Re: Proposal for Updating CRC32C with AVX-512 Algorithm.

Bruce Momjian <bruce@momjian.us>

From: Bruce Momjian <bruce@momjian.us>
To: "Amonson, Paul D" <paul.d.amonson@intel.com>
Cc: Alvaro Herrera <alvherre@alvh.no-ip.org>, Andres Freund <andres@anarazel.de>, "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>, Nathan Bossart <nathandbossart@gmail.com>, "Shankaran, Akash" <akash.shankaran@intel.com>
Date: 2024-06-18T18:00:34Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →
  1. Compute CRC32C using AVX-512 instructions where available

On Tue, Jun 18, 2024 at 01:20:50PM -0400, Bruce Momjian wrote:
> On Tue, Jun 18, 2024 at 05:14:08PM +0000, Amonson, Paul D wrote:
> > > And this bit doesn't look good.  The LICENSE file says:
> > ...
> > > > //    * Redistributions in binary form must reproduce the above
> > > > // copyright notice, this list of conditions and the following
> > > > disclaimer // in the documentation and/or other materials provided
> > > > with the // distribution.
> > ...
> > > The second clause essentially says we would have to add a page to our
> > > "documentation and/or other materials" with the contents of the license file.
> > 
> > According to one of Intel’s lawyers, 55 instances of this clause was found when they searched in the PostgreSQL repository. Therefore, I assume that this obligation has either been satisfied or determined not to apply, given that the second BSD clause already appears in the PostgreSQL source tree. I might have misunderstood the concern, but the lawyer believes this is a non-issue. Could you please provide more clarifying details about the concern?
> 
> Yes, I can confirm that:
> 
> 	grep -Rl 'Redistributions in binary form must reproduce' . | wc -l
> 
> reports 54;  file list attached.

I am somewhat embarrassed by this since we made the Intel lawyers find
something that was in our own source code.

First, the "advertizing clause" in the 4-clause license:

	 3. All advertising materials mentioning features or use of this
	software must display the following acknowledgement: This product
	includes software developed by the University of California,
	Berkeley and its contributors.

and was disavowed by Berkeley on July 22nd, 1999:

	https://elrc-share.eu/static/metashare/licences/BSD-3-Clause.pdf

While the license we are concerned about does not have this clause, it
does have:

	 2. Redistributions in binary form must reproduce the above
	copyright notice, this list of conditions and the following
	disclaimer in the documentation and/or other materials provided
	with the distribution.

I assume that must also include the name of the copyright holder.

I think that means we need to mention The Regents of the University of
California in our copyright notice, which we do.  However several
non-Regents of the University of California copyright holder licenses
exist in our source tree, and accepting this AVX-512 patch would add
another one.  Specifically, I see existing entries for:

	Aaron D. Gifford
	Board of Trustees of the University of Illinois
	David Burren
	Eric P. Allman
	Jens Schweikhardt
	Marko Kreen
	Sun Microsystems, Inc.
	WIDE Project
	
Now, some of these are these names plus Berkeley, and some are just the
names above.

-- 
  Bruce Momjian  <bruce@momjian.us>        https://momjian.us
  EDB                                      https://enterprisedb.com

  Only you can decide what is important to you.