*BSD News Article 84054


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!howland.erols.net!news.mathworks.com!fu-berlin.de!news.belwue.de!news.uni-ulm.de!borchert
From: borchert@turing.mathematik.uni-ulm.de (Andreas Borchert)
Newsgroups: comp.unix.solaris,comp.unix.bsd.misc,comp.unix.internals,comp.unix.osf.osf1
Subject: Re: Solaris 2.6
Date: 2 Dec 1996 12:31:30 GMT
Organization: University of Ulm, SAI, Germany
Lines: 60
Distribution: inet
Message-ID: <57ui72$4li@arktur.rz.uni-ulm.de>
References: <32986299.AC7@mail.esrin.esa.it> <casper.329c06bc@mail.fwi.uva.nl> <57ik5l$12i$1@shade.twinsun.com> <x7917mx5gx.fsf@dumbcat.codewright.com> <57shh0$o3u@web.nmti.com>
NNTP-Posting-Host: turing.mathematik.uni-ulm.de
X-Newsreader: slrn (0.9.0.0 (BETA) UNIX)
Xref: euryale.cc.adfa.oz.au comp.unix.solaris:91089 comp.unix.bsd.misc:1647 comp.unix.internals:11399 comp.unix.osf.osf1:16792

On 1 Dec 1996 18:07:28 GMT, Peter da Silva <peter@nmti.com> wrote:
> In article <x7917mx5gx.fsf@dumbcat.codewright.com>,
> Marco S Hyman  <marc@dumbcat.codewright.com> wrote:
> > Stage 2 (not to be started until the world's C software has been fixed
> > 	 regarding pointers and longs):
> 
> DEC has already taken the hit for this. Anything that has been ported to
> the Alpha has already adjusted to 64 bit longs.

Well, this comparison does not count. DEC started the Alpha immediately
with 64 bits (years later after the SPARC architecture), starting with
no software base. On the other hand, the SPARC-platform is
(as far as I know) the base with greatest number of available
and installed software packages after the Intel processor.

There may be arguments how the transition is done best. But DEC
is surely not a good example for binary compatibility: Not that
they only are trying to support multiple OSes they also offer multiple
hardware platforms as well. This opens always the question how long
they will continue to support the various variants. The future
of Sun is bound to SPARC and Solaris (which is at least now
binary compatible to the previous versions) and customers trust on this
(at least we do).

> I sure hope Sun keeps screwing up like this, it'll mean more broken
> transitions in the future when they finally catch on,

Up to now, I've not seen good points why it should be so much worse
to create a 2nd set of system calls not only on the assembly level
(system call numbers) but also on the level of the C libraries.
The new system call numbers may be hidden from the ABI and the
official documentation but are nevertheless present and used.
They are present in /usr/include/sys/syscall.h (at least on Solaris)
and, for example, directly used by Perl scripts or language
implementations different from C and C++ -- we produce compilers
and libraries which directly access the system calls without
bothering with the C library or ABI specifications.

Sun at least tries to think carefully about its customer needs.
As always, this may not be the best solution for all customers
but probably for a sufficient large fraction of them.
There are always tradeoffs between upward compatibility and
the most elegant solution. The latter is in general achieved
by throwing away everything and starting from scratch -- surely
not an option for Sun now.

> and frankly I'd
> love to see more folks running Digital UNIX instead of Solaris... it's
> a much more pleasant environment to work in.

And that is finally your point?

Andreas.

-- 
Andreas Borchert, Universitaet Ulm, SAI, Helmholtzstr. 18, 89069 Ulm,  Germany
E-Mail: borchert@mathematik.uni-ulm.de
WWW:	http://www.mathematik.uni-ulm.de/sai/borchert/
PGP key available via ``finger borchert@laborix.mathematik.uni-ulm.de''