From owner-freebsd-chat Fri Dec 14 22: 9: 3 2001 Delivered-To: freebsd-chat@freebsd.org Received: from albatross.prod.itd.earthlink.net (albatross.mail.pas.earthlink.net [207.217.120.120]) by hub.freebsd.org (Postfix) with ESMTP id 5776B37B417; Fri, 14 Dec 2001 22:08:59 -0800 (PST) Received: from pool0205.cvx21-bradley.dialup.earthlink.net ([209.179.192.205] helo=mindspring.com) by albatross.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16F7zu-0001SQ-00; Fri, 14 Dec 2001 22:08:35 -0800 Message-ID: <3C1AE8E5.ED0E7BA7@mindspring.com> Date: Fri, 14 Dec 2001 22:08:37 -0800 From: Terry Lambert X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Greg Lehey Cc: "Gary W. Swearingen" , hiten@uk.FreeBSD.org, "Brandon D. Valentine" , Hiten Pandya , chat@FreeBSD.org, phk@FreeBSD.org, Poul-Henning Kamp Subject: Re: IBM's intentions with JFS (was: IBM suing (was: RMS Suing was [SUGGESTION] - JFS for FreeBSD)) References: <3C186381.6AB07090@yahoo.com> <3C1875D6.5DE4F996@mindspring.com> <1id71idej9.71i@localhost.localdomain> <3C1875D6.5DE4F996@mindspring.com> <20011213051012.Y56723-100000@turtle.looksharp.net> <3C186381.6AB07090@yahoo.com> <3C1875D6.5DE4F996@mindspring.com> <3C186381.6AB07090@yahoo.com> <20011214122837.O3448@monorchid.lemis.com> <3C19807D.C441F084@mindspring.com> <20011215103210.G85108@monorchid.lemis.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Greg Lehey wrote: > Correct. But there's this funny agreement that loading modules falls > under the LGPL, not the GPL. I'll leave it to Linux people to > explain; it's used to allow loading proprietary modules into the Linux > kernel. The implication appears to be that, since it's not linked > with the kernel, it's not a derivative work. I think it's a > workaround for an overly restrictive license. I'm well aware of this "funny agreement". It's meant to keep the GPL on the kernel from applying to proprietary modules being loaded, like the NUC (NetWare UNIX Client) code module in Caldera Linux. I'm not certain it applies to GPL'ed KLDs in FreeBSD; it certainly did not apply to GPL'ed LKMs, because of the way the technology used to work; I think KLDs still work this way: LKMs used to link against the kernel using "ld -r" of PIC'ed modules, so that they could be loaded at arbitrary locations in the KVA space, given that location of the allocation was uncertain. The term "significant code" is used quite purposefully, since it dictates what does or does not qualify as a derivative work (it is the "derivation test" used by legal authority), and it's what keeps most code run on Linux from being GPL'ed by virtue of including sometimes complex inline code and macros from Linux header files. The problem is that the modules import static data and symbols at link time (and hence "match" the kernel -- and so can not meet the GPL/LGPL "relink clause" in all cases). FWIW: Jeffrey Hsu did the original PIC generation for the GNU compiler and assembler, and I did the original module loading (LKM) implementation that was picked up by NetBSD, FreeBSD, and OpenBSD. > On Thursday, 13 December 2001 at 23:49:37 -0800, Gary W. Swearingen wrote: > > Greg Lehey writes: > >> The loader might have to be GPL'd. > > > > If the loader, then the kernel too. > > No, of course not. That would imply that anything you run under the > kernel also has to be GPLd. The loader and kernel are two very > different beasts. You can load FreeBSD with the Linux loader. Does > that make FreeBSD GPLd? No. The Linux kernel has a special expection to cause it to be treated as an LGPL'ed library by kernel modules. Please read the Linux license in its entirety. See also the GPL FAQ, which explicitly discusses how to grant such exceptions, yet still apply the GPL to your code. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message