The changing Fedora

[Greg Woods]

On Tue, 2013-01-22 at 18:24 +0000, Patrick O'Callaghan wrote:
> On Tue, 2013-01-22 at 09:45 -0700, Greg Woods wrote:
> > On Tue, 2013-01-22 at 15:29 +0000, Patrick O'Callaghan wrote:
> > > On Tue, 2013-01-22 at 08:02 -0700, Greg Woods wrote:
> > >  The game CD has some stuff written
> > > > beyond the "end" of the disc.
> > 
> > > I would have thought that correct emulation would allow those same
> > > low-level calls in a VM. AFAIK the guest system can read the whole CD as
> > > a raw device (assuming appropriate privileges).
> > 
> > The trick is in defining "correct" emulation and "the whole CD".

> So are you talking about 1) the actual physical CD, or 2) an ISO image
> of the CD? I presumed the former

Yes, I meant the real CD passed through to the VM as a virtual CD-ROM.

>  in which case there *should* be no
> difference (and if there is, then the VM's emulation is incomplete).

Yes, the hypervisor's emulation is incomplete if my original speculation
is correct. But it's not surprising that it's incomplete, since 99.9% of
the time, there is no good reason to be reading blocks beyond the end of
the disc.

>  some copy-protection schemes use
> deliberately damaged blocks so that normal s/w will get read errors
> which an ISO image cannot reproduce.

I hadn't thought of that, but in that case, the emulator is also
incomplete since it isn't passing through the read errors.

This is getting quite far afield now. The original point was only that
there are reasons why a native boot might be needed and a VM won't
always work as a replacement.

--Greg