about it. The step where you replace kernel components with servers and go to a full microkernel design never seems to happen, or it happens so slowly nobody notices.

I think that Matt's got an idea here that, because it's approaching things from the side of building a "microkernel" inside an existing system rather than the other way around, has the potential of ending up with something that has decent performance *and* gives you the advantages of a microkernel design... because when you hit that breathing space, you're already at the finish line.