🚀 go-pugleaf

In article <1991May7.163912.615@bwdls61.bnr.ca>, awhitton@bnr.ca (Alan
Whitton) writes:
> guy@auspex.auspex.com (Guy Harris) writes:
> > >1) You can't do OSF/1 without Mach.  They are one in the same.
> > 
> > The first statement doesn't necessarily imply the second, if in the
> > second you mean "one and the same".  If OSF/1 has stuff in it that Mach,
> > as it comes from CMU, doesn't, the first statement would be true but the
> > second would be false....
> 
> I agree with Guy. I was at an OSF symposium and the authentication suite
> (Authentication Environment Specification) does not specifically mention
> Mach. I in fact asked, "Does this mean, you need not have Mach to be 
> certified as an OSF/1 compliant OS?", and the response I got back was
> "Yes, you don't need Mach to be OSF compliant".

Without wishing to act as the official OSF spokesman, perhaps I can help
clear a few things up.

If you use the source of OSF/1 as supplied by OSF to create your OSF/1
system you most certainly will have the Mach kernel in there.

OSF/1 and Mach are not the same. OSF/1 contains the Mach kernel but very
little else from Mach. If you got Mach 2.5 from CMU, beyond the Mach
kernel it would bear little ressemblance to OSF/1 (other than due to the
fact that the rest of both kernels have a common heritage).

You may implement your system to OSF/1 interfaces without Mach and still
call it OSF/1 compliant if all the required interfaces and there
semantics are supported. These do not include the Mach interfaces (which
is where the confusion seems to be stemming from).

I hope this helps.

Simon.

  Simon Patience
  Open Software Foundation			Phone: +33-76-63-48-72
  Research Institute				FAX:   +33-76-51-05-32
  2 Avenue De Vignate				Email: sp@gr.osf.org
  38610 Gieres, France				       uunet!gr.osf.org!sp