[OpenAFS-devel] Re: OpenAFS: conventions for new @sys? (rs_aix43 s390_linux22)

[Russ Allbery]

Nathan Neulinger <nneul@umr.edu> writes:

> Well, building architecture specific trees of products could be quite
> handy, particularly if you start building versions optimized to
> particular CPUs, etc.

> Or, mapping product trees according to particular environment variable
> choices that would be completely transparent to the calling product.

> Or things that are kernel version specific. 

I think I'm just disagreeing with the specific example rather than the
general point.  Yeah, I can see how that would be useful; I just haven't
found config.guess output to be a particularly useful way of encoding it
except for some specific purposes.

> I would suggest that openafs/afs do anything with it, I'd just suggest
> that the functionality be there for anyone who might want to use it.

Fair enough.

I guess my interest pretty much stops with @sys values, as those are the
only ones I can see myself wanting to use, but I certainly don't object to
the generalization.

It's a touch orthogonal to the problem of a canonical set of @sys
mappings, though.

-- 
Russ Allbery (rra@stanford.edu)             <http://www.eyrie.org/~eagle/>