Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 17 Aug 2001 01:11:47 -0700
From:      Peter Wemm <peter@wemm.org>
To:        Wilko Bulte <wkb@freebie.xs4all.nl>
Cc:        "David O'Brien" <obrien@FreeBSD.org>, alpha@FreeBSD.org, developers@FreeBSD.org
Subject:   Re: dropping compiler support in -current 
Message-ID:  <20010817081147.DCEEA3811@overcee.netplex.com.au>
In-Reply-To: <20010817081929.A2859@freebie.xs4all.nl> 

next in thread | previous in thread | raw e-mail | index | archive | help
Wilko Bulte wrote:
> On Thu, Aug 16, 2001 at 09:58:07PM -0700, David O'Brien wrote:
> 
> 
> Great. Just great. With one halfbaked CPU port after the other sprouting
> out of the project the one single port that *works* [sortof] is being nuked
> by x86-centric developments. Maybe it is about time to *think* instead
> of hacking away blindly. If this whole blindness is not changed into 
> something more productive we might as well throw the ppc, arm, sparc 
> fledgling ports in the can here and now. 
> 
> [Yes, I'm pissed. And no, this is not a stab at David]

Before everybody gets all bent out of shape over this, lets make an
extra effort to stay constructive rather than going off into the weeds
where it does nobody any good.  All is not lost yet.

Lets go over a couple of key things..

- Whether we like it or not, i386 is our reference platform.  By definition
there is going to be a general trend of i386 things happening first and the
other platforms catching up.  Yes, we can make effort to minimize this, but
realistically, it is going to happen to some degree.  Yes, this is
i386-centric, yes it sucks, but sheer inertia will keep it happening. This
is leading to the true measure of the viability of a port to platform X -
ie: Are there sufficient people using that platform who are prepared to
dive in and help keep that platform up to date with i386?  Are there enough
people with sufficient knowledge to dive in when hairy problems turn up?
and so on.  Yelling at a committer because they did something for x86
and forgot platform X does not achieve much except cause wear and tear.
Things will get done much more effectively if people help rather than
criticize.  Yes, we have some fearless souls who slog through this
stuff, but they cannot do everything.

- There are going to be alpha boxes around for another year so so. The
hardware is viable for a while more yet.  The problem is developer time.
Right now, we are at the regular FreeBSD late summer doldrums and we
usually have a slowdown anyway.  Combine that with the fact that we have
a real bleeding-edge -current and nobody is even pretending that we're going
to try and call it "-stable" for another 6 months at least, and we end up
with even less people hanging on for the ride on -alpha-current.  Dont
forget, we're in the middle of major architectural turmoil and the core
infrastructure is bumpy as hell right now.  I'm not sure that its a bad
thing that the pace of the alpha port is cooling off a bit, otherwise
there would be a lot of energy spent on moving targets.  Would it be such
a bad thing if the real effort to stabilize alpha didn't happen till
5.1 time frame when the core kernel services have stabilized?

- Also, platforms die sooner or later.  Eventually there will not be enough
willpower to keep the platform up to date.  I dont think FreeBSD/Alpha is
there yet.  I dont think anybody should be prepared to call a 'time of death'
on the alpha port unless we get to RELENG_5 without a functional system.

- David is annoyed because he cannot test his gcc / toolchain work adequately
on the alpha right now.  Maybe we should be offloading that from him
so that he doesn't have to worry so much about it?  I would not blame
David in the slightest for saying "ok, here is what I think is needed
for the alpha as best as I can tell, you're on your own - let me know how
it went".  And then have somebody else tidy up the loose ends.

There is a similar thing with FreeBSD/Linux.  Whether we like it or not,
Linux is pretty much the reference i386 open source OS platform these
days.  Do we complain to the linux folks that they are not helping us
keep up?  Or do we dive in and do our best?

[And no, I am not criticizing the good people that have been trying to
 do their bit.  I'm only annoyed at the people that dont contribute but
 still complain that it isn't happening to their satisfaction.  There are
 lots of ways to help other than post "alpha is broken again, will
 somebody please fix it NOW!"]

</soapbox mode>

I just *know* I'm going to get hate mail for this...

> Wilko
> 
> > With another 3 hours spent chasing panics and bad kernels today (on top
> > of the lost weeks since May), I am no longer willing to keep the
> > toolchain up to date for the Alpha platform.
> > 
> > We simply have no developers running -current on Alpha hardware anymore.
> > Sorry, but Alpha has just become too much of a dead platform as far as
> > FreeBSD goes it seems.  
> >                                 .
> >                                -|-
> >                                 |
> >                             .-'~~~`-.
> >                           .'         `.
> >                           |  R  I  P  |
> >                           |           |
> >                           |  A  X  P  |
> >                         \\|           |//
> >    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> -- 
> |   / o / /  _   	Arnhem, The Netherlands    	email: wilko@FreeBSD.or
    g
> |/|/ / / /( (_) Bulte	
> 
> 

Cheers,
-Peter
--
Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au
"All of this is for nothing if we don't go to the stars" - JMS/B5


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-alpha" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010817081147.DCEEA3811>