Re: Mach emulation

[OKUJI Yoshinori]

From: Farid Hajji <address@hidden>
Subject: Re: Mach emulation
Date: Wed, 15 Nov 2000 03:37:27 +0100

> Is this really true? The problem is how to have the Hurd use only
> a minimal set of kernel services.

  What are considered as the minimal is different among various
microkernels. For example, I think Pebble is smaller than L4 in some
aspects. In addition, it is not guaranteed that using a minimal set of
kernel services is efficient even on another kernel.

> "connect" L4-ish services with the new ukernel services. But then,
> how likely is this going to be? Nobody knows.

  Yes, but see at least some of existing microkernel implementations.

> Just one more point: The L4 subset maps without problems to Mach.
> A libmom-mach would simply replace L4 syscalls with Mach syscalls,

  That doesn't mean that is optimal, as well as we can emulate Mach on
L4 but this is not optimal.

> Would it be safe to assume that a multiprocessor ukernel multiplexes
> kernel-level threads to the available processors transparently?

  As far as I see L4/Alpha, that isn't.

Okuji