[OpenAFS-devel] DRAFT: New sysname standard

[Derek Atkins]

Considering that @sys is a client matter, and we're talking about new
client ports of OpenAFS...  Why can't we make other @foo variables
that solve the other problems?  Then we could have an @value for the
user-space and an @value for the kernel-space naming....  Then we
could use something like:

	i386_rh62	<- user-space
	i386_linux22_18	<- kernel-space (2.2.18)

And then we could have, say, @sys for the user-space and @kern for the
kernel space.  This way we don't need to have a single @sys value that
is trying to solve two unrelated problems.

-derek

Derrick J Brashear <shadow@dementia.org> writes:

> On Wed, 25 Apr 2001, David Thompson wrote:
> 
> > >Then you start assuming that just because someone runs redhat they are
> > >running redhat's kernel. 
> > 
> > Not necessarily.  But you would be assuming that the environment
> > around that kernel looks like a Red Hat version 'n' or debian version
> > 'q'.  I think that's the right assumption.
> > 
> > You're always going to have to build the new kernel load module.  But
> > no one uses @sys to determine things related to the kernel (??).  The
> > problem we're trying to solve is the userland problem.
> 
> That was my argument, but there are people who expect @sys values to solve
> both problems for them.
> 
> -D
> 
> 
> _______________________________________________
> OpenAFS-devel mailing list
> OpenAFS-devel@openafs.org
> https://lists.openafs.org/mailman/listinfo.cgi/openafs-devel

-- 
       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
       warlord@MIT.EDU                        PGP key available