[lkml]   [1996]   [Sep]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [off-topic] Fully virtualisable CPUs
On Sun, 8 Sep 1996, Graham Mitchell wrote:

> On 7 Sep 96 at 12:26, Craig Milo Rogers wrote:
> > Hmmm... if AMD or Cyrix were to offer this as an architectural
> > extension, I bet they'd sell a *lot* of chips to Linux users. Maybe
> > it could be done in a way that doesn't require much kernel support
> > beyond what's already in place for DOSEMU, too.
> The problem is, that this would probably break Win3, Win95, WinNT,
> SCO and about every other program that relies on the virtualisation
> happening in the current manner.I wouldn't hold my breath waiting for
> them to do this.

No, I don't think it would, as I see it, what we have now is that various
things create exceptions, which we can trap (wether we have to trap
everything, or if we can just trap some things I don't know, I don't have

The only problem is that fact we can't trap things that show the actual
cpu status. We can then return a bogus result, like you can on other
archiectures, in order to hide the fact that whatever isn't actually the
only thing running, and that it's really just being helped along by
something else that is in total control (or not ;).

It would make running any of WinNT, Win95, (Win3.xx), Linux, SCO, (DOS)
under any other trival (well _under_ not those in ()'s), with a little
work to support it (in the OS/microkernel/schedular/... running the show).

> Graham
PGP key pass phrase forgotten, \ Overload -- core meltdown sequence
again :( and I don't care ;) | initiated.
/ This space is intentionally left
| blank, apart from this text ;-)

 \ /
  Last update: 2005-03-22 13:37    [W:0.068 / U:0.008 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site