Lguest (Was re: [9fans] Xen 3.1 w/ Plan 9 domU)

Lguest (Was re: [9fans] Xen 3.1 w/ Plan 9 domU)

[Joel C. Salomon]

On 7/4/07, ron minnich <rminnich@gmail.com> wrote:
> folks, just FYI, for the THX project, I have looked hard and am fixing
> on the new lguest hypervisor as the final plan. Please google lguest
> and see what you think. It's 5K lines of code, the guest support code
> is trivial, and I think overall this is going to be the easiest thing
> to do. The work should actually not take too long this time. Xen can
> be hard to get your brain around.

Looks like a patch on the Linux kernel to run as a hypervisor.  Is
that how you intend to run it, or are you planning on porting the code
to Plan 9, making p9 the hypervisor?  Or am I totally misunderstanding
the blurbs on <http://lguest.ozlabs.org/>?

--Joel

Re: Lguest (Was re: [9fans] Xen 3.1 w/ Plan 9 domU)

[ron minnich]

I'm porting plan 9 to lguest. making plan9 support lguest would not be
impossible, but since the whole issue here is drivers, I don't see any
reason to do it.

ron

Re: Lguest (Was re: [9fans] Xen 3.1 w/ Plan 9 domU)

[lucio]

> I'm porting plan 9 to lguest. making plan9 support lguest would not be
> impossible, but since the whole issue here is drivers, I don't see any
> reason to do it.

Well, it's a chicken and egg situation.  If adding to lguest the
ability to use the standardised Plan 9 driver interface actually
simplifies lguest (it may not, of course, but one can wish), then the
exercise becomes much more worthwhile and lguest developers may then
prefer to use Plan 9 as the platform.  And develop drivers for it.

I have a feeling some of that is beginning to filter into Xen,
although their priorities are necessarily elsewhere presently and for
the foreseeable future.

++L