Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 16 Dec 1997 16:58:33 -0800 (PST)
From:      Jason Evans <jasone@canonware.com>
To:        "Robert S. Sciuk" <rob@ControlQ.com>
Cc:        freebsd-sparc@FreeBSD.ORG
Subject:   Re: Freeloading FreeBSD for Free -- NOT!
Message-ID:  <Pine.BSF.3.96.971216145708.7374S-100000@paladio>
In-Reply-To: <Pine.UW2.3.96.971216171836.29252W-100000@fatlady.controlq.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 16 Dec 1997, Robert S. Sciuk wrote:
> I detect a minor problem in the Sun involvement in this project (see the
> FAQ sent earlier -- Thanks Jason/Dale ). They seem to want to be driving
> the ultrasparc (64Bit) chip sales with FreeBSD, in which case they should
> put out a PCI board and processors priced cheaply enough to allow us to
> hack together our own (preferably SMP) boxes ... and thus justify their
> investment ...

Sun is in the process of doing exactly what you propose.  They currently
have a MB/CPU with PCI (and a number of other things) for about $2900.
This is a bit spendy still.  The next board they put out in the next few
months will be much faster, and it will cost almost $1000 less.  This
upcoming board will AFAIK have a comparable price/performance to P6-based
boards.  As for SMP, that's still spendy, but don't expect it to stay that
way forever.

As for Sun involvement in this project, there are two things to point out:

1) This project was started by a Sun employee (me), so claiming Sun
shouldn't be involved is a bit strange. =)  Actually, if I understand your
statement, you're saying that Sun would be stupid to put effort into
porting FreeBSD without offering an affordable system to run it on.  Yes,
that would be silly, but they're providing the goods.

2) Sun must necessarily focus on ways to make money.  That means selling
new processors.  Such is the way of a free economy.  It happens in this
case that people stand to benefit in a major way (i.e. FreeBSD for the
UltraSPARC).  

Even if I weren't working for Sun, chances are that I would limit my
initial efforts to the UltraSPARC, and perhaps the sun4m, just to keep the
project scope reasonable.  Porting to the UltraSPARC is the hard part.  If
people want to run FreeBSD on the sun4c, they will have lots to work with,
between the UltraSPARC port and what has been done for NetBSD and OpenBSD. 

I should also mention that Ian Logan is planning to put some serious
effort into making sun4m work under FreeBSD.  That doesn't take care of
your IPX, but it does cover most machines still commonly used today.

> FreeBSD users on the other hand, typically have access to miled out legacy
> boxes (I know mine is) ... and I'm too cheap to shell out for the Solaris
> 2.6 release, never mind an ultrasparc server.  To be sure, my evidence is
> anecdotal, my opinion, and may not be indicative of the `typical' FreeBSD
> user. When Sun finds out about me (and my old IPX) ... will they pull the
> plug?? 8-) 

I have contemplated scenarios where Sun might pull the plug.  There are
certain risks, but you running FreeBSD on an IPX is hardly one of them. =)

Here's what I'm doing to ensure my efforts are not yanked from the
community:

* Source repository on an external site.
* Up to date access to my code.
* Assurance from my manager that all of my work can (and should) be
  available to the FreeBSD community under the standard FreeBSD license.

So, even if I get pulled off of the project in an official capacity, all
of my work will still be there without restrictions.  And, even if I get
pulled off, I will continue to work on the project on my own time.

>From now on I'm going to try to avoid spending much time on "public
relations"  mail like this so that I can get more real work done.  I hope
at this point that Sun's (SME's actually) and my motivations for this
project are clear.  

I hope it's clear that my official efforts are necessarily limited to the
UltraSPARC, though there will be others working on sun4m support.
Recognize also that my concentration on the UltraSPARC may well make the
difference between a mere working port and a truly excellent computing
platform. 

Jason

Jason Evans
Email: [jasone@canonware.com]
Home phone: [(650) 856-8204]
Work phone: [(408) 774-8007]
Quote: ["Invention is 1% inspiration, 99% perspiration" - Thomas Edison]









Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.971216145708.7374S-100000>