[OpenAFS-devel] DRAFT: New sysname standard

[David Thompson]

Derek Atkins wrote:
>Phil.Moore@morganstanley.com writes:
>
>> The point is: the sysname values needs to describe the machine on
>> which the cache manage is currently running with sufficient
>> granularity, and not be a more generic specification that describes
>> the set of platforms on which the kernel module can be loaded.
>
>Perhaps for Linux we really want something more akin to:
>	i386_redhat_62
>or	i386_debian_potato

This is the direction we've gone here at Wisconsin, although not the exact 
semantics in Derek's example.  There is no OS called linux.  I also think that 
encoding _only_ the libc version won't take into account the differences 
between linux distributions that might require a different binary to be used.  
There may be two different linux distros I want to support, both at 'kernel 
2.4/glibc 2.2', but different in some aspect that requires me to know which is 
which.  I think you're more likely to get the differentiation you want by 
using i386_(rh|deb|suse|md)_<release>.

--
Dave Thompson  <thomas@cs.wisc.edu>

Associate Researcher                    Department of Computer Science
University of Wisconsin-Madison         http://www.cs.wisc.edu/~thomas
1210 West Dayton Street                 Phone:    (608)-262-1017
Madison, WI 53706-1685                  Fax:      (608)-262-6626
--