Why doesn't Sun release the crypto module of the OpenSPARC?

[David G. Koontz]

[Moderator's note: this seems to be much more about the open source
 wars and such than about crypto and security. I'm not going to
 forward replies on this topic that don't specifically address
 security issues -- those who were not interested in the original
 thread may want to skip this message, too. --Perry]


David G. Koontz wrote:
> zooko wrote:
>> On Jun 12, 2008, at 4:35 PM, David G. Koontz wrote:
>>
>>> There's the aspect of competition.
>>> I've also wondered if a reason they didn't release it is because they
>>> bought
>>> the 'IP' from someone.
>> Those are good guesses, David, and I guessed similar things myself and
>> inquired of various Sun folks if this was the "real" reason.  Nobody
>> could give me any definite answer, however, until Sridhar Vajapey wrote:
>>
>>  "US export control regulations prevent Sun from opensourcing the crypto
>> portion of N2.".
> 
> 
> You've got to admit, that the work load for implementation is quite a bit
> higher without the PCI-E, 10GE MACs, and crypto, for a piece of competitive
> silicon.  All the sudden you don't have that 'Server On a Chip' that Sun
> sells.
> 
> The net result is still that you can't compete directly with Sun, but you
> can still expand the range of applications for Sun processors, and oh by the
> way, Sun's silicon works perfectly well in any new markets.  It still walks
> like a duck.
> 
> For the record I don't begrudge Sun captive markets, it supports a fairly
> decent 64 bit architecture and isn't Intel.  What they have released isn't
> what they sell.  They're demonstrably Rice Christian open source advocates.
> 
> http://en.wikipedia.org/wiki/Rice_Christian
> 

On June 5th, Sun released more of the T2 RTL source, in this case for the
NIU (Network Interface Unit) which includes the 10G Ethernet interfaces.

This was presented by the fellow responsible on several blogs and
announcement on opensparc.net.

http://blogs.sun.com/dv/entry/10g_ethernet_design_open_sourced
http://blogs.sun.com/dwaynelee/entry/latest_opensparc_t1_and_t2
http://www.opensparc.net/2008-06/durgam-vahia-10g-ethernet-design-open-sourced.html
http://www.opensparc.net/opensparc-t2/version-1.1-released.html
announcement in which:

  OpenSPARC T2 System-on-chip (SoC) micro-architecture document has a new
  chapter on NIU. The document now is also divided in two volumes to reduce
  the size of the each book.

This new System-on-chip (SoC) micro-architecture documents are not in
evidence for download, the SoC document available is from December 2007
(without the NIU).

One could note that lack of open release of documentation on the NIU would
not be a major impediment to a Sun partner with a close working relationship
(and access to the UltraSPARC T2 full documentation).  One could likewise
view it to be detrimental to an independent effort.

The reasoning for why I labeled Sun as faint hearted open source advocates
can be summed up by how well they fit into the open source community.

http://www.groklaw.net/article.php?story=20080626160554713
Pamela Jone's Groklaw article entitled  'A Sun Update on the NetApp
Litigation', 26 June 2008:

  Brendan Scott of Open Source Law was interviewed recently and he made
  this point:

        Open source is about community. You need to understand the community
    to operate effectively in it and this means changing your own behaviour.
    For example, if you start a new job and there is a cake morning on
    Monday, then you are going to have to find a cake shop or hone your
    baking skills if you want to be part of the team. Popping out for group
    coffee on a Friday like you did at your last employer won't work. That
    is why it's important to connect with people who are already members of
    the community and know their way around it - people from the Monday set,
    not the Friday set.

  In other words, you have to fit in. Releasing code is not all there is to
  it. Ethics, fairness, honesty -- it's the FOSS culture, and it's the value
  add. Any company that tries to play by the old rules undercuts that
  advantage.

http://www.computerworld.com.au/index.php/id;1993045790;pp;1;fp;16;fpid;1
(Brenden Scott interview)

Mike Dillon, Sun's General Counsel comments on the NetApp v. Sun litigation
over ZFS and on his blog http://blogs.sun.com/dillon/entry/netapp_draft
firmly declares Sun as a FOSS player.

>From what we see with Sun's opensparc.net effort, their part of the Friday
coffee crowd and not part of the Monday effort.  The difference between
getting by as an open source advocate while continuing your proprietary
behaviors, versus really changing and pro actively supporting open source.
Sun has lots of irons in the fire for open source,  but as far as I know the
opensparc effort is furthest along the path as far as using GPL licenses,
favoring the CDDL for various other efforts, particularly historically.

Sun is still hedging their open source efforts to the point of the
appearance of maintaining a proprietary advantage.  This can be demonstrated
by nickel and dime releases, as well as multi year hedges for releasing
things like java under the GPL.  It boils down to an acknowledgment that
what they use internally isn't what the release as open source.  A hedge to
'play by the old rules' in PJ's words, and that makes them Rice Christian
open source advocates.  Essentially clinging to the words of faith during
fair weather, and reverting to the old ways during hard times, they aren't
committed irrevocably to their new 'FOSS' faith.

If it's any conciliation, Sun appears to be moving closer toward that
commitment.






---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com