[Lguest] [patch] core.c i386 disentangle

Jes Sorensen jes at sgi.com
Wed Aug 15 19:43:24 EST 2007


Rusty Russell wrote:
> Hi Jes,
> 
> 	I think you've misunderstood the role of "i386_guest.c".   That's the
> *guest* side of i386.  We need a new file (i386.c? i386_core.c?) which
> has all the host-side of i386.  Some of this, of course, will be moved
> to x86.c when we share with x86_64, but that's ok for later.

Hmmm ok, so much for just aiming for a dumping ground for all that
i386 rubbish :-) I'll try and move it around again, the problem was
also that some of the stuff was called from i386_guest.c, but I'll just
export it.

>> -/*H:030 Let's jump straight to the the main loop which runs the Guest.
>> - * Remember, this is called by the Launcher reading /dev/lguest, and we keep
>> - * going around and around until something interesting happens. */
>> -int run_guest(struct lguest *lg, unsigned long __user *user)
>> -{
> 
> I wonder if we want to move this whole function, or try to hook in arch
> stuff in a sane way, eg:

I looked at arch'ing it, but it ended up being like 90% moved to arch
and I thought it was better to just move the whole thing. We can do an
arch split up if you insist, but I am not sure it's really worth it.

I'll try it out and see how it ends up - be afraid, be very afraid!

Jes



More information about the Lguest mailing list